Alban Hertroys wrote:
We recently (about a week ago) sent a patch to the maintainer of xmlproc, but we didn't receive a reply yet. A look at the site reveals that the last update was somewhere in 2000.
Does anybody know who the current maintainer is (if that changed), or what the status of xmlproc is? We kind of depend on it...
The patch fixes a buffering problem if the XML contains utf-8 codes, which gets especially problematic if one such character pair starts as the last byte in the buffer... Patch attached, in case someone can use it.
This should no longer be an issue with Python 2.4, because the stateful UTF-8 and UTF-16 decoder have been fixed to support incomplete input.
Unfortunately xmlproc doesn't seem to use the stateful decoder but the stateless decoder (and even handcrafted decoders when the codecs module doesn exist). Adding support for this might be a little tricky, because the parser must determine which encoding to use before instantiating the decoder.
Bye, Walter Dörwald -- http://mail.python.org/mailman/listinfo/python-list
