ToString is implemented by many different types than &str, and
represents a serialization into string data. The fact that said data is
returned as owned, is an implementation detail.
If merely copying borrowed string data to owned string data is all that
is desired, ToOwned is a much better choice, because if the user later
refactors the code such that the input is no longer an `&str`, then they
will get a compiler error instead of a mysterious change-in-behavior.
Automatically instaciate trivially instaciable structs in "Generate new" and "Fill struct fields"
As proposed in #12535 this PR changes the "Generate new" and "Fill struct fields" assist/diagnostic to instanciate structs with no fields and enums with a single empty variant.
For example:
```rust
pub enum Bar {
Bar {},
}
struct Foo<T> {
a: usize,
bar: Bar,
_phantom: std::marker::PhantomData<T>,
}
impl<T> Foo<T> {
/* generate new */
fn random() -> Self {
Self { /* Fill struct fields */ }
}
}
```
was previously:
```rust
impl<T> Foo<T> {
fn new(a: usize, bar: Bar, _phantom: std::marker::PhantomData<T>) -> Self {
Self { a, bar, _phantom }
}
fn random() -> Self {
Self {
a: todo!(),
bar: todo!(),
_phantom: todo!(),
}
}
}
```
and is now:
```rust
impl<T> Foo<T> {
fn new(a: usize) -> Self {
Self {
a,
bar: Bar::Bar {},
_phantom: std::marker::PhantomData
}
}
fn random() -> Self {
Self {
a: todo!(),
bar: Bar::Bar {},
_phantom: std::marker::PhantomData,
}
}
}
```
I'd be happy about any suggestions.
## TODO
- [x] deduplicate `use_trivial_constructor` (unclear how to do as it's used in two separate crates)
- [x] write tests
Closes#12535
feat: Show witnesses of non-exhaustiveness in `missing-match-arm` diagnostic
Shamelessly copied from rustc. Thus reporting format is same.
This extends public api `hir::diagnostics::MissingMatchArms` with `uncovered_patterns: String` field. It does not expose data for implementing a quick fix yet.
-----
Worth to note: current implementation does not give a comprehensive list of missing patterns. Also mentioned in [paper](http://moscova.inria.fr/~maranget/papers/warn/warn.pdf):
> One may think that algorithm I should make an additional effort to provide more
> non-matching values, by systematically computing recursive calls on specialized
> matrices when possible, and by returning a list of all pattern vectors returned by
> recursive calls. We can first observe that it is not possible in general to supply the
> users with all non-matching values, since the signature of integers is (potentially)
> infinite.
fix: trailing ':' on empty inactive reasons
## Description
Fixes trailing ':' even when there is no explanation. e.g.
``` sh
code is inactive due to #[cfg] directives:
```
## Issue
Fixes: #12615
Distinguish between
- there is no build data (for some reason?)
- there is build data, but the cargo package didn't build a proc macro dylib
- there is a proc macro dylib, but it didn't contain the proc macro we expected
- the name did not resolve to any macro (this is now an
unresolved_macro_call even for attributes)
I changed the handling of disabled attribute macro expansion to
immediately ignore the macro and report an unresolved_proc_macro,
because otherwise they would now result in loud unresolved_macro_call
errors. I hope this doesn't break anything.
Also try to improve error ranges for unresolved_macro_call / macro_error
by reusing the code for unresolved_proc_macro. It's not perfect but
probably better than before.
