The fact that CP1252 and ISO-8859-1[5] are almost identical (MySQL's "latin1" is a cp1252/iso-8859-1 hybrid, see http://bugs.mysql.com/bug.php?id=11216) and CP1252 is widely used on non Un*x platforms for locales (languages) that use ISO-8859-1[5] on Un*x platforms lead me to believe CP1252 can be safely used in combination with any locale supporting ISO-8859-1 and/or ISO-8859-15. Please add the output of $ for locale in $(grep "ISO-8859-1\(5\)\?\ " localedata/SUPPORTED | cut \ -f 1 -d / | cut -f 1 -d \. | sort -u); do echo "${locale}.CP1252/CP1252 \ \\"; done to localedata/SUPPORTED.
Not all possible locale/charset combinations, especially extremely rarely (in Linux) used ones such as *.CP125{0,2}, need to be in localedata/SUPPORTED. In fact, I'd say that in these days only UTF-8 charset should be added there. glibc includes the (POSIX mandated) localedef utility, so anyone who wishes to use any of the locales in non-default charactersets is just one command away from creating them. Every addition to localedata/SUPPORTED means wasting usually from hundreds to thousands of kilobytes of diskspace.
Hello Jakub, Ok, I will not compile a list of valid locales/charsets for IBM pages then :) . I wrongly assumed the (runtime) information was taken from /usr/share/i18n/locales and charmaps, and didn't realize they are compiled and located in /usr/lib/locale. (This is more of a WONTFIX by the way.) Maybe the comment in localedata/SUPPORTED about reporting bugs to get locales included should be changed to express this policy. After a bit of struggling I seem to have managed to add locales using localedef. Thanks :-) .