improve error message when `global_asm!` uses `asm!` options
specifically, what was
error: expected one of `)`, `att_syntax`, or `raw`, found `preserves_flags`
--> $DIR/bad-options.rs:45:25
|
LL | global_asm!("", options(preserves_flags));
| ^^^^^^^^^^^^^^^ expected one of `)`, `att_syntax`, or `raw`
is now
error: the `preserves_flags` option cannot be used with `global_asm!`
--> $DIR/bad-options.rs:45:25
|
LL | global_asm!("", options(preserves_flags));
| ^^^^^^^^^^^^^^^ the `preserves_flags` option is not meaningful for global-scoped inline assembly
mirroring the phrasing of the [reference](https://doc.rust-lang.org/reference/inline-assembly.html#options).
This is also a bit of a refactor for a future `naked_asm!` macro (for use in `#[naked]` functions). Currently this sort of error can come up when switching from inline to global asm, or when a user just isn't that experienced with assembly. With `naked_asm!` added to the mix hitting this error is more likely.
Improve `extern "<abi>" unsafe fn()` error message
These errors were already reported in #87217, and fixed by #87235 but missed the case of an explicit ABI.
This PR does not cover multiple keywords like `extern "C" pub const unsafe fn()`, but I don't know what a good way to cover this would be. It also seems rarer than `extern "C" unsafe` which I saw happen a few times in workshops.
Remove unnecessary range replacements
This PR removes an unnecessary range replacement in `collect_tokens_trailing_token`, and does a couple of other small cleanups.
r? ````@petrochenkov````
A fully imperative style is easier to read than a half-iterator,
half-imperative style. Also, rename `inner_attr` as `attr` because it
might be an outer attribute.
Imagine you have replace ranges (2..20,X) and (5..15,Y), and these tokens:
```
a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p,q,r,s,t,u,v,w,x
```
If we replace (5..15,Y) first, then (2..20,X) we get this sequence
```
a,b,c,d,e,Y,_,_,_,_,_,_,_,_,_,p,q,r,s,t,u,v,w,x
a,b,X,_,_,_,_,_,_,_,_,_,_,_,_,_,_,_,_,_,u,v,w,x
```
which is what we want.
If we do it in the other order, we get this:
```
a,b,X,_,_,_,_,_,_,_,_,_,_,_,_,p,q,r,s,t,u,v,w,x
a,b,X,_,_,Y,_,_,_,_,_,_,_,_,_,_,_,_,_,_,u,v,w,x
```
which is wrong. So it's true that we need the `.rev()` but the comment
is wrong about why.
The current code is this:
```
self.capture_state.replace_ranges.push((start_pos..end_pos, Some(target)));
self.capture_state.replace_ranges.extend(inner_attr_replace_ranges);
```
What's not obvious is that every range in `inner_attr_replace_ranges`
must be a strict sub-range of `start_pos..end_pos`. Which means, in
`LazyAttrTokenStreamImpl::to_attr_token_stream`, they will be done
first, and then the `start_pos..end_pos` replacement will just overwrite
them. So they aren't needed.
This has been bugging me for a while. I find complex "if any of these
are true" conditions easier to think about than complex "if all of these
are true" conditions, because you can stop as soon as one is true.
Reorder trait bound modifiers *after* `for<...>` binder in trait bounds
This PR suggests changing the grammar of trait bounds from:
```
[CONSTNESS] [ASYNCNESS] [?] [BINDER] [TRAIT_PATH]
const async ? for<'a> Sized
```
to
```
([BINDER] [CONSTNESS] [ASYNCNESS] | [?]) [TRAIT_PATH]
```
i.e., either
```
? Sized
```
or
```
for<'a> const async Sized
```
(but not both)
### Why?
I think it's strange that the binder applies "more tightly" than the `?` trait polarity. This becomes even weirder when considering that we (or at least, I) want to have `async` trait bounds expressed like:
```
where T: for<'a> async Fn(&'a ()) -> i32,
```
and not:
```
where T: async for<'a> Fn(&'a ()) -> i32,
```
### Fallout
No crates on crater use this syntax, presumably because it's literally useless. This will require modifying the reference grammar, though.
### Alternatives
If this is not desirable, then we can alternatively keep parsing `for<'a>` after the `?` but deprecate it with either an FCW (or an immediate hard error), and begin parsing `for<'a>` *before* the `?`.
Adding details, clarifying lots of little things, etc. In particular,
the commit adds details of an example. I find this very helpful, because
it's taken me a long time to understand how this code works.
Currently in `collect_tokens_trailing_token`, `start_pos` and `end_pos`
are 1-indexed by `replace_ranges` is 0-indexed, which is really
confusing. Making them both 0-indexed makes debugging much easier.
The `Option`s within the `ReplaceRange`s within the hashmap are always
`None`. This PR omits them and inserts them when they are extracted from
the hashmap.
There are three places where we currently check `force_collect` and call
`collect_tokens_no_attrs` for `ForceCollect::Yes` and a vanilla parsing
function for `ForceCollect::No`.
