8531 refactor validators

[poikilotherm]

What this PR does / why we need it:
This pull request pays back some technical debt around validators.

• Cleanup some strange code constructs that slipped in over time
• Add way more tests for bean validation and programmatic usage
• Make the programmatic API more comprehensible, adding API docs etc

Which issue(s) this PR closes:

Closes #8531

Special notes for your reviewer:
None. I really put effort into keeping things working. I did draw a line: I did not refactor the JSF validation parts. They are as messy as before (it would make things cleaner by reusing bean validation 🤷 - less duplicated code, less different error messages.). And I did not cleanup all the code in DatasetFieldValueValidator, which could use some love...

Suggestions on how to test this:
There is a large amount of unit tests added with this PR. I tried very hard not to introduce unexpected behavior changes, but better watch out for those.

Does this PR introduce a user interface change? If mockups are available, please link/include them here:
Nope.

Is there a release notes update needed for this change?:
Nope.

Additional documentation:
None so far.

[coveralls]

Coverage decreased (-0.005%) to 18.884% when pulling 43126e1 on poikilotherm:8531-refactor-validators into c5d1df2 on IQSS:develop.

[poikilotherm]

@donsizemore what does "Ansible run terminated abnormally, failing build." on Jenkins mean?

[donsizemore]

it means Ansible exited 1 - EC2 instance fell over, something timed out or generally went splat.

[qqmyers]

If you can click on details, the artifacts tab has the pipeline.log and in the one for this run:
TASK [the root dataverse must be published] ************************************
fatal: [localhost]: FAILED! => {"access_control_allow_headers": "Accept, Content-Type, X-Dataverse-Key", "access_control_allow_methods": "PUT, GET, POST, DELETE, OPTIONS", "access_control_allow_origin": "*", "changed": false, "connection": "close", "content": "{"status":"ERROR","message":"Command edu.harvard.iq.dataverse.engine.command.impl.PublishDataverseCommand@5b1db381 failed: Exception thrown from bean: javax.ejb.EJBTransactionRolledbackException: Exception thrown from bean: javax.ejb.EJBTransactionRolledbackException: Exception thrown from bean: javax.ejb.EJBTransactionRolledbackException: Exception thrown from bean: java.lang.NoClassDefFoundError: org/hibernate/validator/internal/util/CollectionHelper"}", "content_length": "458", "content_type": "application/json;charset=UTF-8", "elapsed": 0, "json": {"message": "Command edu.harvard.iq.dataverse.engine.command.impl.PublishDataverseCommand@5b1db381 failed: Exception thrown from bean: javax.ejb.EJBTransactionRolledbackException: Exception thrown from bean: javax.ejb.EJBTransactionRolledbackException: Exception thrown from bean: javax.ejb.EJBTransactionRolledbackException: Exception thrown from bean: java.lang.NoClassDefFoundError: org/hibernate/validator/internal/util/CollectionHelper", "status": "ERROR"}, "msg": "Status code was 500 and not [200]: HTTP Error 500: Internal Server Error", "redirected": false, "server": "Payara Server 5.2021.6 #badassfish", "status": 500, "url": "http://localhost:8080/api/dataverses/root/actions/:publish", "x_frame_options": "SAMEORIGIN", "x_powered_by": "Servlet/4.0 JSP/2.3 (Payara Server 5.2021.6 #badassfish Java/Red Hat, Inc./11)"}

[poikilotherm]

Thx @qqmyers - fixed with 43126e1

[poikilotherm]

@qqmyers I see a failing test

java.lang.AssertionError: 
Expected: https://dataverse.org/schema/citation/Contact
     but none found
	at edu.harvard.iq.dataverse.api.DatasetsIT.testSemanticMetadataAPIs(DatasetsIT.java:2333)

I don't think this is related to my changes, as I see the same failures at #8532

[sekmiller]

Looks good.
(is there any thought to allowing the min and max username size to be configurable?)

[poikilotherm]

Looks good.

Thx ❤️

(is there any thought to allowing the min and max username size to be configurable?)

Well it hasn't been configurable for a looooong time, so maybe this is not necessary?

Btw this would be hard to do with annotations, as these require constants at compile time.

And IIRC the schema is generating limitations for SQL fields from this, so it might be necessary to create a schema migration.

[sekmiller]

I wasn't saying that this was desirable or necessary - just wondering if the change here was for something beyond readability and maintainability at some point in the future.

[poikilotherm]

Nope, it maintains status quo.

Guess if you'd want to do this at a later point in time, you'd remove the @Size and create a custom annotation that supports retrieving the configured maximum form a setting and makes the model skip any database length restrictions.

[kcondon]

@poikilotherm would you mind refreshing from develop branch? thanks

[poikilotherm]

@poikilotherm would you mind refreshing from develop branch? thanks

✔️

[poikilotherm]

As always thx for merging @kcondon 🤩