Closed Bug 361383 Opened 18 years ago Closed 18 years ago

Patch for bug 348183 can be backed out now that the patch for bug 348357 went in

Categories

(Firefox :: Tabbed Browser, defect)

x86
Windows XP
defect
Not set
normal

Tracking

()

RESOLVED FIXED

People

(Reporter: martijn.martijn, Assigned: martijn.martijn)

References

Details

Attachments

(1 file)

Patch for bug 348183 can be backed out now that the patch for bug 348357 went in.
The patch for bug 348183 was more or less a work-around.
Attached patch patchSplinter Review
Like this.
Attachment #246145 - Flags: review?(sspitzer)
backing out my fix seems reasonable, but given that nsGlobalWindow::Focus() could still return NS_ERROR_SOMETHING (if widget->SetFocus() returns an error, for example [1]) maybe keeping the try / catch isn't a bad idea? 

[1] http://lxr.mozilla.org/seamonkey/source/dom/src/base/nsGlobalWindow.cpp#3624

I'll defer to gavin.
Assignee: nobody → martijn.martijn
Depends on: 348357
Summary: Patch for bug 348183 can be backed out now that the patch for bug 348357 went in → Patch for bug 348183 went can be backed out now that the patch for bug 348357 went in
Comment on attachment 246145 [details] [diff] [review]
patch

while it was a work around, since .focus() could still fail, I don't think keeping the try/catch is a bad idea.  but if you get r=gavin, then I'm ok with it.
Attachment #246145 - Flags: review?(gavin.sharp)
(In reply to comment #2)
> backing out my fix seems reasonable, but given that nsGlobalWindow::Focus()
> could still return NS_ERROR_SOMETHING (if widget->SetFocus() returns an error,
> for example [1]) maybe keeping the try / catch isn't a bad idea? 

When looking at the functions:
http://lxr.mozilla.org/seamonkey/source/widget/src/gtk2/nsWindow.cpp#696
http://lxr.mozilla.org/seamonkey/source/widget/src/os2/nsWindow.cpp#1513
http://lxr.mozilla.org/seamonkey/source/widget/src/gtk/nsWindow.cpp#1347
http://lxr.mozilla.org/seamonkey/source/widget/src/mac/nsWindow.cpp#582
http://lxr.mozilla.org/seamonkey/source/widget/src/beos/nsWindow.cpp#1168
http://lxr.mozilla.org/seamonkey/source/widget/src/windows/nsWindow.cpp#2204
I see that gtk2 and windows indeed can return failure.
But it seems to me that nsGlobalWindow::Focus() should never be able to return an error, should it?
Comment on attachment 246145 [details] [diff] [review]
patch

Those failure cases look to me like they're pretty exceptional (mostly OOM and being called from a non-GUI thread), so I think it's a good idea to remove this. There are many other places where we assume that .focus() on a window won't throw, and it largely hasn't in the past, modulo the relatively uncommon case of bug 348357.
Attachment #246145 - Flags: review?(gavin.sharp) → review+
(In reply to comment #5)
> modulo the relatively uncommon case of bug 348357.

Er, I meant bug 331457 and bug 348183, of course.
Comment on attachment 246145 [details] [diff] [review]
patch

r=sspitzer, all hail G#!
Attachment #246145 - Flags: review?(sspitzer) → review+
Summary: Patch for bug 348183 went can be backed out now that the patch for bug 348357 went in → Patch for bug 348183 can be backed out now that the patch for bug 348357 went in
Checking in tabbrowser.xml;
/cvsroot/mozilla/toolkit/content/widgets/tabbrowser.xml,v  <--  tabbrowser.xml
new revision: 1.215; previous revision: 1.214
done

Checked in on trunk.
Status: NEW → RESOLVED
Closed: 18 years ago
Resolution: --- → FIXED
Flags: in-testsuite-
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: