225820 - Refactor nsFrameFrame to fix bugs and improve code

Status: RESOLVED FIXED

(Core :: Layout, defect, P1)

Description

[Robert O'Callahan (:roc) (email my personal email if necessary)]

Continuation of work in bug 212366

Comment 1

[Robert O'Callahan (:roc) (email my personal email if necessary)]

Attachment: revised patch

Okay... There are only a few changes from the last patch. The nsBlockFrame
change and the cosmetic fix that you requested is in there.

I don't know how visibility:hidden <frame>s worked before. To get it to work
now I had to make <frame>s work like <iframe>s and hook their view managers up
into the parent view hierarchy (nsDocumentViewer.cpp change); this ensures that
when the nsSubDocumentFrame's view is hidden, all the views in the subdocument
are treated as hidden. This change will also make <frame>s transparent by
default. I think that's the right thing. To get printing and print preview to
hide <frame>s I adjusted the logic in nsPrintEngine to generalize the IFRAME
hiding code to all subdocuments.

To get printing and print preview to work properly with IFRAMEs I had to
eliminate their assumptions about getting to the inner frame in nsPrintEngine.

I've tested all the testcases in all the bugs you mentioned, as well as the
IFRAME testcase you posted. They all seem to work. The position:relative bugs
that you were worried about were all about computing the correct overflow area.
That shouldn't affect or be affected by these changes.

Comment 2

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

Comment on attachment 135606 [details] [diff] [review]
revised patch

r+sr=dbaron, although this may be a little risky to land this close to the
freeze

Comment 3

[Robert O'Callahan (:roc) (email my personal email if necessary)]

Yeah. I think I'll save it for 1.7a.

Comment 4

[Robert O'Callahan (:roc) (email my personal email if necessary)]

Fix checked in.