ENH: Better feedback when create-sibling-gitlab is called with non-existing path

[adswa]

When create-sibling-gitlab receives a path to a non-existing or non-dataset location, it does not provide any feedback to the user about this (#6700). This PR makes create-sibling-gitlab yield an impossible result whenever a provided path does not point to a subdataset:

$ datalad create-sibling-gitlab --site https://jugit.fz-juelich.de thisismyproject                    1 !
create_sibling_gitlab(impossible): thisismyproject (dataset) [No dataset found under /tmp/some/thisismyproject]

Changelog

💫 Enhancements and new features

• When create-sibling-gitlab is called on non-existing subdatasets or paths it now returns an impossible result instead of no feedback at all

[codecov]

Codecov Report

Merging #6701 (682c910) into master (734a646) will decrease coverage by 0.73%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master    #6701      +/-   ##
==========================================
- Coverage   91.14%   90.40%   -0.74%     
==========================================
  Files         353      353              
  Lines       44485    45740    +1255     
==========================================
+ Hits        40545    41351     +806     
- Misses       3940     4389     +449     

Impacted Files	Coverage Δ
datalad/distributed/create_sibling_gitlab.py	69.09% <100.00%> (+0.76%)	⬆️
...ad/distributed/tests/test_create_sibling_gitlab.py	98.49% <100.00%> (+0.03%)	⬆️
datalad/version.py	0.00% <0.00%> (-100.00%)	⬇️
datalad/plugin/wtf.py	0.00% <0.00%> (-100.00%)	⬇️
datalad/cmdline/main.py	0.00% <0.00%> (-100.00%)	⬇️
datalad/plugin/addurls.py	0.00% <0.00%> (-100.00%)	⬇️
datalad/interface/clean.py	0.00% <0.00%> (-100.00%)	⬇️
datalad/interface/rerun.py	0.00% <0.00%> (-100.00%)	⬇️
datalad/plugin/__init__.py	0.00% <0.00%> (-100.00%)	⬇️
datalad/plugin/no_annex.py	0.00% <0.00%> (-100.00%)	⬇️
... and 175 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 734a646...682c910. Read the comment docs.

[codeclimate]

Analysis results are not available for those commits

View more on Code Climate.

[bpoldrack]

Hm. This change looks to me as if the actual problem is subdatasets that should yield an impossible or error result when passing such a path. Then here that result could just be yielded upwards. WDYT, @adswa? Do I miss something?

[adswa]

oh true, I haven't thought in that direction at all yet

[adswa]

mhhh, looking into subdatasets seems to suggest that yielding an impossible or error result when a given path does not match a subdataset is against the way the path parameter is used, though. Its docstring says path/name to query for subdatasets. Defaults to the current directory[PY: , or the entire dataset if called as a dataset method PY] and internal logging confirms that it is a constrain for where to search underneath for subdataset, not where specifically a subdataset should be:

        if paths is not None:
            lgr.debug('Query subdatasets underneath paths: %s', paths)

Making subdatasets() yield impossible would probably be a quite severe API change with mostly bad consequences (I suspect this argument is especially useful in large datasets), so I wouldn't go down that road.

[bpoldrack]

Ah, you're right. Apologies. Approved then. :)

[adswa]

Thanks for the review and approval, @bpoldrack! Merging, since no-one has complained for a week :)