Hi Florian,
> I tried adw1-demo.scm with the attached patch to replace
> (current-filename), which would return #f so (dirname) throws.
> It fails because it cannot find namespaces. Guix uses libadwaita
> 1.4.0. Is it too old?
Yes, as announced in the NEWS, the release notes, and on the g-golf
Hi David. It is interesting that pygobject has more C.
I tried adw1-demo.scm with the attached patch to replace
(current-filename), which would return #f so (dirname) throws.
It fails because it cannot find namespaces. Guix uses libadwaita 1.4.0.
Is it too old?
florian@florianhp ~/src/g-golf/e
Hi Florian,
> Thank you for the python code. This works unlike g-golf in Guix.
> ...
Ok, expected, but I wanted to double check - Not an excuse ofc,
but pygobjec is mostly written in C, including their callback and
gclosure marshal(ers), so when they need to either replace or access
such VFunc p
Thank you for the python code. This works unlike g-golf in Guix.
guix shell python python-pygobject gtk --no-grafts -- python3
/tmp/drawing-widget.py
and
guix shell python python-pygobject gtk -- python3 /tmp/drawing-widget.py
both show a window divided in four nicely-colored rectangles.
I a
Hi Florian,
> ...
> pygobject GTK3 examples are working. I will experiment and try to
> write Vala/C code that breaks now.
I thought they had Gtk-4 examples as well. But anyway, I cooked a
pygobject version of this drawing-widget example, attached.
Could you:
install pygobject
Hello Basile, thank you for the tips, but I have since managed to get a
good backtrace. -O is what I used by compiling GTK with configure flag
-Dc_args=-O0. -g did not work with g-golf (this is Guix’
with-debug-info transformation), but stripped binaries with debug info
in a separate output and s
On 5/4/24 11:39 PM, David Pirotte wrote:
Hi Florian,
Hi David, thank you for your continuing responses.
No problem. I am also interested in fixing this bug
I will try to get a backtrace with GTK debug symbols by patching
Guix to add a debug output to GTK, in the hope that I patch the right
David Pirotte writes:
>> (gc-disable) was a smart idea, sadly it does not help, but at least
>> we know now. ...
>
> I am not entirely convinced yet - in one way or another, somewhere,,
> somehow, the klass->snapshot is 'corrupted', either the pointer, or the
> mem it points to ... why, when, wher
Hi Florian,
> (gc-disable) was a smart idea, sadly it does not help, but at least
> we know now. ...
I am not entirely convinced yet - in one way or another, somewhere,,
somehow, the klass->snapshot is 'corrupted', either the pointer, or the
mem it points to ... why, when, where ... ?
> scheme@(
Hi David. (gc-disable) was a smart idea, sadly it does not help, but at
least we know now. It makes Valgrind errors disappear after
(gc-disable) was called.
scheme@(guile-user)> ,use (g-golf)
scheme@(guile-user)> (gc-disable)
scheme@(guile-user)> (chdir "/home/florian/src/g-golf/examples/gtk-4")
Hi Florian,
> For drawing-widget.scm on x86_64, klass->snapshot looks high like a
> code segment address but has a bogus disassembly. Was it clobbered?
can you try the following
scheme@(guile-user)> ,use (g-golf)
scheme@(guile-user)> (gc-disable)
scheme@(guile-user)> (ch
Hi Florian,
> > Hi David, thank you for your continuing responses. I will try to
> > get a backtrace with GTK debug symbols
> I now managed to build gtk with a debug output and meson configure
> flag "-Dc_args=-O0".
Perfect, thanks!
Let me look at it ... i'll get back asap ...
David
pgpn9Q
Hi Florian,
> Hi David, thank you for your continuing responses.
No problem. I am also interested in fixing this bug
> I will try to get a backtrace with GTK debug symbols by patching
> Guix to add a debug output to GTK, in the hope that I patch the right
> GTK.
It is definitely not a GLib/GObj
"pelzflorian (Florian Pelz)" writes:
> Hi David, thank you for your continuing responses. I will try to get a
> backtrace with GTK debug symbols
I now managed to build gtk with a debug output and meson configure flag
"-Dc_args=-O0".
