Full wide numbers

[Mckenna, Mike]

ICU made the right choice in allowing both.  I would think normalize from fullwide to normal first before parsing would be a good tack to use.


Mike McKenna
Internationalization Technology Architect
+1-408-967-3631 (desk), +1-510-332-7820 (mobile)
PayPal
2211  N. First Street, San Jose CA 95131 - USA


From: CLDR-Users <cldr-users-bounces at unicode.org> on behalf of Rafael Xavier <rxaviers at gmail.com>
Date: Monday, February 13, 2017 at 8:07 AM
To: "cldr-users at unicode.org" <cldr-users at unicode.org>
Subject: Full wide numbers

Hi everyone,
I have a question for you about numbering systems...

Chinese uses latn for the default nu (zh), hanidec for the native nu (zh-u-nu-native), hans for the traditional nu (zh-u-nu-traditio), and hansfin for finance nu (zh-u-nu-finance). CLDR also includes data for the numeric cited numbering systems, i.e., latn and hanidec (e.g., decimalFormats-numberSystem-latn<https://github.com/unicode-cldr/cldr-numbers-modern/blob/30.0.3/main/zh/numbers.json#L82> and decimalFormats-numberSystem-hanidec<https://github.com/unicode-cldr/cldr-numbers-modern/blob/30.0.3/main/zh/numbers.json#L47>). So far so good...
My question is, what decimalFormats (percentFormats, currencyFormats, etc) should be used for another arbitrary numbering system? For example, using fullwide numbering system in Chinese (zu-u-nu-fullwide). There's no decimalFormats-numberSystem-fullwide. I didn't find anything in UTS Part 3: Numbers about<http://www.unicode.org/reports/tr35/tr35-numbers.html#Numbering_Systems> defining what to do in such case.
ICU seems to handle that case fine though: http://demo.icu-project.org/icu4jweb/flexTest.jsp?pat=yMd&_=zh%40numbers%3Dfullwide

Is there any recommendation implementations should follow?
Thanks

--
+55 (16) 98138-1583, skype: rxaviers
http://rafael.xavier.blog.br
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://unicode.org/pipermail/cldr-users/attachments/20170213/8272215e/attachment.html>