On Sun, 27 Jun 2004 08:49:57 +0200, Klaus Reimer wrote: > >_2utf8: Can't convert LÃnder from US-ASCII to UTF8: Illegal byte sequence > >I think Webdruid tries to convert the translated strings to UTF-8 for the >HTML-Output and it can't do this because someone tells webdruid (or the >_2utf8 function) that the strings are coming in in US-ASCII encoding. But >it's still strange that this error does only occur under Cygwin, not under >Linux. >
Because Linux has a working locale implementation, whereas Cygwin does not. >Is there another environment variable which can be used to tell this _2utf8 >function that the input is iso-8859-1? Haven't heard of OUTPUT_CHARSET before >so I just tried INPUT_CHARSET, but doesn't do anything. > OUTPUT_CHARSET is a gettext thing, a provision to handle platforms with broken/incomplete locale, such as Cygwin. I came across it while skimming over gettext code. It seems to be an interim option because a platform is eventually expected to have a working locale, or not have it at all. AFAIK, there's no INPUT_CHARSET. I guess _2utf8() is calling setlocale(LC_ALL, "") or nl_langinfo(CODESET), both of which return "C" locale in Cygwin regardless of your env. settings. -- A. Alper Atici OpenPGP KeyID: 0xB824F550 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/