Bug 315204
| Summary: | REGRESSION(294355@main): [import-w3c-tests] --tip-of-tree option doesn't seems to actually pick the tip of the tree anymore | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Sam Sneddon [:gsnedders] <gsnedders> |
| Component: | Tools / Tests | Assignee: | Sam Sneddon [:gsnedders] <gsnedders> |
| Status: | RESOLVED FIXED | ||
| Severity: | Normal | CC: | koivisto, webkit-bug-importer |
| Priority: | P2 | Keywords: | InRadar |
| Version: | WebKit Nightly Build | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
Sam Sneddon [:gsnedders]
From Antti:
> The cloned wpt repo's HEAD was at 1d3aae1b59 (2026-05-06). 271dd5a9d3 ("Disallow <url> type in attr() in the tests", PR #59516) landed 2026-05-14 — 8 days after the snapshot the import used. Today is 2026-05-20, so the clone was 14 days behind master. After I ran git fetch origin just now, the new commits including 271dd5a are present.
>
> The wpt "temp dir" is actually a symlink to my canonical clone at /Users/antti/webkit/wpt. That clone's local master was at 1d3aae1b59 (May 6) and the import script's fetch step apparently didn't refresh origin/master — so when it checked out origin/master, it got the stale 14-day-old commit.
>
> The import script's git.fetch(git.default_branch) at webkitpy/w3c/test_downloader.py:112 is buggy — it passes 'master' as the remote arg, so the auto-fetch is a no-op. That's why my first run was 14 days stale. Worth filing.
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Radar WebKit Bug Importer
<rdar://problem/177532662>
Sam Sneddon [:gsnedders]
Pull request: https://github.com/WebKit/WebKit/pull/65317
EWS
Committed 313618@main (8b9b74668e91): <https://commits.webkit.org/313618@main>
Reviewed commits have been landed. Closing PR #65317 and removing active labels.