Add go-to-definition inference for shadowed namespaces #62813
+5,530
−1,453
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Outdated suggestions cannot be applied.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.
Written as a proposed solution to #55593.
Note that like 95% of the changes for this PR come from updating test baselines, since my solution involves the use of a timeout inside the session.
Solves the issue of namespace shadowing by checking for "typey" definitions in comparison with the kind of the node selected. If its not possible to narrow to one definition, defaults to the current behavior of showing all options. Additionally, if the same symbol is go-to-definitioned on twice within 5 seconds, this will show all definitions as a backoff system.
Tested locally + with
hereby runtests