[Debian-science-sagemath] Doctests failing with "ECL says: THROW: The catch MACSYMA-QUIT is undefined."

[Julien Puydt]

Hi,

On 27/10/2016 23:18, Tobias Hansen wrote:
> after updating our package to sage 7.4 the doctests are failing with
> errors like [1] (see end of message).
>
> sage/interfaces/maxima_lib.py, line 177 is this:
>
> for l in init_code:
>     ecl_eval("#$%s$"%l)
>
> where
>
> init_code = ['besselexpand : true', 'display2d : false', 'domain :
> complex', 'keepfloat : true',
>             'load(to_poly_solve)', 'load(simplify_sum)',
>             'load(abs_integrate)', 'load(diag)', 'nolabels : true']
>
> The errors can be reproduced by running in ecl:
>
> (require 'maxima "/usr/lib/ecl/maxima.fas");
> (in-package :maxima);
> #$load(to_poly_solve)$;
>
> which gives:
>
> append: argument must be a non-atomic expression; found false
>  -- an error. To debug this try: debugmode(true);
>
> Condition of type: SIMPLE-CONTROL-ERROR
> THROW: The catch MAXIMA::MACSYMA-QUIT is undefined.
> Available restarts:
>
> I'm not sure if this is expected to happen here, since sage runs other
> commands before to set maxima-in-ecl up properly. However the error
> message does not tell me anything and I have no idea what the problem
> is. I also could not find any ecl or maxima related changes between sage
> 7.3 and 7.4 that seemed relevant.
>
> Does someone have an idea?

I have met those errors when I tried to make maxima+ECL work for sage a 
few years ago, and I think the problem was with maxima+ecl : there was a 
problem, but since it happens in init code, the real error about it 
doesn't get catched properly, and that triggers the MACSYMA-QUIT error, 
which doesn't get catched properly since it's init code.

You end up chasing at the wrong place ; I don't remember how I got 
around it, but you might check the old Debian bugs against maxima for 
attached patches in my messages.

I hope that will help,

Snark on #debian-science