[Webkit-unassigned] [Bug 13759] REGRESSION (r20182-r20184): Incorrect rendering of multipart images

[bugzilla-daemon at webkit.org]

http://bugs.webkit.org/show_bug.cgi?id=13759





------- Comment #4 from mitz at webkit.org  2007-05-17 08:38 PDT -------
Created an attachment (id=14593)
 --> (http://bugs.webkit.org/attachment.cgi?id=14593&action=view)
Band-aid fix

The way the loader currently works, only when a new part arrives, the previous
part is delivered to the client. This means that at time t, Safari always shows
you part t-1. This is in fact the behavior of shipping Safari as well. You can
see it by changing the timeouts in the test case to make them uneven: change
the sleep(1) after abe to sleep(4) and run the test. Notice how it is the green
square, rather than abe, that is displayed for 4 seconds in each cycle. In
contrast, in Firefox it is indeed abe that is displayed for 4 seconds.

This patch doesn't fix the above behavior, but accounting for it, it ensures
that the client gets its own copy of the data so that it can use it after the
resource data changes to contain (parts of) the next part.

I think it's worth applying this fix for now, until the aforementioned bug is
addressed, so I will try to make a test to go with this patch.


-- 
Configure bugmail: http://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.