Chromium has a bug that it always sends a same challenge value due to inability to generate a random value in a worker process. We need to add a test to confirm whether the bug is fixed or not. Original bug: http://crbug.com/47390
Created attachment 61368 [details] Patch
Comment on attachment 61368 [details] Patch I don't think that requiring strict comparison is formally correct. Is there anything in specs that disallows MessageEvent.data return a String object? r=me, since it's at least testing for our current behavior, and the behavior is unlikely to change.
Comment on attachment 61368 [details] Patch LGTM
Comment on attachment 61368 [details] Patch Clearing flags on attachment: 61368 Committed r63298: <http://trac.webkit.org/changeset/63298>
All reviewed patches have been landed. Closing bug.
http://trac.webkit.org/changeset/63298 might have broken Chromium Mac Release The following changes are on the blame list: http://trac.webkit.org/changeset/63298 http://trac.webkit.org/changeset/63299 http://trac.webkit.org/changeset/63300
(In reply to comment #6) > http://trac.webkit.org/changeset/63298 might have broken Chromium Mac Release > The following changes are on the blame list: > http://trac.webkit.org/changeset/63298 > http://trac.webkit.org/changeset/63299 > http://trac.webkit.org/changeset/63300 63298 and 63299 add test only, so 63300 should be blamed.