The behavior is reasonable if losing events is reasonable.

There should have been a previous event invalidating the entire window.

It might be reasonable to synthesize such an event when the window has been
idle for a while?

Thank,

-- 
Raul

On Friday, November 15, 2019, bill lam <[email protected]> wrote:

> I suspect the behavior is reasonable. refresh event is
> smart enough to update not the entire windows, but only
> portions that it thinks should be updated.
>
> Thu, 14 Nov 2019, Raul Miller написал(а):
> > I have experimented some more with this issue (it still occurs under
> > the current beta-p with the current ide/qt implementation, and it
> > happens reliably for me, on long running J instances).
> >
> > And an interesting defect is that if my cursor was partway down in the
> > jqt evaluation window, I can start typing, and  the window goes from
> > completely transparent to having normal J content from that point in
> > the window on down. However, the part of the screen *above* the cursor
> > remains "behind" other content.
> >
> > Specifically:  I can drag a terminal window over the jqt window (and
> > it covers the jqt window), and then when I click on an exposed part of
> > the jqt window, the part of the window above where my  cursor was when
> > I started typing into jqt remains *behind* the terminal window, while
> > the part of the window from where my cursor was on down to the bottom
> > becomes above that terminal window. If you'd like me to upload a
> > screenshot of this to some location, I can, but hopefully my
> > description here is adequate?
> >
> > (Is it relevant that the terminal window is set to 95% opacity when
> > it's not in focus?)
> >
> > Thanks,
> >
> > --
> > Raul
> >
> > On Wed, Nov 6, 2019 at 12:07 PM Raul Miller <[email protected]>
> wrote:
> > >
> > > This shouldn't matter, because the application was idle (sitting at
> > > the prompt waiting for commands).
> > >
> > > That said, I've also upgraded to j901 beta-p, so maybe that will have
> > > addressed something relevant to this issue.
> > >
> > > Thanks,
> > >
> > > --
> > > Raul
> > >
> > > On Tue, Nov 5, 2019 at 5:57 PM bill lam <[email protected]> wrote:
> > > >
> > > > jqt is a single threaded application, ie it has only the main thread
> (UI
> > > > thread). J engine also runs on the main thread so that it can block
> UI
> > > > events such as paint.
> > > >
> > > > You will also notice there is not session output if running a tight
> loop. A
> > > > workaround is to run wd'msgs' at some interval to let pending UI
> events has
> > > > a chance to be processed.
> > > >
> > > > On Wed, Nov 6, 2019, 2:35 AM Raul Miller <[email protected]>
> wrote:
> > > >
> > > > > Engine: j901/j64avx2/darwin
> > > > > Beta-m: commercial/2019-10-02T11:28:59
> > > > > Library: 9.01.11
> > > > > Qt IDE: 1.8.4/5.12.5
> > > > > Platform: Darwin 64
> > > > > Installer: J901 install
> > > > >
> > > > > When I come back to a long running JQt session the editor has...
> > > > > transparency.
> > > > >
> > > > > Some lines of text in the editor will be visible, but the rest
> shows
> > > > > content from other windows.
> > > > >
> > > > > I can scroll the window to make the content reappear (once I
> recognize
> > > > > what's going on - before that, I don't know if it's J that's
> messing
> > > > > up or something else), and -- since it's a long running process --
> I'm
> > > > > often not focussing on J at the time.
> > > > >
> > > > > Anyways, this smells like the Qt infrastructure having a timing
> issue
> > > > > where it can lose refresh events. But, as a workaround, I would
> like
> > > > > to propose that if there's more than 24 hours between paint events
> (or
> > > > > maybe something more aggressive, like an hour?) that J try to do a
> > > > > complete paint when that happens (assuming that's possible... I
> don't
> > > > > know enough about Qt internals and I *am* guessing, here.)
> > > > >
> > > > > Thanks,
> > > > >
> > > > > --
> > > > > Raul
> > > > > ------------------------------------------------------------
> ----------
> > > > > For information about J forums see http://www.jsoftware.com/
> forums.htm
> > > > >
> > > > ------------------------------------------------------------
> ----------
> > > > For information about J forums see http://www.jsoftware.com/
> forums.htm
> > ----------------------------------------------------------------------
> > For information about J forums see http://www.jsoftware.com/forums.htm
>
> --
> regards,
> ====================================================
> GPG key 1024D/4434BAB3 2008-08-24
> gpg --keyserver subkeys.pgp.net --armor --export 4434BAB3
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
>
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to