Implement PartialOrd for Cursor
#1236
Merged
Conversation
This file contains 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
Fixes dtolnay#1235. As seen in that issue, a syntax node can actually cross the border of a None-delimited group, due to such groups being transparent to the parser in most cases. In the cases that this can occur, the presence of the group is known to be semantically irrelevant, so we just explicitly ignore the presence of the group. It's important that we only ignore None-delimited groups when the verbatim end requires entering the group, as such groups can have semantic impact (such as grouping binary operator chains) in other positions.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
New version of #1196 now that #1223 has changed token buffers to a flat layout, making the comparison trivial to implement.
As additional motivation and since it was first discovered in #1196, fixes #1235. The issue has nothing to do with
eqcomparing the scope pointer; in fact, the involved pointers actually have the same scope. Properly fixing the issue requires the ability to compare cursors in order to detect the case where the verbatim end is inside of aNone-delimited group we previously would have stepped over.I removed the check for scope equality here, as it is much more useful to be able to be able to compare cursors which don't necessarily have the same scope end (e.g. to determine which of multiple speculative parses got the furthest). Additionally, we actually now rely on comparing two cursors at the same location with different scopes in
verbatim::betweentreating them as equal, as we explicitly enter theNone-delimited group which the end cursor implicitly entered.If exposing this is undesirable, we can reintroduce the scope check to
Cursorcomparison (treating cursors not with the same scope as unordered and nonequal) and introduce aCursor::position(&self) -> impl Ordwhich can be used to order cursor's position ignoring the scope. (This would also provide a nice place to document how cursors are ordered.)I'm happy to implement + document that alternative approach if it's desired, but since just dropping the scope-matters in comparison is simpler, I'm providing this solution first.