[Unicode]   Common Locale Data Repository : Bug Tracking Home | Site Map | Search

CLDR Ticket #7769(accepted unittest)

Opened 4 years ago

Last modified 3 years ago

Vet RBNF against plural categories

Reported by: mark Owned by: grhoten
Component: rbnf Data Locale:
Phase: rc Review:
Weeks: Data Xpath:


write a unit test that checks for differences between RBNF and plural categories, for RBNFs that result in digits.


Change History

comment:1 Changed 4 years ago by emmons

  • Status changed from new to assigned
  • Component changed from unknown to test
  • Priority changed from assess to medium
  • Milestone changed from UNSCH to 27rc
  • Owner changed from anybody to grhoten
  • Type changed from unknown to enhancement

comment:2 Changed 4 years ago by mark

This would be for those cases that make sense, like ordinals.

comment:3 Changed 4 years ago by grhoten

This is a good idea. I could write a test to verify that enough of the prefix/suffix is the same. I'm not sure what I'd do if the grammar works like roman numerals. I'll worry about that when I actually see a verification problem.

comment:4 Changed 4 years ago by markus

  • Phase set to rc
  • Milestone changed from 27rc to 27

comment:5 Changed 3 years ago by emmons

  • Milestone changed from 27 to 28

comment:6 Changed 3 years ago by markus

  • Type changed from enhancement to unittest
  • Component changed from test to rbnf

comment:7 Changed 3 years ago by srl

  • Status changed from assigned to accepted

comment:8 Changed 3 years ago by grhoten

  • Milestone changed from 28 to UNSCH

While it's a good idea, it's not trivial to implement.

  • ICU can't parse the CLDR xml natively. So I can only test ICU.
  • The CLDR syntax doesn't support the full range of ICU syntax, like lenient parsing. It has to go through a full ICU integration process.
  • Languages like Polish have some grammatical variants that are different from the default plural categories. That's because some of the plural categories are optional. The same goes for Arabic.

While they tend to have a common relationship between some languages, it's not relevant between all languages. That makes reliable interoperability testing difficult.

So I could parse the CompactDecimalFormat data with RBNF (maybe with lenient on), but I couldn't do it the other way. This would only verify that RBNF is a superset of CompactDecimalFormat. Given the current state of affairs between ICU and CLDR, I'd have to write this test in ICU and not CLDR.


Add a comment

Modify Ticket

as accepted

E-mail address and user name can be saved in the Preferences.

Note: See TracTickets for help on using tickets.