Bug#818912: apt-cudf shouldn't exit non-zero if it isn't crashing

[Pietro Abate <pietro.abate@inria.fr>]

On 21/03/16 18:07, David Kalnischkies wrote:
> $ apt install exim4 postfix -s --solver aspcud
> [… busy working for a while …]
> [ currently a useless apt unmet dependency error ]
> E: Sub-process aspcud returned an error code (1)

yes. and its seems there is something going on between apt-get and
apt-cudf. If you dump the edsp as in

export APT_EDSP_DUMP_FILENAME=/tmp/apt-get.edsp
apt install exim4 postfix -s --solver dump

and then pass it to apt-cudf directly

$apt-cudf -v -v  /tmp/apt-get.edsp -s aspcud
(N)StdUtils: Setting Native Architecture to amd64 and Foreign
Architectures to amd64,i386
(N)CudfSolver: Exec: /usr/bin/aspcud /tmp/tmp.apt-cudf.d24f86/in-cudf
/tmp/tmp.apt-cudf.d24f86/out-cudf -count(removed),-count(changed)
[...]
UNSATISFIABLE
Reading from /tmp/gringo.outYvYww7
clasp version 3.1.4
Error: aspcud "-count(removed),-count(changed)"
Message: (UNSAT) No Solutions according to the given preferences

and this is exactly the answer you were looking for ...

> "When the external solver is *unable to find a solution* (and is aware
> of that), it will write an error to standard output and then exit with
> an exit code of 0. An exit code other than 0 will be interpreted as
> a solver crash […]"

BUT, as you noticed, for some reason the exitcode is 1 and not 0 ...

I'll fix this problem shortly to conform to the spec.

pietro