On Mar 13 08:40, Yeo Kai Wei via Cygwin wrote:
> Hi All,
>
> May I ask if there's a bug with Cygwin and Windows currency?
The en_ZM locale isn't yet supported by Cygwin. This will change with
Cygwin 3.5.0. You can install the latest Cygwin test release
3.5.0-0.231.g93f70d7849b8 and retry. You'll get the correct output:
Region: en_AU.utf-8 Currency symbol: $ International currency symbol: AUD
Region: en_CA.utf-8 Currency symbol: $ International currency symbol: CAD
Region: en_GB.utf-8 Currency symbol: £ International currency symbol: GBP
Region: en_US.utf-8 Currency symbol: $ International currency symbol: USD
Region: en_NZ.utf-8 Currency symbol: $ International currency symbol: NZD
Region: en_ZM.utf-8 Currency symbol: K International currency symbol: ZMW
Btw., you don't even need a test application, just call
$ LC_MONETARY="en_ZM.utf-8" locale -ck LC_MONETARY
> The correct answer should be "Region: en_ZM.utf-8 Currency symbol: K
> International currency symbol: ZMK"
>
> Supposedly, the code works on Linux.
>
> Is this an issue with Windows?
We're fetching most locale information from Windows, this includes the
LC_MONETARY information. However, the locale support up to Cygwin 3.4.x
is restricted by an issue in Windows:
Originally, locales were handled in Windows by using so called locale
identifiers, LCID, 2 byte numbers. These LCIDs are used by Cygwin
internally.
In the meantime, Windows switched from LCIDs to locale string
identifiers following RFC 4646 resp. RFC 5646. Now, the problem is this:
Cygwin didn't yet follow suite, still using LCIDs internally, but
Windows introduced locales for which no LCID has ever been assigned.
Asking for the LCID for "en-ZW" returns 0.
For Cygwin 3.5.0, I converted Cygwin to use RFC 5646 locale strings
internally, too, so the next major version of Cygwin will be able to
support all locales Windows supports.
> while (regions[i])
> {
> setlocale(LC_MONETARY, regions[i]);
You're not checking for errors here. If the locale isn't supported,
setlocale should return NULL...
> const struct lconv* loc = localeconv();
...and the locale information stays unchanged. As such, if en_ZM isn't
supported, this call will still fetch the information of the en_NZ.utf-8
locale.
HTH,
Corinna
--
Problem reports: https://cygwin.com/problems.html
FAQ: https://cygwin.com/faq/
Documentation: https://cygwin.com/docs.html
Unsubscribe info: https://cygwin.com/ml/#unsubscribe-simple
