Page MenuHomePhabricator

Improve Remarkup parsing performance for certain large input blocks
ClosedPublic

Authored by epriestley on Feb 4 2020, 10:37 PM.
Tags
None
Referenced Files
F13269949: D20968.diff
Wed, May 29, 8:19 AM
F13258793: D20968.id49962.diff
Sun, May 26, 12:56 PM
F13238712: D20968.diff
Tue, May 21, 8:53 PM
Unknown Object (File)
Mon, May 6, 2:25 AM
Unknown Object (File)
Sat, May 4, 8:44 PM
Unknown Object (File)
May 1 2024, 7:29 AM
Unknown Object (File)
Apr 24 2024, 12:46 AM
Unknown Object (File)
Apr 21 2024, 11:02 PM
Subscribers
None

Details

Summary

Fixes T13487. In PHI1628, an install has a 4MB remarkup corpus which takes a long time to render. This is broadly expected, but a few reasonable improvements fell out of running it through the profiler.

Test Plan
  • Saw local cold-cache end-to-end rendering time drop from 12s to 4s for the highly secret input corpus.
  • Verified output has the same hashes before/after.
  • Ran all remarkup unit tests.

Diff Detail

Repository
rP Phabricator
Branch
remarkup1
Lint
Lint Passed
SeverityLocationCodeMessage
Advicesrc/infrastructure/markup/rule/PhabricatorObjectRemarkupRule.php:200XHP14Misuse of `preg_quote`
Advicesrc/infrastructure/markup/rule/PhabricatorObjectRemarkupRule.php:213XHP14Misuse of `preg_quote`
Unit
Tests Passed
Build Status
Buildable 23787
Build 32712: Run Core Tests
Build 32711: arc lint + arc unit

Event Timeline

src/infrastructure/markup/remarkup/PhutilRemarkupEngine.php
250–254

This piece was pretty wild. It meant "Are both blocks default blocks (plain old paragraphs)?" but tested that in a very convoluted way.

This revision was not accepted when it landed; it landed in state Needs Review.Feb 4 2020, 11:07 PM
This revision was automatically updated to reflect the committed changes.