Benedikt, Thanks for you insight. I am not a i18n or charsets guru but I believe that the encoding mapserver expects deep in C code only depends on glibc which is unicode compliant (at least on modern distros, dunno abot ye ol slackware 3 box ;-) ). The behaviour is controlled by the LANG and LC_* environment variables.
Please review my last post here which solved a similar issue: http://thread.gmane.org/gmane.comp.gis.mapserver.user/19248/focus=19249 Best regards, Umberto On 4/18/06, listuser HH <[EMAIL PROTECTED]> wrote: > Benedikt Rothe wrote: > > >Hi > > > >The C-Wrapper "mapscript_wrap.c" converts Java-Strings to > >"char*" with the JNI-Function "GetStringUTFChars". > >This UTF-8-byte-sequence is passed to the mapserver-kernel. > >But mapserver-kernel expects an one-byte-encoding. (ISO-8859-1 or CP437 > >(!?)) > > > >I wrote a JNI-sample-program to illustrate. > >The sample-program converts to UTF-8, ISO-8859-1 and CP437. > >On my Windows-Command-Prompt the CP437-result was correct. > > > >I think mapscript_wrap.c has to do more than UTF-8-Conversion. > > > >Benedikt > > > > > Hi all, > > is there a way to do the converion in java dircetly? > > Thanks in advance > > Norbert >
