For slightly complex data structures like `rustc_serialize::json::Json`, it is often convenient to have helper methods like `Json::as_string(&self) -> Option<&str>` that return a borrow of some component of `&self`.
However, when `RefCell`s are involved, keeping a `Ref` around is required to hold a borrow to the insides of a `RefCell`. But `Ref` so far only references the entirety of the contents of a `RefCell`, not a component. But there is no reason it couldn’t: `Ref` internally contains just a data reference and a borrow count reference. The two can be dissociated.
This adds a `map_ref` function that creates a new `Ref` for some other data, but borrowing the same `RefCell` as an existing `Ref`.
Example:
```rust
struct RefCellJson(RefCell<Json>);
impl RefCellJson {
fn as_string(&self) -> Option<Ref<str>> {
map_ref(self.borrow(), |j| j.as_string())
}
}
```
r? @alexcrichton
The first commit simply forwards `io::Error`'s `cause` implementation to the inner error.
The second commit adds accessor methods for the inner error. Method names mirror those used elsewhere like `BufReader`.
r? @alexcrichton
`core::cell::Cell<T>` and `core::cell::RefCell<T>` currently implement `PartialEq` when `T` does, and just defer to comparing `T` values. There is no reason the same shouldn’t apply to `Eq`.
This enables `#[derive(Eq, PartialEq)]` on e.g. structs that have a `RefCell` field.
r? @alexcrichton
I’m unsure what to do with `#[stable]` attributes on `impl`s. `impl`s generated by `#[derive]` don’t have them.
collections: Make BinaryHeap panic safe in sift_up / sift_down
Use a struct called Hole that keeps track of an invalid location
in the vector and fills the hole on drop.
I include a run-pass test that the current BinaryHeap fails, and the new
one passes.
NOTE: The BinaryHeap will still be inconsistent after a comparison fails. It will
not have the heap property. What we fix is just that elements will be valid
values.
This is actually a performance win -- the new code does not bother to write in `zeroed()`
values in the holes, it just leaves them as they were.
Net result is something like a 5% decrease in runtime for `BinaryHeap::from_vec`. This
can be further improved by using unchecked indexing (I confirmed it makes a difference,
not a surprise with the non-sequential access going on), but let's leave that for another PR.
Safety first 😉Fixes#25842
Use a struct called Hole that keeps track of an invalid location
in the vector and fills the hole on drop.
I include a run-pass test that the current BinaryHeap fails, and the new
one passes.
Fixes#25842
Windows tests can often deadlock if a child thread continues after the main
thread and then panics, and a `println!` executed in a child thread after the
main thread has exited is at risk of panicking.
Currently, there are two conditional blocks that exist to check for "clang or gcc"
On line 866:
```
if [ -z "$CFG_ENABLE_CLANG" -a -z "$CFG_GCC" ]
then
err "either clang or gcc is required"
fi
```
and on line 1019:
```
if [ -z "$CC" -a -z "$CFG_ENABLE_CLANG" -a -z "$CFG_GCC" ]
then
err "either clang or gcc is required"
fi
```
Given the order of the clauses, this results in the "either clang or gcc is required" error from the earlier block, (even) when CC is set.
The expected behaviour is to honour user-flags, in this case CC.
Aside from removing all hand-holdy compiler checks in favour of actual compiler *feature* checks, this change removes the redundant former block in favour of the latter block, which appears designed to allow the expected behaviour.
The changes scaled back in 4cc025d8 were a little too aggressive and broke a
bunch of cross compilations by not defining the `LINK_$(1)` variable for all
targets. This commit ensures that the variable is defined for all targets by
defaulting it to the normal compiler if it's not already defined (it's only
defined specially for MSVC).
Closes#25723Closes#25802
Windows tests can often deadlock if a child thread continues after the main
thread and then panics, and a `println!` executed in a child thread after the
main thread has exited is at risk of panicking.
The current codegen tests only compare IR line counts between similar
rust and C programs, the latter getting compiled with clang. That looked
like a good idea back then, but actually things like lifetime intrinsics
mean that less IR isn't always better, so the metric isn't really
helpful.
Instead, we can start doing tests that check specific aspects of the
generated IR, like attributes or metadata. To do that, we can use LLVM's
FileCheck tool which has a number of useful features for such tests.
To start off, I created some tests for a few things that were recently
added and/or broken.
The current codegen tests only compare IR line counts between similar
rust and C programs, the latter getting compiled with clang. That looked
like a good idea back then, but actually things like lifetime intrinsics
mean that less IR isn't always better, so the metric isn't really
helpful.
Instead, we can start doing tests that check specific aspects of the
generated IR, like attributes or metadata. To do that, we can use LLVM's
FileCheck tool which has a number of useful features for such tests.
To start off, I created some tests for a few things that were recently
added and/or broken.
Currently, for `use` declarations with multiple paths, only the `use` item itself is saved in the AST map, not the individual path nodes. This can lead to a problem when a span of a specific path node is needed.
For example, #24818 caused an ICE because of this, in `ImportResolver::check_for_conflicting_import()`.
Fixes#25763.
