More on those at http://cldr.unicode.org/index/bcp47-extension Mark *— Il meglio è l’inimico del bene —* On Fri, Feb 4, 2011 at 08:35, Ian Abbott <abbotti@mev.co.uk> wrote:
On 04/02/11 16:00, yoshito_umaoka@us.ibm.com wrote:
Unicode CLDR defines locale identifiers based on BCP47 language tag (http://www.ietf.org/rfc/bcp/bcp47.txt) BCP47 provides a mechanism for extending language tags for use in various applications.
We recently published RFC6067 (http://datatracker.ietf.org/doc/rfc6067/) to allocate BCP47 extension letter "u" for the use of Unicode Locale Identifiers. Unicode Locale Identifier has its own subtag for specifying time zone IDs. Unfortunately, zone IDs in the tz database cannot be used in language tag because each subtag in extension must be 2 to 8 alpha/numeric characters. For this reason, we defined "short ID" for zones based on UN/LOCODE [http://www.unece.org/cefact/locode/]. Most of exemplar cities by the tz database are covered by LOCODE, with small exceptions.
For example, zone "America/New_York" is represented by "usnyc". The mapping data is provided in the file - http://www.unicode.org/repos/cldr/trunk/common/bcp47/timezone.xml
With the extension, you can exchange locale data including time zone information with a language tag - for example, "en-US-u-tz-usnyc"
I'm wondering if zone.tab can include LOCODE - for example, LOCODE consist from ISO3166 2-letter country code + 3-letter location code.
#locode coordinates TZ comments AD ALV +4230+00131 Europe/Andorra AE DXB +2518+05518 Asia/Dubai AF KBL +3431+06912 Asia/Kabul AG ANU +1703-06148 America/Antigua ...
-Yoshito Umaoka (ICU / CLDR project)
Including LOCODE in zone.tab may or may not be a good idea, but I'm trying hard to think why you'd want to encode a time zone in a language tag.
-- -=( Ian Abbott @ MEV Ltd. E-mail: <abbotti@mev.co.uk> )=- -=( Tel: +44 (0)161 477 1898 FAX: +44 (0)161 718 3587 )=-