[webkit-dev] best practice for failing Windows tests?
ojan at chromium.org
Mon Mar 15 12:11:56 PDT 2010
I committed http://trac.webkit.org/changeset/55913, which to my surprise
caused three Windows tests to fail.
I tested manually on a Windows nightly and discovered that we have a bug on
Windows where we don't smart-delete in some cases depending on where on the
word you are double-clicking. The bug obviously isn't a result of r55913
since there were no code changes.
What's the right thing to do here to get the build green?
1. Rollback r55913. This seems silly since it's just exposing an existing
2. Update the windows expected results and file a bug.
3. Skip the tests on Windows.
Option 2 seems best to me.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the webkit-dev