Commit Graph

80 Commits

Author SHA1 Message Date
Nicholas Nethercote
cee88f7a3f Prepare for invisible delimiters.
Current places where `Interpolated` is used are going to change to
instead use invisible delimiters. This prepares for that.
- It adds invisible delimiter cases to the `can_begin_*`/`may_be_*`
  methods and the `failed_to_match_macro` that are equivalent to the
  existing `Interpolated` cases.
- It adds panics/asserts in some places where invisible delimiters
  should never occur.
- In `Parser::parse_struct_fields` it excludes an ident + invisible
  delimiter from special consideration in an error message, because
  that's quite different to an ident + paren/brace/bracket.
2024-11-21 08:22:11 +11:00
Michael Goulet
c682aa162b Reformat using the new identifier sorting from rustfmt 2024-09-22 19:11:29 -04:00
Michael Goulet
97910580aa Add initial support for raw lifetimes 2024-09-06 10:32:48 -04:00
Michael Goulet
c61f85b6dd Don't make pattern nonterminals match statement nonterminals 2024-08-26 18:30:15 -04:00
Matthias Krüger
3acd910036
Rollup merge of #126697 - vincenzopalazzo:macros/find_the_expression_tok, r=eholk,compiler-errors
[RFC] mbe: consider the `_` in 2024 an expression

This commit is adding the possibility to parse the `_` as an expression inside the esition 2024.

Link: https://rust-lang.zulipchat.com/#narrow/stream/404510-wg-macros/topic/supporting.20.60_.60.20expressions

Issue https://github.com/rust-lang/rust/issues/123742

r? `@eholk`
2024-07-31 23:20:09 +02:00
Michael Goulet
79ef91e879
tweak comment on NonterminalKind::Expr
Co-authored-by: Eric Holk <eric@theincredibleholk.org>
2024-07-31 15:37:55 -04:00
Vincenzo Palazzo
276fa19c0a rustc_parser: consider the in 2024 an expression
This commit is adding the possibility to parse the `_` as
an expression inside the esition 2024.

Link: https://rust-lang.zulipchat.com/#narrow/stream/404510-wg-macros/topic/supporting.20.60_.60.20expressions
Co-authored-by: Eric Holk <eric@theincredibleholk.org>
Signed-off-by: Vincenzo Palazzo <vincenzopalazzodev@gmail.com>
2024-07-31 18:34:22 +00:00
Nicholas Nethercote
84ac80f192 Reformat use declarations.
The previous commit updated `rustfmt.toml` appropriately. This commit is
the outcome of running `x fmt --all` with the new formatting options.
2024-07-29 08:26:52 +10:00
Nicholas Nethercote
9d908a2877 Use ForceCollect in parse_attr_item.
Instead of a `bool`. Because `ForceCollect` is used in this way
everywhere else.
2024-07-19 08:24:54 +10:00
Michael Goulet
d0a1851ec2 Deny keyword lifetimes pre-expansion 2024-07-16 12:06:25 -04:00
Nicholas Nethercote
e2aa38e6ab Rework pattern and expression nonterminal kinds.
Merge `PatParam`/`PatWithOr`, and `Expr`/`Expr2021`, for a few reasons.

- It's conceptually nice, because the two pattern kinds and the two
  expression kinds are very similar.

- With expressions in particular, there are several places where both
  expression kinds get the same treatment.

- It removes one unreachable match arm.

- Most importantly, for #124141 I will need to introduce a new type
  `MetaVarKind` that is very similar to `NonterminalKind`, but records a
  couple of extra fields for expression metavars. It's nicer to have a
  single `MetaVarKind::Expr` expression variant to hold those extra
  fields instead of duplicating them across two variants
  `MetaVarKind::{Expr,Expr2021}`. And then it makes sense for patterns
  to be treated the same way, and for `NonterminalKind` to also be
  treated the same way.

I also clarified the comments, because I have long found them a little
hard to understand.
2024-06-23 15:57:24 +10:00
Michael Goulet
3e8898a4e1 Allow naming expr_2021 in all editions 2024-06-19 12:37:49 -04:00
Vincenzo Palazzo
36d5fc9a64
Avoid checking the edition as much as possible
Inside #123865, we are adding support for the new semantics
for expr2024, but we have noted a performance issue.

We realized there is a redundant check for each
token regarding an edition. This commit moves the edition
check to the end, avoiding some extra checks that
can slow down compilation time.

