[GRASS-dev] testsuite error messages analysed: some issues remaining

[Vaclav Petras]

On Sat, Dec 20, 2014 at 12:20 PM, Markus Neteler <neteler at osgeo.org> wrote:

> Hi devs,
>
> I spent some time today to check all failing test in the test suite:
> http://fatra.cnr.ncsu.edu/grassgistests/summary_report/index.html
>
>
> #### NO IDEA problem
>
> http://fatra.cnr.ncsu.edu/grassgistests/reports_for_date-2014-12-20-08-00/report_for_nc_spm_08_grass7_nc/raster/r.profile/test_profile_ncspm/index.html
> --> I don't know that the issue is
>
> No idea here neither.

>
> ###### JUNK REMAINED issue:
>
>
> http://fatra.cnr.ncsu.edu/grassgistests/reports_for_date-2014-12-20-08-00/report_for_nc_spm_08_grass7_nc/raster/r.in.poly/test_rinpoly/index.html
> --> ERROR: option <output>: <test_rinpoly> exists.
>
> Some overwrite flag needed?
>
> overwrite flag should not be used with the tests.

Currently the test should try to clean up after itself using standard
means. This is advantageous in case of running in the user's current
mapset. This might be somehow changes in the future, e.g. some API
introduced, but generally we probably will not get rid of it.

In case of automatic running of all tests, each test file has it's own
mapset which is deleted afterwards, so no junk should remain.

This is might caused by two different test methods or test case classes
using the maps with the same names. This should be avoided.


> But when I run it locally here:
>
> Traceback (most recent call last):
>   File "raster/r.in.poly/testsuite/test_rinpoly.py", line 107, in
> testLabels
>     msg="Labels do not match")
> AssertionError: Labels do not match
> '-8\t\n3\tlabel2\n10.01\tlabel1' != '-8\t\n3\tlabel2\n10.01'
>
> The order of test execution is not defined (in test file driven by Python
unittest) and can be different each time. Test methods should not depend on
each other.

>
> ###### MISSING MAPS problems (in the piemonte location the names
> differ, so they should be locally renamed):
>
>
> http://fatra.cnr.ncsu.edu/grassgistests/reports_for_date-2014-12-20-08-00/report_for_piemonte_utm32_wgs84_grass7_stdmaps/lib/python/pygrass/vector/test_geometry/index.html
> --> "roadmajor" does not exist in the server dataset
>
>
> http://fatra.cnr.ncsu.edu/grassgistests/reports_for_date-2014-12-20-08-00/report_for_piemonte_utm32_wgs84_grass7_stdmaps/lib/python/pygrass/vector/test_vector/index.html
> --> OpenError: Map <points_of_interest> not found.
>
>
> http://fatra.cnr.ncsu.edu/grassgistests/reports_for_date-2014-12-20-08-00/report_for_piemonte_utm32_wgs84_grass7_stdmaps/lib/python/pygrass/vector/test_doctests/index.html
> --> WARNING: <census> not found
>
>
> http://fatra.cnr.ncsu.edu/grassgistests/reports_for_date-2014-12-20-08-00/report_for_piemonte_utm32_wgs84_grass7_stdmaps/vector/v.what/test_vwhat_ncspm/index.html
> --> ERROR: Vector map <bridges> not found
>
>
> http://fatra.cnr.ncsu.edu/grassgistests/reports_for_date-2014-12-20-08-00/report_for_piemonte_utm32_wgs84_grass7_stdmaps/raster/r.profile/test_profile_ncspm/index.html
>
> http://fatra.cnr.ncsu.edu/grassgistests/reports_for_date-2014-12-20-08-00/report_for_piemonte_utm32_wgs84_grass7_stdmaps/raster/r.slope.aspect/test_r_slope_aspect/index.html
> --> ERROR: Raster map <elevation> not found
>
>
I'm afraid, I don't want to maintain my own version of the dataset, so I
will wait for the official one or at least the official table for renaming.

###### MISSING MAPS problems NC dataset:
>
> http://fatra.cnr.ncsu.edu/grassgistests/reports_for_date-2014-12-20-08-00/report_for_nc_spm_08_grass7_nc/lib/python/pygrass/vector/test_vector/index.html
> --> OpenError: Map <points_of_interest> not found.
>
>
> http://fatra.cnr.ncsu.edu/grassgistests/reports_for_date-2014-12-20-08-00/report_for_nc_spm_08_grass7_nc/lib/python/pygrass/vector/test_doctests/index.html
> --> WARNING: <census> not found
>
> Unfortunately, same as above and the following is blocker for this:

http://lists.osgeo.org/pipermail/grass-dev/2014-December/072600.html
http://osgeo-org.1560.x6.nabble.com/Heading-towards-unified-dataset-td5165053.html
http://comments.gmane.org/gmane.comp.gis.grass.devel/60326

####### DESIGN wish:
>
> Each test page contains a title like this:
> ./vector/v.what – test_vwhat_layers
>
> Could please the path to the file be added as
> Testsuite: vector/v.what/testsuite/test_vwhat_layers.py
>
> for easy copy-pasting in the local computer? Would be more joy for
> fixing tests... thanks.
>
> I'm all for joy. Done in r63795 in a messy way for now. It is part of the
table at the beginning. However, note that you are also interested in the
files in the tested directory and it would be more challenging to add
these. Another ideas are links to Trac tickets, revisions or files.

Create tickets if you have some concrete ideas and feel free to discuss the
incomplete ones. However, I cannot promise I can work on that but perhaps
somebody else could.

Vaclav

http://trac.osgeo.org/grass/changeset/63795


> All fixes I have backported as well.
>
> Markus
> _______________________________________________
> 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/20141227/7ebb1f07/attachment-0001.html>