Forum OpenACS Development: ns_xml namespaces and encoding support

I am working on some changes to the news-aggregator package, and we are having some problems with RSS feeds where encoding="iso-8859-1" is specified in the processing instruction, and which actually contains ISO-8859-1 specific characters, like this: Det strømmer ind med henvendelser i forbindelse med efterforskningen af drabet på den 12-årige Mia Teglgaard Sprotte. The ø is supposed to be ø.

To me this looks like a conversion back and forth from UTF-8. Before I start reading through all of the news-aggregator package to see if we are using UTF-8 unsafe functions, I would like to ask people whether there are known problems with non-UTF encodings in ns_xml. Is this the case?

Another problem is that I cannot find anything in the docs on how to use namespaces in ns_xml. How do I determine the namespace of an element in ns_xml?

"[The parsefile command] reads the XML data directly out of the file with the filename filename and parses that data. This is done with low level file operations. The XML data must be in ISO-8859-1, UTF-8 or UTF-16 encoding. If applicable, this is the fastest way, to parse XML data."

And it looks like it supports namespaces as well. Perfect. Thanks!