mikros/rust
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
91aae58568
rust/tests/ui/proc-macro/allowed-attr-stmt-expr.stdout

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

324 lines
9.5 KiB
Plaintext
Raw Normal View History

Improve `print_tts` by changing `tokenstream::Spacing`. `tokenstream::Spacing` appears on all `TokenTree::Token` instances, both punct and non-punct. Its current usage: - `Joint` means "can join with the next token *and* that token is a punct". - `Alone` means "cannot join with the next token *or* can join with the next token but that token is not a punct". The fact that `Alone` is used for two different cases is awkward. This commit augments `tokenstream::Spacing` with a new variant `JointHidden`, resulting in: - `Joint` means "can join with the next token *and* that token is a punct". - `JointHidden` means "can join with the next token *and* that token is a not a punct". - `Alone` means "cannot join with the next token". This *drastically* improves the output of `print_tts`. For example, this: ``` stringify!(let a: Vec<u32> = vec![];) ``` currently produces this string: ``` let a : Vec < u32 > = vec! [] ; ``` With this PR, it now produces this string: ``` let a: Vec<u32> = vec![] ; ``` (The space after the `]` is because `TokenTree::Delimited` currently doesn't have spacing information. The subsequent commit fixes this.) The new `print_tts` doesn't replicate original code perfectly. E.g. multiple space characters will be condensed into a single space character. But it's much improved. `print_tts` still produces the old, uglier output for code produced by proc macros. Because we have to translate the generated code from `proc_macro::Spacing` to the more expressive `token::Spacing`, which results in too much `proc_macro::Along` usage and no `proc_macro::JointHidden` usage. So `space_between` still exists and is used by `print_tts` in conjunction with the `Spacing` field. This change will also help with the removal of `Token::Interpolated`. Currently interpolated tokens are pretty-printed nicely via AST pretty printing. `Token::Interpolated` removal will mean they get printed with `print_tts`. Without this change, that would result in much uglier output for code produced by decl macro expansions. With this change, AST pretty printing and `print_tts` produce similar results. The commit also tweaks the comments on `proc_macro::Spacing`. In particular, it refers to "compound tokens" rather than "multi-char operators" because lifetimes aren't operators.
2023-08-07 20:43:44 -05:00
PRINT-ATTR INPUT (DISPLAY): struct ItemWithSemi;
Add additional semicolon test
2020-11-25 15:58:10 -06:00
PRINT-ATTR INPUT (DEBUG): TokenStream [
Ident {
ident: "struct",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:49:1: 49:7 (#0),
Add additional semicolon test
2020-11-25 15:58:10 -06:00
},
Ident {
ident: "ItemWithSemi",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:49:8: 49:20 (#0),
Add additional semicolon test
2020-11-25 15:58:10 -06:00
},
Punct {
ch: ';',
spacing: Alone,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:49:20: 49:21 (#0),
Add additional semicolon test
2020-11-25 15:58:10 -06:00
},
]
Pretty print empty blocks as {}
2021-12-01 13:45:14 -06:00
PRINT-ATTR INPUT (DISPLAY): #[rustc_dummy] struct Foo {}
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
PRINT-ATTR INPUT (DEBUG): TokenStream [
Punct {
ch: '#',
spacing: Alone,
Hide `compiler_builtins` in the prelude This crate is a private implementation detail. We only need to insert it into the crate graph for linking and should not expose any of its public API. Fixes #113533
2023-07-10 17:16:28 -05:00
span: $DIR/allowed-attr-stmt-expr.rs:35:9: 35:10 (#10),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
Group {
delimiter: Bracket,
stream: TokenStream [
Ident {
Add additional semicolon test
2020-11-25 15:58:10 -06:00
ident: "rustc_dummy",
Hide `compiler_builtins` in the prelude This crate is a private implementation detail. We only need to insert it into the crate graph for linking and should not expose any of its public API. Fixes #113533
2023-07-10 17:16:28 -05:00
span: $DIR/allowed-attr-stmt-expr.rs:35:11: 35:22 (#10),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
],
Hide `compiler_builtins` in the prelude This crate is a private implementation detail. We only need to insert it into the crate graph for linking and should not expose any of its public API. Fixes #113533
2023-07-10 17:16:28 -05:00
span: $DIR/allowed-attr-stmt-expr.rs:35:10: 35:23 (#10),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
Ident {
ident: "struct",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:53:16: 53:22 (#0),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
Ident {
ident: "Foo",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:53:23: 53:26 (#0),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
Group {
delimiter: Brace,
stream: TokenStream [],
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:53:27: 53:29 (#0),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
]
