NEW76727
Forcing style recalc changes result of multicol rendering
https://bugs.webkit.org/show_bug.cgi?id=76727
Summary Forcing style recalc changes result of multicol rendering
James Robinson
Reported 2012-01-20 12:25:50 PST
The two attached testcases render differently, even though the only difference between the two is the presence or absence of a 'document.body.offsetTop;' This probably means there's some bug in the incremental style recalculation logic. The patch at https://bugs.webkit.org/show_bug.cgi?id=46761 exposes this on a few layout tests, but this bug exists with or without that patch.
Attachments
testcase without the document.body.offsetTop (1.92 KB, text/html)
2012-01-20 12:26 PST, James Robinson
no flags
testcase with document.body.offsetTop after insertBefore() line (1.95 KB, text/html)
2012-01-20 12:26 PST, James Robinson
no flags
testcase without the document.body.offsetTop (1.42 KB, text/html)
2012-01-20 16:28 PST, James Robinson
no flags
testcase with document.body.offsetTop after insertBefore() line (1.44 KB, text/html)
2012-01-20 16:28 PST, James Robinson
no flags
James Robinson
Comment 1 2012-01-20 12:26:04 PST
Created attachment 123358 [details] testcase without the document.body.offsetTop
James Robinson
Comment 2 2012-01-20 12:26:23 PST
Created attachment 123359 [details] testcase with document.body.offsetTop after insertBefore() line
James Robinson
Comment 3 2012-01-20 16:28:14 PST
Created attachment 123403 [details] testcase without the document.body.offsetTop This version makes it slightly easier to see where various bits of text are coming from
James Robinson
Comment 4 2012-01-20 16:28:26 PST
Created attachment 123404 [details] testcase with document.body.offsetTop after insertBefore() line
Levi Weintraub
Comment 5 2012-03-20 16:11:01 PDT
The tests marking as failing here are suddenly randomly passing, or failing in different ways... I'm simply updating the test_expectations file for now.
Note You need to log in before you can comment on or make changes to this bug.