Right now, the encoding argument added to
xml.dom.minidom.DOMObject.toxml() in Python 2.3 seems fairly useless. 
It has to be UTF-8.  But a one-line change to the implementation of
toprettyxml would make it useful; instead of the encoding error method
being "strict", make it "xmlcharrefreplace".  So change

    writer = codecs.lookup(encoding)[3](writer)

to

    writer = codecs.lookup(encoding)[3](writer, "xmlcharrefreplace")

The method "toxml" is actually on xml.dom.minidom.Node.

Is this still relevant for 3.2?

This is how toprettyxml looks in 3.1/2 which seems to meet the OP's need, I'll close in a few days time unless someone objects.

def toprettyxml(self, indent="\t", newl="\n", encoding=None):
    # indent = the indentation string to prepend, per level
    # newl = the newline string to append
    use_encoding = "utf-8" if encoding is None else encoding
    writer = codecs.getwriter(use_encoding)(io.BytesIO())
    if self.nodeType == Node.DOCUMENT_NODE:
        # Can pass encoding only to document, to put it into XML header
        self.writexml(writer, "", indent, newl, encoding)
    else:
        self.writexml(writer, "", indent, newl)
    if encoding is None:
        return writer.stream.getvalue().decode(use_encoding)
    else:
        return writer.stream.getvalue()

Closing as no response to msg111607.