New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Firefox does not gain focus when started from Windows 7 with latest snapshots #4296
Comments
Attachment firefox no focus.log added by briang1 on 2014-07-15 21:53 |
Comment 1 by briang1 on 2014-07-15 22:46 There seems to be no ill effects from doing this. |
Comment 2 by jteh on 2014-07-16 01:05 |
Comment 3 by briang1 on 2014-07-16 07:12 |
Attachment firefox no focus master.log added by briang1 on 2014-07-16 07:15 |
Comment 4 by briang1 on 2014-07-17 22:36 IO - speech.speak (23:13:56): IO - speech.speak (23:14:05): So far this does not seem to cause Firefox a problem, but I've probably not tested it enough. |
Comment 5 by msuch (in reply to comment description) on 2015-01-22 08:32
|
Comment 6 by jteh (in reply to comment 5) on 2015-01-22 21:31
For you, does it occur only in next or also in master/release? I can understand it occurring in next, but not the others. |
Comment 7 by msuch (in reply to comment 6) on 2015-01-22 21:57
Yes, with latest next versions, it occurs on all my machines. With all previous version for a long time, it only uccured on my slow netbook. |
Comment 8 by msuch (in reply to comment 7) on 2015-01-23 09:42
|
Comment 9 by briang1 on 2015-01-23 11:10 INFO - main (10:54:02): DEBUGWARNING - core.Notify (10:54:16): DEBUGWARNING - core.Notify (10:54:17): DEBUGWARNING - core.Notify (10:54:18): DEBUGWARNING - IAccessibleHandler.accessibleObjectFromEvent (10:54:19): IO - speech.speak (10:54:44): I don't quite see what is happening but it looks like it never detects the fact that it needs to create a buffer until I press a key and make Firefox magically appear to nvda, which otherwise is confused. |
Comment 10 by msuch (in reply to comment 9) on 2015-01-23 13:39 Replying to briang1:
|
Comment 11 by briang1 on 2015-01-23 22:50 As for the other issue mentioned, I don't see that very much here. |
@Brian1Gaff, @MichelSuch, can you still reproduce this on non-XP systems? |
Yes intermittently I can. With auto say all on, quite often the screen will
not read on its own. One has to open a menu item then close it again and
then do a say all. Even cursoring seems to be as if the page is empty.
Windows 7 and a four core Intel and 8 gig of memory.
It is more common on a first open of the browser and there is some core
frozen in stack in the log but whether this is a red herring I'm not so
sore.
Its been an issue for some years.
I think it could be Firefox, but how can one prove it.
Any version from 48 to the 54 I am currently using seem to be the same.
Brian
bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
|
Not on xp demo.
I launched Firefox via the File Hippo update checker applet. This displayed
the page but the page was incomplete, ie the files needing to be updated
were not displayed. I did an f5 to refresh the page. This happened.
