mikros/rust
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
8ec49ad19a
rust/tests/ui/numeric/numeric-cast.stderr

2038 lines
55 KiB
Plaintext
Raw Normal View History

Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:23:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<usize>(x_u64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^^ expected `usize`, found `u64`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u64` to a `usize` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<usize>(x_u64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:25:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<usize>(x_u32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^^ expected `usize`, found `u32`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u32` to a `usize` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<usize>(x_u32.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:27:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<usize>(x_u16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^^ expected `usize`, found `u16`
| |
| arguments to this function are incorrect
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert a `u16` to a `usize`
|
LL | foo::<usize>(x_u16.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:29:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<usize>(x_u8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^ expected `usize`, found `u8`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert a `u8` to a `usize`
|
LL | foo::<usize>(x_u8.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:31:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<usize>(x_isize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^^^^ expected `usize`, found `isize`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `isize` to a `usize` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<usize>(x_isize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:33:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<usize>(x_i64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^^ expected `usize`, found `i64`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i64` to a `usize` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<usize>(x_i64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:35:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<usize>(x_i32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^^ expected `usize`, found `i32`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i32` to a `usize` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<usize>(x_i32.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:37:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<usize>(x_i16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^^ expected `usize`, found `i16`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i16` to a `usize` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<usize>(x_i16.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:39:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<usize>(x_i8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^ expected `usize`, found `i8`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i8` to a `usize` and panic if the converted value doesn't fit
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
LL | foo::<usize>(x_i8.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:44:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<isize>(x_usize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^^^^ expected `isize`, found `usize`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `usize` to an `isize` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<isize>(x_usize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:46:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<isize>(x_u64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^^ expected `isize`, found `u64`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u64` to an `isize` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<isize>(x_u64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:48:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<isize>(x_u32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^^ expected `isize`, found `u32`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u32` to an `isize` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<isize>(x_u32.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:50:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<isize>(x_u16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^^ expected `isize`, found `u16`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u16` to an `isize` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<isize>(x_u16.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:52:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<isize>(x_u8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^ expected `isize`, found `u8`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert a `u8` to an `isize`
|
LL | foo::<isize>(x_u8.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:55:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<isize>(x_i64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^^ expected `isize`, found `i64`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i64` to an `isize` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<isize>(x_i64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:57:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<isize>(x_i32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^^ expected `isize`, found `i32`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i32` to an `isize` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<isize>(x_i32.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:59:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<isize>(x_i16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^^ expected `isize`, found `i16`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert an `i16` to an `isize`
|
LL | foo::<isize>(x_i16.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:61:18
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<isize>(x_i8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ------------ ^^^^ expected `isize`, found `i8`
| |
| arguments to this function are incorrect
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert an `i8` to an `isize`
|
LL | foo::<isize>(x_i8.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:66:16
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<u64>(x_usize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^ expected `u64`, found `usize`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `usize` to a `u64` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u64>(x_usize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:69:16
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<u64>(x_u32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `u64`, found `u32`
| |
| arguments to this function are incorrect
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert a `u32` to a `u64`
|
LL | foo::<u64>(x_u32.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:71:16
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<u64>(x_u16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `u64`, found `u16`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert a `u16` to a `u64`
|
LL | foo::<u64>(x_u16.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:73:16
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<u64>(x_u8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^ expected `u64`, found `u8`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert a `u8` to a `u64`
|
LL | foo::<u64>(x_u8.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:75:16
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<u64>(x_isize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^ expected `u64`, found `isize`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `isize` to a `u64` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u64>(x_isize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:77:16
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<u64>(x_i64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `u64`, found `i64`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i64` to a `u64` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u64>(x_i64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:79:16
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<u64>(x_i32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `u64`, found `i32`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i32` to a `u64` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u64>(x_i32.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:81:16
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<u64>(x_i16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `u64`, found `i16`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i16` to a `u64` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u64>(x_i16.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:83:16
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Update UI tests
2018-02-22 18:42:32 -06:00
LL | foo::<u64>(x_i8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^ expected `u64`, found `i8`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i8` to a `u64` and panic if the converted value doesn't fit
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
|
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
LL | foo::<u64>(x_i8.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:88:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i64>(x_usize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^ expected `i64`, found `usize`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `usize` to an `i64` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i64>(x_usize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:90:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i64>(x_u64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `i64`, found `u64`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u64` to an `i64` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i64>(x_u64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:92:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i64>(x_u32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `i64`, found `u32`
| |
| arguments to this function are incorrect
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert a `u32` to an `i64`
|
LL | foo::<i64>(x_u32.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:94:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i64>(x_u16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `i64`, found `u16`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert a `u16` to an `i64`
|
