On Tuesday, 21. January 2003 08:58, Andrzej Talarczyk wrote: > I've had a look at the package AxKit::XSP::SAXParser in XSP.pm and I've > noticed that the difference between my two boxes is in the function > process_node(). The string this function gets as a parameter is already > UTF-8-encoded but it is further processed by encodeToUTF8(). On the > Debian box encodeToUTF8() does nothing, i.e. the input string is > identical to the output string. On the FreeBSD box, however, the input > string (already UTF-8) is converted once again, resulting in garbage > where non-ASCII characters were present. I made a quick and dirty patch > which fixes my problem by removing encodeToUTF-8() call in this place, > but I'm not sure if this doesn't break something else. Unfortunately, my > time resources are limited at the moment to perform a more thorough > query but I'm posting this info here in case someone is interested/has a > similar problem.
As I said, this might be a version problem with your perl and/or libxml. It should just do no conversion (as in the Debian case) if the input is already UTF-8. It obviously doesn't. The next try would be to check if perl or libxml breaks things. For example, check what value the "encoding" parameter of encodeToUTF8 call has, and what's in the string immediately before the call and immediately after the call. Check if the same stuff happens with "äöüß" (or any other national chars) when passed directly in the source code. That should get us much further. -- CU Joerg PGP Public Key at http://ich.bin.kein.hoschi.de/~trouble/public_key.asc PGP Key fingerprint = D34F 57C4 99D8 8F16 E16E 7779 CDDC 41A4 4C48 6F94 --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
