Request for ICU JIRA Access and Bug Report: Czech Language Pluralization in ICU4J

[Suraj Muraleedharan]

Hello ICU Support Team,

I am a Java developer working with ICU4J in one of my private code bases, and I recently encountered a potential issue when testing pluralization rules for the Czech language.

Here is a minimal reproducible example:

PluralRules czechRules = PluralRules.forLocale(Locale.forLanguageTag("cs"));
System.out.println("for cs 3 -> " + czechRules.select(3));
System.out.println("for cs 3.38 -> " + czechRules.select(3.38));

PluralRules cZRules = PluralRules.forLocale(Locale.forLanguageTag("cs_CZ"));
System.out.println("for cs_CZ 3 -> " + cZRules.select(3));
System.out.println("for cs_CZ 3.38 -> " + cZRules.select(3.38));

Observed output:

for cs 3 -> few
for cs 3.38 -> many
for cs_CZ 3 -> other
for cs_CZ 3.38 -> other

Expectation:
From my understanding, cs should be the parent of cs_CZ, and unless specified otherwise in the CLDR data, PluralRules.forLocale(Locale.forLanguageTag("cs_CZ")) should inherit the same rules as cs.

Could you please confirm whether this is expected behavior or a potential bug?

Additionally, I would like to request access to the ICU JIRA so I can file this issue directly and potentially contribute a fix. While I’m not yet certain of the complexity, I am happy to investigate the ICU4J source code and submit a pull request if it’s a suitable candidate.

My details for JIRA access:

• Name: Suraj Muraleedharan

• Email: sur...@gmail.com

• GitHub Username: surajcm

Thank you for your assistance, and I look forward to your guidance.

Best regards,

Suraj Muraleedharan