Bug 47644 – Add a LINUX_LUCID flag to Chromium's test_expectations.txt

• Status: RESOLVED
• Resolution: DUPLICATE
• Component: New Bugs

[Adam Barth]

Add a test_expectations.txt file for Chromium Linux on Lucid

[Adam Barth]

Created attachment 70699 [details]
Patch

[James Robinson]

Comment on attachment 70699 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=70699&action=review

> LayoutTests/platform/chromium-linux-lucid/test_expectations.txt:4
> +BUGXXX : fast/replaced/css-content-and-webkit-mask-box-image-crash.html = TIMEOUT PASS

this has nothing to do with lucid. if it's failing please file a separate bug about it

> LayoutTests/platform/chromium-linux-lucid/test_expectations.txt:36
> +BUGXXX : fast/filesystem/op-copy.html = TIMEOUT
> +BUGXXX : fast/filesystem/op-get-entry.html = TIMEOUT
> +BUGXXX : fast/filesystem/op-get-parent.html = TIMEOUT
> +BUGXXX : fast/filesystem/op-move.html = TIMEOUT
> +BUGXXX : fast/filesystem/op-read-directory.html = TIMEOUT
> +BUGXXX : fast/filesystem/op-remove.html = TIMEOUT
> +BUGXXX : fast/filesystem/op-restricted-chars.html = TIMEOUT
> +BUGXXX : fast/filesystem/op-restricted-names.html = TIMEOUT
> +BUGXXX : fast/filesystem/simple-persistent.html = TIMEOUT
> +BUGXXX : fast/filesystem/simple-temporary.html = TIMEOUT
> +BUGXXX : http/tests/media/video-referer.html = TIMEOUT

Please don't include these. the fact that the filesystem and media test time out has nothing to do with lucid or hardy and shouldn't be captured here.

> LayoutTests/platform/chromium-linux-lucid/test_expectations.txt:38
> +BUGXXX : platform/gtk/accessibility/aria-table-hierarchy.html = MISSING

nothing to do with lucid

[Evan Martin]

Do we depend on the CPU arch too?

[Adam Barth]

(In reply to comment #3)
> Do we depend on the CPU arch too?

Dunno.  The machine I was using here was 64 bit.

I'm probably going to abandon this path because I don't think the test_expectations.txt files are maintained quickly enough for the commit-queue.  It seems like the commit-queue would often be blocked on a human updating the file.  We're going to try to do something more automatic.

[Dirk Pranke]

We should probably still have Lucid support, since Hardy will be deprecated soon and I'd like to be able to run cleanly on Lucid. Taking ownership. Note that since the refactoring of the test expectations modifiers, this should be real easy to do.

[Evan Martin]

For reference, we hope to drop Hardy support in the next few months, at which point we'll only support lucid.  But I think for future-proofing the next version it's nice to put "lucid" in the relevant file names anyway.

[Dirk Pranke]

I'm gonna close this and roll it into bug 55527, where we are adding support for the Chromium Linux 64-bit tester. Depending on what we do there we can add the necessary combination of 'LUCID', and 'HARDY' flags.

*** This bug has been marked as a duplicate of bug 55527 ***