LL | foo::<i64>(x_u16.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:96:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i64>(x_u8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^ expected `i64`, found `u8`
| |
| arguments to this function are incorrect
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert a `u8` to an `i64`
|
LL | foo::<i64>(x_u8.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:98:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i64>(x_isize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^ expected `i64`, found `isize`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `isize` to an `i64` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i64>(x_isize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:101:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i64>(x_i32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `i64`, found `i32`
| |
| arguments to this function are incorrect
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert an `i32` to an `i64`
|
LL | foo::<i64>(x_i32.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:103:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i64>(x_i16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `i64`, found `i16`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert an `i16` to an `i64`
|
LL | foo::<i64>(x_i16.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:105:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i64>(x_i8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^ expected `i64`, found `i8`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert an `i8` to an `i64`
|
LL | foo::<i64>(x_i8.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Remove licenses
2018-12-25 09:56:47 -06:00
--> $DIR/numeric-cast.rs:110:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u32>(x_usize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^ expected `u32`, found `usize`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `usize` to a `u32` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u32>(x_usize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:112:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u32>(x_u64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `u32`, found `u64`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u64` to a `u32` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u32>(x_u64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:115:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u32>(x_u16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `u32`, found `u16`
| |
| arguments to this function are incorrect
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert a `u16` to a `u32`
|
LL | foo::<u32>(x_u16.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:117:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u32>(x_u8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^ expected `u32`, found `u8`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert a `u8` to a `u32`
|
LL | foo::<u32>(x_u8.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:119:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u32>(x_isize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^ expected `u32`, found `isize`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `isize` to a `u32` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u32>(x_isize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:121:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u32>(x_i64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `u32`, found `i64`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i64` to a `u32` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u32>(x_i64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:123:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u32>(x_i32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `u32`, found `i32`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i32` to a `u32` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u32>(x_i32.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:125:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u32>(x_i16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `u32`, found `i16`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i16` to a `u32` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u32>(x_i16.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:127:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u32>(x_i8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^ expected `u32`, found `i8`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i8` to a `u32` and panic if the converted value doesn't fit
Update UI tests
2018-02-22 18:42:32 -06:00
|
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
LL | foo::<u32>(x_i8.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:132:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i32>(x_usize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^ expected `i32`, found `usize`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `usize` to an `i32` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i32>(x_usize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:134:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i32>(x_u64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `i32`, found `u64`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u64` to an `i32` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i32>(x_u64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:136:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i32>(x_u32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `i32`, found `u32`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u32` to an `i32` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i32>(x_u32.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:138:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i32>(x_u16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `i32`, found `u16`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert a `u16` to an `i32`
|
LL | foo::<i32>(x_u16.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:140:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i32>(x_u8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^ expected `i32`, found `u8`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert a `u8` to an `i32`
|
LL | foo::<i32>(x_u8.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:142:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i32>(x_isize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^ expected `i32`, found `isize`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `isize` to an `i32` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i32>(x_isize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:144:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i32>(x_i64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `i32`, found `i64`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i64` to an `i32` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i32>(x_i64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:147:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i32>(x_i16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `i32`, found `i16`
| |
| arguments to this function are incorrect
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert an `i16` to an `i32`
|
LL | foo::<i32>(x_i16.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:149:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i32>(x_i8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^ expected `i32`, found `i8`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert an `i8` to an `i32`
|
LL | foo::<i32>(x_i8.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:154:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u16>(x_usize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^ expected `u16`, found `usize`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `usize` to a `u16` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u16>(x_usize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:156:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u16>(x_u64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `u16`, found `u64`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u64` to a `u16` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u16>(x_u64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:158:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u16>(x_u32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `u16`, found `u32`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u32` to a `u16` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u16>(x_u32.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:161:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u16>(x_u8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^ expected `u16`, found `u8`
| |
| arguments to this function are incorrect
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert a `u8` to a `u16`
|
LL | foo::<u16>(x_u8.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:163:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u16>(x_isize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^ expected `u16`, found `isize`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `isize` to a `u16` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u16>(x_isize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:165:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u16>(x_i64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `u16`, found `i64`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i64` to a `u16` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u16>(x_i64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:167:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u16>(x_i32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `u16`, found `i32`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i32` to a `u16` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u16>(x_i32.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:169:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u16>(x_i16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `u16`, found `i16`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i16` to a `u16` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u16>(x_i16.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:171:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u16>(x_i8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^ expected `u16`, found `i8`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i8` to a `u16` and panic if the converted value doesn't fit
Update UI tests
2018-02-22 18:42:32 -06:00
|
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
