Metaschema XSLT Inspector

[wendellpiez]

Committer Notes

Working towards #72

All Submissions:

• Have you followed the guidelines in our Contributing document?
• Have you checked to ensure there aren't other open Pull Requests for the same update/change?
• Have you squashed any non-relevant commits and commit messages? [instructions]
• Do all automated CI/CD checks pass?

Changes to Core Features:

• Have you added an explanation of what your changes do and why you'd like us to include them?
• Have you written new tests for your core changes, as applicable?
• Have you included examples of how to use your new feature(s)?
• Have you updated all website and readme documentation affected by the changes you made? Changes to the website can be made in the website/content directory of your branch.

[wendellpiez]

See wiki page here: https://github.com/usnistgov/metaschema-xslt/wiki/InspectorXSLT

Discussion board to come (watch for link).

[aj-stein-nist]

Initial feedback, but more to follow with an approval later today or tomorrow when I experiment with it more and make sure I don't find any major problems. Looks good so far. I know we discussed README organization and separation. I would think it is better to make it in a different repo to not confuse consumers of this repo, but you are the maintainer. Not my repo! :-)

[wendellpiez]

Closes #72

[wendellpiez]

Holding back pending

• Support for all paths in OSCAL (cf Some complex but valid Metapath in constraint targets do not produce functional Inspector XSLTs. #91 Update InspectorXSLT to handle Metapath not XPath #92)
• Support for updated allowed-values semantics (cf Clarify allowed-values constraints metaschema#413)
• Integrating PR Some XProc for running XSpec #94

[aj-stein-nist]

@wendellpiez, I am re-reviewing and approving at the time of this message to signal I am happy with the current version of the CI additions, now with more STDOUT and STDERR logging to move us forward. If you want me to re-review after you add relevant documentation, feel free to do so, but I already reviewed the other XSLT, XSpec, XProc, and shell script changes here as needed to advance this work.

[wendellpiez]

@aj-stein-nist please feel free to re-review changes -- mainly in TESTING.md documentation (not perfect yet but better), plus also a couple of tweaks elsewhere.

The remaining desideratum for this Issue is to have CI on a PR report failure not only for broken runs but also for successful runs reporting test failures. I have vacillated but I think this is more or less essential.

A cheap/dirty way to do this is by inspecting the outputs for key (regex) match patterns.

And there could be better options. (Maybe our testing could produce a format that Github Actions could use?) Happy for any thoughts.

[wendellpiez]

In the current 'integration tests' (smoke tests) we have failures successfully reported by at least one XSpec -

https://github.com/usnistgov/metaschema-xslt/actions/runs/7791678616/job/21248238501?pr=73#step:5:144

We have the option of making either script or XSLT runtime error out, but it makes sense to catch this gracefully, non? (Unless an actual reporting-XML concept takes hold.)