feedparser.py

    def unknown_starttag(self, tag, attrs):
        # called for each start tag
        # attrs is a list of (attr, value) tuples
        # e.g. for <pre class='screen'>, tag='pre', attrs=[('class', 'screen')]
        if _debug: sys.stderr.write('_BaseHTMLProcessor, unknown_starttag, tag=%s\n' % tag)
        uattrs = []
        strattrs=''
        if attrs:
            for key, value in attrs:
                value=value.replace('>','&gt;').replace('<','&lt;').replace('"','&quot;')
                value = self.bare_ampersand.sub("&amp;", value)
                # thanks to Kevin Marks for this breathtaking hack to deal with (valid) high-bit attribute values in UTF-8 feeds
                if type(value) != type(u''):
                    try:
                        value = unicode(value, self.encoding)
                    except:
                        value = unicode(value, 'iso-8859-1')
                uattrs.append((unicode(key, self.encoding), value))
            strattrs = u''.join([u' %s="%s"' % (key, value) for key, value in uattrs])
            if self.encoding:
                try:
                    strattrs=strattrs.encode(self.encoding)
                except:
                    pass
        if tag in self.elements_no_end_tag:
            self.pieces.append('<%(tag)s%(strattrs)s />' % locals())
        else:
            self.pieces.append('<%(tag)s%(strattrs)s>' % locals())