When the parent window of a sheet is resized, the content of that window is never resized along with it resulting in visual corruption. The reason for this is that since the sheet parent window is not enabled, Gecko NS_SIZE events are not passed along to its contents.
Created attachment 170890 [details] [diff] [review] patch This fixes it by allowing events through even if the window is disabled. I haven't noticed any ill effects, but more testing would probably be helpful.
Attachment #170890 - Flags: review?(sfraser_bugs)
Comment on attachment 170890 [details] [diff] [review] patch I don't have an OS X box to test on, so I can only offer guidelines. Ari's patch effectively hamstrings Mozilla's attempt to disable a window. Perhaps that doesn't hurt anything with sheets on OS X; perhaps the OS takes care of disabling the parent window. Event filtering was part of Mozilla's homespun window modality necessitated by OS 9. I'd rather see the filter modified to allow paint events into the disabled window, rather than all events as this patch does. But if OS X finally takes it upon itself to handle modality, and *not just for sheets* (because ideas like bug 136422 come and go), then feel free to chuck event filtering. If you go that route, I'd rather see a cleaner patch that doesn't leave little bits of the old code dangling useless.
Attachment #170890 - Flags: review?(sfraser_bugs) → review-
QA Contact: mac
*** Bug 334081 has been marked as a duplicate of this bug. ***
You need to log in before you can comment on or make changes to this bug.