Improve `print_tts` by changing `tokenstream::Spacing`. `tokenstream::Spacing` appears on all `TokenTree::Token` instances, both punct and non-punct. Its current usage: - `Joint` means "can join with the next token *and* that token is a punct". - `Alone` means "cannot join with the next token *or* can join with the next token but that token is not a punct". The fact that `Alone` is used for two different cases is awkward. This commit augments `tokenstream::Spacing` with a new variant `JointHidden`, resulting in: - `Joint` means "can join with the next token *and* that token is a punct". - `JointHidden` means "can join with the next token *and* that token is a not a punct". - `Alone` means "cannot join with the next token". This *drastically* improves the output of `print_tts`. For example, this: ``` stringify!(let a: Vec<u32> = vec![];) ``` currently produces this string: ``` let a : Vec < u32 > = vec! [] ; ``` With this PR, it now produces this string: ``` let a: Vec<u32> = vec![] ; ``` (The space after the `]` is because `TokenTree::Delimited` currently doesn't have spacing information. The subsequent commit fixes this.) The new `print_tts` doesn't replicate original code perfectly. E.g. multiple space characters will be condensed into a single space character. But it's much improved. `print_tts` still produces the old, uglier output for code produced by proc macros. Because we have to translate the generated code from `proc_macro::Spacing` to the more expressive `token::Spacing`, which results in too much `proc_macro::Along` usage and no `proc_macro::JointHidden` usage. So `space_between` still exists and is used by `print_tts` in conjunction with the `Spacing` field. This change will also help with the removal of `Token::Interpolated`. Currently interpolated tokens are pretty-printed nicely via AST pretty printing. `Token::Interpolated` removal will mean they get printed with `print_tts`. Without this change, that would result in much uglier output for code produced by decl macro expansions. With this change, AST pretty printing and `print_tts` produce similar results. The commit also tweaks the comments on `proc_macro::Spacing`. In particular, it refers to "compound tokens" rather than "multi-char operators" because lifetimes aren't operators.
2023-08-07 20:43:44 -05:00
PRINT-ATTR INPUT (DISPLAY): #[expect_let] let string = "Hello, world!";
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
PRINT-ATTR INPUT (DEBUG): TokenStream [
Punct {
ch: '#',
spacing: Alone,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:56:5: 56:6 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Group {
delimiter: Bracket,
stream: TokenStream [
Ident {
ident: "expect_let",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:56:7: 56:17 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
],
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:56:6: 56:18 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Ident {
ident: "let",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:57:5: 57:8 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Ident {
ident: "string",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:57:9: 57:15 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Punct {
ch: '=',
spacing: Alone,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:57:16: 57:17 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Literal {
kind: Str,
symbol: "Hello, world!",
suffix: None,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:57:18: 57:33 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Punct {
ch: ';',
spacing: Alone,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:57:33: 57:34 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
]
Add spacing information to delimiters. This is an extension of the previous commit. It means the output of something like this: ``` stringify!(let a: Vec<u32> = vec![];) ``` goes from this: ``` let a: Vec<u32> = vec![] ; ``` With this PR, it now produces this string: ``` let a: Vec<u32> = vec![]; ```
2023-10-11 23:36:14 -05:00
PRINT-ATTR INPUT (DISPLAY): #[expect_my_macro_stmt] my_macro!("{}", string);
Don't insert spaces before most semicolons in `print_tts`. This gives better output for code produced by proc macros.
2023-11-09 21:34:56 -06:00
PRINT-ATTR RE-COLLECTED (DISPLAY): #[expect_my_macro_stmt] my_macro! ("{}", string);
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
PRINT-ATTR INPUT (DEBUG): TokenStream [
Punct {
ch: '#',
spacing: Alone,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:60:5: 60:6 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Group {
delimiter: Bracket,
stream: TokenStream [
Ident {
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
ident: "expect_my_macro_stmt",
span: $DIR/allowed-attr-stmt-expr.rs:60:7: 60:27 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
],
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:60:6: 60:28 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Ident {
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
ident: "my_macro",
span: $DIR/allowed-attr-stmt-expr.rs:61:5: 61:13 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Punct {
ch: '!',
spacing: Alone,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:61:13: 61:14 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Group {
delimiter: Parenthesis,
stream: TokenStream [
Literal {
kind: Str,
symbol: "{}",
suffix: None,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:61:15: 61:19 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Punct {
ch: ',',
spacing: Alone,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:61:19: 61:20 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Ident {
ident: "string",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:61:21: 61:27 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
],
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:61:14: 61:28 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Punct {
ch: ';',
spacing: Alone,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:61:28: 61:29 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
]
Add spacing information to delimiters. This is an extension of the previous commit. It means the output of something like this: ``` stringify!(let a: Vec<u32> = vec![];) ``` goes from this: ``` let a: Vec<u32> = vec![] ; ``` With this PR, it now produces this string: ``` let a: Vec<u32> = vec![]; ```
2023-10-11 23:36:14 -05:00
PRINT-ATTR INPUT (DISPLAY): second_make_stmt!(#[allow(dead_code)] struct Bar {});
Don't insert spaces before most semicolons in `print_tts`. This gives better output for code produced by proc macros.
