Re: [PATCH] [BZ #20862] mo_MD@cyrillic: new language locale

[Rafal Luzynski <digitalfreak at lingonborough dot com>]

Hi,

First of all: please check your email address, it seems not to
accept replies.

I don't speak Moldovan also I'm not a core glibc developer so please
read my comments as tips rather than any promise that once you fix
the problems the patch will be committed.


2.01.2017 23:26 Vlad Nenea <cybernenea11@cyberdude.com> wrote:
>
>
>
> comment_char %
> escape_char /
> %
> % Moldovan Cyrillic language locale for Moldova
> % Source:
> % Address:
> % Contact: cybernenea11
> % Email: cybernenea11@cyberdude.com
> % Tel:
> % Fax:
> % Language: mo
> % Territory: MD
> % Revision: 0.1
> % Date: 2016-11-01
> % Application: general
> % Users: general
> % Charset: UTF-8
> % Distribution and use is free, also
> % for commercial purposes.

Those two lines ("Distribution and use is free...") have been removed
from all locale data.  Instead another paragraph has been added,
starting with "This file is part of the GNU C Library and contains..."
See:

https://sourceware.org/ml/libc-locales/2013-q1/msg00048.html
https://sourceware.org/git/?p=glibc.git;a=commitdiff;h=a4cea54

> [...]
> LC_MESSAGES
> yesexpr "<U005E><U005B><U0414><U0434><U0059><U0079><U005D><U002E><U002A>"
> noexpr "<U005E><U005B><U041D><U043D><U004E><U006E><U005D><U002E><U002A>"

This says:

    yesexpr "^[ДдYy].*"
    noexpr  "^[НнNn].*"

Note that trailing ".*" have been removed as not necessary.  Then [+1]
have been added to the yes expression and [-0] have been added
to the no expression.  Please consider doing the same.
So these lines should be:

yesexpr "<U005E><U005B><U002B><U0031><U0414><U0434><U0059><U0079><U005D>"
noexpr "<U005E><U005B><U002D><U0030><U041D><U043D><U004E><U006E><U005D>"

See: https://sourceware.org/git/?p=glibc.git;a=commitdiff;h=ebf01e8
https://sourceware.org/git/?p=glibc.git;a=commitdiff;h=f982160

While at this, wouldn't you like to include [Dd] to the yes expression?

> [...]
> abday "<U0414><U0443>.";/
> "<U041B><U0443>.";/
> "<U041C><U0430>.";/
> "<U041C><U0435>.";/
> "<U0416><U043E>.";/
> "<U0412><U0438>.";/
> "<U0421><U044B>."

All characters in the locale data are encoded as U + hex code
even if they are trivial.  For consequence, please consider using
"<U002E>" instead of the literal ".".

> [...]
> abmon "<U044F><U043D>.";/
> "<U0444><U0435><U0431>.";/
> "<U043C><U0430><U0440>.";/
> "<U0430><U043F><U0440>.";/
> "<U043C><U0430><U0439>";/
> "<U044E><U043D><U0438><U0435>";/
> "<U044E><U043B><U0438><U0435>";/
> "<U0430><U0443><U0433>.";/
> "<U0441><U0435><U043F><U0442>.";/
> "<U043E><U043A><U0442>.";/
> "<U043D><U043E><U0435><U043C>.";/
> "<U0434><U0435><U0447>."

Same comment here.

Out of the curiosity, why all weekdays and months names start with
uppercase and only abmon data start with lowercase?  What is the
source of this information?  Moldovan is missing from CLDR.

> d_t_fmt "<U0025><U0061><U0020><U0025><U0064><U0020><U0025><U0062>/
> <U0020><U0025><U0059><U0020><U0025><U0054>"
> d_fmt "<U0025><U0064><U002E><U0025><U006D><U002E><U0025><U0059>"

Those two format strings use "%d" as the day format.  This
generates a leading zero for the days 1-9 (01, 02, ..., 09, 10...).
Is this what you want?  You may want to use "%-d" if you want to
remove the leading zero.

> [...]
> date_fmt "<U0025><U0061><U0020><U0025><U0062><U0020><U0025><U0065>/
> <U0020><U0025><U0048><U003A><U0025><U004D><U003A><U0025><U0053><U0020>/
> <U0025><U005A><U0020><U0025><U0059>"

This starts with "%a %b %e" - a month before a day, is that what
you want?  Also "%e" generates a leading space before the day number
1-9, you will get "Ду. ян.  1" - is this what you want?  Again,
you may want "%-e" or "%-d" to remove any leading characters.

Thanks for your contribution.

Regards,

Rafal