1. Forum
  2. Usenet
  3. LINUX.DEBIAN.MAINT.I18N

  • Re: [backintime] =?UTF-8?Q?sr=5FLatn=3A=20can=27t=20guess=20langu?= =?U

    From [email protected]@21:1/5 to All on Sat Mar 1 13:00:01 2025
    Hello Holger,
    thank you for the quick reply.

    Am 01.03.2025 10:28 schrieb Holger Wansing:
    Looking at [1] sr@latin seems to be common for such language,
    and since [2] does not show such error that you see for backintime,
    I think this would be the solution.

    [1] <https://www.debian.org/international/l10n/po/>
    [2] <https://i18n.debian.org/l10n-pkg-status/a/avahi.html>

    I am not convinced. Do we have an ISO or another standard for this we
    can refer, too.
    Another project (e.g. avahi) is not a good reference. I can show you
    some other projects using "sr_Latn", too. Also Weblate does ofer this
    code.

    So I am confused.

    I would like to follow an official standard.
    It could also be the case that gettext is wrong in this case.

    I also don't understand why Debian does this language check. Why does
    Debian need to determin the language?
    For example python-babel is able to determine the correct language name
    from the code "sr_Latn".

    Regards,
    Christian

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Holger Wansing@21:1/5 to All on Sat Mar 1 13:50:01 2025
    Hi,

    Am 1. März 2025 12:52:30 MEZ schrieb [email protected]:
    Hello Holger,
    thank you for the quick reply.

    Am 01.03.2025 10:28 schrieb Holger Wansing:
    Looking at [1] sr@latin seems to be common for such language,
    and since [2] does not show such error that you see for backintime,
    I think this would be the solution.

    [1] <https://www.debian.org/international/l10n/po/>
    [2] <https://i18n.debian.org/l10n-pkg-status/a/avahi.html>

    I am not convinced. Do we have an ISO or another standard for this we can refer, too.
    Another project (e.g. avahi) is not a good reference. I can show you some other projects using "sr_Latn", too. Also Weblate does ofer this code.

    So I am confused.

    I would like to follow an official standard.
    It could also be the case that gettext is wrong in this case.

    I also don't understand why Debian does this language check. Why does Debian need to determin the language?
    For example python-babel is able to determine the correct language name from the code "sr_Latn".

    My proposal is mostly based on the i18n.debian.org link, you gave.
    So, that's a debian-only topic in the first place.
    The error message you referred to, is only relevant for the creation of the i18n.debian.org
    page, and its translation statistics pages.

    For that, it is required to determine, to which language a po file belongs to.

    Apparently this logic does not know about the " sr_Latn" notation, but
    relies on "sr@latin", what seems to be something like a Debian-specific standard.

    I don't know, who decided to use this notation, and not another one, or why...


    You can ignore this error message, if you want IMHO.
    The only impact would be, that Serbian translators probably don't get aware of this
    file, and so maybe you get no translation (updates).



    Holger

    --
    Sent from /e/ OS on Fairphone3

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Holger Wansing@21:1/5 to All on Sat Mar 1 10:30:03 2025
    Hi Christian,

    Am 1. März 2025 09:37:40 MEZ schrieb [email protected]:
    Hello,
    I am upstream maintainer of Back In Time [1]. The Debian Tracker system gives me an error about the sr_Latn.po file [2].

    I checked several official lists and would say that "sr_Latn" is a usual and standardized identifier.

    My Debian Package maintainer is not responsive to anything. Two other persons involved in the packaging have very limited ressources. So I am asking you how to handle the problem and if I can fix something at upstream to solve the problem?

    Regards,
    Christian

    [1] -- <https://github.com/bit-team/backintime>
    [2] -- <https://i18n.debian.org/l10n-pkg-status/b/backintime.html>

    Looking at [1] sr@latin seems to be common for such language,
    and since [2] does not show such error that you see for backintime,
    I think this would be the solution.


    Greetings
    Holger



    [1] <https://www.debian.org/international/l10n/po/>
    [2] <https://i18n.debian.org/l10n-pkg-status/a/avahi.html>



    --
    Sent from /e/ OS on Fairphone3

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From [email protected]@21:1/5 to [email protected] on Wed Mar 5 03:20:02 2025
    On 2025-03-01 13:43 Holger Wansing <[email protected]> wrote:
    The error message you referred to, is only relevant for the creation
    of the i18n.debian.org page, and its translation statistics pages.

    [..]

    Apparently this logic does not know about the " sr_Latn" notation, but
    relies on "sr@latin", what seems to be something like a
    Debian-specific standard.

    I don't know, who decided to use this notation, and not another one,
    or why...

    Thank you for your feedback. In this case I'll open this as a bug
    ticket for "dl10n" package. I got no reply on their mailing list.

    The only impact would be, that Serbian translators probably don't get
    aware of this file, and so maybe you get no translation (updates).

    Why is that? My project is well translated and use
    translated.codeberg.org for translations.

    I never understood why a Distro (even if Debian) is involved in
    translating non-distro-related projects.

    If a translator using the Debian infrastructure to provide translation
    to "backintime", how does this contribution arrive at my upstream
    project?

    Regards,
    Christian Buhtz

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)