fix: diff outputs for queries participating in a cycle #980
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Outdated suggestions cannot be applied.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.
The title is a bit misleading but it's my intention to have a fix. For now, this only adds a regression test.
The problematic scenario is a query participating in a cycle (not the cycle head)
that creates some tracked structs only in later iterations. The issue here is that
we run
diff_outputs
after each iteration, which will delete any tracked structthat hasn't been created in the first iteration... However, other queries might depend
on those tracked structs already because salsa called
maybe_changed_after
on them.The fix is that we only call
diff_outputs
once the value is finalized, the same as for cycle heads.What makes this difficult is that finalization happens lazily.