Marked version:

18.0.3 / 69257e45

Markdown flavor: CommonMark

Description

This caches the indentation-sensitive regexes used while scanning list items. These regexes are rebuilt for each list item today even though the effective indentation is bounded to Math.min(3, indent - 1), so the same few regex shapes are reused repeatedly during list parsing.

A CPU profile of a CommonMark parse loop showed samples in nextBulletRegex, fencesBeginRegex, headingBeginRegex, and htmlBeginRegex. After caching the bounded indentation regexes, those helper functions no longer showed self samples in the same profile pass.

Local benchmark notes from my machine:

• npm run bench: marked improved from 1655 ms to 1513 ms on the CommonMark benchmark run.
• A list-heavy parse benchmark improved from roughly 488-502 ms to 376-466 ms across seven rounds.
• A repeated CommonMark parse loop improved from roughly 244-260 ms to 229-245 ms across seven rounds.

Contributor

• Test(s) exist to ensure functionality and minimize regression: existing spec/unit coverage for list parsing, plus full npm test passed locally.
• no tests required for this PR.
• If submitting new feature, it has been documented in the appropriate places. n/a, this is not a new feature.

Committer

In most cases, this should be a different person than the contributor.

• CI is green (no forced merge required).
• Squash and Merge PR following conventional commit guidelines.