Currently, if the charset is not Latin-1, the language
names Français and Portuguęs come out incorrectly.
This patch fixes this, by:
1. Representing the language names as Unicode strings
in source code.
2. Converting them to the target character set when
creating the userform
3. Adding a new function i18n.htmlQuote, which works
like cgi.escape, but also encodes Unicode strings. If
the Unicode string cannot be encoded in config.charset,
it uses HTML character references.
None of the currenlty-English language names (Hebrew,
Japanese, etc.) have been converted; if this is done,
the encoding of the file needs to change from Latin-1
to UTF-8. Since Python 2.x for x<3 only supports
Latin-1 in Unicode literals, the native-language names
of the language need to be changed like so: Replace
unicode("<Hebrew name of Hebrew>", "utf-8")
(A local function u() might come handy for that)
Log in to post a comment.