Repository navigation
fix(spec): compare truncate width in code points when projecting strings - #3634
Open
compiler-segfault wants to merge 2 commits into
Open
compiler-segfault wants to merge 2 commits into
compiler-segfault wants to merge 2 commits into
Conversation
The truncate transform truncates strings by Unicode code points, but the projection compared the literal length in UTF-8 bytes. A prefix of two code points and three bytes looked as wide as truncate[3], so STARTS WITH was rewritten to an equality predicate and scans dropped matching rows.
This branch has not been deployed
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Which issue does this PR close?
What changes are included in this PR?
truncate[N]truncates strings by Unicode code points, but the projection compared the literal length in UTF-8 bytes. A prefix of two code points and three bytes (Év) therefore looked as wide astruncate[3],STARTS WITHwas rewritten to=, and scans silently returned no rows.NOT STARTS WITHwas affected the same way.The four comparisons in
Transform::strict_projectandTransform::adjust_operatornow count code points for string literals; binary literals keep byte counting. The other evaluators (inclusive_metrics_evaluator,row_group_metrics_evaluator,page_index_evaluator) already usechars().count().Are these changes tested?
Two unit tests in
transform/truncate.rscover the inclusive and the strict projection for prefixes of two, three and four code points overtruncate[3]. Both fail before the change (city = "Év"instead ofcity STARTS WITH "Év") and pass with it.cargo test -p iceberg --libpasses: 2135 tests.AI Disclosure
Diagnosis and patch were written with an AI assistant. I read the projection code paths and ran the full
iceberglib suite locally. Points worth a reviewer's attention:strict_projectreturnsNoneforSTARTS WITHwhen the prefix is wider than the truncate width (left unchanged, as it was), and binary literals are still compared in bytes, which matchestruncate_binary.