[Biopython-dev] Continuous integration server

[Peter]

2010/11/4 Tiago Antão <tiagoantao at gmail.com>:
> 2010/11/3 Peter <biopython at maubp.freeserve.co.uk>:
>> Certainly we can't scale this up to many machines running regular
>> testing without limiting the network access somewhat.
>
> As we discussed before, I was thinking in adding an option to
> run_tests.py (like --offline) and change the tests that access the
> Internet to honour that flag. I was thinking in coding this myself and
> then send to the list for approval (I am not going to make big changes
> to the test framework myself without passing them through here).

Yep, that sounds good.

The previous discussion is here if anyone missed it:
http://lists.open-bio.org/pipermail/biopython-dev/2010-October/008295.html

>>> 6. Discuss policies: One test a day? Full tests or updates? Full
>>> network tests (probably sporadically)? Send emails?
>>
>> Right now triggering tests after each commit isn't easy to do
>> is it (due to limited git support in builtbot)? That might be nice
>> but in the short term running the tests once a day is a big step
>> forward.
>
> It is actually quite easy (with an hook on github), but I would
> suggest leaving this for version 2: lets put the fundamental working
> and the add bells and whistles.

I agree.

>> I'd suggest we do network tests once a week (or fortnight?).
>
> OK, I will go ahead and do some changes to run_tests.py as per above.
>
>> That isn't possible - some applications are not available on Windows,
>> and some libraries are not available on Jython or Python 3 (yet).
>
> OK, we just have to be sure (manually) that all applications that need
> tested are tested.

Yes, that will be a manual task. When we document the slave setup
process we can list which applications we ideally want people to install
on each OS. Having a slight range in versions would actually be a good
thing here.

>>> 8. While I have volunteer Windows testing myself, I will not be able
>>> to maintain it regularly.
>>
>> I have access to a Windows machine (which I use to build the
>> Biopython installers) but currently it is only online intermittently.
>> I'd have to reorganise machines due to limited network ports in
>> the office, but it could in principle be used as a builtbot slave.
>
> Regarding Mac and Windows, I will email again as soon as we have the
> network issue sorted out. Before that we would be doing maybe too much
> traffic as we have no way to stop the network access for now.
>
>> What is wrong with your Linux Python 3.1 slave? It seems that
>> 2to3 is failing on the doctest conversion.
>
> I do not have time to evaluate this now, I will trace this issue over
> the weekend.

Sure.

And once the --offline switch is working, we can start adding slaves
(and documenting how to do it to assist future volunteers).

Good work Tiago :)

Peter