But we can instead just pass in `force_collect` and let
`collect_tokens_trailing_token` do the appropriate thing.
Fix ICE in suggestion caused by `⩵` being recovered as `==`
The second suggestion shown here would previously incorrectly assume that the span corresponding to `⩵` was 2 bytes wide composed by 2 1 byte wide chars, so a span pointing at `==` could point only at one of the `=` to remove it. Instead, we now replace the whole thing (as we should have the whole time):
```
error: unknown start of token: \u{2a75}
--> $DIR/unicode-double-equals-recovery.rs:1:16
|
LL | const A: usize ⩵ 2;
| ^
|
help: Unicode character '⩵' (Two Consecutive Equals Signs) looks like '==' (Double Equals Sign), but it is not
|
LL | const A: usize == 2;
| ~~
error: unexpected `==`
--> $DIR/unicode-double-equals-recovery.rs:1:16
|
LL | const A: usize ⩵ 2;
| ^
|
help: try using `=` instead
|
LL | const A: usize = 2;
| ~
```
Fix#127823.
The second suggestion shown here would previously incorrectly assume that the span corresponding to `⩵` was 2 bytes wide composed by 2 1 byte wide chars, so a span pointing at `==` could point only at one of the `=` to remove it. Instead, we now replace the whole thing (as we should have the whole time):
```
error: unknown start of token: \u{2a75}
--> $DIR/unicode-double-equals-recovery.rs:1:16
|
LL | const A: usize ⩵ 2;
| ^
|
help: Unicode character '⩵' (Two Consecutive Equals Signs) looks like '==' (Double Equals Sign), but it is not
|
LL | const A: usize == 2;
| ~~
error: unexpected `==`
--> $DIR/unicode-double-equals-recovery.rs:1:16
|
LL | const A: usize ⩵ 2;
| ^
|
help: try using `=` instead
|
LL | const A: usize = 2;
| ~
```
Remove `TrailingToken`.
It's used in `Parser::collect_tokens_trailing_token` to decide whether to capture a trailing token. But the callers actually know whether to capture a trailing token, so it's simpler for them to just pass in a bool.
Also, the `TrailingToken::Gt` case was weird, because it didn't result in a trailing token being captured. It could have been subsumed by the `TrailingToken::MaybeComma` case, and it effectively is in the new code.
r? `@petrochenkov`
It's used in `Parser::collect_tokens_trailing_token` to decide whether
to capture a trailing token. But the callers actually know whether to
capture a trailing token, so it's simpler for them to just pass in a
bool.
Also, the `TrailingToken::Gt` case was weird, because it didn't result
in a trailing token being captured. It could have been subsumed by the
`TrailingToken::MaybeComma` case, and it effectively is in the new code.
More accurate span for anonymous argument suggestion
Use smaller span for suggesting adding `_:` ahead of a type:
```
error: expected one of `(`, `...`, `..=`, `..`, `::`, `:`, `{`, or `|`, found `)`
--> $DIR/anon-params-denied-2018.rs:12:47
|
LL | fn foo_with_qualified_path(<Bar as T>::Baz);
| ^ expected one of 8 possible tokens
|
= note: anonymous parameters are removed in the 2018 edition (see RFC 1685)
help: explicitly ignore the parameter name
|
LL | fn foo_with_qualified_path(_: <Bar as T>::Baz);
| ++
```
Some parser improvements
I was looking closely at attribute handling in the parser while debugging some issues relating to #124141, and found a few small improvements.
``@spastorino``
Use smaller span for suggesting adding `_:` ahead of a type:
```
error: expected one of `(`, `...`, `..=`, `..`, `::`, `:`, `{`, or `|`, found `)`
--> $DIR/anon-params-denied-2018.rs:12:47
|
LL | fn foo_with_qualified_path(<Bar as T>::Baz);
| ^ expected one of 8 possible tokens
|
= note: anonymous parameters are removed in the 2018 edition (see RFC 1685)
help: explicitly ignore the parameter name
|
LL | fn foo_with_qualified_path(_: <Bar as T>::Baz);
| ++
```
It only has two call sites, and it extremely similar to
`Parser::parse_expr_dot_or_call_with`, in both name and behaviour. The
only difference is the latter has an `attrs` argument and an
`ensure_sufficient_stack` call. We can pass in an empty `attrs` as
necessary, as is already done at some `parse_expr_dot_or_call_with` call
sites.
Make parse error suggestions verbose and fix spans
Go over all structured parser suggestions and make them verbose style.
When suggesting to add or remove delimiters, turn them into multiple suggestion parts.
Fix `DebugParser`.
I tried using this and it didn't work at all. `prev_token` is never eof, so the accumulator is always false, which means the `then_some` always returns `None`, which means `scan` always returns `None`, and `tokens` always ends up an empty vec. I'm not sure how this code was supposed to work.
(An aside: I find `Iterator::scan` to be a pretty wretched function, that produces code which is very hard to understand. Probably why this is just one of two uses of it in the entire compiler.)
This commit changes it to a simpler imperative style that produces a valid `tokens` vec.
r? `@workingjubilee`
