extend NLL with preliminary support for free regions on functions
This PR extends https://github.com/rust-lang/rust/pull/45538 with support for free regions. This is pretty preliminary and will no doubt want to change in various ways, particularly as we add support for closures, but it's enough to get the basic idea in place:
- We now create specific regions to represent each named lifetime declared on the function.
- Region values can contain references to these regions (represented for now as a `BTreeSet<RegionIndex>`).
- If we wind up trying to infer that `'a: 'b` must hold, but no such relationship was declared, we report an error.
It also does a number of drive-by refactorings.
r? @arielb1
cc @spastorino
new rules for merging expected and supplied types in closure signatures
As uncovered in #38714, we currently have some pretty bogus code for combining the "expected signature" of a closure with the "supplied signature". To set the scene, consider a case like this:
```rust
fn foo<F>(f: F)
where
F: for<'a> FnOnce(&'a u32) -> &'a u32
// ^ *expected* signature comes from this where-clause
{
...
}
fn main() {
foo(|x: &u32| -> &u32 { .. }
// ^^^^^^^^^^^^^^^^^ supplied signature
// comes from here
}
```
In this case, the supplied signature (a) includes all the parts and (b) is the same as the expected signature, modulo the names used for the regions. But often people supply only *some* parts of the signature. For example, one might write `foo(|x| ..)`, leaving *everything* to be inferred, or perhaps `foo(|x: &u32| ...)`, which leaves the return type to be inferred.
In the current code, we use the expected type to supply the types that are not given, but otherwise use the type the user gave, except for one case: if the user writes `fn foo(|x: _| ..)` (i.e., an underscore at the outermost level), then we will take the expected type (rather than instantiating a fresh type variable). This can result in nonsensical situations, particularly with bound regions that link the types of parameters to one another or to the return type. Consider `foo(|x: &u32| ...)` -- if we *literally* splice the expected return type of `&'a u32` together with what the user gave, we wind up with a signature like `for<'a> fn(&u32) -> &'a u32`. This is not even permitted as a type, because bound regions like `'a` must appear also in the arguments somewhere, which is why #38714 leads to an ICE.
This PR institutes some new rules. These are not meant to be the *final* set of rules, but they are a kind of "lower bar" for what kind of code we accept (i.e., we can extend these rules in the future to be smarter in some cases, but -- as we will see -- these rules do accept some things that we then would not be able to back off from).
These rules are derived from a few premises:
- First and foremost, anonymous regions in closure annotation are mostly requests for the code to "figure out the right lifetime" and shouldn't be read too closely. So for example when people write a closure signature like `|x: &u32|`, they are really intended for us to "figure out" the right region for `x`.
- In contrast, the current code treats this supplied type as being more definitive. In particular, writing `|x: &u32|` would always result in the region of `x` being bound in the closure type. In other words, the signature would be something like `for<'a> fn(&'a u32)` -- this is derived from the fact that `fn(&u32)` expands to a type where the region is bound in the fn type.
- This PR takes a different approach. The "binding level" for reference types appearing in closure signatures can be informed in some cases by the expected signature. So, for example, if the expected signature is something like `(&'f u32)`, where the region of the first argument appears free, then for `|x: &u32|`, the new code would infer `x` to also have the free region `'f`.
- This inference has some limits. We don't do this for bindings that appear within the selected types themselves. So e.g. `|x: fn(&u32)|`, when combined with an expected type of `fn(fn(&'f u32))`, would still result in a closure that expects `for<'a> fn(&'a u32)`. Such an annotation will ultimately result in an error, as it happens, since `foo` is supplying a `fn(&'f u32)` to the closure, but the closure signature demands a `for<'a> fn(&'a u32)`. But still we choose to trust it and have the user change it.
- I wanted to preserve the rough intuition that one can copy-and-paste a type out of the fn signature and into the fn body without dramatically changing its meaning. Interestingly, if one has `|x: &u32|`, then regardless of whether the region of `x` is bound or free in the closure signature, it is also free in the region body, and that is also true when one writes `let x: &u32`, so that intuition holds here. But the same would not be true for `fn(&u32)`, hence the different behavior.
- Second, we must take either **all** the references to bound regions from the expected type or **none**. The current code, as we saw, will happily take a bound region in the return type but drop the other place where it is used, in the parameters. Since bound regions are all about linking multiple things together, I think it's important not to do that. (That said, we could conceivably be a bit less strict here, since the subtyping rules will get our back, but we definitely don't want any bound regions that appear only in the return type.)
- Finally, we cannot take the bound region names from the supplied types and "intermix" them with the names from the expected types.
- We *could* potentially do some alpha renaming, but I didn't do that.
- Ultimately, if the types the user supplied do not match expectations in some way that we cannot recover from, we fallback to deriving the closure signature solely from those expected types.
