| Age | Commit message (Collapse) | Author | Lines |
|---|
|
|
|
When suggesting to import an item, also suggest changing the path if appropriate
When we don't find an item we search all of them for an appropriate
import and suggest `use`ing it. This is sometimes done for expressions
that have paths with more than one segment. We now also suggest changing
that path to work with the `use`.
Fix #95413
|
|
Fix unit struct/enum variant in destructuring assignment
See https://github.com/rust-lang/rfcs/blob/master/text/2909-destructuring-assignment.md#guide-level-explanation, "including **unit** and tuple structs"
Fixes #94319
|
|
Mitigate impact of subtle invalid call suggestion logic
There's some subtle interaction between inferred expressions being
passed as an argument to fn calls with fewer than expected arguments. To
avoid the ICE, I'm changing indexing operations with `.get(idx)`, but
the underlying logic still needs to be audited as it was written with
the assumption that `final_arg_types` and `provided_args` have the right
length.
Address #96638.
|
|
ken-matsui:fix-invalid-keyword-order-for-function-declarations, r=davidtwco
Fix invalid keyword order for function declarations
Closes: https://github.com/rust-lang/rust/issues/94879
|
|
Add a regression test for #92305
Closes #92305
r? ``@oli-obk`` as per https://github.com/rust-lang/rust/issues/92305#issuecomment-1114043188
|
|
Use source callsite in check_argument_types suggestion
This makes the "remove extra arguement" suggestion valid when the function argument is a macro.
Additionally, this may fix #96225, but the only way I can reproduce that issue is using the playground, so we will need to wait until after this is merged to ensure it's fixed.
|
|
Switch settings menu to full js
Since the settings can only be set when the JS is enabled, it's not really a problem. It also fixes a debate we had around the themes not being accessible easily before.
You can test it [here](https://rustdoc.crud.net/imperio/settings-js/doc/foo/index.html).
r? ``@jsha``
|
|
|
|
When we don't find an item we search all of them for an appropriate
import and suggest `use`ing it. This is sometimes done for expressions
that have paths with more than one segment. We now also suggest changing
that path to work with the `use`.
Fix #95413
|
|
There's some subtle interaction between inferred expressions being
passed as an argument to fn calls with fewer than expected arguments. To
avoid the ICE, I'm changing indexing operations with `.get(idx)`, but
the underlying logic still needs to be audited as it was written with
the assumption that `final_arg_types` and `provided_args` have the right
length.
Address 96638.
|
|
|
|
Tweak the vec-calloc runtime check to only apply to shortish-arrays
r? `@Mark-Simulacrum`
`@nbdd0121` pointed out in https://github.com/rust-lang/rust/pull/95362#issuecomment-1114085395 that LLVM currently doesn't constant-fold the `IsZero` check for long arrays, so that seems like a reasonable justification for limiting it.
It appears that it's based on length, not byte size, (https://godbolt.org/z/4s48Y81dP), so that's what I used in the PR. Maybe it's a ["the number of inlining shall be three"](https://youtu.be/s4wnuiCwTGU?t=320) sort of situation.
Certainly there's more that could be done here -- that generated code that checks long arrays byte-by-byte is highly suboptimal, for example -- but this is an easy, low-risk tweak.
|
|
Move some tests to more reasonable places
cc https://github.com/rust-lang/rust/issues/73494
r? `@petrochenkov`
16602 -> `codegen` because of https://github.com/rust-lang/rust/issues/16602#issuecomment-53806665
|
|
|
|
Add a bathroom stall to weird expressions test
|
|
|
|
rustdoc: Resolve doc links referring to `macro_rules` items
cc https://github.com/rust-lang/rust/issues/81633
UPD: the fallback to considering *all* `macro_rules` in the crate for unresolved names is not removed in this PR, it will be removed separately and will be run through crater.
|
|
Move Derefer before Retag
_Follow up work to #96116 #95857 #95649_
This moves `Derefer` before `Retag` and creates a new `LocalInfo` called `Temp` to avoid retagging created temp values.
Zulip discussion [link](https://rust-lang.zulipchat.com/#narrow/stream/189540-t-compiler.2Fwg-mir-opt/topic/deref.20as.20first.20and.20only.20projection)
r? `@oli-obk`
|
|
Add `do yeet` expressions to allow experimentation in nightly
Two main goals for this:
- Ensure that trait restructuring in https://github.com/rust-lang/rust/issues/84277#issuecomment-1066120333 doesn't accidentally close us off from the possibility of doing this in future, as sketched in https://rust-lang.github.io/rfcs/3058-try-trait-v2.html#possibilities-for-yeet
- Experiment with the *existence* of syntax for this, to be able to weight the syntax-vs-library tradeoffs better than we can right now. Notably the syntax (with `do`) and name in this PR are not intended as candidates for stabilization, but they make a good v0 PR for adding this with minimal impact to compiler maintenance or priming one possible name choice over another.
r? `@oli-obk`
The lang `second` for doing this: https://github.com/rust-lang/lang-team/issues/160#issuecomment-1107896716
Tracking issues
- Lang, https://github.com/rust-lang/rust/issues/96373
- Libs-api, https://github.com/rust-lang/rust/issues/96374
|
|
|
|
Also report the call site of PME errors locally.
