[GRASS-dev] Travis-ci and tests of grass 7

[Rainer M Krug]

Vaclav Petras <wenzeslaus at gmail.com> writes:

> Hi Rainer,
>
> On Thu, Sep 3, 2015 at 11:24 AM, Rainer M Krug <Rainer at krugs.de> wrote:
>>
>> Hi
>>
>> is anything happening in regards to the travis-ci integration? I did not
>> get any feedback or how I can submit it to the svn.
>
> Ideally open a ticket a submit a diff. That's what I do (although I can
> actually commit) for things which are unclear, unfinished or need
> discussion, see e.g. #2732.
>
> https://trac.osgeo.org/grass/ticket/2732

Done.

see https://trac.osgeo.org/grass/ticket/2733

>
>> In regards to testing: I know about the python gunit tests [1]. Are
>> there any other tests which should be included in the scripts? Should
>> detailed results generated by the tests be uploaded somewhere in case of
>> an error? Or always? If yes, where?
>
> The tests should run as well. We probably cannot upload the detailed
> results anywhere. There is some basic text output however. Perhaps we
> should add option to gunittest to not generate the detailed output at all.

From the travis-ci side, this is not a problem as far as I know, but it
obviously would be good to avoid upload of unnecessary data.

Yes - a summary outout and, in the case of failed tests, uploading detailed
results of this individual test, would be an option? 

>
> The other issues that it takes much more time then compilation and it needs
> to download data.

The download of the data is already in the script - so that is no
problem. Running time is, as far as I know, not that restricted, but I
would have to investigate.
>
> Finally, the tests are still not at 100%, although we are pretty close:
>
> http://fatra.cnr.ncsu.edu/grassgistests/summary_report/nc/index.html

This looks impressive.

>
>> Also, should the same tests be done for other versions?
>
> Probably yes. I don't know how difficult would be to sync Git repo also for
> release branch.

No idea - but I assume only a script which is mirroring the grass svn to
github. At the moment it seems only trunc is mirrored.

There is also the option to not mirror anything at all, but get the
branch from the svn when the tests on travis-ci are started, and only to
have a github repo for the actual travis files - but the mirroring would
be cleaner.

> I'm also not sure about differences for Travis between Git solution in
> comparison to Homebrew but I suppose you do :)

What do you mean by "Git solution"? Homebrew downloads the grass repo
(svn or github, at the moment from svn) when run and installs it.

For the normal travis-ci (Linux) approach, the github repo is used
(which is a mirror of the svn repo).


Cheers,

Rainer

>
> Vaclav

-- 
Rainer M. Krug, PhD (Conservation Ecology, SUN), MSc (Conservation Biology, UCT), Dipl. Phys. (Germany)

Centre of Excellence for Invasion Biology
Stellenbosch University
South Africa

Tel :       +33 - (0)9 53 10 27 44
Cell:       +33 - (0)6 85 62 59 98
Fax :       +33 - (0)9 58 10 27 44

Fax (D):    +49 - (0)3 21 21 25 22 44

email:      Rainer at krugs.de

Skype:      RMkrug

PGP: 0x0F52F982
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 480 bytes
Desc: not available
URL: <http://lists.osgeo.org/pipermail/grass-dev/attachments/20150907/2e3f7ca9/attachment.pgp>