Commit Graph

117 Commits

Author SHA1 Message Date
Aleksey Kladov
dbb702cfc1 internal: remove accidental code re-use
FragmentKind played two roles:

* entry point to the parser
* syntactic category of a macro call

These are different use-cases, and warrant different types. For example,
macro can't expand to visibility, but we have such fragment today.

This PR introduces `ExpandsTo` enum to separate this two use-cases.

I suspect we might further split `FragmentKind` into `$x:specifier` enum
specific to MBE, and a general parser entry point, but that's for
another PR!
2021-09-05 22:36:36 +03:00
Lukas Wirth
36a5ce9790 minor: fix some clippy lints 2021-09-03 16:00:50 +02:00
Aleksey Kladov
81602f8a5d internal: reduce coupling
tt is a data structure, data structures cant' go wrong, they shouldn't
have the knowledge that the world outside of them has all kinds of
errors.
2021-08-31 19:14:33 +03:00
Dezhi Wu
ba0947dded switch log crate to tracing 2021-08-30 15:11:42 +08:00
Lukas Wirth
c5059e0623 Return all ranges corresponding to a token id in TokenMap 2021-08-29 00:49:57 +02:00
bors[bot]
fae440c32a
Merge #10025
10025: Don't mutate syntax trees when preparing proc-macro input r=Veykril a=Veykril

Fixes #10013

Co-authored-by: Lukas Wirth <lukastw97@gmail.com>
2021-08-28 14:14:54 +00:00
bors[bot]
97409e5fc8
Merge #9970
9970: feat: Implement attribute input token mapping, fix attribute item token mapping r=Veykril a=Veykril

![image](https://user-images.githubusercontent.com/3757771/130328577-4c1ad72c-51b1-47c3-8d3d-3242ec44a355.png)

The token mapping for items with attributes got overwritten partially by the attributes non-item input, since attributes have two different inputs, the item and the direct input both.
This PR gives attributes a second TokenMap for its direct input. We now shift all normal input IDs by the item input maximum(we maybe wanna swap this see below) similar to what we do for macro-rules/def. For mapping down we then have to figure out whether we are inside the direct attribute input or its item input to pick the appropriate mapping which can be done with some token range comparisons.

Fixes https://github.com/rust-analyzer/rust-analyzer/issues/9867

Co-authored-by: Lukas Wirth <lukastw97@gmail.com>
2021-08-27 19:30:36 +00:00
Lukas Wirth
1195cb50c2 Add simple test for syntax_node_to_token_tree_censored 2021-08-25 19:57:18 +02:00
Lukas Wirth
d6134b6802 Don't mutate syntax trees when preparing proc-macro input 2021-08-25 18:57:24 +02:00
Frank Steffahn
5f5d45468d Fix two more “a”/“an” typos (this time the other way) 2021-08-22 17:36:58 +02:00
Lukas Wirth
5fb8c0ddfd Remove MappedSubtree 2021-08-21 18:19:18 +02:00
Lukas Wirth
177c70128c Map attribute input tokens correctly 2021-08-21 18:13:41 +02:00
Aleksey Kladov
78c7940f5c internal: remove dead code 2021-08-14 20:29:46 +03:00
Aleksey Kladov
9aa6be71a5 internal: remove useless helpers
We generally avoid "syntax only" helper wrappers, which don't do much:
they make code easier to write, but harder to read. They also make
investigations harder, as "find_usages" needs to be invoked both for the
wrapped and unwrapped APIs
2021-08-09 15:58:21 +03:00
Aleksey Kladov
86720f2953 minor: drop dummy authors field 2021-07-05 14:19:41 +03:00
Jonas Schievink
c6669776e1 Rewrite convert_tokens to use an explicit stack 2021-06-23 00:21:11 +02:00
Jonas Schievink
6504c3c32a Move subtree collection out of TokenConvertor 2021-06-23 00:19:54 +02:00
Clemens Wasser
47747cd412 Apply some clippy suggestions 2021-06-21 16:40:21 +02:00
Lukas Wirth
95c8c65139 Nest all the or-patterns! 2021-06-17 17:37:14 +02:00
bors[bot]
5a8ddb4b2d
Merge #9260
9260: tree-wide: make rustdoc links spiky so they are clickable r=matklad a=lf-

Rustdoc was complaining about these while I was running with --document-private-items and I figure they should be fixed.

Co-authored-by: Jade <software@lfcode.ca>
2021-06-14 07:16:48 +00:00
Jade
20b325c7d5 tree-wide: make rustdoc links spiky so they are clickable 2021-06-13 21:58:05 -07:00
Maan2003
aabd41cafc
clippy::redundant_field_names 2021-06-13 09:40:22 +05:30
Maan2003
c9b4ac5be4
clippy::redudant_borrow 2021-06-13 09:24:16 +05:30
Aleksey Kladov
0463d76a1f internal: cross-crate cov-marks 2021-06-12 23:40:52 +03:00
Jonas Schievink
1d6eef1350 Update ungrammar 2021-06-11 18:34:30 +02:00
Clemens Wasser
629e8d1ed0 Apply more clippy suggestions and update generated 2021-06-03 12:46:56 +02:00
Jonas Schievink
4c4bcaebbe Treat pat_param like pat fragments 2021-05-29 18:32:32 +02:00
Jonas Schievink
c8f40b1503 Fixup 2021-05-24 21:47:01 +02:00
Jonas Schievink
489ae7a800 Make TokenTextRange private 2021-05-24 20:29:48 +02:00
Jonas Schievink
27bf62b70e Move TokenMap to its own file 2021-05-24 18:43:42 +02:00
Aleksey Kladov
dc1577d58d Add even more docs 2021-05-22 17:20:22 +03:00
bors[bot]
2ace128dd4
Merge #8560
8560: Escape characters in doc comments in macros correctly r=jonas-schievink a=ChayimFriedman2

