fixes https://github.com/zazuko/rdf-validate-shacl/issues/136

[s-tittel]

No description provided.

[changeset-bot]

🦋 Changeset detected

Latest commit: 6041552

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 1 package

Name	Type
rdf-validate-shacl	Patch

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

[tpluscode]

Hey, thank you for your contribution!

Would you please add your failing test case to test/data/data-shapes/custom/ and to the manifest.ttl?

[tpluscode]

Also, I noticed that the CI wasn't set up correctly. You may want to rebase master in order to run the tests on GH

[s-tittel]

Hey, I have rebased master, added the test case and fixed a bug in my code that erroneously let all tests pass. Now I get:

132 passing (3s)
3 pending
6 failing

The failing tests are:

1. Test of validation report for shape xone-duplicate by property constraints
2. Test of sh:property at property shape 001
3. Test of sh:qualifiedMinCount with disjoint shapes at property shape 001
4. Test of sh:qualifiedValueShape at property shape 001
5. Test of sh:qualifiedValueShapesDisjoint at property shape 001
6. Test of validation report for shape shared by property constraints

I am not sure if my code really breaks something or if it just changes the order in the expected test reports.

[tpluscode]

Thank you for you update.

Unfortunately, there appear to be multiple types of issues here. It is most definitely not just ordering :)

These three produce less validation results, related to logical constraints, I think.

• Test of validation report for shape xone-duplicate by property constraints
• Test of sh:property at property shape 001
• Test of validation report for shape shared by property constraints

On the other hand these two produces duplicate results. Probably somehow related.

• Test of sh:qualifiedMinCount with disjoint shapes at property shape 001
• Test of sh:qualifiedValueShapesDisjoint at property shape 001

Finally, this one succeeds validation where it should have failed

• Test of sh:qualifiedValueShape at property shape 001`

[s-tittel]

Yes, unfortunately. The first 3 mentioned test cases could be fixed by reordering / deduplication of validation results (see latest commit).
Regarding the qualified value shape tests I currently have no clue.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 98.77%. Comparing base (1d43b2f) to head (6041552).

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #137   +/-   ##
=======================================
  Coverage   98.76%   98.77%           
=======================================
  Files          12       12           
  Lines        1866     1881   +15     
=======================================
+ Hits         1843     1858   +15     
  Misses         23       23           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[s-tittel]

The latest commit is a bit quirky, but fixes validation for qualified value shapes by resetting the list of already checked focus nodes.

[s-tittel]

Ok, still 3 tests failing. I'll see what I can do.

[s-tittel]

the latest commit lets all tests pass. it's not a perfect solution, but I couldn't come up with something better. at least the qualified value shape tests obviously need some degree of recursion, so what I do is check for a max recursion depth. what do you think?

[tpluscode]

Looks good, thanks.

For final touches, please see the comment suggestion and I'd make the maxRecursionDepth a field coming from the constructor like maxErrors.