Add CorpusStorage::find_extra returning the query alternative for each match - #342
Merged
thomaskrause merged 3 commits intoAug 11, 2026
Conversation
# Conflicts: # CHANGELOG.md
The current benchmarks suggest there is a about 6% worse performance for find() queries and these benchmarks for find_extra() allow to compare the two methods and isolate the source of the increased workload. Since find_extra() and find() have roughly the same execution time, the source could be the added usize added as value (which increases the memory stored in the temporary BTreeMap) or some issue with refactoring out some of the work to closures or sub-functions which might not be inlined.
thomaskrause
approved these changes
Aug 11, 2026
thomaskrause
left a comment
thomaskrause
left a comment
Member
There was a problem hiding this comment.
Thanks for the PR! I had some questions about storing the alternatives as vectors, because I assumed that they always would be just using the index as value which was not the case so I deleted the comments.
I also added some benchmarks to compare find() with find_extra() and it seems the overhead is marginal. Also, benchmarking in comparision with the main branch did not find any difference outside what I would consider noise (in the first benchmark run I forgot to disable dynamic CPU throttling and got quite some noise, but this improved after running the benchmark correctly.
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This adds a
CorpusStorage::find_extramethod that returns for each match which query alternative produced it. In case multiple alternatives produce the same match, it gets attributed to the first alternative.The new method has the same signature as
CorpusStorage::find, except that it returns aVec<MatchExtra>instead of aVec<String>. TheMatchExtrastruct is marked#[non_exhaustive]so that adding more fields later isn't a breaking change. The namesfind_extraandMatchExtrafollow the convention established forcount_extraandCountExtra.Performance: the alternative index is a plain
usizecarried alongside the numeric match group, so it stays out of the sort key and the uniqueness check, and match IDs are still resolved to strings at the end. I did extract the common parts offindandfind_extra: The added cost forfindis thatusizeper match in the temporary BTree index plus one iterator indirection; it showed no measurable effect on the GUM test corpus.I didn't extend the FFI surface for now.
Closes #341.