LL | foo::<u16>(x_i8.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:176:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i16>(x_usize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^ expected `i16`, found `usize`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `usize` to an `i16` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i16>(x_usize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:178:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i16>(x_u64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `i16`, found `u64`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u64` to an `i16` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i16>(x_u64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:180:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i16>(x_u32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `i16`, found `u32`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u32` to an `i16` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i16>(x_u32.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:182:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i16>(x_u16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `i16`, found `u16`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u16` to an `i16` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i16>(x_u16.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:184:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i16>(x_u8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^ expected `i16`, found `u8`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert a `u8` to an `i16`
|
LL | foo::<i16>(x_u8.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:186:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i16>(x_isize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^ expected `i16`, found `isize`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `isize` to an `i16` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i16>(x_isize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:188:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i16>(x_i64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `i16`, found `i64`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i64` to an `i16` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i16>(x_i64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:190:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i16>(x_i32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `i16`, found `i32`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i32` to an `i16` and panic if the converted value doesn't fit
Update UI tests
2018-02-22 18:42:32 -06:00
|
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
LL | foo::<i16>(x_i32.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:193:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
LL | foo::<i16>(x_i8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^ expected `i16`, found `i8`
| |
| arguments to this function are incorrect
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert an `i8` to an `i16`
|
LL | foo::<i16>(x_i8.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:198:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u8>(x_usize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^^^^ expected `u8`, found `usize`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `usize` to a `u8` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u8>(x_usize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:200:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u8>(x_u64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^^ expected `u8`, found `u64`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u64` to a `u8` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u8>(x_u64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:202:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u8>(x_u32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^^ expected `u8`, found `u32`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u32` to a `u8` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u8>(x_u32.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:204:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u8>(x_u16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^^ expected `u8`, found `u16`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u16` to a `u8` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u8>(x_u16.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:207:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u8>(x_isize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^^^^ expected `u8`, found `isize`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `isize` to a `u8` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u8>(x_isize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:209:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u8>(x_i64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^^ expected `u8`, found `i64`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i64` to a `u8` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u8>(x_i64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:211:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u8>(x_i32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^^ expected `u8`, found `i32`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i32` to a `u8` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u8>(x_i32.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:213:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u8>(x_i16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^^ expected `u8`, found `i16`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i16` to a `u8` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<u8>(x_i16.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:215:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u8>(x_i8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^ expected `u8`, found `i8`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i8` to a `u8` and panic if the converted value doesn't fit
Update UI tests
2018-02-22 18:42:32 -06:00
|
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
LL | foo::<u8>(x_i8.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:220:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i8>(x_usize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^^^^ expected `i8`, found `usize`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `usize` to an `i8` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i8>(x_usize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:222:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i8>(x_u64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^^ expected `i8`, found `u64`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u64` to an `i8` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i8>(x_u64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:224:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i8>(x_u32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^^ expected `i8`, found `u32`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u32` to an `i8` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i8>(x_u32.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:226:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i8>(x_u16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^^ expected `i8`, found `u16`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u16` to an `i8` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i8>(x_u16.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:228:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i8>(x_u8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^ expected `i8`, found `u8`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert a `u8` to an `i8` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i8>(x_u8.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:230:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i8>(x_isize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^^^^ expected `i8`, found `isize`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `isize` to an `i8` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i8>(x_isize.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:232:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i8>(x_i64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^^ expected `i8`, found `i64`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i64` to an `i8` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i8>(x_i64.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:234:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i8>(x_i32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^^ expected `i8`, found `i32`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i32` to an `i8` and panic if the converted value doesn't fit
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<i8>(x_i32.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:236:15
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i8>(x_i16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| --------- ^^^^^ expected `i8`, found `i16`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Say "doesn't" instead of "wouldn't" in convert message
2020-09-29 21:27:58 -05:00
help: you can convert an `i16` to an `i8` and panic if the converted value doesn't fit
Update UI tests
2018-02-22 18:42:32 -06:00
|
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
LL | foo::<i8>(x_i16.try_into().unwrap());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++++++++++++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:242:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f64>(x_usize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^ expected `f64`, found `usize`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Remove incorrect newline from float cast suggestion
2021-12-02 19:00:09 -06:00
help: you can cast a `usize` to an `f64`, producing the floating point representation of the integer, rounded if necessary
|
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
LL | foo::<f64>(x_usize as f64);
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:244:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f64>(x_u64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `f64`, found `u64`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Remove incorrect newline from float cast suggestion
2021-12-02 19:00:09 -06:00
help: you can cast a `u64` to an `f64`, producing the floating point representation of the integer, rounded if necessary
|
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
LL | foo::<f64>(x_u64 as f64);
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:246:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f64>(x_u32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `f64`, found `u32`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Add article after "to" Also added missing backtick in "you can cast" message.