IO - inputCore.InputManager.executeGesture (09:19:44.298):
Input: kb(desktop):f5
DEBUGWARNING - NVDAObjects.IAccessible.IAccessible._get_IA2Attributes
(09:19:45.154):
IAccessibleObject.attributes COMError (-2147220995, 'Object is not connected
to server', (None, None, None, 0, None))
DEBUG - treeInterceptorHandler.killTreeInterceptor (09:19:45.155):
Killed treeInterceptor: <virtualBuffers.gecko_ia2.Gecko_ia2 object at
0x05A4C970>
DEBUG - treeInterceptorHandler.update (09:19:45.164):
Adding new treeInterceptor to runningTable:
<virtualBuffers.gecko_ia2.Gecko_ia2 object at 0x05B345D0>
DEBUGWARNING - RPC process 2064 (firefox.exe) (09:19:45.236):
Thread 4904, build\x86\vbufBase\storage.cpp,
VBufStorage_buffer_t::findNodeByAttributes, 897:
offset 661 is past end of buffer, returning NULL
DEBUGWARNING - RPC process 2064 (firefox.exe) (09:19:45.237):
Thread 4904, build\x86\vbufBase\storage.cpp,
VBufStorage_buffer_t::locateTextFieldNodeAtOffset, 788:
Offset 661 out of range. Returnning NULL
DEBUGWARNING - RPC process 2064 (firefox.exe) (09:19:45.239):
Thread 4904, build\x86\vbufBase\storage.cpp,
VBufStorage_buffer_t::locateControlFieldNodeAtOffset, 808:
Could not locate node at offset, returning NULL
DEBUGWARNING - IAccessibleHandler.accessibleObjectFromEvent (09:19:45.239):
oleacc.AccessibleObjectFromEvent with window 0, objectID -4 and childID 0:
[Error -2147467259] Unspecified error
DEBUGWARNING - browseMode.BrowseModeDocumentTreeInterceptor._set_selection
(09:19:45.239):
Invalid NVDAObjectAtStart
IO - speech.speak (09:19:45.239):
Speaking [LangChangeCommand ('en_GB'), u'FileHippo.com - Download Free
Software busy']
DEBUG - queueHandler.registerGeneratorObject (09:19:45.240):
Adding generator 4
IO - speech.speak (09:19:45.240):
Speaking [IndexCommand(1)]
DEBUGWARNING - RPC process 2064 (firefox.exe) (09:19:45.503):
Thread 4904, build\x86\vbufBase\storage.cpp,
VBufStorage_buffer_t::locateTextFieldNodeAtOffset, 788:
Offset 661 out of range. Returnning NULL
DEBUGWARNING - RPC process 2064 (firefox.exe) (09:19:45.506):
Thread 4904, build\x86\vbufBase\storage.cpp,
VBufStorage_buffer_t::locateControlFieldNodeAtOffset, 808:
Could not locate node at offset, returning NULL
DEBUGWARNING - IAccessibleHandler.accessibleObjectFromEvent (09:19:45.506):
oleacc.AccessibleObjectFromEvent with window 0, objectID -4 and childID 0:
[Error -2147467259] Unspecified error
DEBUGWARNING - RPC process 2064 (firefox.exe) (09:19:45.539):
Thread 3776, build\x86\vbufBase\storage.cpp,
VBufStorage_buffer_t::locateTextFieldNodeAtOffset, 788:
Offset 661 out of range. Returnning NULL
DEBUGWARNING - RPC process 2064 (firefox.exe) (09:19:45.540):
Thread 3776, build\x86\vbufBase\storage.cpp,
VBufStorage_buffer_t::locateControlFieldNodeAtOffset, 808:
Could not locate node at offset, returning NULL
DEBUG - NVDAObjects.IAccessible.IAccessible._get_role (09:19:45.822):
IARole: span
DEBUG - queueHandler.pumpAll (09:19:47.842):
generator 4 finished
IO - inputCore.InputManager.executeGesture (09:20:20.441):
Input: kb(desktop):NVDA+control+f1
IO - speech.speak (09:20:20.450):
Speaking [LangChangeCommand ('en_GB'), u'Currently running application is
firefox.exe and currently loaded module is firefox']
IO - inputCore.InputManager.executeGesture (09:20:30.273):
Input: kb(desktop):NVDA+f1
INFO - globalCommands.GlobalCommands.script_navigatorObject_devInfo
(09:20:30.345):
Developer info for navigator object:
name: None
role: ROLE_UNKNOWN
states:
isFocusable: False
hasFocus: False
Python object: <NVDAObjects.IAccessible.IAccessible object at 0x05A4C690>
Python class mro: (<class 'NVDAObjects.IAccessible.IAccessible'>, <class
'NVDAObjects.window.Window'>, <class 'NVDAObjects.NVDAObject'>, <class
'baseObject.ScriptableObject'>, <class 'baseObject.AutoPropertyObject'>,
<type 'object'>)
description: None
location: None
value: None
appModule: <'firefox' (appName u'firefox', process ID 2064) at address
5a36150>
appModule.productName: u'Firefox'
appModule.productVersion: u'55.0'
TextInfo: <class 'NVDAObjects.NVDAObjectTextInfo'>
