Fix JSP failures with scx

[macchiati]

Fixes #192

scx=deva has not been working; the results were essentially the same as sc=deva

As it turns out, the code really never accounted for multivalued properties. So when testing for values, an scx value like "Devanagari,Bengali" would be skipped.

The code change was simple; to change the code so that a UnicodeProperty could be marked as multivalued, and when comparing values there would be an extra level: splitting the property value if there was a ",", and otherwise testing like a single value. Now, this is done on each unique value rather than on each code point, so it doesn't affect performance much (especially as there are not many unique multivalue property values).

Ideally we should revamp UnicodeProperty to support multivalued properties more directly, but that would be a much bigger overhaul.

UnicodeJsps/src/main/java/org/unicode/jsp/XPropertyFactory.java

• Set scx to be multivalued
• Add other examples of multivalued (namely exemplars, with new addExamplarProperty())
• The rest was Eclipse cleanup of @Override. There are a few other similar Eclipse cleanups in other files.

UnicodeJsps/src/test/java/org/unicode/jsptest/TestMultivalued.java

• Added tests
• Also verifies that \p{scx=beng,deva} doesn't work

UnicodeJsps/src/test/java/org/unicode/jsptest/TestUnicodeSet.java

• Removes test for Arab,Deva (since that is now disabled)

unicodetools/src/main/java/org/unicode/props/UnicodeProperty.java

• Makes small changes to enable multivalued properties, with comma between values
• Disables values like \p{scx=beng,deva}. These never worked right, and can always be computed with [\p{scx=beng}&\p{scx=deva}] if one wants restriction, and [\p{scx=beng}\p{scx=deva}] if one wants inclusion.

[macchiati]

Not sure why "@github-actions build.md / Check UCA data (pull_request) Failing after 1m" is failing; didn't do anything with UCA

[eggrobin]

I’ll take a look on Monday. The UCA check has been failing for a while because our UCD is ahead of our UCA, nothing to do with this PR.

[macchiati]

No urgency

…

On Sat, Nov 25, 2023 at 8:08 AM Robin Leroy ***@***.***> wrote: I’ll take a look on Monday. The UCA check has been failing for a while because our UCD is ahead of our UCA, nothing to do with this PR. — Reply to this email directly, view it on GitHub <#615 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ACJLEMAIE22QX5XVJ6WIT4TYGIJWZAVCNFSM6AAAAAA7ZUEPL6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMRWGM3DOMZTGY> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[markusicu]

please squash-and-merge