Previously they were escaped twice, both by `.escape_default()` and the debug view of strings (`{:?}`). This leads to things like newlines or tabs in documentation comments being `\\n`, but we unescape literals only once, ending up with `\n`.

This was hard to spot because CMark unescaped them (at least for `'` and `"`), but it did not do so in code blocks.

This also was the root cause of #7781. This issue was solved by using `.escape_debug()` instead of `.escape_default()`, but the real issue remained.
We can bring the `.escape_default()` back by now, however I didn't do it because it is probably slower than `.escape_debug()` (more work to do), and also in order to change the code the least.

Example (the keyword and primitive docs are `include!()`d at https://doc.rust-lang.org/src/std/lib.rs.html#570-578, and thus originate from macro):

Before:
![image](https://user-images.githubusercontent.com/24700207/115130096-40544300-9ff5-11eb-847b-969e7034e8a4.png)

After:
![image](https://user-images.githubusercontent.com/24700207/115130143-9cb76280-9ff5-11eb-9281-323746089440.png)


Co-authored-by: Chayim Refael Friedman <chayimfr@gmail.com>
2021-04-18 02:14:27 +00:00
Chayim Refael Friedman
f92be7eaab Escape characters in doc comments in macros correctly
Previously they were escaped twice, both by `.escape_default()` and the debug view of strings (`{:?}`). This leads to things like newlines or tabs in documentation comments being `\\n`, but we unescape literals only once, ending up with `\n`.

This was hard to spot because CMark unescaped them (at least for `'` and `"`), but it did not do so in code blocks.

This also was the root cause of #7781. This issue was solved by using `.escape_debug()` instead of `.escape_default()`, but the real issue remained.
We can bring the `.escape_default()` back by now, however I didn't do it because it is probably slower than `.escape_debug()` (more work to do), and also in order to change the code the least.
2021-04-18 03:16:38 +03:00
Edwin Cheng
c4173bb468 Handle extended key value attr in mbe 2021-04-17 14:31:52 +08:00
Jonas Schievink
3abcdc03ba Make ast_to_token_tree infallible
It could never return `None`, so reflect that in the return type
2021-04-04 01:46:45 +02:00
Edwin Cheng
20d55ce44d Allow include! an empty content file 2021-04-03 12:50:55 +08:00
Jonas Schievink
eaffdae300 Allow , to delimit macro 2.0 rules 2021-04-03 03:08:31 +02:00
Matthias Krüger
202b51bc7b a lot of clippy::style fixes 2021-03-21 16:15:41 +01:00
Matthias Krüger
ae7e55c1dd clippy::complexity simplifications related to Iterators 2021-03-21 13:13:34 +01:00
Kevin Mehall
0a7f28620a Fix and test edge cases of _ as ident 2021-03-20 17:54:57 -06:00
Kevin Mehall
0a0e22235b Make bare underscore token an Ident rather than Punct in proc-macro 2021-03-20 12:28:44 -06:00
Edwin Cheng
49b876de09 Reorganize mbe tests 2021-03-18 17:22:52 +08:00
Matthias Krüger
048dad8c2e don't clone types that are copy (clippy::clone_on_copy) 2021-03-17 01:56:31 +01:00
Matthias Krüger
c5d654d513 use if let Some(x) instead of if x.is_some() and x.unwrap() (clippy::unnecessary-unwrap) 2021-03-17 01:39:58 +01:00
Matthias Krüger
966c23f529 avoid converting types into themselves via .into() (clippy::useless-conversion)
example: let x: String = String::from("hello world").into();
2021-03-17 01:27:56 +01:00
Edwin Cheng
8e07b23b84 Fix macro expansion for statements w/o semicolon 2021-03-16 13:44:50 +08:00
Laurențiu Nicola
88cee24c6c Enable thread-local coverage marks 2021-03-15 16:02:50 +02:00
Matthias Krüger
cad617bba0 some clippy::performance fixes
use vec![] instead of Vec::new() +  push()
avoid redundant clones
use chars instead of &str for single char patterns in ends_with() and starts_with()
allocate some Vecs with capacity to avoid unneccessary resizing
2021-03-15 10:19:59 +01:00
Edwin Cheng
7279749bbb Simpify mbe bindings builder 2021-03-14 11:54:19 +08:00
Edwin Cheng
b3b91046dd Make sure ill-form macro handle propely 2021-03-14 11:24:55 +08:00