Repository navigation
Prefilter long decimal scans before keyword searches - #1325
Merged
charliermarsh merged 1 commit intoOct 18, 2026
Merged
Conversation
charliermarsh
deleted the
charlie/codex-prefilter-oversized-decimal-scans-20261018
branch
October 18, 2026 02:26
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
We prefilter the decimal-limit prescan using aligned blocks of digits and underscores, before searching for soft keywords. Every qualifying digit run contains a complete block; candidates retain the existing precise scan and error ordering.
Matched project profiles use 1.13-1.29% fewer instructions for Mypy across the sampled Python 3.12-3.15 configurations, and 0.58% fewer for SymPy on Python 3.15 O2. The 177-case instruction geometric mean improves by 0.06%, with individual changes from -0.93% to +0.63%. All measured project outputs are byte-identical.
Mypy native CPU time decreases 1.09%, with peak RSS effectively flat; those samples retain two timing warnings. SymPy remains inconclusive across two studies: pooled CPU changes +0.42% and +0.30%, with opposite directions by order; peak RSS changes -0.92% and +0.16%. We make no SymPy native speed or memory claim.
Compatibility checks pass across Python 3.12 through 3.15 at all three optimization levels, with unchanged results.