[ty] Add assertions to ensure that we never call KnownClass::Tuple.to_instance() or similar
#21027
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Outdated suggestions cannot be applied.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.
Summary
This came up in #20988. It's usually a mistake to call
KnownClass::Tuple.to_instance(); it will usually be more efficient and more correct to callType::homogeneous_tuple()orType::heterogeneous_tuple()instead. Add some debug assertions to ensure that we don't make this mistake.I considered making these non-debug assertions, as the equality check should be pretty cheap, but
KnownClass::to_instance()is a pretty hot method, and I think it's exceedingly likely that debug assertions will catch this bug cropping up in the future.I also removed the private helper method
Type::non_tuple_instance()frominstance.rs. It wasn't a zero-cost abstraction, it didn't save that much code, and it had to have a big warning doc-comment above it telling us never to make it public to code outside the module. All things considered, it was no longer worth it to keep it around.Test Plan
cargo test -p ty_python_semantic