Re: [Denemo-devel] Scheme error trap is broken

[Richard Shann]

On Wed, 2014-01-22 at 14:23 +0100, Éloi Rivard wrote:
> With your last commit and mine, do you steel see warnings ?

I was getting no warnings, but on running denemo -n I saw that this was
because the program was looping trying to set the enharmonic set; fixing
this it then went further and started warning about the button boxes
when changing movements, fixing that I got to this one ...

#6  0x00007ffff5e3642a in gtk_action_group_get_action ()
   from /usr/lib/x86_64-linux-gnu/libgtk-3.so.0
#7  0x000000000044086f in lookup_action_from_name (

here there is no gtk_action_group it would seem, so even looking up an
action is not possible. I really don't know if a non-interactive denemo
is a serious possibility.
I have pushed the fixes up until this last point ...

Richard


> 
> 
> 2014/1/21 Richard Shann <address@hidden>
>         On Tue, 2014-01-21 at 16:39 +0100, Éloi Rivard wrote:
>         > I would like to set up a non regression with your issue. Do
>         you think
>         > running denemo with the following arguments, and checks that
>         it is not
>         > killed, would be relevant ?
>         >
>         > denemo -n -a "(define foo \"bar\")"
>         
>         
>         Interesting, this segfaults on the current git denemo (it is
>         the -n that
>         causes the segfault, the -a stuff is not needed)
>         
>         
>         (gdb) run -n  -a "(define foo \"bar\")"
>         Starting program: /home/rshann/local/bin/denemo -n  -a
>         "(define foo
>         \"bar\")"
>         [Thread debugging using libthread_db enabled]
>         Using host libthread_db library
>         "/lib/x86_64-linux-gnu/libthread_db.so.1".
>         Denemo - MESSAGE : Loading preference
>         file: /home/rshann/.denemo-1.1.1/denemorc
>         Denemo - MESSAGE : Denemo version 1.1.1
>         Denemo - MESSAGE : Loaded
>         keymap /home/rshann/.denemo-1.1.1/actions/Default.commands
>         Success 1
>            Gtk - CRITICAL: gtk_widget_hide: assertion `GTK_IS_WIDGET
>         (widget)'
>         failed
>         
>         [...]
>         
>         Program received signal SIGSEGV, Segmentation fault.
>         event_queue_write_immediate (queue=0x0,
>             address@hidden "\024\360\177\177\b\b\003
>         \177\177",
>         '@' <repeats 12 times>, "\177", address@hidden)
>             at ../../denemo/src/eventqueue.c:132
>         132       if (!queue->immediate || jack_ringbuffer_write_space
>         (queue->immediate) < length)
>         
>         Those Gtk -CRITICAL messages are the ones I see when the
>         thumbnailer
>         runs on terminating an interactive session.
>         
>         Richard
>         
>         
>         
>         
>         >
>         >
>         >
>         > 2013/12/23 Richard Shann <address@hidden>
>         >         I have a revert of the change to the scheme trap
>         handler code
>         >         - is it ok
>         >         with you that I push this to git, as at the moment
>         Denemo
>         >         crashes for
>         >         some bad scheme uses?
>         >         (I have been pushing without this commit so far,
>         which means
>         >         that I am
>         >         not actually building and running quite what I push,
>         and now I
>         >         have
>         >         changes to view.c this becomes impossible, so it
>         would help me
>         >         continue
>         >         if I push this too: the make check will then fail on
>         the check
>         >         scheme
>         >         error test, unfortunately).
>         >
>         >         Richard
>         >
>         >
>         >         On Wed, 2013-12-18 at 11:08 +0000, Richard Shann
>         wrote:
>         >
>         >         > The scheme trapping of errors is broken
>         >         > If I put this in the CLI box
>         >         > (define abc "hello")
>         >         >
>         >         > it responds with "syntax error" in the console on
>         a version
>         >         from some
>         >         > time back and just kills Denemo instantly on the
>         current
>         >         version
>         >         > (current =  git checkout
>         >         a32f50f80539d05c5356d11d0bef5ee433dfe857)
>         >         >
>         >         > There must be two bugs here - the CLI box is
>         broken for
>         >         entering strings
>         >         > (I suspect) and the trapping is fatal to Denemo
>         instead of
>         >         returning.
>         >         >
>         >         > Richard
>         >         >
>         >         >
>         >         >
>         >         > _______________________________________________
>         >         > Denemo-devel mailing list
>         >         > address@hidden
>         >         >
>         https://lists.gnu.org/mailman/listinfo/denemo-devel
>         >
>         >
>         >
>         >         _______________________________________________
>         >         Denemo-devel mailing list
>         >         address@hidden
>         >         https://lists.gnu.org/mailman/listinfo/denemo-devel
>         >
>         >
>         >
>         >
>         > --
>         > Éloi Rivard - address@hidden
>         >
>         > « On perd plus à être indécis qu'à se tromper. »
>         >
>         
>         
>         
> 
> 
> 
> -- 
> Éloi Rivard - address@hidden
>         
> « On perd plus à être indécis qu'à se tromper. »
>