- For example, if the expected type is `u32` but the user wrote `i32`.
- Or, more subtle, if the user wrote e.g. `&'x u32` for some named lifetime `'x`, but the expected type includes a bound lifetime (`for<'a> (&'a u32)`). In that case, preferring the type that the user explicitly wrote would hide an appearance of a bound name from the expected type, and we try to never do that.
The detailed rules that I came up with are found in the code, but for ease of reading I've also [excerpted them into a gist](https://gist.github.com/nikomatsakis/e69252a2b57e6d97d044c2f254c177f1). I am not convinced they are correct and would welcome feedback for alternative approaches.
(As an aside, the way I think I would ultimately *prefer* to think about this is that the conversion from HIR types to internal types could be parameterized by an "expected type" that it uses to guide itself. However, since that would be a pain, I opted *in the code* to first instantiate the supplied types as `Ty<'tcx>` and then "merge" those types with the `Ty<'tcx>` from the expected signature.)
I think we should probably FCP this before landing.
cc @rust-lang/lang
r? @arielb1
Copy all `AsciiExt` methods to the primitive types directly in order to deprecate it later
**EDIT:** [this PR is ready now](https://github.com/rust-lang/rust/pull/44042#issuecomment-333883548). I edited this post to reflect the current status of discussion, which is (apart from code review) pretty much settled.
---
This is my current progress in order to prepare stabilization of #39658. As discussed there (and in #39659), the idea is to deprecated `AsciiExt` and copy all methods to the type directly. Apparently there isn't really a reason to have those methods in an extension trait¹.
~~This is **work in progress**: copy&pasting code while slightly modifying the documentation isn't the most exciting thing to do. Therefore I wanted to already open this WIP PR after doing basically 1/4 of the job (copying methods to `&[u8]`, `char` and `&str` is still missing) to get some feedback before I continue. Some questions possibly worth discussing:~~
1. ~~Does everyone agree that deprecating `AsciiExt` is a good idea? Does everyone agree with the goal of this PR?~~ => apparently yes
2. ~~Are my changes OK so far? Did I do something wrong?~~
3. ~~The issue of the unstable-attribute is currently set to 0. I would wait until you say "Ok" to the whole thing, then create a tracking issue and then insert the correct issue id. Is that ok?~~
4. ~~I tweaked `eq_ignore_ascii_case()`: it now takes the argument `other: u8` instead of `other: &u8`. The latter was enforced by the trait. Since we're not bound to a trait anymore, we can drop the reference, ok?~~ => I reverted this, because the interface has to match the `AsciiExt` interface exactly.
¹ ~~Could it be that we can't write `impl [u8] {}`? This might be the reason for `AsciiExt`. If that is the case: is there a good reason we can't write such an impl block? What can we do instead?~~ => we couldn't at the time this PR was opened, but Simon made it possible.
/cc @SimonSapin @zackw
Add a nicer error message for missing in for loop, fixes#40782.
As suggested by @estebank in issue #40782, this works in the same way as #42578: if the in keyword is missing, we continue parsing the expression and if this works correctly an adapted error message is produced. Otherwise we return the old error.
A specific test case has also been added.
This is my first PR on rust-lang/rust so any feedback is very welcome.
Many AsciiExt imports have become useless thanks to the inherent ascii
methods added in the last commits. These were removed. In some places, I
fully specified the ascii method being called to enforce usage of the
AsciiExt trait. Note that some imports are not removed but tagged with
a `#[cfg(stage0)]` attribute. This is necessary, because certain ascii
methods are not yet available in stage0. All those imports will be
removed later.
Additionally, failing tests were fixed. The test suite should exit
successfully now.
`unreachable-pub` lint (as authorized by RFC 2126)
To whom it may concern:
RFC 2126 commissions the creation of a lint for `pub` items that are not visible from crate root (#45521). We understand (but seek confirmation from more knowledgable compiler elders) that this can be implemented by linting HIR items that are _not_ ~~`cx.access_levels.is_exported`~~ `cx.access_levels.is_reachable` but have a `vis` (-ibility) field of `hir::Visibility::Public`.
The lint, tentatively called ~~`unexported-pub`~~ `unreachable-pub` (with the understanding that much could be written on the merits of various names, as it is said of the colors of bicycle-sheds), suggests `crate` as a replacement for `pub` if the `crate_visibility_modifier` feature is enabled (see #45388), and `pub(crate)` otherwise. We also use help messaging to suggest the other potential fix of exporting the item; feedback is desired as to whether this may be confusing or could be worded better.
As a preview of what respecting the proposed lint would look like (and to generate confirmatory evidence that this implementation doesn't issue false positives), ~~we take its suggestions for `libcore`~~ (save one, which is deferred to another pull request because it brings up an unrelated technical matter). I remain your obedient servant.
r? @petrochenkov
typeck: suggest use of match_default_bindings feature
Fixes#45383.
