290290 - nsObjectFrame's PaintSelf call does the wrong thing for paint layers

Status: RESOLVED WORKSFORME

(Core :: Layout, defect)

Description

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

nsObjectFrame's PaintSelf call does the wrong thing for paint layers.

Testcase to come later, but patch now, so I don't lose it.

Comment 1

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Attachment: patch

This obviously isn't tested very well yet.

Perhaps we should put this display check inside of PaintSelf itself.  It'll
mean calling it more, but I maybe that's OK.

Comment 2

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Actually, we may want to drop the PaintSelf completely due to the shared style
contexts.  And I don't think this has much to do with Acid2.

Comment 3

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

(In reply to comment #1)
> Perhaps we should put this display check inside of PaintSelf itself.  It'll
> mean calling it more, but I maybe that's OK.

Actually, probably not, since we need it for bug 78087.

Comment 4

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Attachment: testcase (draft)

Comment 5

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Attachment: testcase (draft)

Comment 6

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Hmm, this patch doesn't actually seem to help any of the testcases.

Comment 7

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Attachment: patch

This is cleaned up a little, but it doesn't actually work.

Comment 8

[Christian :Biesinger (don't email me, ping me on IRC)]

the code that's patch here will go away in bug 309521

Comment 9

[Boris Zbarsky [:bzbarsky]]

So is this still an issue now that bug 309521 is fixed?  That is, is there a bug when rendering a plug-in in an nsObjectFrame?

Comment 10

[Gábor Stefanik]

This is completely fixed in the latest trunk build of Firefox, possibly because bug 317375 have been fixed.

Comment 11

[Gábor Stefanik]

The testcase seems to have regressed badly in Firefox 4.0b8. Should this bug be reopened?

Comment 12

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

regressed how?

Comment 13

[Gábor Stefanik]

Sorry, not really a regression - I was referring to red visible in the testcase, which is apparently the correct rendering - IE8 and Chrome render the same (though Chrome shows some additional rendering issues).