2020-09-29 12:30:42 -05:00
help: you can convert a `u32` to an `f64`, producing the floating point representation of the integer
Update UI tests
2018-02-22 18:42:32 -06:00
|
Update UI tests
2018-02-24 17:01:39 -06:00
LL | foo::<f64>(x_u32.into());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:248:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f64>(x_u16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `f64`, found `u16`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Add article after "to" Also added missing backtick in "you can cast" message.
2020-09-29 12:30:42 -05:00
help: you can convert a `u16` to an `f64`, producing the floating point representation of the integer
Update UI tests
2018-02-22 18:42:32 -06:00
|
Update UI tests
2018-02-24 17:01:39 -06:00
LL | foo::<f64>(x_u16.into());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:250:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f64>(x_u8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^ expected `f64`, found `u8`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Add article after "to" Also added missing backtick in "you can cast" message.
2020-09-29 12:30:42 -05:00
help: you can convert a `u8` to an `f64`, producing the floating point representation of the integer
Update UI tests
2018-02-22 18:42:32 -06:00
|
Update UI tests
2018-02-24 17:01:39 -06:00
LL | foo::<f64>(x_u8.into());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:252:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f64>(x_isize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^ expected `f64`, found `isize`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Add article after "to" Also added missing backtick in "you can cast" message.
2020-09-29 12:30:42 -05:00
help: you can convert an `isize` to an `f64`, producing the floating point representation of the integer, rounded if necessary
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<f64>(x_isize as f64);
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:254:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f64>(x_i64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `f64`, found `i64`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Add article after "to" Also added missing backtick in "you can cast" message.
2020-09-29 12:30:42 -05:00
help: you can convert an `i64` to an `f64`, producing the floating point representation of the integer, rounded if necessary
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<f64>(x_i64 as f64);
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:256:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f64>(x_i32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `f64`, found `i32`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Add article after "to" Also added missing backtick in "you can cast" message.
2020-09-29 12:30:42 -05:00
help: you can convert an `i32` to an `f64`, producing the floating point representation of the integer
Update UI tests
2018-02-22 18:42:32 -06:00
|
Update UI tests
2018-02-24 17:01:39 -06:00
LL | foo::<f64>(x_i32.into());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:258:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f64>(x_i16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `f64`, found `i16`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Add article after "to" Also added missing backtick in "you can cast" message.
2020-09-29 12:30:42 -05:00
help: you can convert an `i16` to an `f64`, producing the floating point representation of the integer
Update UI tests
2018-02-22 18:42:32 -06:00
|
Update UI tests
2018-02-24 17:01:39 -06:00
LL | foo::<f64>(x_i16.into());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:260:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f64>(x_i8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^ expected `f64`, found `i8`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Add article after "to" Also added missing backtick in "you can cast" message.
