[GNC] Saving window position (and size)

Wed Jul 13 14:30:26 EDT 2022

I did a little research on this because GnuCash 4.8 in Lubuntu 22.04 with
Lxqt desktop in VirtualBox doesn't have that window placement problem.   It
seems that one of the barriers to greater overall acceptance of Wayland is
the fact that some older applications have code that is incompatible with
some aspects of Wayland. Thus there is a motivation to review GnuCash code
for compatibility sometime in the future.

Unfortunately,  I don't have appropriate skills to help,  so I can only ask
politely for developers to consider making it a future goal.

On Wed, Jul 13, 2022, 9:35 AM Tracy <gnulist at vbot.org> wrote:

> Good morning John,
>
> I get what you are saying, and I agree that tracking down the exact problem
> would be a huge undertaking. However, in this case, I can't just point the
> finger at GnuCash - there are other apps that were doing the same thing
> (one example being LibreOffice Calc). However, there were other apps
> (notably Pidgin, Firefox, and Thunderbird) that were working correctly.
> Which, of course, further supports your theory that it is a buried problem
> that would take an inordinate amount of time to dig out, between the
> various levels of interactions. Which is why I said I doubted that I would
> get any traction on a bug report - I just don't have the time (or the
> skills) to dig through all the layers to find the underlying problem.
>
> But as it happens, I am fine with using X11 (I prefer it, I was just giving
> Wayland a trial since they made it default), so I also don't really have
> the motivation to dig either.
>
>
> On Tue, Jul 12, 2022, 23:46 john <jralls at ceridwen.us> wrote:
>
> > Setting aside the tone of Adrien's comments, I think that filing a bug
> > report against the Wayland project would be seriously misplaced. They're
> > just publishing a spec and a model implementaiton.  The actual
> > implementation for Gnome is https://gitlab.gnome.org/GNOME/mutter. KDE
> > has a different compositor called Plasma. Compositors take bitmaps for
> > different logical areas of the screen and blend them together to
> determine
> > what color dot goes at what coordinate, then passes the result to the
> > hardware. Mutter handles both Wayland and X11 protocols for sending the
> > bitmaps and it's possible but unlikely that it's the source of the
> problem.
> >
> > It's far more likely that the problem is in Gtk's gdk-wayland backend.
> Gdk
> > is the layer that converts the Window abstraction into drawing
> instructions
> > that it passes to cairo to execute; it's what decides that an x-by-y
> window
> > will be represented by what set of colored dots at some position on the
> > screen. So in theory that would be the right place to aim your bug
> report.
> > In theory.
> >
> > In practice, if you file a bug with Gtk saying "GnuCash remembers where
> to
> > put my window on X11 but not on Wayland" they won't know what to do with
> > it, so unless you're able to reduce it to a simple example they're not
> > going to be able to figure out what's wrong with their code--if anything
> is.
> >
> > There's another problem, and it's that GnuCash's Gtk code is still warmed
> > over Gtk2 code with the absolute minimum done to keep it working in Gtk3.
> > There's tons of deprecated code that we haven't made time to update,
> partly
> > for lack of time and partly because we support older distributions. It's
> > entirely possible, likely even, that the reason that the window doesn't
> get
> > put in the right place on Wayland is that GnuCash is calling a deprecated
> > function that's implemented for X11 but not for Wayland. In a perfect
> world
> > you should file a bug with GnuCash and let us devs figure out where the
> > problem is and whether to push it upstream. Unfortunately there are at
> > present only two core devs who know GUI coding enough to do that, Bob
> > Fewell and me. I can't really speak for Bob but I have far too big a pile
> > of higher priority work to put in the several hours of debugging that
> > problem. It won't do any harm to file that bug but I can't offer much
> hope
> > that it will get acted upon.
> >
> > Regards,
> > John Ralls
> >
> >
> > > On Jul 12, 2022, at 7:25 PM, Adrien Monteleone <
> > adrien.monteleone at lusfiber.net> wrote:
> > >
> > > RATS. I hate it when I proof a post but don't stop to think before
> > hitting 'send.'
> > >
> > > My apologies to the GnuCash team, the moderator, and any readers of
> this
> > thread.
> > >
> > > While I stand behind my comments, this is not the forum for such a
> > discussion. My sentiments aren't about 'using GnuCash' and don't advance
> > the original question posted. (though I do think a Wayland-related bug
> > report is in order, if it doesn't exist.)
> > >
> > > I'll strive to pause and reflect more before posting replies rather
> than
> > just proofreading.
> > >
> > > Tracy, I'm glad you found the cause and have a work around. Thanks to
> > Martin for providing a path for investigation, and thank you again Tracy
> > for reporting back your findings. These posts are what this User list is
> > all about.
> > >
> > > Regards,
> > > Adrien
> > >
> > > On 7/12/22 9:14 PM, Adrien Monteleone wrote:
> > >> For the simple reason they'll be annoyed and try to ignore the bug
> > report, I'd file it. You never know who else has the same problem with
> > various apps. Heck, I'd bet the bug is already filed, maybe even closed
> as
> > 'won't fix'. That's even better. File another and make them spend their
> > time marking it as a duplicate and closing it. (when their time could be
> > better spent by FIXING *their* bug.)
> > >> Normally, I'd never suggest such action, but these folks bring it on
> > themselves with their dismissive and haughty attitudes, and you're doing
> > nothing wrong reporting or reviving an obvious bug that is indeed, their
> > fault. (why introduce Wayland as 'ready' when it isn't? If it works, why
> > doesn't it work?)
> > >> I'm sure I'll get admonished for the above, but really the
> admonishment
> > belongs on sour attitudes. Such are not conducive to the Open Source
> model,
> > community, and users providing much needed feedback.
> > >> ** Special Note **
> > >> GnuCash devs are *not* like this. They'll (at their discretion of
> > course) spend an inordinate amount of their personal, volunteer time
> > helping users fix something, even if they've helped 10k people before you
> > apply the same fix. These folks are golden and have more patience than
> the
> > rest of the known universe. I have the utmost respect and appreciation
> for
> > the GnuCash dev team.
> > >
> > >> Please remember to CC this list on all your replies.
> > >> You can do this by using Reply-To-List or Reply-All.
> > >
> > >
> > > _______________________________________________
> > > gnucash-user mailing list
> > > gnucash-user at gnucash.org
> > > To update your subscription preferences or to unsubscribe:
> > > https://lists.gnucash.org/mailman/listinfo/gnucash-user
> > > If you are using Nabble or Gmane, please see
> > https://wiki.gnucash.org/wiki/Mailing_Lists for more information.
> > > -----
> > > Please remember to CC this list on all your replies.
> > > You can do this by using Reply-To-List or Reply-All.
> >
> > _______________________________________________
> > gnucash-user mailing list
> > gnucash-user at gnucash.org
> > To update your subscription preferences or to unsubscribe:
> > https://lists.gnucash.org/mailman/listinfo/gnucash-user
> > If you are using Nabble or Gmane, please see
> > https://wiki.gnucash.org/wiki/Mailing_Lists for more information.
> > -----
> > Please remember to CC this list on all your replies.
> > You can do this by using Reply-To-List or Reply-All.
> >
> _______________________________________________
> gnucash-user mailing list
> gnucash-user at gnucash.org
> To update your subscription preferences or to unsubscribe:
> https://lists.gnucash.org/mailman/listinfo/gnucash-user
> If you are using Nabble or Gmane, please see
> https://wiki.gnucash.org/wiki/Mailing_Lists for more information.
> -----
> Please remember to CC this list on all your replies.
> You can do this by using Reply-To-List or Reply-All.
>