Note this does not produce a full stack all the way to the first call that specifies all monomorphic parameters, it's just shallowly mentioning the last call site.
previous work: https://github.com/rust-lang/rust/pull/85633
tracking issue: https://github.com/rust-lang/rust/issues/85155
r? `@lqd`
I figured we could get some improvement for traces in local crates without going into the backtrace hell you landed in last time
|
|
Make [e]println macros eagerly drop temporaries (for backport)
This PR extracts the subset of #96455 which is only the parts necessary for fixing the 1.61-beta regressions in #96434.
My larger PR #96455 contains a few other changes relative to the pre-#94868 behavior; those are not necessary to backport into 1.61.
argument position | before #94868 | after #94868 | after this PR
--- |:---:|:---:|:---:
`write!($tmp, "…", …)` | :rage: | :rage: | :rage:
`write!(…, "…", $tmp)` | :rage: | :rage: | :rage:
`writeln!($tmp, "…", …)` | :rage: | :rage: | :rage:
`writeln!(…, "…", $tmp)` | :rage: | :rage: | :rage:
`print!("…", $tmp)` | :rage: | :rage: | :rage:
`println!("…", $tmp)` | :smiley_cat: | :rage: | :smiley_cat:
`eprint!("…", $tmp)` | :rage: | :rage: | :rage:
`eprintln!("…", $tmp)` | :smiley_cat: | :rage: | :smiley_cat:
`panic!("…", $tmp)` | :smiley_cat: | :smiley_cat: | :smiley_cat:
|
|
This reverts commit ac5c657a0801db84b29ea9b3ae322107756575b0.
|
|
|
|
|
|
Using an obviously-placeholder syntax. An RFC would still be needed before this could have any chance at stabilization, and it might be removed at any point.
But I'd really like to have it in nightly at least to ensure it works well with try_trait_v2, especially as we refactor the traits.
|
|
This way links referring to `macro_rules` items are resolved correctly
|
|
This makes the "remove extra arguement" suggestion valid when the
function argument is a macro
|
|
Erase type params when suggesting fully qualified path
When suggesting the use of a fully qualified path for a method call that
is ambiguous because it has multiple candidates, erase type params in
the resulting code, as they would result in an error when applied. We
replace them with `_` in the output to rely on inference. There might be
cases where this still produces slighlty incomplete suggestions, but it
otherwise produces many more errors in relatively common cases.
Fix #96292
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Note this does not produce a full stack all the way to the first call that specifies all monomorphic parameters, it's just shallowly mentioning the last call site.
|
|
Enforce static lifetimes in consts during late resolution
This PR moves the handling of implicitly and explicitly static lifetimes in constants from HIR to the AST.
|
|
|
|
the right path
|
|
|
|
Rollup of 6 pull requests
Successful merges:
- #96390 (Switch JS code to ES6 - part 2)
- #96527 (RustWrapper: explicitly don't handle DXILPointerTyID)
- #96536 (rustdoc: fix missing method list for primitive deref target)
- #96559 (Use the correct lifetime binder for elided lifetimes in path.)
- #96560 (Remove unnecessary environment variable in cf-protection documentation)
- #96562 (Fix duplicate directory separator in --remap-path-prefix.)
Failed merges:
r? `@ghost`
`@rustbot` modify labels: rollup
|
|
Fix duplicate directory separator in --remap-path-prefix.
The compiler will currently emit duplicate directory separators when `--remap-path-prefix` has an exact match of the working directory and it is invoked with a relative path to the main source file. For example
```bash
rustc src/main.rs -Cdebuginfo=2 --remap-path-prefix="$(pwd)=abc"
```
will produce the path `abc//src/main.rs` in debuginfo. This is because `some_path.join("")` will append a directory separator to `some_path` and then LLVM does not check if the working directory already ends a directory separator before concatenating it with the relative path.
|
|
Use the correct lifetime binder for elided lifetimes in path.
Fixes https://github.com/rust-lang/rust/issues/96540
|
|
rustdoc: fix missing method list for primitive deref target
This change makes it so that local impls count when listing primitives that need retained.
Fixes #95325
|
|
Enforce Copy bounds for repeat elements while considering lifetimes
fixes https://github.com/rust-lang/rust/issues/95477
this is a breaking change in order to fix a soundness bug.
Before this PR we only checked whether the repeat element type had an `impl Copy`, but not whether that impl also had the appropriate lifetimes. E.g. if the impl was for `YourType<'static>` and not a general `'a`, then copying any type other than a `'static` one should have been rejected, but wasn't.
r? `@lcnr`
|
|
|
|
|
|
|
|
|