2023-11-09 21:34:56 -06:00
PRINT-ATTR RE-COLLECTED (DISPLAY): second_make_stmt! (#[allow(dead_code)] struct Bar {});
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
PRINT-ATTR INPUT (DEBUG): TokenStream [
Ident {
ident: "second_make_stmt",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:5: 64:21 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Punct {
ch: '!',
spacing: Alone,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:21: 64:22 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Group {
delimiter: Parenthesis,
stream: TokenStream [
Punct {
ch: '#',
spacing: Alone,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:23: 64:24 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Group {
delimiter: Bracket,
stream: TokenStream [
Ident {
ident: "allow",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:25: 64:30 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Group {
delimiter: Parenthesis,
stream: TokenStream [
Ident {
ident: "dead_code",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:31: 64:40 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
],
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:30: 64:41 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
],
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:24: 64:42 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Ident {
ident: "struct",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:43: 64:49 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Ident {
ident: "Bar",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:50: 64:53 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Group {
delimiter: Brace,
stream: TokenStream [],
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:54: 64:56 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
],
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:22: 64:57 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Punct {
ch: ';',
spacing: Alone,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:57: 64:58 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
]
Pretty print empty blocks as {}
2021-12-01 13:45:14 -06:00
PRINT-ATTR INPUT (DISPLAY): #[rustc_dummy] #[allow(dead_code)] struct Bar {}
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
PRINT-ATTR INPUT (DEBUG): TokenStream [
Punct {
ch: '#',
spacing: Alone,
Hide `compiler_builtins` in the prelude This crate is a private implementation detail. We only need to insert it into the crate graph for linking and should not expose any of its public API. Fixes #113533
2023-07-10 17:16:28 -05:00
span: $DIR/allowed-attr-stmt-expr.rs:35:9: 35:10 (#31),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
Group {
delimiter: Bracket,
stream: TokenStream [
Ident {
Add additional semicolon test
2020-11-25 15:58:10 -06:00
ident: "rustc_dummy",
Hide `compiler_builtins` in the prelude This crate is a private implementation detail. We only need to insert it into the crate graph for linking and should not expose any of its public API. Fixes #113533
2023-07-10 17:16:28 -05:00
span: $DIR/allowed-attr-stmt-expr.rs:35:11: 35:22 (#31),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
],
Hide `compiler_builtins` in the prelude This crate is a private implementation detail. We only need to insert it into the crate graph for linking and should not expose any of its public API. Fixes #113533
2023-07-10 17:16:28 -05:00
span: $DIR/allowed-attr-stmt-expr.rs:35:10: 35:23 (#31),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
Punct {
ch: '#',
spacing: Alone,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:23: 64:24 (#0),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
Group {
delimiter: Bracket,
stream: TokenStream [
Ident {
ident: "allow",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:25: 64:30 (#0),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
Group {
delimiter: Parenthesis,
stream: TokenStream [
Ident {
ident: "dead_code",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:31: 64:40 (#0),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
],
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:30: 64:41 (#0),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
],
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:24: 64:42 (#0),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
Ident {
ident: "struct",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:43: 64:49 (#0),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
Ident {
ident: "Bar",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:50: 64:53 (#0),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
Group {
delimiter: Brace,
stream: TokenStream [],
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:64:54: 64:56 (#0),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
]
Pretty print empty blocks as {}
2021-12-01 13:45:14 -06:00
PRINT-ATTR INPUT (DISPLAY): #[rustc_dummy] struct Other {}
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
PRINT-ATTR INPUT (DEBUG): TokenStream [
Punct {
ch: '#',
spacing: Alone,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:67:5: 67:6 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Group {
delimiter: Bracket,
stream: TokenStream [
Ident {
ident: "rustc_dummy",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:67:7: 67:18 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
