Hi Hristo,
On Tue, Dec 13, 2005 at 16:04:43 +0200, Hristo Simenov Hristov wrote:
> > I'm not satisfied with the trailing (bg), personally I don't care, but
> > most users will not know what it stands for. Isn't there another way to
> > create identifiers that would distinguish them from other possible
> > Cyrillic numbering, without such artificial workaround?
> Someone proposed the table with letter to be made localizable but I do not
> know is it possible and how to do it.
That would involve much more work, it needed to define a new API and add
locale data structures. And it wouldn't solve much, since selecting
unique and persistant identifiers that are also pleasant to the user
still would have to be done.
> I do not know how to take the localized
> strings. I supose this I can look in other localization string how is it
> done. Is there a variable or object from which to get the current locale?
That wouldn't help. The list of supported numbering types is offered
regardless of the current locale, all types are listed and selectable,
if CJK and/or CTL aren't excluded per configuration. Furthermore, you
normally don't want your document change its numbering style just
because it is loaded in another locale. And making it dependent on the
UI language, having a string resource localization, would be even worse.
> It is not very good way to connect it with the locale but I do not see
> other way.
I suggest to create identifiers that contain the difference in usage,
for example, if xx uses J and yy doesn't:
xx: A,B,C,...
yy: A,B,C,...,I,K,...
Similar if they use different characters, like in the Cyrillic case.
Since the identifiers have to be persistant, once chosen, because
they're also used in the file format, it should be carefully evaluated
which different numberings are possible and needed before finalizing
anything.
> I tryed with Unicode codes (I took them from the table which I made for
> numbering just replaced 0x with \x) but the same result.
Which we finally solved in IRC yesterday evening :)
For the records: a char string constant encoded in UTF-8 instead of
UCS2, together with a
return OUString( aSupportedTypes[index].cSymbol, strlen(
aSupportedTypes[index].cSymbol), RTL_TEXTENCODING_UTF8);
does it.
Eike
--
OOo/SO Calc core developer. Number formatter stricken i18n transpositionizer.
GnuPG key 0x293C05FD: 997A 4C60 CE41 0149 0DB3 9E96 2F1A D073 293C 05FD
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