For drawing-widget.scm on x86_64, klass->snapshot looks high l
Hi David, thank you for your continuing responses. I will try to get a
backtrace with GTK debug symbols by patching Guix to add a debug output
to GTK, in the hope that I patch the right GTK.
However, my believe is that the error is in G-Golf
closures/callbacks/vfuncs (even though the hl-api tests
Hi Florian,
> > The true error location was given by the gdb backtrace, the
> > gtk-widget-snapshot-child call, presumably with a bad pointer or so
> > ...
> Hmm my assumption was that all callbacks are broken, not vfuncs
> specifically ...
Very unlikely, but the important thing is to get a pr
David Pirotte writes:
>> Hi David. I’m trying to find the true error location.
>
> The true error location was given by the gdb backtrace, the
> gtk-widget-snapshot-child call, presumably with a bad pointer or so ...
Hmm my assumption was that all callbacks are broken, not vfuncs
specifically, w
Hi Florian,
> > Could you tell me if, when you run such commands in guile in your
> > Debian setup, you get similar output? First with tracepoint:
> With all due respect, this will get you nowhere, please upload the
> full gdb backtrace ... thanks
I patched the drawing-widget example so use
Hi Florian,
> ...
> examples/gtk-4/simple-paintable.scm fails at (add-to-load-path
> (dirname (current-filename)). I have not yet spent time to debug
> this one further.
So we can make progress, till you find why (current-filename) fails -
note that it is a syntax, not a procedure, and the doc c
Hi Florian,
> Hi David. I’m trying to find the true error location.
The true error location was given by the gdb backtrace, the
gtk-widget-snapshot-child call, presumably with a bad pointer or so ...
Could you please upload a full gdb backtrace somewhere tor-friendly
(and zero javascript), so i
Hi David. I’m trying to find the true error location. Could you tell
me if, when you run such commands in guile in your Debian setup, you
get similar output? First with tracepoint:
scheme@(guile-user)> (load
"/home/florian/src/g-golf/examples/gtk-4/drawing-widget.scm")
Hello David,
David Pirotte writes:
> fwiw, g-golf doesn't load/import/invoke/call nor the GdkPixbuf
> typelib, nor any of its functions/methods 'on its own', nor
> does any example, this occurs as part of the Gtk/Gdk/Gsk
> 'engine' - it is not g-golf that 'runs g_typelib_symbol on
> GdkPixbuf fun
Hello Florian,
> To develop with G-golf and GTK as in G-Golf’
> examples/adw-1/hello-world.scm or examples/gtk-4/hello-world.scm, the
> Guix commands must (at the time of writing) be used with the
> “--no-grafts” option, because G-Golf first when loading runs
> g_typelib_symbol on GdkPixbuf functi
Hello again, one more message on the current status of G-Golf on
GNU Guix, which has two real issues:
To develop with G-golf and GTK as in G-Golf’
examples/adw-1/hello-world.scm or examples/gtk-4/hello-world.scm, the
Guix commands must (at the time of writing) be used with the
“--no-grafts” option
Hello David,
David Pirotte writes:
> Hello Florian,
>
> Before i get to the other points of your email, let me exceptionally
> raise the last point first:
>
> > https://ci.guix.gnu.org/build/3767854/details
> This refers to g-golf-0.8.0-a.1
>
> Can you re-ensure me and confirm y
Hello Florian,
Before i get to the other points of your email, let me exceptionally
raise the last point first:
> https://ci.guix.gnu.org/build/3767854/details
This refers to g-golf-0.8.0-a.1
Can you re-ensure me and confirm you are using the latest
version, 0.8.0
Hello again. This is a bug in Guix territory; not your g-golf. I have
successfully run gtk-4/hello-world.scm, fully responsive, apparently by
running “guix shell --no-grafts” with with-debug-info transformations,
instead of “guix shell” as before, and then broke it again, when I
changed guix.scm
Thank you for the work, David. I have a question; how do you debug when
Guile crashes to terminal printing “Segmentation fault”`?
The long story so far:
I want to develop a GTK app with G-Golf. I try to run it with
GNU Guix [1] (not Debian like you do, as far as I remember), because now
the Gui
28 matches
Mail list logo