],
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:67:6: 67:19 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Ident {
ident: "struct",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:68:5: 68:11 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Ident {
ident: "Other",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:68:12: 68:17 (#0),
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
},
Group {
delimiter: Brace,
stream: TokenStream [],
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:68:18: 68:20 (#0),
Add additional semicolon test
2020-11-25 15:58:10 -06:00
},
]
Improve `print_tts` by changing `tokenstream::Spacing`. `tokenstream::Spacing` appears on all `TokenTree::Token` instances, both punct and non-punct. Its current usage: - `Joint` means "can join with the next token *and* that token is a punct". - `Alone` means "cannot join with the next token *or* can join with the next token but that token is not a punct". The fact that `Alone` is used for two different cases is awkward. This commit augments `tokenstream::Spacing` with a new variant `JointHidden`, resulting in: - `Joint` means "can join with the next token *and* that token is a punct". - `JointHidden` means "can join with the next token *and* that token is a not a punct". - `Alone` means "cannot join with the next token". This *drastically* improves the output of `print_tts`. For example, this: ``` stringify!(let a: Vec<u32> = vec![];) ``` currently produces this string: ``` let a : Vec < u32 > = vec! [] ; ``` With this PR, it now produces this string: ``` let a: Vec<u32> = vec![] ; ``` (The space after the `]` is because `TokenTree::Delimited` currently doesn't have spacing information. The subsequent commit fixes this.) The new `print_tts` doesn't replicate original code perfectly. E.g. multiple space characters will be condensed into a single space character. But it's much improved. `print_tts` still produces the old, uglier output for code produced by proc macros. Because we have to translate the generated code from `proc_macro::Spacing` to the more expressive `token::Spacing`, which results in too much `proc_macro::Along` usage and no `proc_macro::JointHidden` usage. So `space_between` still exists and is used by `print_tts` in conjunction with the `Spacing` field. This change will also help with the removal of `Token::Interpolated`. Currently interpolated tokens are pretty-printed nicely via AST pretty printing. `Token::Interpolated` removal will mean they get printed with `print_tts`. Without this change, that would result in much uglier output for code produced by decl macro expansions. With this change, AST pretty printing and `print_tts` produce similar results. The commit also tweaks the comments on `proc_macro::Spacing`. In particular, it refers to "compound tokens" rather than "multi-char operators" because lifetimes aren't operators.
2023-08-07 20:43:44 -05:00
PRINT-ATTR INPUT (DISPLAY): #[rustc_dummy] struct NonBracedStruct;
Add additional semicolon test
2020-11-25 15:58:10 -06:00
PRINT-ATTR INPUT (DEBUG): TokenStream [
Punct {
ch: '#',
spacing: Alone,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:73:5: 73:6 (#0),
Add additional semicolon test
2020-11-25 15:58:10 -06:00
},
Group {
delimiter: Bracket,
stream: TokenStream [
Ident {
ident: "rustc_dummy",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:73:7: 73:18 (#0),
Add additional semicolon test
2020-11-25 15:58:10 -06:00
},
],
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:73:6: 73:19 (#0),
Add additional semicolon test
2020-11-25 15:58:10 -06:00
},
Ident {
ident: "struct",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:74:5: 74:11 (#0),
Add additional semicolon test
2020-11-25 15:58:10 -06:00
},
Ident {
ident: "NonBracedStruct",
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:74:12: 74:27 (#0),
Add additional semicolon test
2020-11-25 15:58:10 -06:00
},
Punct {
ch: ';',
spacing: Alone,
Use custom macro instead of println Loading a macro from libstd causes us to load serialized `SyntaxContext`s in a platform-dependent way, causing the printed spans to differ between platforms.
2020-11-26 20:44:09 -06:00
span: $DIR/allowed-attr-stmt-expr.rs:74:27: 74:28 (#0),
Properly handle attributes on statements We now collect tokens for the underlying node wrapped by `StmtKind` instead of storing tokens directly in `Stmt`. `LazyTokenStream` now supports capturing a trailing semicolon after it is initially constructed. This allows us to avoid refactoring statement parsing to wrap the parsing of the semicolon in `parse_tokens`. Attributes on item statements (e.g. `fn foo() { #[bar] struct MyStruct; }`) are now treated as item attributes, not statement attributes, which is consistent with how we handle attributes on other kinds of statements. The feature-gating code is adjusted so that proc-macro attributes are still allowed on item statements on stable. Two built-in macros (`#[global_allocator]` and `#[test]`) needed to be adjusted to support being passed `Annotatable::Stmt`.
2020-11-17 13:27:44 -06:00
},
Invoke attributes on the statement for statement items
2020-11-24 13:47:49 -06:00
]
Reference in New Issue Copy Permalink