umm. doesn't that doctype point to an SGML DTD? even if minidom did fetch external DTD's (I don't think it does), it would probably choke on that DTD.
running your documents through "tidy -asxml -numeric" before parsing them as XML might be a good idea...
http://tidy.sourceforge.net/ (command-line binaries, library) http://utidylib.berlios.de/ (python bindings)
</F>
Thanks, but the problem is, that I can't use the numeric representations of these special chars. I will probably simply play find&replace before feeding the document into minidom and change the output back afterwards :-)
MfG, Horst -- http://mail.python.org/mailman/listinfo/python-list
