ICU-22326 Integrate CLDR release-44-beta5 to ICU

[pedberg-icu]

Checklist

• Required: Issue filed: https://unicode-org.atlassian.net/browse/ICU-22326
• Required: The PR title must be prefixed with a JIRA Issue number.
• Required: The PR description must include the link to the Jira Issue, for example by completing the URL in the first checklist item
• Required: Each commit message must be prefixed with a JIRA Issue number.
• Issue accepted (done by Technical Committee after discussion)
• Tests included, if applicable
• API docs and/or User Guide docs changed or added, if applicable

Integrate CLDR release-44-beta5 to ICU. The only change on the CLDR side is to add English locale display names for 3 additional languages. However, in the integration process we also pick up changes for several additional CLDR locales (including 3 for which English display names were added):

• blo Anii Moderate
• csw Swampy Cree Basic
• kxv Kuvi Basic
• kxv_Deva Kuvi (Devanagari) Basic
• kxv_Orya Kuvi (Odia) Basic
• kxv_Telu Kuvi (Telugu) Basic
• tok Toki Pona Basic
• vec Venetian Moderate
• xnr Kangri Basic

The commits are in 4 groups:

1. Binary data, and binary-like source data (No review needed)
2. Data generated or copied from CLDR (Can spot check)
3. Source and test code changes
4. manually add kxv_Latn* stubs

ALLOW_MULTIPLE_COMMITS=true

[richgillam]

Spot-checked. LGTM.

[markusicu]

Shouldn't there also be an empty kxv_Latn.txt like we have an empty zh_Hans.txt and an empty sr_Cyrl.txt (in each locale data tree)? There is a kxv_Latn.txt in the main "locales" tree but not in each of the trees.

[pedberg-icu]

Hmm, checking to see what if anything is different in how we handle kxv_Latn vs e.g. sr_Cyrl

[pedberg-icu]

@markusicu @srl295 OK, in the CLDR tool GenerateAliases.java I see this which seems relevant:

        static final Set<String> HAS_MULTIPLE_SCRIPTS =
                org.unicode.cldr.util.Builder.with(new HashSet<String>())
                        .addAll("ha", "ku", "zh", "sr", "uz", "sh")
                        .freeze();

I can update that and re-integrate. But that will add another round of PR approvals etc. Can we approve and merge this PR and then handle that as a separate pass?

[pedberg-icu]

No, that is not it, because we have plenty of other multiscript locales for which a stub for the default script is generated in all the subtrees:

az_Latn
bs_Latn
ff_Latn
ks_Arab
mni_Beng
pa_Guru
sat_Olck
sd_Arab
shi_Tfng
su_Latn
vai_Vaii
yue_Hant

Must be something else.

[pedberg-icu]

One option could be to add the stubs for kxv_Latn manually to this PR and then file a separate ticket to figure out why they are not getting generated automatically... (I note that kxv is not in CLDR coverage, nor in languageInfo.xml).

[pedberg-icu]

Or just to take this PR without the stubs for kxv_Latn in other trees, and mark that as a known issue...

[markusicu]

One option could be to add the stubs for kxv_Latn manually to this PR and then file a separate ticket to figure out why they are not getting generated automatically...

This seems ok, and better than omitting data files that the code might rely on.

(I note that kxv is not in CLDR coverage, nor in languageInfo.xml).

Not in coverage? Wasn't the point that these locales had gotten enough coverage to be useful?

[pedberg-icu]

Not in coverage?

Sorry, I meant the display name for the language is not in modern coverage for other languages.

[srl295]

• adding stubs makes sense. Gotta get rid of those hard coded lists :(
• @markusicu different overload of "coverage". kxv is in generated coverageLevels.txt because it achieved. It's not in the curated coverageLevels.xml because of a process hole.

[pedberg-icu]

OK, I added the stubs here manually as a part-4 commit (can squash into part 3 if desired, it has both binary and source data files). Filed https://unicode-org.atlassian.net/browse/ICU-22557 to fix the underlying problem.

[srl295]

Spot lgtm

[markusicu]

Thanks!
I am fine with the fourth commit. I added it to the list in the PR description.