CLDR-19238 Add Bashkir Plurals and Ordinals#5427
CLDR-19238 Add Bashkir Plurals and Ordinals#5427conradarcturus wants to merge 3 commits intomainfrom
Conversation
Deploying with
|
| Status | Name | Latest Commit | Preview URL | Updated (UTC) |
|---|---|---|---|---|
| ✅ Deployment successful! View logs |
cldr-next | 27a3650 | Commit Preview URL Branch Preview URL |
Mar 08 2026, 05:40 PM |
common/supplemental/ordinals.xml
Outdated
|
|
||
| <!-- 3: one,few,other --> | ||
| <pluralRules locales="ba"> | ||
| <pluralRule count="one">n = 0 or n % 10 = 1,2,4,5,7,8 or n % 100 = 20,50,70,80 or n % 1000 = 0 and n % 1000000 != 0</pluralRule> |
There was a problem hiding this comment.
Just to double-check: does "and" have higher precedence than "or"? I assumed the following logic in the ticket, but now I’m having second thoughts:
(n = 0) or (n % 10 = 1,2,4,5,7,8) or (n % 100 = 20,50,70,80) or (n % 1000 = 0 and n % 1000000 != 0)
There was a problem hiding this comment.
Thanks for checking. The logical groups are first split on the ors.
I have a plural rule debugger on my computer that I'm working on, maybe it helps to visualize the rules:
Your "one" rule
Your "few" rule
A full table
|
It looks like I need to update the minimal pair examples in ba.xml to make this succeed. For example these are the Kazakh examples: Can you provide the equivalent examples for Bashkir? |
|
I meant to say @stdbug can you provide plural minimal pair examples? |
conradarcturus
force-pushed
the
CLDR-19238-ba-plurals
branch
from
191d1d5 to
a191ab5
Compare
|
Hooray! The files in the branch are the same across the force-push. 😃 ~ Your Friendly Jira-GitHub PR Checker Bot |
<pluralMinimalPairs count="one">Арбала {0} алма бар</pluralMinimalPairs>
<pluralMinimalPairs count="other">Арбала {0} алма бар</pluralMinimalPairs>
<ordinalMinimalPairs ordinal="one">{0}-се сатта уң яҡҡа боролоғоҙ</ordinalMinimalPairs>
<ordinalMinimalPairs ordinal="few">{0}-сө сатта уң яҡҡа боролоғоҙ</ordinalMinimalPairs>
<ordinalMinimalPairs ordinal="other">{0}-сы сатта уң яҡҡа боролоғоҙ</ordinalMinimalPairs> |
|
There appears to be something wrong with the examples.
The samples have to *mean the same* except for changes due to the effect of
the numerical value, but *not* be identical. That seems paradoxical. So you
have to look very carefully at the documentation.
https://cldr.unicode.org/index/cldr-spec/plural-rules
The examples for the Kazakh language also appear suspect on first reading.
…On Wed, Mar 4, 2026, 23:55 Mansur ***@***.***> wrote:
*stdbug* left a comment (unicode-org/cldr#5427)
<#5427 (comment)>
I meant to say @stdbug <https://github.com/stdbug> can you provide plural
minimal pair examples?
Sure, here are the examples:
<pluralMinimalPairs count="one">Арбала {0} алма бар</pluralMinimalPairs>
<pluralMinimalPairs count="other">Арбала {0} алма бар</pluralMinimalPairs>
<ordinalMinimalPairs ordinal="one">{0}-се сатта уң яҡҡа боролоғоҙ</ordinalMinimalPairs>
<ordinalMinimalPairs ordinal="few">{0}-сө сатта уң яҡҡа боролоғоҙ</ordinalMinimalPairs>
<ordinalMinimalPairs ordinal="other">{0}-сы сатта уң яҡҡа боролоғоҙ</ordinalMinimalPairs>
—
Reply to this email directly, view it on GitHub
<#5427 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACJLEMEHKLCMOGWAH5BXG5L4PEXHBAVCNFSM6AAAAACWG5Y5BGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHM2DAMBTGA2TMNRWGQ>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
|
|
Understood. This version demonstrates the grammatical differences: <pluralMinimalPairs count="one">Миндә {0} алма бар, уны һиңә бирәйемме?</pluralMinimalPairs>
<pluralMinimalPairs count="other">Миндә {0} алма бар, уларҙы һиңә бирәйемме?</pluralMinimalPairs>
<ordinalMinimalPairs ordinal="one">{0}-се сатта уң яҡҡа боролоғоҙ</ordinalMinimalPairs>
<ordinalMinimalPairs ordinal="few">{0}-сө сатта уң яҡҡа боролоғоҙ</ordinalMinimalPairs>
<ordinalMinimalPairs ordinal="other">{0}-сы сатта уң яҡҡа боролоғоҙ</ordinalMinimalPairs>The first two translate to: "I have {0} apple(s), should I give it/them to you?" One clarification: in which cases should the |
|
Good question. The normal approach is to supply whatever would work with the digit 0. Because the 0 case is often reworded, many implementations will also provide special handling for that. The code that normally used with the categories provides for that. |
|
Then it should be fine to stick with the current |
`java -jar tools/cldr-code/target/cldr-code.jar GeneratedPluralSamples`
conradarcturus
force-pushed
the
CLDR-19238-ba-plurals
branch
from
a191ab5 to
27a3650
Compare
|
Notice: the branch changed across the force-push!
~ Your Friendly Jira-GitHub PR Checker Bot |
|
Ah, yikes. Now it looks like we have a problem since Bashkir changed from 1 cardinal plural state (count="other) to 2 cardinals (count="one" or "other"). What's the order of operations to fix this? Should I add a placeholder to be fixed in the survey tool? @macchiati do we have any utilities to help expand plural states? |
3 participants
CLDR-19238
Adding plural rules for Bashkir
ALLOW_MANY_COMMITS=true