Updates #42640.
r? @nikomatsakis
cc @tschottdorf
This needs a UI test, but thought I'd get some early feedback.
Improve display of error E0308
Ref. Forgetting to call a variant constructor causes a confusing error message #35241.
This PR modifies [`note_type_err`](b7041bfab3/src/librustc/infer/error_reporting/mod.rs (L669-L674)) to display a `help` message when a `TyFnPtr` or `TyFnDef` are found and the return type, of the function or function pointer, is the same as the type that is expected.
The output of compiling
```rust
struct Foo(u32);
fn test() -> Foo { Foo }
fn main() {}
```
is now
```bash
$ rustc src/test/ui/issue-35241.rs
error[E0308]: mismatched types
--> src/test/ui/issue-35241.rs:13:20
|
13 | fn test() -> Foo { Foo }
| --- ^^^ expected struct `Foo`, found fn item
| |
| expected `Foo` because of return type
|
= help: did you mean `Foo { /* fields */ }`?
= note: expected type `Foo`
found type `fn(u32) -> Foo {Foo::{{constructor}}}`
error: aborting due to previous error
```
Suggest renaming import if names clash
Closes https://github.com/rust-lang/rust/issues/32354.
The output for the example in the issue looks like this:
```
~/p/local-rust-dev-testing ❯❯❯ cargo +local-s1 build
Compiling local-rust-dev-testing v0.1.0 (file:///home/cldfire/programming_projects/local-rust-dev-testing)
error[E0252]: the name `ConstructorExtension` is defined multiple times
--> src/main.rs:49:5
|
48 | use extension1::ConstructorExtension;
| -------------------------------- previous import of the trait `ConstructorExtension` here
49 | use extension2::ConstructorExtension;
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ `ConstructorExtension` reimported here
|
= note: `ConstructorExtension` must be defined only once in the type namespace of this module
help: You can use `as` to change the binding name of the import
|
49 | use extension2::ConstructorExtension as OtherConstructorExtension;
| ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
...
```
This is my first PR that touches the compiler in any way, so if there's something else I need to do here (e.g. add a test), please let me know :).
This multiline string literal was missing a backslash, leaving an awkward
newline and 35 spaces in the middle of the message.
But while we're here, the existing message seems kind of long in comparison to
similar notes: to cut it down, we excise the mentions of doc comments, which
seems sensible because we know that this erroneous attribute is not a doc
comment (notice the `is_sugared_doc: false` at the end of the function; if it
had been a doc comment, that error would get set in the `token::DocComment`
match branch of `parse_outer_attributes`).
Fix duplicate display of error E0502
Ref. Repeated "mutable/immutable borrow" error messages #42106.
This PR modifies the return type of [`report_error_if_loan_conflicts_with_restriction`](0f0f5db465/src/librustc_borrowck/borrowck/check_loans.rs (L398-L403)) so the result can be checked in [`report_error_if_loans_conflict`](0f0f5db465/src/librustc_borrowck/borrowck/check_loans.rs (L377-L396)). This is done to prevent displaying a duplicate of the error message E0502 which is referenced in #42106.
The output of compiling:
```rust
fn do_something<T>(collection: &mut Vec<T>) {
let _a = &collection;
collection.swap(1, 2);
}
fn main() {}
```
is now
```bash
$ rustc src/test/compile-fail/issue-42106.rs
error[E0502]: cannot borrow `*collection` as mutable because `collection` is also borrowed as immutable
--> src/test/compile-fail/issue-42106.rs:13:5
|
12 | let _a = &collection;
| ---------- immutable borrow occurs here
13 | collection.swap(1, 2);
| ^^^^^^^^^^ mutable borrow occurs here
14 | }
| - immutable borrow ends here
error: aborting due to 2 previous errors
```
r? @estebank
Improve diagnostics when list of tokens has incorrect separators
Make `parse_seq_to_before_tokens` more resilient to error conditions. Where possible it is better if it can consume up to the final bracket before returning. This change improves the diagnostics in a couple of situations:
```
struct S(pub () ()); // omitted separator
use std::{foo. bar}; // used a similar but wrong separator
```
Fixes#44339
r? @petrochenkov
Don't emit the same compiler diagnostic twice.
This PR makes the compiler filter out diagnostic messages that have already been emitted during the same compilation session.
don't issue "expected statement after outer attr." after inner attr.
While an inner attribute here is in fact erroneous, that error ("inner
attribute is not permitted in this context") successfully gets set earlier;
this further admonition is nonsensical.
Resolves#45296.
rustc: Remove `used_mut_nodes` from `TyCtxt`
This updates the borrowck query to return a result, and this result is then used
to incrementally check for unused mutable nodes given sets of all the used
mutable nodes.
Closes#42384