Link: https://github.com/rust-lang/rust/pull/123865
Co-Developed-by: @eholk
Signed-off-by: Vincenzo Palazzo <vincenzopalazzodev@gmail.com>
2024-06-02 09:42:21 +02:00
bors
9b75a43881 Auto merge of #123865 - eholk:expr_2021, r=fmease
Update `expr` matcher for Edition 2024 and add `expr_2021` nonterminal

This commit adds a new nonterminal `expr_2021` in macro patterns, and `expr_fragment_specifier_2024` feature flag.

This change also updates `expr` so that on Edition 2024 it will also match `const { ... }` blocks, while `expr_2021` preserves the current behavior of `expr`, matching expressions without `const` blocks.

Joint work with `@vincenzopalazzo.`

Issue #123742
2024-05-17 21:54:14 +00:00
Eric Holk
3986ea0ea5
Update compiler/rustc_parse/src/parser/nonterminal.rs
Co-authored-by: León Orell Valerian Liehr <me@fmease.dev>
2024-05-17 11:55:38 -07:00
Nicholas Nethercote
95e519ecbf Remove NtIdent and NtLifetime.
The extra span is now recorded in the new `TokenKind::NtIdent` and
`TokenKind::NtLifetime`. These both consist of a single token, and so
there's no operator precedence problems with inserting them directly
into the token stream.

The other way to do this would be to wrap the ident/lifetime in invisible
delimiters, but there's a lot of code that assumes an interpolated
ident/lifetime fits in a single token, and changing all that code to work with
invisible delimiters would have been a pain. (Maybe it could be done in a
follow-up.)

This change might not seem like much of a win, but it's a first step toward the
much bigger and long-desired removal of `Nonterminal` and
`TokenKind::Interpolated`. That change is big and complex enough that it's
worth doing this piece separately. (Indeed, this commit is based on part of a
late commit in #114647, a prior attempt at that big and complex change.)
2024-05-14 08:19:58 +10:00
Vincenzo Palazzo
a55d06323a
Macros: match const { ... } with expr nonterminal in edition 2024
Co-authored-by: Eric Holk <eric@theincredibleholk.org>
Signed-off-by: Vincenzo Palazzo <vincenzopalazzodev@gmail.com>
2024-05-13 11:55:26 -07:00
Eric Holk
ef6478ba5f
Add expr_2021 nonterminal and feature flag
This commit adds a new nonterminal `expr_2021` in macro patterns, and
`expr_fragment_specifier_2024` feature flag. For now, `expr` and
`expr_2021` are treated the same, but in future PRs we will update
`expr` to match to new grammar.

Co-authored-by: Vincezo Palazzo <vincenzopalazzodev@gmail.com>
2024-05-13 11:27:26 -07:00
Nicholas Nethercote
9a63a42cb7 Remove a Span from TokenKind::Interpolated.
This span records the declaration of the metavariable in the LHS of the macro.
It's used in a couple of error messages. Unfortunately, it gets in the way of
the long-term goal of removing `TokenKind::Interpolated`. So this commit
removes it, which degrades a couple of (obscure) error messages but makes
things simpler and enables the next commit.
2024-05-13 10:30:30 +10:00
León Orell Valerian Liehr
3cbc9e9560
Rename ModSep to PathSep 2024-04-04 19:44:04 +02:00
Nicholas Nethercote
a94bb2a013 Streamline NamedMatch.
This commit combines `MatchedTokenTree` and `MatchedNonterminal`, which
are often considered together, into a single `MatchedSingle`. It shares
a representation with the newly-parameterized `ParseNtResult`.

This will also make things much simpler if/when variants from
`Interpolated` start being moved to `ParseNtResult`.
2024-03-21 10:18:34 +11:00
Nicholas Nethercote
c14d9ae23e Fix some formatting. 2024-03-21 10:18:34 +11:00
clubby789
06d6c62f80 Add newtype for raw idents 2024-02-20 13:13:29 +00:00
Nicholas Nethercote
cea683c08f Use .into_diagnostic() less.
This commit replaces this pattern:
```
err.into_diagnostic(dcx)
```
with this pattern:
```
dcx.create_err(err)
```
in a lot of places.

It's a little shorter, makes the error level explicit, avoids some
`IntoDiagnostic` imports, and is a necessary prerequisite for the next
commit which will add a `level` arg to `into_diagnostic`.

