Re-name semantic::Policy to Semantic#944
Conversation
Yeah, for sure. I have such a patch in my massive tree and it's super annoying to deal with. Would appreciate getting it into |
|
Docs and fmt CI failures are real. |
Tighten up the linting by enabling `use_self` and fix all warnings. I got claude to do the fixes. It eventually found its way to using cargo clippy --fix --allow-dirty --allow-staged --workspace \ --all-targets --all-features -- -D clippy::use-self
All these issues were introduced by the `use_self` change. Fixed separately to make review of that change easier since its 12000 lines long.
Rename the `semantic::Policy` to `semantic::Semantic`. Keep the re-export at `Policy/mod.rs` (was an alias). In the `semantic` module unit tests leave `StringPolicy` as it is. Outside the `semantic` module remove the stutter; use `Semantic` instead of `semantic::Semantic`.
tcharding
force-pushed
the
push-rwptozvxwzuv
branch
from
d92a96a to
0c4a888
Compare
|
I threw a patch in the middle to do the mechanical formatting fixes instead of changing the |
|
Where do we "retain the aliases" (from Andrew's comment) for deprecation purposes? I don't see any |
|
|
||
| pub use self::concrete::Policy as Concrete; | ||
| pub use self::semantic::Policy as Semantic; | ||
| pub use self::semantic::Semantic; |
There was a problem hiding this comment.
@trevorj this is 'retaining the alias' - bit weird I know. Good attention to detail BTW.
|
0c4a888 needs rebase |
3 participants
First patch introduces
use_selflint. Its massive, are you ok with that?Second patch does the rename. For reference, this was prompted by the first item in #937 (comment)
Patch two is only 100 lines of red/green.