diff options
| author | bors[bot] <26634292+bors[bot]@users.noreply.github.com> | 2019-11-26 11:05:05 +0000 |
|---|---|---|
| committer | GitHub <[email protected]> | 2019-11-26 11:05:05 +0000 |
| commit | 500e022f7decbee29a693b0f0dd2f63789a99e5a (patch) | |
| tree | 2f7839288ce5676a89c6d6062cbaf70544e0beed /crates/ra_ide_api/src/completion/completion_item.rs | |
| parent | 5901cc736074bbc4d780a8e45079d405ab2cec4b (diff) | |
| parent | e5eadb339039e21718d382c0b3d02a4bf053b3f4 (diff) | |
Merge #2398
2398: WIP: introduce hir::Type r=matklad a=matklad
This introduces `hir::Type` wrapper over `hir::Ty`, with two purposes:
* bind `Ty` and it's corresponding environment
* Am I correct that `Ty` without an env doesn't make much sense, because the meaning of type parameters is unclear
* Am I correct that we can safely re-use the same environment for all types derived from the given type?
* hide representation defails of `Ty`. Specifically, I want to change `Ty::Adt` to use `hir_def::AdtId` instead of `hir::Adt`, but IDE doesn't know about underlying IDs. More generally, I feel like IDE shouldn't know that `Ty` is enum.
@flodiebold what do you think about this?
Co-authored-by: Aleksey Kladov <[email protected]>
Diffstat (limited to 'crates/ra_ide_api/src/completion/completion_item.rs')
0 files changed, 0 insertions, 0 deletions