276100 - GetOriginToView has unused PresContext argument

Status: RESOLVED FIXED

(Core :: Layout, defect)

Description

[Bernd]

due to bug 276015 the PresContext argument is not used anymore. We should remove
it. This opens the way to remove the prescontext from 
nsContainerFrame::GetFrameForPointUsing and
http://lxr.mozilla.org/seamonkey/ident?i=GetFrameForPoint

Comment 1

[Vidar Haarr (not reading bugmail)]

Attachment: removes prescontext from GetOriginToViewOffset and GetFrameForPoint(Using)

diff -up7, removes aPresContext argument from GetOriginToViewOffset,
GetFrameForPoint and GetFrameForPointUsing.

Comment 2

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

You should remove the aPresContext from the static GetFrameFromLine function in
nsBlockFrame.cpp too.  This makes the argument unused, and passing the pres
context for that argument is the only reason you needed to add GetPresContext().

Comment 3

[Vidar Haarr (not reading bugmail)]

Attachment: update to comment #2

Also includes missing mathml changes; I've enabled mathml in my builds now :)

Comment 4

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

Comment on attachment 169957 [details] [diff] [review]
update to comment #2

When this is checked in, watch the warnings on brad tinderbox and make sure
that there aren't additional method-hiding warnings.

>Index: layout/generic/nsBlockFrame.cpp

Please don't unwrap the lines -- they should be wrapped at less than 80
characters.  Feel free to wrap lines *that you're already changing* if they
weren't properly wrapped, in this file and elsewhere (but you don't need to,
and it's a good bit of work).

>Index: svg/base/src/nsSVGForeignObjectFrame.cpp
>   nsPresContext *presContext = GetPresContext();

Remove this line too.

r+sr=dbaron with those changes

Comment 5

[Vidar Haarr (not reading bugmail)]

Attachment: removes the stray nsPresContext and fixes some 80 breaks

Carrying over r/sr from dbaron

Comment 6

[Vidar Haarr (not reading bugmail)]

Attachment 169974 [details] [diff] checked in by neil with no new warnings:
<http://tinderbox.mozilla.org/SeaMonkey/warn1104598200.12912.html#neil>

Thanks for the review!

marking FIXED.

Comment 7

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

brad tinderbox hasn't cycled since it was checked in, so I'm not sure how you
can say "no new warnings" yet.

Comment 8

[Vidar Haarr (not reading bugmail)]

Heh, I guess I should get a refill on my coffee ;)
Thanks for the heads up.

Comment 9

[Vidar Haarr (not reading bugmail)]

If I'm not mistaken, it has cycled now:
<http://tinderbox.mozilla.org/SeaMonkey/warn1104607500.5979.html#neil>

Comment 10

[Bill Gianopoulos [:WG9s]]

I suspect this check-in has caused the regression reported in bug 276719.

Comment 11

[Vidar Haarr (not reading bugmail)]

I just found an oddity here that I can't explain:
1. Leftclick on the arrow of a html dropdown
2. Leftclick again.

Moz 1.7.3 and 1.8a6 results on Linux:
(1) Drops down
(2) Pulls up

Recent CVS FF or Seamonkey:
(1) Drops down
(2) Drops down

Quite strange; might be a related regression ?

Comment 12

[Vidar Haarr (not reading bugmail)]

The behaviour in comment #11 was also fixed by the regression fix in bug 276719.