[GRASS-dev] [GRASS GIS] #2409: last call for options keys consolidation

[Vaclav Petras]

On Fri, Nov 28, 2014 at 9:05 AM, Moritz Lennert <
mlennert at club.worldonline.be> wrote:

> On 28/11/14 14:51, GRASS GIS wrote:
>
>> #2409: last call for options keys consolidation
>> ----------------------------------+-------------------------
>> ----------------
>>   Reporter:  martinl               |       Owner:  grass-dev@…
>>       Type:  task                  |      Status:  new
>>   Priority:  blocker               |   Milestone:  7.0.0
>> Component:  Default               |     Version:  unspecified
>>   Keywords:  standardized options  |    Platform:  Unspecified
>>        Cpu:  Unspecified           |
>> ----------------------------------+-------------------------
>> ----------------
>>
>> Comment(by huhabla):
>>
>>   I am not against renaming the options, but IMHO its completely
>>   unnecessary. However, if the majority thinks it is necessary then i am
>>   fine with that.
>>
>>   But the recent changes in the option naming broke the temporal framework
>>   and as a result many temporal modules and tests.
>>
>>   Why??!!
>>
>>   If such invasive changes are performed, then i would expect that all the
>>   consequences of these changes are recognized and fixed before the
>> commit!
>>
>>   We have a wonderful testsuite, so please use it to detect possible
>> issues.
>>   Please do not simply break GRASS and expect that other developers will
>> fix
>>   it.
>>
>
> Yes, I've already stumbled upon one test which fails because of changed
> type name (r.slope.aspect test calls g.remove with type="rast").
>
> I agree with Soeren that we have to be careful about these changes,
> especially if we want to release in a not too far future...
>
> To run all tests before a commit run:

python -m grass.gunittest.main --location nc_spm_grass7 --location-type nc

inside GRASS session. nc_spm_grass7 should be the name of NC sample
location in your current GISDBASE (grassdata). The command will also work
in any subdirectory.

To run just one test file, cd to testsuite directory with the test file and
run

python test_something.py

inside GRASS session. This will use the current location and mapset.

See more at:

http://grass.osgeo.org/grass71/manuals/libpython/gunittest_testing.html
http://grass.osgeo.org/grass71/manuals/libpython/gunittest_running_tests.html

To see the results of tests which currently run once a day on a server, see:

http://fatra.cnr.ncsu.edu/grassgistests/

The most interesting is the report for nc location:

http://fatra.cnr.ncsu.edu/grassgistests/summary_report/nc/index.html

Which currently ends with:
...
2014-11-26 03:01:42    62954    nc_spm_08_grass7    85%    98%
2014-11-27 03:01:47    63174    nc_spm_08_grass7    38%    80%
2014-11-28 03:01:44    63217    nc_spm_08_grass7    39%    81%

The recent report for individual test files is:

http://fatra.cnr.ncsu.edu/grassgistests/reports_for_date-2014-11-28-08-00/report_for_nc_spm_08_grass7_nc/testfiles.html

When running the test yourself, you will get the same output as you can see
online for one day. So, your index.html should look like:

http://fatra.cnr.ncsu.edu/grassgistests/reports_for_date-2014-11-28-08-00/report_for_nc_spm_08_grass7_nc/index.html

You will find this index.html file in a testreport directory after running
the tests using `python -m grass.gunittest.main ...`.

Vaclav

Moritz
>
> _______________________________________________
> grass-dev mailing list
> grass-dev at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/grass-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/grass-dev/attachments/20141128/bcb4d5b2/attachment-0001.html>