buildbot needs to understand whether NRWT exited early after having too many failures Requested by abarth on #webkit.
The results.json needs to include that information too.
> The results.json needs to include that information too. It might already. If not, it's easy to add.
Created attachment 99750 [details] Patch
Comment on attachment 99750 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=99750&action=review > Tools/Scripts/webkitpy/layout_tests/layout_package/manager.py:212 > + results['interrupted'] = interrupted # Does results.html have enough information to compute this itself? (by checking total number of results vs. total number of tests?) I doubt it. You can restrict the test set in a bunch of different ways.
Comment on attachment 99750 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=99750&action=review > Tools/ChangeLog:15 > + itself. It's possible Ojan could come up with a better fix. Nope. This is roughly how I would have done it. >> Tools/Scripts/webkitpy/layout_tests/layout_package/manager.py:212 >> + results['interrupted'] = interrupted # Does results.html have enough information to compute this itself? (by checking total number of results vs. total number of tests?) > > I doubt it. You can restrict the test set in a bunch of different ways. Agree.
Comment on attachment 99750 [details] Patch Clearing flags on attachment: 99750 Committed r90413: <http://trac.webkit.org/changeset/90413>
All reviewed patches have been landed. Closing bug.