Developers of Sympa

[IKEDA Soji <address@concealed>]

Hi developers,

If beta release of Sympa-6.2 will be postponed, in this week I want
to work on following issue: convention of language/locales ---

  * Complicated use of "lang" and "locale": Sympa takes former as
    just a canonicalized form of latter.
    As a result, there are difficulty to convert them between
    appropriate language codes.

  * Current convention of locale, "ll_RR" form, is not flexible
    enough:

    - Currently Arabic locale is "ar_SY" which may specify the
      Arabic language spoken around Damascus and so on.  "ar" would
      be better to specify Standard Arabic (al-Fushā).

    - It is impossible to specify variants in script, region etc.
      For example: Serbian language written by Latin script instead
      of Cyrillic ("sr@latin"); Valencian variant of Catalan
      language ("ca_ES@valencian").

My plans are:

  (a) Use IETF language tags (RFC 5646) anywhere except (b) and (c):
        fr, zh-CN, ar, sr-Latn, ca-ES-valencian, ...

      Configuration parameters related to language settings also
      would be changed.  Compatibility with old format should be
      kept for the time being.

  (b) Use gettext-style locale names only for names of po files (and
      "Language:" field of their headers):
        fr, zh_CN, ar, sr@latin, ca_ES@valencian, ...

  (c) Use "ll_RR" style names only for language subdirectories
      of template paths:
        fr_FR, zh_CN, ar_SY, (others are not supported).

      Then in nearly later release, rename these directories using
      IETF language tags.

Regards,

--- Soji

-- 
株式会社 コンバージョン  セキュリティ&OSSソリューション部   池田荘児
〒231-0004　神奈川県横浜市中区元浜町3-21-2 ヘリオス関内ビル7F
e-mail address@concealed  TEL 045-640-3550
http://www.conversion.co.jp/