This requires adding `track_caller` on `create_err` to avoid mucking up
the output of `tests/ui/track-diagnostics/track4.rs`. It probably should
have been there already.
2023-12-18 20:46:13 +11:00
Nicholas Nethercote
73bac456d4 Rename Parser::span_diagnostic as Parser::dcx. 2023-12-18 16:06:21 +11:00
Nicholas Nethercote
31ac4efb31 Use Session::diagnostic in more places. 2023-12-02 09:01:35 +11:00
Esteban Küber
4e418805da More detail when expecting expression but encountering bad macro argument
Partially address #71039.
2023-11-16 16:19:04 +00:00
Michael Goulet
b2d2184ede Format all the let chains in compiler 2023-10-13 08:59:36 +00:00
Nicholas Nethercote
9e22351c74 Rename NtOrTt as ParseNtResult.
It's more descriptive, and future-proofs it if/when additional variants
get added.
2023-08-18 16:50:41 +10:00
Nicholas Nethercote
e46caaf84b Simplify a match.
`may_be_ident` is true for `NtPath` and `NtMeta`, so we don't need to
check for them separately.
2023-08-17 09:06:03 +10:00
Nicholas Nethercote
9a3c907bdb Make some matches exhaustive in nonterminal.rs.
For ones matching more than one or two variants, this is easier to think
about.
2023-08-17 09:05:37 +10:00
Nicholas Nethercote
f8a21a5df0 Use Nonterminal::* in nonterminal.rs.
It makes the code more readable.
2023-08-17 09:05:07 +10:00
Nicholas Nethercote
acd3a5e35f Remove unnecessary braces on PatWithOr patterns. 2023-08-17 09:04:54 +10:00
Nicholas Nethercote
04cf6b4ac5 Rename parse_no_question_mark_recover.
Adding a `ty_` makes its purpose much clearer, and consistent with other
`parse_ty_*` functions.
2023-08-17 09:01:12 +10:00
Nicholas Nethercote
434bfc3162 Remove outdated comment.
All nonterminals collect and store tokens now. (Unless they are very
simple, e.g. single-token, and can precisely recover them without
collecting.)
2023-08-17 08:26:56 +10:00
r0cky
8c8af6cf99 Avoid too many expected symbols and reduce Nones 2023-08-03 08:56:31 +00:00
Mu001999
89b2fe7750 Keep the suggestion for wrong arbitrary self types 2023-08-03 00:00:56 +08:00
Matthias Krüger
bcc15bba95 use matches! macro in more places 2023-04-16 12:08:30 +02:00
Xiretza
87ef37dbd7 rustc_parse: migrate more to diagnostic structs 2023-02-01 21:50:34 +01:00
Maybe Waffle
616df0f03b rustc_parse: remove ref patterns 2022-11-22 18:49:16 +00:00
Rageking8
6d7beafc87 slight vertical formatting 2022-09-15 23:51:43 +08:00
Vadim Petrochenkov
f2b7fa4847 ast: Introduce some traits to get AST node properties generically
And use them to avoid constructing some artificial `Nonterminal` tokens during expansion
2022-05-11 12:43:27 +03:00
Vadim Petrochenkov
2733ec1be3 rustc_ast: Harmonize delimiter naming with proc_macro::Delimiter 2022-04-28 10:04:29 +03:00
Nicholas Nethercote
d2b9bbbf78 Inline Parser::nonterminal_may_begin_with. 2022-04-20 14:13:49 +10:00
Nicholas Nethercote
f9235db37e Inline Parser::parse_nonterminal. 2022-04-20 14:08:59 +10:00
Nicholas Nethercote
d9592c2d9f Shrink Nonterminal.
By heap allocating the argument within `NtPath`, `NtVis`, and `NtStmt`.
This slightly reduces cumulative and peak allocation amounts, most
notably on `deep-vector`.
2022-04-07 12:51:50 +10:00
Nicholas Nethercote
364b908d57 Remove Nonterminal::NtTT.
It's only needed for macro expansion, not as a general element in the
AST. This commit removes it, adds `NtOrTt` for the parser and macro
expansion cases, and renames the variants in `NamedMatch` to better
match the new type.
2022-03-28 10:03:02 +11:00
Esteban Kuber
004f2ed219 Do not recover from Ty? in macro parsing
Follow up to #92746. Address #94510.
2022-03-04 02:03:55 +00:00
Esteban Kuber
f42b4f595e Tweak diagnostics
* Recover from invalid `'label: ` before block.
* Make suggestion to enclose statements in a block multipart.
* Point at `match`, `while`, `loop` and `unsafe` keywords when failing
  to parse their expression.
* Do not suggest `{ ; }`.
* Do not suggest `|` when very unlikely to be what was wanted (in `let`
  statements).
2022-02-28 18:22:45 +00:00
Léo Lanteri Thauvin
fde1b76b4b Use if-let guards in the codebase 2021-08-25 20:24:35 +02:00