Closed Bug 270030 Opened 20 years ago Closed 20 years ago

Tabbing through document is broken in certain cases

Categories

(Core :: DOM: UI Events & Focus Handling, defect)

x86
All
defect
Not set
normal

Tracking

()

VERIFIED FIXED

People

(Reporter: mcsmurf, Assigned: aaronlev)

References

()

Details

(Keywords: access, regression)

Attachments

(2 files)

happens with current cvs trunk build
To reproduce:
1. Go to URL
2. Try to tab through the document

Observe that it hangs at the checkbox. If you press the tab button once again
(when the selection is at the checkbox), you see that the statusbar updates (it
displays the url of the link), but the dotted border for the selection doesn't
move on. With the build from 2004-08-25-08 it works, with 2004-09-02-06 it
doesn't. Between those builds i couldn't test, because there are no nightly
builds available for that timeframe for Windows on
http://archive.mozilla.org/pub/mozilla/nightly/. But someone with Linux might
want to test some builds and so decrease the regression window.
I can see bug in the latest nightlies trunk. 
It works for me with a 2004-08-31 (7:12am) trunk build.
Status: UNCONFIRMED → NEW
Ever confirmed: true
Broke between 2004-08-31 21:00:00 and 2004-09-02 00:00:00 
Bonsai link: 
http://bonsai.mozilla.org/cvsquery.cgi?treeid=default&module=SeaMonkeyAll&branch=HEAD&branchtype=match&dir=&file=&filetype=match&who=&whotype=match&sortby=Date&hours=2&date=explicit&mindate=2004-08-31+20%
3A00%3A00&maxdate=2004-09-02+01%3A00%3A00&cvsroot=%2Fcvsroot 
OS->All (tested on Linux for nightly builds)
OS: Windows 2000 → All
What date range were you trying to give for bonsai? That's over a month's worth
of checkins.
(In reply to comment #5)
> What date range were you trying to give for bonsai? That's over a month's worth
> of checkins.

Maybe you need to manually enter the checkin dates into bonsai, i copy&pasted
that link from Konqueror, which didn't quite work as it seems. One last attempt:
http://bonsai.mozilla.org/cvsquery.cgi?treeid=default&module=SeaMonkeyAll&branch=HEAD&branchtype=match&dir=&file=&filetype=match&who=&whotype=match&sortby=Date&hours=2&date=explicit&mindate=2004-08-31+20%3A00%3A00&maxdate=2004-09-02+01%3A00%3A00&cvsroot=%2Fcvsroot
Surprising this didn't get noticed before, I don't know how it ever worked  in
these cases.
Attachment #166241 - Flags: superreview?(neil.parkwaycc.co.uk)
Attachment #166241 - Flags: review?(neil.parkwaycc.co.uk)
Comment on attachment 166241 [details] [diff] [review]
Allow selection offsets of 0

Aren't offsets always positive (i.e. negative offsets are excluded by range
code)? And I'm fortunately not allowed to do content r+sr :-P
Attachment #166241 - Flags: superreview?(neil.parkwaycc.co.uk)
Attachment #166241 - Flags: review?(neil.parkwaycc.co.uk)
(In reply to comment #9)
Neil, aren't you allowed to review stuff in esm? That's focus related and by now
you know as much as anyone.

> Aren't offsets always positive 
False, zero is allowed

> (i.e. negative offsets are excluded by range code)? 
True, but zero is still allowed

For example SetStartBefore on a first child sets the start to parent, 0


*** Bug 268983 has been marked as a duplicate of this bug. ***
(In reply to comment #10)
>(In reply to comment #9)
>Neil, aren't you allowed to review stuff in esm?
I said r+sr, I'll do one of your choice.

>>(i.e. negative offsets are excluded by range code)? 
>True, but zero is still allowed
What I meant was, will endOffset ever be negative in normal use, i.e. should you
be asserting it is nonnegative in debug, rather than testing it?
Comment on attachment 166241 [details] [diff] [review]
Allow selection offsets of 0

You're right, those can be changed to assertions. New patch necessary?
Attachment #166241 - Flags: superreview?(neil.parkwaycc.co.uk)
Comment on attachment 166241 [details] [diff] [review]
Allow selection offsets of 0

sr=me with those conditions removed in release builds (I don't care about debug
builds).
Attachment #166241 - Flags: superreview?(neil.parkwaycc.co.uk) → superreview+
Attachment #166241 - Flags: review?(ginn.chen)
Comment on attachment 166241 [details] [diff] [review]
Allow selection offsets of 0

r=me
Attachment #166241 - Flags: review?(ginn.chen) → review+
Attachment #166241 - Flags: approval1.8a5?
Comment on attachment 166241 [details] [diff] [review]
Allow selection offsets of 0

a=asa for 1.8a5 checkin.
Attachment #166241 - Flags: approval1.8a5? → approval1.8a5+
Checking in nsEventStateManager.cpp;
/cvsroot/mozilla/content/events/src/nsEventStateManager.cpp,v  <-- 
nsEventStateManager.cpp
new revision: 1.544; previous revision: 1.543
done
Status: NEW → RESOLVED
Closed: 20 years ago
Resolution: --- → FIXED
Keywords: access
V.
With a current CVS trunk build the testcases work.
Status: RESOLVED → VERIFIED
Component: Keyboard: Navigation → User events and focus handling
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: