Re: [OOo-Hebrew] ODP not saving directionality information

Yitzchak Gale wrote:
> Hi Shachar,
>
>>>> Locale - LANG=en_US, LC_CTYPE=he_IL
>
>>> In my opinion, those are the correct locale settings for a
>>> speaker of U.S. English living in Israel.
>
>> Actually, the correct setting for a user who wishes to get English
>> interface and lives in Israel would be LANG=he_IL, LC_MESSAGES=en_US.
>
> I am not an expert, but I don't think that is correct either.
>
> LC_MESSAGES only defines certain very specific responses -
> like "yes" or "no".
No, LC_MESSAGES defines the language in which messages are displayed. In 
other words, LC_MESSAGES defines the language in which the program talks 
to you.
> LC_CTYPE only defines which letters are uppercase
> and lowercase, officially.
No. LC_CTYPE defines the character encoding.
> (see definitions in the Unix standard at www.unix.org)
Care to give a link actually pointing at the information you quote? I 
got 
http://www.opengroup.org/onlinepubs/000095399/basedefs/xbd_chap08.html 
which 100% agrees with my interpretation.
> LANG defines the language.
No. LANG defines the default value for the rest of them. The idea is 
that an application, in order to decide what the value for, say, 
LC_MESSAGES should be, checks the following environment variables in 
order. LC_ALL, LC_MESSAGES, LC_LANG. The first on to be defined sets 
what language (in the text's wording, "the language and cultural 
conventions in which messages should be written").
> If you have LANG=he_IL,
> you should get the Hebrew UI, no matter what
> LC_MESSAGES is set to.
Both LC_ALL and, to a lesser extent, LC_MESSAGES override LANG in that 
respect.
> Maybe what we really need is an "en_IL" locale,
> which would mean "the language is English,
> but use iso-8859-8 encoding and other Israeli
> l10n stuff."
No need, the current settings already allow it.
> Also, as UTF-8 and UTF-16 catch on more and
> more, I think this issue will become less
> important as time goes on. Not solved, but
> less of a show-stopper.
UTF-8 is an encoding. When you define LANG to be en_US.UTF-8, the only 
var that the "UTF-8" part affects is LC_CTYPE. This is how I read the 
specs, and is also how actual programs behave.
> There is no locale variable that officially
> specifies the encoding, but traditionally most
> apps use LC_CTYPE for this.
No, that's not what your specs say. "This environment variable 
determines the interpretation of sequences of bytes of text data as 
characters".
> -Yitzchak

Shachar


-- 
Shachar Shemesh
Lingnu Open Source Consulting ltd.
Have you backed up today's work? http://www.lingnu.com/backup.html

--
Hebrew OpenOffice Mailing List [email protected]
To unsubscribe see: http://openoffice.org.il/mailman/listinfo/hebrew