Add layout tests to verify assignment to items of NodeList
Created attachment 47526 [details] Patch
Created attachment 47528 [details] First take
(In reply to comment #2) > Created an attachment (id=47528) [details] > First take It passes on both WebKit and Chromium's ToTs.
Is this a regression test for some previously fixed bug? Why 13 iterations?
(In reply to comment #4) > Is this a regression test for some previously fixed bug? Why 13 iterations? Yes, precisely: http://code.google.com/p/chromium/issues/detail?id=27967 Technically for current V8 2 is enough, but I thought that 13 is a reasonable compromise. If you prefer to see smaller constant, np, I'd only ask it to be >= 2, ideally >= 3
Comment on attachment 47528 [details] First take r=me I like to have a reference to the bug fixed in test output, but that's not really important, one can always do svn log.
Created attachment 47607 [details] Reference to Chromium bug added
(In reply to comment #6) > (From update of attachment 47528 [details]) > r=me > > I like to have a reference to the bug fixed in test output, but that's not > really important, one can always do svn log. I've added it into description. Is it fine?
Comment on attachment 47607 [details] Reference to Chromium bug added It is a little strange to see a link on a Web page that's plain text, and not clickable (one can use HTML tags in JS test description). But it's not a practical issue, no need to further change the patch.
Comment on attachment 47607 [details] Reference to Chromium bug added Thanks a lot for review, Alexey.
Comment on attachment 47607 [details] Reference to Chromium bug added Clearing flags on attachment: 47607 Committed r54004: <http://trac.webkit.org/changeset/54004>
All reviewed patches have been landed. Closing bug.