Provide an extended framework for type visit, for use in rust-analyzer #149856
+399
−55
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Collaborator
|
Some changes occurred in src/tools/cargo cc @ehuss |
rustbot
added
S-waiting-on-review
This comment has been minimized.
This comment has been minimized.
rust-analyzer needs to be able to visit types when treating not only `Ty`, `Const`, `Region` and `Predicate` specifically, but *all* rust-analyzer-made types specifically (excluding e.g. `TraitRef`, that is declared in rustc_type_ir). This is needed to implement garbage collection. To support this, we introduce a second, rust-analyzer-only visit trait, named, without much thought, `CustomizableTypeVisitable`. It's simpler than `TypeVisitable` (for example, it does not have a trait for the visitor, and does not support early-returning) because this is what rust-analyzer needs, but its most distinguished feature is that the visitor is a generic of the *trait* instead of the *method*. This way, specific types can treat specific visitor types specifically and call their methods. In rustc_type_ir we implement it for a bunch of basic types, and using a derive macro for the rest. The macro and trait are completely disabled when compiling for rustc (`feature = "nightly"`), so not even a compile time penalty will be paid.
ChayimFriedman2
force-pushed
the
no-copy-solver-v3
branch
from
dcbc91b to
8ed8db8
Compare
Member
|
I'm going to nominate this. I have to catch up with all the discussion here, but this is tiny compared to everything I've seen and overall something I think we should land if it all works out. bikeshed, but I think a better name is |
jackh726
added
the
I-types-nominated
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
rust-analyzer needs to be able to visit types when treating not only
Ty,Const,RegionandPredicatespecifically, but all rust-analyzer-made types specifically (excluding e.g.TraitRef, that is declared in rustc_type_ir). This is needed to implement garbage collection.To support this, we introduce a second, rust-analyzer-only visit trait, named, without much thought,
CustomizableTypeVisitable. It's simpler thanTypeVisitable(for example, it does not have a trait for the visitor, and does not support early-returning) because this is what rust-analyzer needs, but its most distinguished feature is that the visitor is a generic of the trait instead of the method. This way, specific types can treat specific visitor types specifically and call their methods.In rustc_type_ir we implement it for a bunch of basic types, and using a derive macro for the rest. The macro and trait are completely disabled when compiling for rustc (
feature = "nightly"), so not even a compile time penalty will be paid.r? types
This is a replacement to other efforts to support non-
Copytype in the solver, replacing them with a GC in r-a, as decided by @rust-lang/rust-analyzer. The code is tiny in comparison, and I believe T-types will have no problem maintaining it, which mostly means adding the derive on new things when they are added and things break on the r-a side.