Fixing return of incorrect subprocess

[roigcarlo]

This fixes the run tests script not returning any error code while failing to execute a subprocess due to the command being invalid. Additionally I added to return an error if the subprocess can be initiated but the command fails due to wrong number of arguments (which should never happen)

I also fixed a warning with the imp lib being deprecated. This should increase compatibility with python 2.x and 3.3 or lower.

Finally, now all executions timed or not, go through the same pipeline to prevent future errors like this.

[philbucher]

Nice work!
Are you working on more updates of the testing? I am asking bcs I am also currently trying to better integrate the c-tests (remember the issue we have).
Just asking to prevent merging struggles

[roigcarlo]

Hi @philbucher, we plan to integrate support for mpi, but honestly I don't see it done before christmas

[philbucher]

ok thx

[philbucher]

This is no longer general I'd say => GetModulePath should now be named GetKratosMultiphysicsPath

[pooyan-dadvand]

Just to say that it works for me and now reports correctly an error in subprocess

[philbucher]

will this be merged this year?
Asking bcs I have some developments touching the same files

[philbucher]

@roigcarlo I used the new functions from #3738

Also I noticed that if there is an error in one of the python-tests, running of all tests is aborted! (unrelated to what I changed)
E.g. if there is an error in the core it doesn't run the tests in the apps
or if there is an error in an app it won't run the tests in the remaining apps

[roigcarlo]

@philbucher Agree with the changes.

Also, the premature exit is caused in one of the changes I introduced in this PR, but honestly I don't remember if it was our intention or a bug, because its even stated that the program should exit. But I find it odd too

This is the change:

if self.exitCode != 0:
    # If thread terminated with an error, propagate and terminate
    sys.exit(self.exitCode)

Let me speak with @pooyan-dadvand before reverting it

[roigcarlo]

Hi @philbucher, I spoke with @pooyan-dadvand and he agreed to let the other tests continue, I will change it this afternoon

[philbucher]

ping

[philbucher]

ping :)

[pooyan-dadvand]

I assume that the ping is for @roigcarlo

[philbucher]

I assume that the ping is for @roigcarlo

indeed it is :)

[roigcarlo]

Give me one sec

[roigcarlo]

Done

[philbucher]

Thanks @roigcarlo I am testing it now

[philbucher]

works again, thx!