Last Comment Bug 188517 - Switching between tabs when form text field is active leaves caret showing
: Switching between tabs when form text field is active leaves caret showing
Status: VERIFIED DUPLICATE of bug 306173
: pp
Product: Core
Classification: Components
Component: Layout: Form Controls (show other bugs)
: Trunk
: PowerPC Mac OS X
: P4 trivial with 3 votes (vote)
: Future
Assigned To: layout.form-controls
: Terri Preston
Mentors:
Depends on: 167801
Blocks:
  Show dependency treegraph
 
Reported: 2003-01-10 07:57 PST by Stewart Gordon
Modified: 2005-11-01 17:45 PST (History)
12 users (show)
See Also:
Crash Signature:
(edit)
QA Whiteboard:
Iteration: ---
Points: ---
Has Regression Range: ---
Has STR: ---


Attachments

Description Stewart Gordon 2003-01-10 07:57:35 PST
User-Agent:       Mozilla/5.0 (Macintosh; U; PPC Mac OS X; en-US; rv:1.2) Gecko/20021126
Build Identifier: Mozilla/5.0 (Macintosh; U; PPC Mac OS X; en-US; rv:1.2) Gecko/20021126

If I have multiple tabs open in a window, have the cursor in a form text field
and then switch to another tab in the same window, half the time a 'cursor'
appears at that point in the tab that I have switched to.

Reproducible: Always

Steps to Reproduce:
1. Open a new tab.
2. Load a web page that has a form text field in it.
3. Switch to the text field.
4. At a moment when the text cursor's blinking is in the 'on' phase, click on
another tab.
Actual Results:  
At the point where the cursor was in the text field, a steady vertical line, the
size and shape of a cursor, appears on the switched-to tab, in the inverse of
the background colour it is on.

Expected Results:  
Displayed the page in the switched-to tab normally, without the extra line.
Comment 1 Erik Neuenschwander 2003-01-10 18:13:54 PST
Confirming on OS X 10.1.5, Mozilla 2003-01-10-08

This also occurs with TEXTAREA, and it is necessary to _click_ the tab -- if you
use the keyboard, the cursor appears in the new tab, but then blinks off at the
normal rate.  If you click the new tab, the cursor never blinks off.

If you scroll the new tab, the faux-cursor stays in the same relative position,
moving with the text of the page.  If you click on a 3rd tab, it doesn't persist.
Comment 2 Greg K. 2003-02-20 12:06:24 PST
Marking confirmed per comment 1. See also bug 121704.
Comment 3 Stewart Gordon 2003-02-21 02:47:06 PST
Re: comment 1

I think what actually happens is that the faux-cursor _is_ the blinking off, since blinking 
on/off the cursor consists of drawing the little line in XOR mode.  When switching tabs, it 
thinks the cursor is still 'on', so blinks it off, producing the line.
Comment 4 sairuh (rarely reading bugmail) 2003-04-14 13:11:33 PDT
*** Bug 199758 has been marked as a duplicate of this bug. ***
Comment 5 sairuh (rarely reading bugmail) 2003-04-14 13:15:39 PDT
seems to be OS X-specific. able to repro on 10.2.5 (mac), but unable to do so on
win2k or linux rh8.0.
Comment 6 sairuh (rarely reading bugmail) 2003-04-14 13:17:04 PDT
also affects mac embed apps such as camino.
Comment 7 Mina 2003-08-31 01:24:53 PDT
Confirming using Build ID: 2003083002, OS X 10.2.6
Comment 8 Derek Schrock 2003-11-17 00:46:16 PST
*** Bug 225939 has been marked as a duplicate of this bug. ***
Comment 9 Jasper 2004-03-20 09:55:46 PST
*** Bug 207210 has been marked as a duplicate of this bug. ***
Comment 10 Rob Hazlewood 2004-09-27 04:35:08 PDT
confirmed on OS X 10.3.5 firefox 1.0 preview release
It also happens on gentoo linux, although I haven't tried with 1.0PR
Comment 11 Rob Hazlewood 2004-11-28 00:09:58 PST
I have found an easier way to replicate this bug.
1. Add a bookmark with url about:blank to the bookmark bar
2. Go to www.google.com - the text field should be flashing
3. Command-click (mac-click) the about:blank bookmark entry
a new tab will appear with a flashing cursor.

A similar problem occurred a few days ago when i could actually enter data into
a text field in a non-active tab. This is a critical security issue as I am sure
it could be exploited for some web page to grab form data from another tab.

Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.7.5) Gecko/20041107
Firefox/1.0
Comment 12 Hiro 2004-11-29 22:01:39 PST
*** Bug 265616 has been marked as a duplicate of this bug. ***
Comment 13 Jasper 2004-11-30 01:19:38 PST
*** Bug 265844 has been marked as a duplicate of this bug. ***
Comment 14 Jolita Seckute 2004-12-04 12:58:42 PST
Confirming on Camino 2004120308 (v0.8+), Mac OS X 10.3.6.

As a blinking cursor remained when I went to a new page, it seemed to multiply
in a vertical line when I scrolled up and down. Those extra cursor images
disappeared after a couple of seconds, so there seems to be a lag time in their
redrawing.
Comment 15 Jasper 2005-01-11 08:34:03 PST
*** Bug 276873 has been marked as a duplicate of this bug. ***
Comment 16 Tim 2005-02-01 06:02:28 PST
(In reply to comment #14)
> Confirming on Camino 2004120308 (v0.8+), Mac OS X 10.3.6.
> 
> As a blinking cursor remained when I went to a new page, it seemed to multiply
> in a vertical line when I scrolled up and down. Those extra cursor images
> disappeared after a couple of seconds, so there seems to be a lag time in their
> redrawing.

Yes, that happens frequently with me as well.
Comment 17 Samuel Sidler (old account; do not CC) 2005-04-03 04:53:47 PDT
*** Bug 181983 has been marked as a duplicate of this bug. ***
Comment 18 Simon Fraser 2005-06-27 10:36:53 PDT
One case of this issue was fixed in Camino by the checkin for bug 182783 (an
activate/focus state issue on window activate/deactivate).
Comment 19 Uri Bernstein (Google) 2005-10-27 07:29:54 PDT
This problem disappeared on the Suite sometime between 1.8a2 (2004-07-14) and 1.8a3 (2004-08-17).
It then re-appeared, on both the Suite and Firefox, between 2005-08-23 and 2005-08-24. At that time, when I noticed the regression, I filed bug 306173 for it (I wasn't aware of this bug report).

As far I could tell, this bug never happened on Firefox prior to 2005-08-24.
Note that comment #11 is reporting a different bug, with different steps to reproduce, and a different result (a blinking caret). That bug does exist in Firefox builds prior to 2005-08-24 (including Firefox 1.0).

I also couldn't reproduce the problem (using the steps from the original report) on Camino, so I'm assuming the Camino problems are also something else altogether.

Since bug 306173 already has a proposed patch and some discussion, I suggest marking this bug WORKSFORME (as it would have been if anybody would have tested it between August 2004 and August 2005), so that the work on fixing it could be done in 306173.
Comment 20 Stewart Gordon 2005-10-27 07:41:02 PDT
This isn't WORKSFORME.  It's a dupe.  Marking it WORKSFORME would imply that someone is supposed to reopen it if the problem transpires again.

*** This bug has been marked as a duplicate of 306173 ***

Note You need to log in before you can comment on or make changes to this bug.