2020-09-29 12:30:42 -05:00
help: you can convert an `i8` to an `f64`, producing the floating point representation of the integer
Update UI tests
2018-02-22 18:42:32 -06:00
|
Update UI tests
2018-02-24 17:01:39 -06:00
LL | foo::<f64>(x_i8.into());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:263:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f64>(x_f32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `f64`, found `f32`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert an `f32` to an `f64`
|
LL | foo::<f64>(x_f32.into());
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:266:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f32>(x_usize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^ expected `f32`, found `usize`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Remove incorrect newline from float cast suggestion
2021-12-02 19:00:09 -06:00
help: you can cast a `usize` to an `f32`, producing the floating point representation of the integer, rounded if necessary
|
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
LL | foo::<f32>(x_usize as f32);
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:268:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f32>(x_u64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `f32`, found `u64`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Remove incorrect newline from float cast suggestion
2021-12-02 19:00:09 -06:00
help: you can cast a `u64` to an `f32`, producing the floating point representation of the integer, rounded if necessary
|
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
LL | foo::<f32>(x_u64 as f32);
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:270:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f32>(x_u32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `f32`, found `u32`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Remove incorrect newline from float cast suggestion
2021-12-02 19:00:09 -06:00
help: you can cast a `u32` to an `f32`, producing the floating point representation of the integer, rounded if necessary
|
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
LL | foo::<f32>(x_u32 as f32);
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:272:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f32>(x_u16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `f32`, found `u16`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Add article after "to" Also added missing backtick in "you can cast" message.
2020-09-29 12:30:42 -05:00
help: you can convert a `u16` to an `f32`, producing the floating point representation of the integer
Update UI tests
2018-02-22 18:42:32 -06:00
|
Update UI tests
2018-02-24 17:01:39 -06:00
LL | foo::<f32>(x_u16.into());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:274:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f32>(x_u8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^ expected `f32`, found `u8`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Add article after "to" Also added missing backtick in "you can cast" message.
2020-09-29 12:30:42 -05:00
help: you can convert a `u8` to an `f32`, producing the floating point representation of the integer
Update UI tests
2018-02-22 18:42:32 -06:00
|
Update UI tests
2018-02-24 17:01:39 -06:00
LL | foo::<f32>(x_u8.into());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:276:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f32>(x_isize);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^ expected `f32`, found `isize`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Add article after "to" Also added missing backtick in "you can cast" message.
2020-09-29 12:30:42 -05:00
help: you can convert an `isize` to an `f32`, producing the floating point representation of the integer, rounded if necessary
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<f32>(x_isize as f32);
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:278:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f32>(x_i64);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `f32`, found `i64`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Add article after "to" Also added missing backtick in "you can cast" message.
2020-09-29 12:30:42 -05:00
help: you can convert an `i64` to an `f32`, producing the floating point representation of the integer, rounded if necessary
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<f32>(x_i64 as f32);
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:280:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f32>(x_i32);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `f32`, found `i32`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Add article after "to" Also added missing backtick in "you can cast" message.
2020-09-29 12:30:42 -05:00
help: you can convert an `i32` to an `f32`, producing the floating point representation of the integer, rounded if necessary
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
|
LL | foo::<f32>(x_i32 as f32);
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| ++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:282:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f32>(x_i16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `f32`, found `i16`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Add article after "to" Also added missing backtick in "you can cast" message.
2020-09-29 12:30:42 -05:00
help: you can convert an `i16` to an `f32`, producing the floating point representation of the integer
Update UI tests
2018-02-22 18:42:32 -06:00
|
Update UI tests
2018-02-24 17:01:39 -06:00
LL | foo::<f32>(x_i16.into());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:284:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<f32>(x_i8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^ expected `f32`, found `i8`
| |
| arguments to this function are incorrect
Increase spacing for suggestions in diagnostics Make the spacing between the code snippet and verbose structured suggestions consistent with note and help messages.
2019-10-24 00:20:58 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
note: function defined here
--> $DIR/numeric-cast.rs:6:4
|
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Add article after "to" Also added missing backtick in "you can cast" message.
2020-09-29 12:30:42 -05:00
help: you can convert an `i8` to an `f32`, producing the floating point representation of the integer
Update UI tests
2018-02-22 18:42:32 -06:00
|
Update UI tests
2018-02-24 17:01:39 -06:00
LL | foo::<f32>(x_i8.into());
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
| +++++++
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:289:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<u32>(x_u8 as u16);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^^^^^^^ expected `u32`, found `u16`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert a `u16` to a `u32`
|
LL | foo::<u32>((x_u8 as u16).into());
| + ++++++++
Fix into() cast paren check precedence As discussed in #47699 the logic for determining if an expression needs parenthesis when suggesting an `.into()` cast is incorrect. Two broken examples from nightly are: ``` error[E0308]: mismatched types --> main.rs:4:10 | 4 | test(foo as i8); | ^^^^^^^^^ expected i32, found i8 help: you can cast an `i8` to `i32`, which will sign-extend the source value | 4 | test(foo as i8.into()); | ``` ``` error[E0308]: mismatched types --> main.rs:4:10 | 4 | test(*foo); | ^^^^ expected i32, found i8 help: you can cast an `i8` to `i32`, which will sign-extend the source value | 4 | test(*foo.into()); | ``` As suggested by @petrochenkov switch the precedence check to PREC_POSTFIX. This catches both `as` and unary operators. Fixes #47699.
