Skip to content
Skip the list of specifications

Highlight API

94 of 106 files pass (88.7%): a file passes when every subtest in it passes. These are web-platform-tests files, unchanged.

UnitPassFailNot judged
FilesPass94Fail12Not judged10
Subtests, as wpt.fyi countsPass104Fail14—

1 of the failing files differs from the reference only by anti-aliasing; we count it as failing. Of the 290 subtests wpt.fyi runs in this directory, we judge 118; the others we do not run, and no figure here speaks for them.

How the browsers do on this suiteThe suite on GitHub

Every test

104 files are judged by comparing renders and 10 are not judged (they are listed, with why), by the directory the suite keeps them in; the 2 judged by assertions follow. Open a directory to see what each test checks, and a test for our render, the reference and the pixels that differ.

114 tests in 4 directories

Judged by assertions

2 files state what they expect as assertions in a script rather than as a second rendering. Each is one file in the table above, passing only when all its subtests pass.

FileSubtests passing
custom-highlight-universal-parsing-and-computed-style.tentative.html5 of 8
highlight-pseudo-parsing.html6 of 6

Results of the run of 8 November 2026, engine build 1d7a91105. The data behind every page is open.