GNOME Bugzilla – Bug 466955
[blocked] Add ARIA slider/progressbar support
Last modified: 2007-10-31 14:20:32 UTC
Add support for sliders and progressbars now that the Mozilla guys have IValue support implemented see https://bugzilla.mozilla.org/show_bug.cgi?id=389851 for these accessibles. Test cases: sliders and progressbars linked from Mozilla http://developer.mozilla.org/en/docs/Accessible_DHTML#Supported_roles Dojo sliders: http://archive.dojotoolkit.org/nightly/dojotoolkit/dijit/tests/form/test_Slider.html Dojo progressbars: http://archive.dojotoolkit.org/nightly/dojotoolkit/dijit/tests/test_ProgressBar.html
Progressbar and slider survey: Dojo Progressbars: not functional. Always getting IValue.currentValue = 0. Dojo bug filed here http://trac.dojotoolkit.org/ticket/4185 Sliders: seem to be good. Mozilla examples: Progressbar: seem to be good. Sliders: seem to be good. Simple slider returns half the value that is shown on slider but this is a markup issue and not an issue with AT-SPI or Orca. UIUC examples Progressbar: no example. sliders: fail miserably. I am occasionally getting the following AT-SPI exception on object:property-change:accessible-parent events: Traceback (most recent call last):
+ Trace 155671
debug.printException(debug.LEVEL_WARNING)
Will, I am not sure why I would get such an exception. Can you provide any insight into this? These are the last two examples on the Mozilla link under 'sliders'
> UIUC examples > Progressbar: no example. > sliders: fail miserably. I am occasionally getting the following AT-SPI > exception on object:property-change:accessible-parent events: > > Traceback (most recent call last): > File "/usr/local/lib/python2.4/site-packages/orca/atspi.py", line 278, in > notifyEvent > debug.printException(debug.LEVEL_WARNING) > RuntimeError: maximum recursion depth exceeded > > Will, I am not sure why I would get such an exception. Can you provide any > insight into this? These are the last two examples on the Mozilla link under > 'sliders' Odd. Which branch are you developing against (trunk, pyatspi, gnome-2-20), and can you provide more of the stack trace?
I am developing against trunk. Sorry, that is it on the stack trace. I am sure you can reproduce the problem using http://test.cita.uiuc.edu/aria/slider/html.php?title=Slider%20Example%201&ginc=includes/slider1.inc&gcss=css/slider1.css&gjs=../js/globals.js,../js/enable.js,js/slider1.js Sometimes it takes several seconds to get the error as it recurses. Just tab to one of the sliders and use the arrow keys. You will eventually see the exception.
In using the firefox I obtained just a couple minutes ago (Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9a8pre) Gecko/2007082004 Minefield/3.0a8pre) and Orca from trunk, I don't see the stack trace. In addition, the sliders from http://test.cita.uiuc.edu/aria/slider/html.php?title=Slider%20Example%201&ginc=includes/slider1.inc&gcss=css/slider1.css&gjs=../js/globals.js,../js/enable.js,js/slider1.js seem to present nicely. Firefox seems to have some issues today, though, in that the visual indicators for page tabs are not being painted. Instead, I just get a big gray horizontal bar where the page tabs should be. I just have to imagine where the tab might be and then click.
Dojo progressbars are still reporting 0.0 for currentValue. A bug has being issued but that progressbar is reporting 0.0 too :(
Note to self - Don't forget enableProgressBarUpdates and progressBarUpdateInterval in user_settings.py.
Dojo progressbars are looking good. Marking bug as fixed.