diff options
| author | bors <bors@rust-lang.org> | 2022-12-03 18:42:33 +0000 |
|---|---|---|
| committer | bors <bors@rust-lang.org> | 2022-12-03 18:42:33 +0000 |
| commit | 957b4bb2164e57a58d7e6cb5b4915b298dab9391 (patch) | |
| tree | e14d3c1d2454393507c6e821a16906455e9ac6b6 /src/test/codegen/src-hash-algorithm/src-hash-algorithm-md5.rs | |
| parent | 04a2ac2de20565e87c561bdb3c3814dfeec1513c (diff) | |
| parent | de591f08d6a38f1b70365622f62352583c9c06ef (diff) | |
| download | rust-957b4bb2164e57a58d7e6cb5b4915b298dab9391.tar.gz rust-957b4bb2164e57a58d7e6cb5b4915b298dab9391.zip | |
Auto merge of #13717 - lowr:fix/proc-macro-ident-is-raw, r=Veykril
Handle raw identifiers in proc macro server Fixes #13706 When proc macros create `proc_macro::Ident`s, they pass an identifier text without "r#" prefix and a flag `is_raw` to proc macro server. Our `tt::Ident` currently stores the text *with* "r#" so we need to adjust them somewhere. Rather than following rustc and adding `is_raw` field to our `tt::Ident`, I opted for adjusting the representation of identifiers in proc macro server, because we don't need the field outside it. It's hard to write regression test for this, but at least I: - ran `cargo +nightly t --features sysroot-abi` and all the tests passed - built proc macro server with `cargo +nightly b -r --bin rust-analyzer-proc-macro-srv --features sysroot-abi` and made sure #13706 resolved - For the record, the nightly versions used are `rustc 1.67.0-nightly (32e613bba 2022-12-02)` and `cargo 1.67.0-nightly (e027c4b5d 2022-11-25)`.
Diffstat (limited to 'src/test/codegen/src-hash-algorithm/src-hash-algorithm-md5.rs')
0 files changed, 0 insertions, 0 deletions