[Gambas-user] Format$ thousand separator

[Benoit Minisini]

On dimanche 23 décembre 2007, Epíleg wrote:
> En/na Benoit Minisini ha escrit:
> > On jeudi 20 décembre 2007, Epíleg wrote:
> >> En/na Benoit Minisini ha escrit:
> >>> On mercredi 19 décembre 2007, Epíleg wrote:
> >>>> En/na Benoit Minisini ha escrit:
> >>>>> On mercredi 19 décembre 2007, Benoit Minisini wrote:
> >>>>>> On mercredi 19 décembre 2007, Epíleg wrote:
> >>>>>>> Hello,
> >>>>>>>
> >>>>>>> I do not get the thousand separator whit the «format$» function,
> >>>>>>> some examples:
> >>>>>>>
> >>>>>>>
> >>>>>>> PRINT Format$(1000000, "$$,#.00")  =>  EUR 1 000 000,00
> >>>>>>>
> >>>>>>> PRINT Format$(1000000, "$,#.00")  =>  € 1 000 000,00
> >>>>>>>
> >>>>>>> PRINT Format$(1000000, ",#.00")  =>  1000000,00
> >>>>>>>
> >>>>>>>
> >>>>>>> What I'm doing wrong?
> >>>>>>>
> >>>>>>> These tests are done in a Ubuntu Gutsy (7.10) with «ca_ES.UTF-8»
> >>>>>>> locale
> >>>>>>>
> >>>>>>> Regards,
> >>>>>>> Epíleg.
> >>>>>>
> >>>>>> Apparently, this is a bug!
> >>>>>
> >>>>> The last revision now takes thousand separators into account.
> >>>>>
> >>>>> But be careful that in the GNU locales, many languages have thousand
> >>>>> separators only for currencies.
> >>>>>
> >>>>> So, when a locale does not have any thousands separator, a space is
> >>>>> used, and the value is splitted each three digits. Maybe this is a
> >>>>> bad idea that will be removed in the future!
> >>>>>
> >>>>> Regards,
> >>>>
> >>>> Thanks!
> >>>>
> >>>> In an Ubuntu 7.10 with locale en_US.UTF-8 and en_GB.UTF-8, the
> >>>> thousand separator works fine also with numbers, but with locale
> >>>> ca_ES.UTF-8 and es_ES.UTF-8 do not works with numbers (spaces added).
> >>>> In both cases the currencies thousand separator works fine.
> >>>>
> >>>> Excuse my ignorance but in an Open Office Spread document, format
> >>>> function works correctly with numbers and currencies in all of above
> >>>> locale and I don't know why.
> >>>>
> >>>> Thanks a lot,
> >>>> Epíleg.
> >>>
> >>> The locale information is stored in /usr/share/locale/XXX where XXX is
> >>> the contents of the LC_* environment variable.
> >>>
> >>> Gambas uses this information through the standard Unix API. If I set my
> >>> environment to es_ES, then I have no standard thousands separator, and
> >>> a point for currency thousands separator.
> >>>
> >>> Maybe Open Office does not use this information. Maybe it uses it, but
> >>> patches it as Gambas does.
> >>>
> >>> For example, KDE uses its own locale information, in
> >>> the /usr/share/locale/l10n/xxx/entry.desktop file.
> >>>
> >>> Regards,
> >>
> >> Thank You for Your answer.
> >>
> >> From My absolutely ignorance and understanding that this is a locale
> >> problem, isn't a solution to use the currency thousand separator in
> >> numbers when this info is missing?
> >>
> >> Regards,
> >> Epíleg.
> >
> > Yes, but if the locale tells that there is no separators with
> > non-currency numbers, maybe it is right? I must know if it is a bug or
> > not to take a decision!
>
> In my system, Ubuntu 7.10, I got this:
>
> $ LANG=ca_ES.UTF-8 locale -k LC_NUMERIC
> decimal_point=","
> thousands_sep=""
> grouping=-1;-1
> numeric-decimal-point-wc=44
> numeric-thousands-sep-wc=0
> numeric-codeset="UTF-8"
>
> $ LANG=es_ES.UTF-8 locale -k LC_NUMERIC
> decimal_point=","
> thousands_sep=""
> grouping=-1;-1
> numeric-decimal-point-wc=44
> numeric-thousands-sep-wc=0
> numeric-codeset="UTF-8"
>
> $ LANG=es_AR.UTF-8 locale -k LC_NUMERIC
> decimal_point=","
> thousands_sep="."
> grouping=3;3
> numeric-decimal-point-wc=44
> numeric-thousands-sep-wc=46
> numeric-codeset="UTF-8"
>
> $ LANG=en_US.UTF-8 locale -k LC_NUMERIC
> decimal_point="."
> thousands_sep=","
> grouping=3;3
> numeric-decimal-point-wc=46
> numeric-thousands-sep-wc=44
> numeric-codeset="UTF-8"
>
> $ LANG=fr_FR.UTF-8 locale -k LC_NUMERIC
> decimal_point=","
> thousands_sep=" "
> grouping=3;3
> numeric-decimal-point-wc=44
> numeric-thousands-sep-wc=32
> numeric-codeset="UTF-8"
>
> $ LANG=de_DE.UTF-8 locale -k LC_NUMERIC
> decimal_point=","
> thousands_sep="."
> grouping=3;3
> numeric-decimal-point-wc=44
> numeric-thousands-sep-wc=46
> numeric-codeset="UTF-8"
>
> Gambas2 do as locale informs so isn't a Gambas2 bug.
>
> For me is clear that this is a bug in the data of Catalan and Spanish
> locale. There are 3 fields with wrong thousands data.
>
> Do You know the way to solve this bug?
>
> Regards,
> Epíleg.
>

No idea. Maybe we should look at the GNU libc web site. They should have a 
bugzilla.

Regards,

-- 
Benoit Minisini