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

[Richard Shann]

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. »
>