2018-01-24 03:31:36 -06:00
error[E0308]: mismatched types
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
--> $DIR/numeric-cast.rs:291:16
Update UI tests
2018-02-22 18:42:32 -06:00
|
LL | foo::<i32>(-x_i8);
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
| ---------- ^^^^^ expected `i32`, found `i8`
| |
| arguments to this function are incorrect
|
note: function defined here
--> $DIR/numeric-cast.rs:6:4
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
|
Implementation for 65853 This attempts to bring better error messages to invalid method calls, by applying some heuristics to identify common mistakes. The algorithm is inspired by Levenshtein distance and longest common sub-sequence. In essence, we treat the types of the function, and the types of the arguments you provided as two "words" and compute the edits to get from one to the other. We then modify that algorithm to detect 4 cases: - A function input is missing - An extra argument was provided - The type of an argument is straight up invalid - Two arguments have been swapped - A subset of the arguments have been shuffled (We detect the last two as separate cases so that we can detect two swaps, instead of 4 parameters permuted.) It helps to understand this argument by paying special attention to terminology: "inputs" refers to the inputs being *expected* by the function, and "arguments" refers to what has been provided at the call site. The basic sketch of the algorithm is as follows: - Construct a boolean grid, with a row for each argument, and a column for each input. The cell [i, j] is true if the i'th argument could satisfy the j'th input. - If we find an argument that could satisfy no inputs, provided for an input that can't be satisfied by any other argument, we consider this an "invalid type". - Extra arguments are those that can't satisfy any input, provided for an input that *could* be satisfied by another argument. - Missing inputs are inputs that can't be satisfied by any argument, where the provided argument could satisfy another input - Swapped / Permuted arguments are identified with a cycle detection algorithm. As each issue is found, we remove the relevant inputs / arguments and check for more issues. If we find no issues, we match up any "valid" arguments, and start again. Note that there's a lot of extra complexity: - We try to stay efficient on the happy path, only computing the diagonal until we find a problem, and then filling in the rest of the matrix. - Closure arguments are wrapped in a tuple and need to be unwrapped - We need to resolve closure types after the rest, to allow the most specific type constraints - We need to handle imported C functions that might be variadic in their inputs. I tried to document a lot of this in comments in the code and keep the naming clear.
2022-01-21 22:50:54 -06:00
LL | fn foo<N>(_x: N) {}
| ^^^ -----
Use smaller spans for some structured suggestions Use more accurate suggestion spans for * argument parse error * fully qualified path * missing code block type * numeric casts * E0212
2021-08-10 05:53:43 -05:00
help: you can convert an `i8` to an `i32`
|
LL | foo::<i32>((-x_i8).into());
| + ++++++++
Fix into() cast paren check precedence As discussed in #47699 the logic for determining if an expression needs parenthesis when suggesting an `.into()` cast is incorrect. Two broken examples from nightly are: ``` error[E0308]: mismatched types --> main.rs:4:10 | 4 | test(foo as i8); | ^^^^^^^^^ expected i32, found i8 help: you can cast an `i8` to `i32`, which will sign-extend the source value | 4 | test(foo as i8.into()); | ``` ``` error[E0308]: mismatched types --> main.rs:4:10 | 4 | test(*foo); | ^^^^ expected i32, found i8 help: you can cast an `i8` to `i32`, which will sign-extend the source value | 4 | test(*foo.into()); | ``` As suggested by @petrochenkov switch the precedence check to PREC_POSTFIX. This catches both `as` and unary operators. Fixes #47699.
2018-01-24 03:31:36 -06:00
Suggest try_into when possible
2019-04-21 17:44:23 -05:00
error: aborting due to 113 previous errors
Suggest casting on numeric type error
2018-01-06 23:10:51 -06:00
update tests
2018-03-03 08:59:40 -06:00
For more information about this error, try `rustc --explain E0308`.
Reference in New Issue Copy Permalink