Re: [orca-list] Please test master in case I broke the universe

[Jérémy Prego <jeremy pregonetwork net>]

hello,

I confirm that this bug has been present in orca for several months.

the bug occurs when you get from an application to a terminal, but it
doesn't happen systematically, that's why I never reported it.

to fix the bug and for orca to realize that it is now in a terminal, I
just need to press f10 twice to open and close the menu. it should
update something and orca goes back to reading the terminal, as before.

if my description can help.

Jerem
Le 14/06/2021 à 00:12, Joanmarie Diggs a écrit :
> Seems that I broke the universe earlier than I thought. *smile* 
>
> Will take a look tomorrow at this issue because it looks like a case
> where fixing something uncovered a bug elsewhere in Orca.
>
> Thanks again!!
> --joanie
>
> On Mon, 2021-06-14 at 01:03 +0300, Alexander Epaneshnikov wrote:
> > 14.06.2021 00:46, Alexander Epaneshnikov пишет:
> > > 14.06.2021 00:43, Joanmarie Diggs пишет:
> > > > Hey Kyle and Alexander.
> > > >
> > > > In both of your debug.out files, I'm not seeing any messages
> > > > about
> > > > event floods. That makes me think that the problem is either a
> > > > regression in the caret-moved web-related events changes I made
> > > > earlier, or it's not actually a regression but an existing bug in
> > > > Orca
> > > > that thorough testing has revealed. To figure out if something
> > > > regressed, it would be helpful to know which commit introduced
> > > > the
> > > > regression. Is this something you can do?
> > > >
> > > > Thanks again very much for testing and for your prompt replies!
> > > > --joanie
> > > >
> > > > On Mon, 2021-06-14 at 00:24 +0300, Alexander Epaneshnikov wrote:
> > > > > 14.06.2021 00:10, Joanmarie Diggs пишет:
> > > > > > Hey Kyle. Thanks for testing!! If you can get some reliable
> > > > > > steps
> > > > > > to
> > > > > > reproduce the problem along with a full debug.out, I'll take
> > > > > > a
> > > > > > look.
> > > > > > The goal is for Orca to handle event floods but continue to
> > > > > > process
> > > > > > valid events correctly. Having to restart Orca is never good.
> > > > > > Some
> > > > > > special handling for terminals may be in order.
> > > > > >
> > > > > > Thanks again!
> > > > > > --joanie
> > > > > hello Joanie. for me an every time reproducer is just switching
> > > > > to
> > > > > Firefox
> > > > > window. I attaching a debug.out.
> > > > >
> > > > > I switched to terminal
> > > > > executed: echo "everthing is ok"
> > > > > switched to firefox
> > > > > moved in page with arrows
> > > > > switched back to the terminal
> > > > > tried to type something. orca stays silent, braille not
> > > > > updating.
> > > > > created a new terminal tab
> > > > > executed: echo "ok now all is good again" orca works and
> > > > > braille
> > > > > updates.
> > > > >
> > > > > then I replaced orca.
> > > > >
> > > > > I hope it helps.
> > > ok will try to bisect. give me a half an hour.
> > ok. d261ce21e is the first bad commit. and judging by the
> > description, this
> > is highly expected.
> > correctly.
>
> _______________________________________________
> orca-list mailing list
> orca-list gnome org
> https://mail.gnome.org/mailman/listinfo/orca-list
> Orca wiki: https://wiki.gnome.org/Projects/Orca
> Orca documentation: https://help.gnome.org/users/orca/stable/
> GNOME Universal Access guide: https://help.gnome.org/users/gnome-help/stable/a11y.html