windowHandle: 328586
windowClassName: u'MozillaWindowClass'
windowControlID: 0
windowStyle: 399441920
windowThreadID: 3776
windowText: u'FileHippo.com - Download Free Software - Mozilla Firefox'
displayText: exception: 'NoneType' object is not iterable
IAccessibleObject: <POINTER(IAccessible) ptr=0x83e33e0 at 5b39bc0>
IAccessibleChildID: 0
IAccessible event parameters: windowHandle=328586, objectID=-4, childID=-318
IAccessible accName: exception: (-2147417842, 'The application called an
interface that was marshalled for a different thread.', (None, None, None,
0, None))
IAccessible accRole: exception: (-2147417842, 'The application called an
interface that was marshalled for a different thread.', (None, None, None,
0, None))
IAccessible accState: exception: (-2147417842, 'The application called an
interface that was marshalled for a different thread.', (None, None, None,
0, None))
IAccessible accDescription: exception: (-2147417842, 'The application called
an interface that was marshalled for a different thread.', (None, None,
None, 0, None))
IAccessible accValue: exception: (-2147417842, 'The application called an
interface that was marshalled for a different thread.', (None, None, None,
0, None))
When I woke it up by getting the above log and then went back. the page
displayed OK but the actual focus had moved near to the bottom of the page.
Bringing it to the top and using list I found the info displayed.
This sort of part of a page missing is common in firefox intermittently.
You either get the info missing altogether or a secction in the page saying
server unreachable or unavailable. A refresh normally fixes it but often
nvda needs to be sorted by either pulling down a menu or switching between
windows open to get it to read.
Does this help?
Windows 7 pro64
Any old recent Firefox version up to the latest of course I'm using a 32 bit
version, by the way. Not tested before version 48 though.
Latest master snapshot.
Though this was launched via an application, I've seen it happen on links to
other sites from a working system. I did wonder if it was the ad blocker,
but turning it off seemed to make no real difference on the number of times
this sort of thing occured.
My feeling is that firefox is missing bits of a page and whether that is
due to nvda delaying something or is just how firefox is, or even if its my
internet connection, I have no way of knowing. However its far less likely
to occur in IE.
I have had this issue for a fair time now, and it is of course worse in XP
presumably due to the slower processors in those machines.
bglists@blueyonder.co.uk
Sent via blueyonder.
Please address personal email to:-
briang1@blueyonder.co.uk, putting 'Brian Gaff'
in the display name field.
|
@Brian1Gaff: How about the most recent next snapshots that include #7535? |
I suggest closing this in favor of #8389 and referenced issues because they are reproducible on Windows 7 and higher. |
@LeonarddeR, @Brian1Gaff what do you think about my Suggestion above? |
1 similar comment
@LeonarddeR, @Brian1Gaff what do you think about my Suggestion above? |
I am closing this in favor of #8389 because this issue is very general and occurs in several situations regardless of the application. The discussion in that issue goes further and will continue there. |
Reported by briang1 on 2014-07-15 21:51
Firstly, I see another ticket with similar problems, but to me at least this is different enough to need a new ticket.
How to get the effect.
Boot up xp and launch firefox 31 from the desktop shortcut.
Most of the times Firefox never puts in any apparence. However it is there, as can be found by just pressing alt, or space or a cursor key.
Nvda seems to think its still on the desktop.
I am going to attach a log I generated with comments earlier.
Note that this effect is not in 2014.2, only in master and next snaps for some versions now.
Sometimes it will work, but not all that often. There appears to be some kind of looping going on somewhere.
The text was updated successfully, but these errors were encountered: