Commits

Anonymous committed 623d5bb

Move modules to new directory

  • Participants
  • Parent commits b9e82eb

Comments (0)

Files changed (4)

File feedparser.py

-#!/usr/bin/env python
-"""Universal feed parser
-
-Handles RSS 0.9x, RSS 1.0, RSS 2.0, CDF, Atom 0.3, and Atom 1.0 feeds
-
-Visit http://feedparser.org/ for the latest version
-Visit http://feedparser.org/docs/ for the latest documentation
-
-Required: Python 2.1 or later
-Recommended: Python 2.3 or later
-Recommended: CJKCodecs and iconv_codec <http://cjkpython.i18n.org/>
-"""
-
-__version__ = "4.2-pre-" + "$Revision: 262 $"[11:14] + "-svn"
-__license__ = """Copyright (c) 2002-2007, Mark Pilgrim, All rights reserved.
-
-Redistribution and use in source and binary forms, with or without modification,
-are permitted provided that the following conditions are met:
-
-* Redistributions of source code must retain the above copyright notice,
-  this list of conditions and the following disclaimer.
-* Redistributions in binary form must reproduce the above copyright notice,
-  this list of conditions and the following disclaimer in the documentation
-  and/or other materials provided with the distribution.
-
-THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS 'AS IS'
-AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
-IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
-ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
-LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
-CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
-SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
-INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
-CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
-ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
-POSSIBILITY OF SUCH DAMAGE."""
-__author__ = "Mark Pilgrim <http://diveintomark.org/>"
-__contributors__ = ["Jason Diamond <http://injektilo.org/>",
-                    "John Beimler <http://john.beimler.org/>",
-                    "Fazal Majid <http://www.majid.info/mylos/weblog/>",
-                    "Aaron Swartz <http://aaronsw.com/>",
-                    "Kevin Marks <http://epeus.blogspot.com/>",
-                    "Sam Ruby <http://intertwingly.net/>"]
-_debug = 0
-
-# HTTP "User-Agent" header to send to servers when downloading feeds.
-# If you are embedding feedparser in a larger application, you should
-# change this to your application name and URL.
-USER_AGENT = "UniversalFeedParser/%s +http://feedparser.org/" % __version__
-
-# HTTP "Accept" header to send to servers when downloading feeds.  If you don't
-# want to send an Accept header, set this to None.
-ACCEPT_HEADER = "application/atom+xml,application/rdf+xml,application/rss+xml,application/x-netcdf,application/xml;q=0.9,text/xml;q=0.2,*/*;q=0.1"
-
-# List of preferred XML parsers, by SAX driver name.  These will be tried first,
-# but if they're not installed, Python will keep searching through its own list
-# of pre-installed parsers until it finds one that supports everything we need.
-PREFERRED_XML_PARSERS = ["drv_libxml2"]
-
-# If you want feedparser to automatically run HTML markup through HTML Tidy, set
-# this to 1.  Requires mxTidy <http://www.egenix.com/files/python/mxTidy.html>
-# or utidylib <http://utidylib.berlios.de/>.
-TIDY_MARKUP = 0
-
-# List of Python interfaces for HTML Tidy, in order of preference.  Only useful
-# if TIDY_MARKUP = 1
-PREFERRED_TIDY_INTERFACES = ["uTidy", "mxTidy"]
-
-# If you want feedparser to automatically resolve all relative URIs, set this
-# to 1.
-RESOLVE_RELATIVE_URIS = 1
-
-# If you want feedparser to automatically sanitize all potentially unsafe
-# HTML content, set this to 1.
-SANITIZE_HTML = 1
-
-# ---------- required modules (should come with any Python distribution) ----------
-import sgmllib, re, sys, copy, urlparse, time, rfc822, types, cgi, urllib, urllib2
-try:
-    from cStringIO import StringIO as _StringIO
-except:
-    from StringIO import StringIO as _StringIO
-
-# ---------- optional modules (feedparser will work without these, but with reduced functionality) ----------
-
-# gzip is included with most Python distributions, but may not be available if you compiled your own
-try:
-    import gzip
-except:
-    gzip = None
-try:
-    import zlib
-except:
-    zlib = None
-
-# If a real XML parser is available, feedparser will attempt to use it.  feedparser has
-# been tested with the built-in SAX parser, PyXML, and libxml2.  On platforms where the
-# Python distribution does not come with an XML parser (such as Mac OS X 10.2 and some
-# versions of FreeBSD), feedparser will quietly fall back on regex-based parsing.
-try:
-    import xml.sax
-    xml.sax.make_parser(PREFERRED_XML_PARSERS) # test for valid parsers
-    from xml.sax.saxutils import escape as _xmlescape
-    _XML_AVAILABLE = 1
-except:
-    _XML_AVAILABLE = 0
-    def _xmlescape(data,entities={}):
-        data = data.replace('&', '&amp;')
-        data = data.replace('>', '&gt;')
-        data = data.replace('<', '&lt;')
-        for char, entity in entities:
-            data = data.replace(char, entity)
-        return data
-
-# base64 support for Atom feeds that contain embedded binary data
-try:
-    import base64, binascii
-except:
-    base64 = binascii = None
-
-# cjkcodecs and iconv_codec provide support for more character encodings.
-# Both are available from http://cjkpython.i18n.org/
-try:
-    import cjkcodecs.aliases
-except:
-    pass
-try:
-    import iconv_codec
-except:
-    pass
-
-# chardet library auto-detects character encodings
-# Download from http://chardet.feedparser.org/
-try:
-    import chardet
-    if _debug:
-        import chardet.constants
-        chardet.constants._debug = 1
-except:
-    chardet = None
-
-# reversable htmlentitydefs mappings for Python 2.2
-try:
-  from htmlentitydefs import name2codepoint, codepoint2name
-except:
-  import htmlentitydefs
-  name2codepoint={}
-  codepoint2name={}
-  for (name,codepoint) in htmlentitydefs.entitydefs.iteritems():
-    if codepoint.startswith('&#'): codepoint=unichr(int(codepoint[2:-1]))
-    name2codepoint[name]=ord(codepoint)
-    codepoint2name[ord(codepoint)]=name
-
-# BeautifulSoup parser used for parsing microformats from embedded HTML content
-# http://www.crummy.com/software/BeautifulSoup/.  At the moment, it appears
-# that there is a version incompatibility, so the import is replaced with
-# a 'None'.  Restoring the try/import/except/none will renable the MF tests.
-BeautifulSoup = None
-
-# ---------- don't touch these ----------
-class ThingsNobodyCaresAboutButMe(Exception): pass
-class CharacterEncodingOverride(ThingsNobodyCaresAboutButMe): pass
-class CharacterEncodingUnknown(ThingsNobodyCaresAboutButMe): pass
-class NonXMLContentType(ThingsNobodyCaresAboutButMe): pass
-class UndeclaredNamespace(Exception): pass
-
-sgmllib.tagfind = re.compile('[a-zA-Z][-_.:a-zA-Z0-9]*')
-sgmllib.special = re.compile('<!')
-sgmllib.charref = re.compile('&#(\d+|x[0-9a-fA-F]+);')
-
-if sgmllib.endbracket.search(' <').start(0):
-    class EndBracketMatch:
-        endbracket = re.compile('''([^'"<>]|"[^"]*"(?=>|/|\s|\w+=)|'[^']*'(?=>|/|\s|\w+=))*(?=[<>])|.*?(?=[<>])''')
-        def search(self,string,index=0):
-            self.match = self.endbracket.match(string,index)
-            if self.match: return self
-        def start(self,n):
-            return self.match.end(n)
-    sgmllib.endbracket = EndBracketMatch()
-
-SUPPORTED_VERSIONS = {'': 'unknown',
-                      'rss090': 'RSS 0.90',
-                      'rss091n': 'RSS 0.91 (Netscape)',
-                      'rss091u': 'RSS 0.91 (Userland)',
-                      'rss092': 'RSS 0.92',
-                      'rss093': 'RSS 0.93',
-                      'rss094': 'RSS 0.94',
-                      'rss20': 'RSS 2.0',
-                      'rss10': 'RSS 1.0',
-                      'rss': 'RSS (unknown version)',
-                      'atom01': 'Atom 0.1',
-                      'atom02': 'Atom 0.2',
-                      'atom03': 'Atom 0.3',
-                      'atom10': 'Atom 1.0',
-                      'atom': 'Atom (unknown version)',
-                      'cdf': 'CDF',
-                      'hotrss': 'Hot RSS'
-                      }
-
-try:
-    UserDict = dict
-except NameError:
-    # Python 2.1 does not have dict
-    from UserDict import UserDict
-    def dict(aList):
-        rc = {}
-        for k, v in aList:
-            rc[k] = v
-        return rc
-
-class FeedParserDict(UserDict):
-    keymap = {'channel': 'feed',
-              'items': 'entries',
-              'guid': 'id',
-              'date': 'updated',
-              'date_parsed': 'updated_parsed',
-              'description': ['subtitle', 'summary'],
-              'url': ['href'],
-              'modified': 'updated',
-              'modified_parsed': 'updated_parsed',
-              'issued': 'published',
-              'issued_parsed': 'published_parsed',
-              'copyright': 'rights',
-              'copyright_detail': 'rights_detail',
-              'tagline': 'subtitle',
-              'tagline_detail': 'subtitle_detail'}
-    def __getitem__(self, key):
-        if key == 'category':
-            return UserDict.__getitem__(self, 'tags')[0]['term']
-        if key == 'enclosures':
-            norel = lambda link: FeedParserDict([(name,value) for (name,value) in link.items() if name!='rel'])
-            return [norel(link) for link in UserDict.__getitem__(self, 'links') if link['rel']=='enclosure']
-        if key == 'license':
-            for link in UserDict.__getitem__(self, 'links'):
-                if link['rel']=='license' and link.has_key('href'):
-                    return link['href']
-        if key == 'categories':
-            return [(tag['scheme'], tag['term']) for tag in UserDict.__getitem__(self, 'tags')]
-        realkey = self.keymap.get(key, key)
-        if type(realkey) == types.ListType:
-            for k in realkey:
-                if UserDict.has_key(self, k):
-                    return UserDict.__getitem__(self, k)
-        if UserDict.has_key(self, key):
-            return UserDict.__getitem__(self, key)
-        return UserDict.__getitem__(self, realkey)
-
-    def __setitem__(self, key, value):
-        for k in self.keymap.keys():
-            if key == k:
-                key = self.keymap[k]
-                if type(key) == types.ListType:
-                    key = key[0]
-        return UserDict.__setitem__(self, key, value)
-
-    def get(self, key, default=None):
-        if self.has_key(key):
-            return self[key]
-        else:
-            return default
-
-    def setdefault(self, key, value):
-        if not self.has_key(key):
-            self[key] = value
-        return self[key]
-        
-    def has_key(self, key):
-        try:
-            return hasattr(self, key) or UserDict.has_key(self, key)
-        except AttributeError:
-            return False
-        
-    def __getattr__(self, key):
-        try:
-            return self.__dict__[key]
-        except KeyError:
-            pass
-        try:
-            assert not key.startswith('_')
-            return self.__getitem__(key)
-        except:
-            raise AttributeError, "object has no attribute '%s'" % key
-
-    def __setattr__(self, key, value):
-        if key.startswith('_') or key == 'data':
-            self.__dict__[key] = value
-        else:
-            return self.__setitem__(key, value)
-
-    def __contains__(self, key):
-        return self.has_key(key)
-
-def zopeCompatibilityHack():
-    global FeedParserDict
-    del FeedParserDict
-    def FeedParserDict(aDict=None):
-        rc = {}
-        if aDict:
-            rc.update(aDict)
-        return rc
-
-_ebcdic_to_ascii_map = None
-def _ebcdic_to_ascii(s):
-    global _ebcdic_to_ascii_map
-    if not _ebcdic_to_ascii_map:
-        emap = (
-            0,1,2,3,156,9,134,127,151,141,142,11,12,13,14,15,
-            16,17,18,19,157,133,8,135,24,25,146,143,28,29,30,31,
-            128,129,130,131,132,10,23,27,136,137,138,139,140,5,6,7,
-            144,145,22,147,148,149,150,4,152,153,154,155,20,21,158,26,
-            32,160,161,162,163,164,165,166,167,168,91,46,60,40,43,33,
-            38,169,170,171,172,173,174,175,176,177,93,36,42,41,59,94,
-            45,47,178,179,180,181,182,183,184,185,124,44,37,95,62,63,
-            186,187,188,189,190,191,192,193,194,96,58,35,64,39,61,34,
-            195,97,98,99,100,101,102,103,104,105,196,197,198,199,200,201,
-            202,106,107,108,109,110,111,112,113,114,203,204,205,206,207,208,
-            209,126,115,116,117,118,119,120,121,122,210,211,212,213,214,215,
-            216,217,218,219,220,221,222,223,224,225,226,227,228,229,230,231,
-            123,65,66,67,68,69,70,71,72,73,232,233,234,235,236,237,
-            125,74,75,76,77,78,79,80,81,82,238,239,240,241,242,243,
-            92,159,83,84,85,86,87,88,89,90,244,245,246,247,248,249,
-            48,49,50,51,52,53,54,55,56,57,250,251,252,253,254,255
-            )
-        import string
-        _ebcdic_to_ascii_map = string.maketrans( \
-            ''.join(map(chr, range(256))), ''.join(map(chr, emap)))
-    return s.translate(_ebcdic_to_ascii_map)
- 
-_cp1252 = {
-  unichr(128): unichr(8364), # euro sign
-  unichr(130): unichr(8218), # single low-9 quotation mark
-  unichr(131): unichr( 402), # latin small letter f with hook
-  unichr(132): unichr(8222), # double low-9 quotation mark
-  unichr(133): unichr(8230), # horizontal ellipsis
-  unichr(134): unichr(8224), # dagger
-  unichr(135): unichr(8225), # double dagger
-  unichr(136): unichr( 710), # modifier letter circumflex accent
-  unichr(137): unichr(8240), # per mille sign
-  unichr(138): unichr( 352), # latin capital letter s with caron
-  unichr(139): unichr(8249), # single left-pointing angle quotation mark
-  unichr(140): unichr( 338), # latin capital ligature oe
-  unichr(142): unichr( 381), # latin capital letter z with caron
-  unichr(145): unichr(8216), # left single quotation mark
-  unichr(146): unichr(8217), # right single quotation mark
-  unichr(147): unichr(8220), # left double quotation mark
-  unichr(148): unichr(8221), # right double quotation mark
-  unichr(149): unichr(8226), # bullet
-  unichr(150): unichr(8211), # en dash
-  unichr(151): unichr(8212), # em dash
-  unichr(152): unichr( 732), # small tilde
-  unichr(153): unichr(8482), # trade mark sign
-  unichr(154): unichr( 353), # latin small letter s with caron
-  unichr(155): unichr(8250), # single right-pointing angle quotation mark
-  unichr(156): unichr( 339), # latin small ligature oe
-  unichr(158): unichr( 382), # latin small letter z with caron
-  unichr(159): unichr( 376)} # latin capital letter y with diaeresis
-
-_urifixer = re.compile('^([A-Za-z][A-Za-z0-9+-.]*://)(/*)(.*?)')
-def _urljoin(base, uri):
-    uri = _urifixer.sub(r'\1\3', uri)
-    try:
-        return urlparse.urljoin(base, uri)
-    except:
-        uri = urlparse.urlunparse([urllib.quote(part) for part in urlparse.urlparse(uri)])
-        return urlparse.urljoin(base, uri)
-
-class _FeedParserMixin:
-    namespaces = {'': '',
-                  'http://backend.userland.com/rss': '',
-                  'http://blogs.law.harvard.edu/tech/rss': '',
-                  'http://purl.org/rss/1.0/': '',
-                  'http://my.netscape.com/rdf/simple/0.9/': '',
-                  'http://example.com/newformat#': '',
-                  'http://example.com/necho': '',
-                  'http://purl.org/echo/': '',
-                  'uri/of/echo/namespace#': '',
-                  'http://purl.org/pie/': '',
-                  'http://purl.org/atom/ns#': '',
-                  'http://www.w3.org/2005/Atom': '',
-                  'http://purl.org/rss/1.0/modules/rss091#': '',
-                  
-                  'http://webns.net/mvcb/':                               'admin',
-                  'http://purl.org/rss/1.0/modules/aggregation/':         'ag',
-                  'http://purl.org/rss/1.0/modules/annotate/':            'annotate',
-                  'http://media.tangent.org/rss/1.0/':                    'audio',
-                  'http://backend.userland.com/blogChannelModule':        'blogChannel',
-                  'http://web.resource.org/cc/':                          'cc',
-                  'http://backend.userland.com/creativeCommonsRssModule': 'creativeCommons',
-                  'http://purl.org/rss/1.0/modules/company':              'co',
-                  'http://purl.org/rss/1.0/modules/content/':             'content',
-                  'http://my.theinfo.org/changed/1.0/rss/':               'cp',
-                  'http://purl.org/dc/elements/1.1/':                     'dc',
-                  'http://purl.org/dc/terms/':                            'dcterms',
-                  'http://purl.org/rss/1.0/modules/email/':               'email',
-                  'http://purl.org/rss/1.0/modules/event/':               'ev',
-                  'http://rssnamespace.org/feedburner/ext/1.0':           'feedburner',
-                  'http://freshmeat.net/rss/fm/':                         'fm',
-                  'http://xmlns.com/foaf/0.1/':                           'foaf',
-                  'http://www.w3.org/2003/01/geo/wgs84_pos#':             'geo',
-                  'http://www.georss.org/georss':                         'georss',
-                  'http://www.opengis.net/gml':                           'gml',
-                  'http://postneo.com/icbm/':                             'icbm',
-                  'http://purl.org/rss/1.0/modules/image/':               'image',
-                  'http://www.itunes.com/DTDs/PodCast-1.0.dtd':           'itunes',
-                  'http://example.com/DTDs/PodCast-1.0.dtd':              'itunes',
-                  'http://purl.org/rss/1.0/modules/link/':                'l',
-                  'http://search.yahoo.com/mrss':                         'media',
-                  'http://madskills.com/public/xml/rss/module/pingback/': 'pingback',
-                  'http://prismstandard.org/namespaces/1.2/basic/':       'prism',
-                  'http://www.w3.org/1999/02/22-rdf-syntax-ns#':          'rdf',
-                  'http://www.w3.org/2000/01/rdf-schema#':                'rdfs',
-                  'http://purl.org/rss/1.0/modules/reference/':           'ref',
-                  'http://purl.org/rss/1.0/modules/richequiv/':           'reqv',
-                  'http://purl.org/rss/1.0/modules/search/':              'search',
-                  'http://purl.org/rss/1.0/modules/slash/':               'slash',
-                  'http://schemas.xmlsoap.org/soap/envelope/':            'soap',
-                  'http://purl.org/rss/1.0/modules/servicestatus/':       'ss',
-                  'http://hacks.benhammersley.com/rss/streaming/':        'str',
-                  'http://purl.org/rss/1.0/modules/subscription/':        'sub',
-                  'http://purl.org/rss/1.0/modules/syndication/':         'sy',
-                  'http://purl.org/rss/1.0/modules/taxonomy/':            'taxo',
-                  'http://purl.org/rss/1.0/modules/threading/':           'thr',
-                  'http://purl.org/rss/1.0/modules/textinput/':           'ti',
-                  'http://madskills.com/public/xml/rss/module/trackback/':'trackback',
-                  'http://wellformedweb.org/commentAPI/':                 'wfw',
-                  'http://purl.org/rss/1.0/modules/wiki/':                'wiki',
-                  'http://www.w3.org/1999/xhtml':                         'xhtml',
-                  'http://www.w3.org/XML/1998/namespace':                 'xml',
-                  'http://www.w3.org/1999/xlink':                         'xlink',
-                  'http://schemas.pocketsoap.com/rss/myDescModule/':      'szf'
-}
-    _matchnamespaces = {}
-
-    can_be_relative_uri = ['link', 'id', 'wfw_comment', 'wfw_commentrss', 'docs', 'url', 'href', 'comments', 'icon', 'logo']
-    can_contain_relative_uris = ['content', 'title', 'summary', 'info', 'tagline', 'subtitle', 'copyright', 'rights', 'description']
-    can_contain_dangerous_markup = ['content', 'title', 'summary', 'info', 'tagline', 'subtitle', 'copyright', 'rights', 'description']
-    html_types = ['text/html', 'application/xhtml+xml']
-    
-    def __init__(self, baseuri=None, baselang=None, encoding='utf-8'):
-        if _debug: sys.stderr.write('initializing FeedParser\n')
-        if not self._matchnamespaces:
-            for k, v in self.namespaces.items():
-                self._matchnamespaces[k.lower()] = v
-        self.feeddata = FeedParserDict() # feed-level data
-        self.encoding = encoding # character encoding
-        self.entries = [] # list of entry-level data
-        self.version = '' # feed type/version, see SUPPORTED_VERSIONS
-        self.namespacesInUse = {} # dictionary of namespaces defined by the feed
-
-        # the following are used internally to track state;
-        # this is really out of control and should be refactored
-        self.infeed = 0
-        self.inentry = 0
-        self.incontent = 0
-        self.intextinput = 0
-        self.inimage = 0
-        self.inauthor = 0
-        self.incontributor = 0
-        self.inpublisher = 0
-        self.insource = 0
-        
-        # georss
-        self.ingeometry = 0
-        
-        self.sourcedata = FeedParserDict()
-        self.contentparams = FeedParserDict()
-        self._summaryKey = None
-        self.namespacemap = {}
-        self.elementstack = []
-        self.basestack = []
-        self.langstack = []
-        self.baseuri = baseuri or ''
-        self.lang = baselang or None
-        self.svgOK = 0
-        if baselang:
-            self.feeddata['language'] = baselang.replace('_','-')
-
-    def unknown_starttag(self, tag, attrs):
-        if _debug: sys.stderr.write('start %s with %s\n' % (tag, attrs))
-        # normalize attrs
-        attrs = [(k.lower(), v) for k, v in attrs]
-        attrs = [(k, k in ('rel', 'type') and v.lower() or v) for k, v in attrs]
-        
-        # track xml:base and xml:lang
-        attrsD = dict(attrs)
-        baseuri = attrsD.get('xml:base', attrsD.get('base')) or self.baseuri
-        self.baseuri = _urljoin(self.baseuri, baseuri)
-        lang = attrsD.get('xml:lang', attrsD.get('lang'))
-        if lang == '':
-            # xml:lang could be explicitly set to '', we need to capture that
-            lang = None
-        elif lang is None:
-            # if no xml:lang is specified, use parent lang
-            lang = self.lang
-        if lang:
-            if tag in ('feed', 'rss', 'rdf:RDF'):
-                self.feeddata['language'] = lang.replace('_','-')
-        self.lang = lang
-        self.basestack.append(self.baseuri)
-        self.langstack.append(lang)
-        
-        # track namespaces
-        for prefix, uri in attrs:
-            if prefix.startswith('xmlns:'):
-                self.trackNamespace(prefix[6:], uri)
-            elif prefix == 'xmlns':
-                self.trackNamespace(None, uri)
-
-        # track inline content
-        if self.incontent and self.contentparams.has_key('type') and not self.contentparams.get('type', 'xml').endswith('xml'):
-            # element declared itself as escaped markup, but it isn't really
-            self.contentparams['type'] = 'application/xhtml+xml'
-        if self.incontent and self.contentparams.get('type') == 'application/xhtml+xml':
-            # Note: probably shouldn't simply recreate localname here, but
-            # our namespace handling isn't actually 100% correct in cases where
-            # the feed redefines the default namespace (which is actually
-            # the usual case for inline content, thanks Sam), so here we
-            # cheat and just reconstruct the element based on localname
-            # because that compensates for the bugs in our namespace handling.
-            # This will horribly munge inline content with non-empty qnames,
-            # but nobody actually does that, so I'm not fixing it.
-            if tag.find(':') <> -1:
-                prefix, tag = tag.split(':', 1)
-                namespace = self.namespacesInUse.get(prefix, '')
-                if tag=='math' and namespace=='http://www.w3.org/1998/Math/MathML':
-                    attrs.append(('xmlns',namespace))
-                if tag=='svg' and namespace=='http://www.w3.org/2000/svg':
-                    attrs.append(('xmlns',namespace))
-            if tag == 'svg': self.svgOK = 1
-            return self.handle_data('<%s%s>' % (tag, self.strattrs(attrs)), escape=0)
-
-        # match namespaces
-        if tag.find(':') <> -1:
-            prefix, suffix = tag.split(':', 1)
-        else:
-            prefix, suffix = '', tag
-        prefix = self.namespacemap.get(prefix, prefix)
-        if prefix:
-            prefix = prefix + '_'
-
-        # special hack for better tracking of empty textinput/image elements in illformed feeds
-        if (not prefix) and tag not in ('title', 'link', 'description', 'name'):
-            self.intextinput = 0
-        if (not prefix) and tag not in ('title', 'link', 'description', 'url', 'href', 'width', 'height'):
-            self.inimage = 0
-        
-        # call special handler (if defined) or default handler
-        methodname = '_start_' + prefix + suffix
-        try:
-            method = getattr(self, methodname)
-            return method(attrsD)
-        except AttributeError:
-            return self.push(prefix + suffix, 1)
-
-    def unknown_endtag(self, tag):
-        if _debug: sys.stderr.write('end %s\n' % tag)
-        # match namespaces
-        if tag.find(':') <> -1:
-            prefix, suffix = tag.split(':', 1)
-        else:
-            prefix, suffix = '', tag
-        prefix = self.namespacemap.get(prefix, prefix)
-        if prefix:
-            prefix = prefix + '_'
-        if suffix == 'svg': self.svgOK = 0
-
-        # call special handler (if defined) or default handler
-        methodname = '_end_' + prefix + suffix
-        try:
-            method = getattr(self, methodname)
-            method()
-        except AttributeError:
-            self.pop(prefix + suffix)
-
-        # track inline content
-        if self.incontent and self.contentparams.has_key('type') and not self.contentparams.get('type', 'xml').endswith('xml'):
-            # element declared itself as escaped markup, but it isn't really
-            self.contentparams['type'] = 'application/xhtml+xml'
-        if self.incontent and self.contentparams.get('type') == 'application/xhtml+xml':
-            tag = tag.split(':')[-1]
-            self.handle_data('</%s>' % tag, escape=0)
-
-        # track xml:base and xml:lang going out of scope
-        if self.basestack:
-            self.basestack.pop()
-            if self.basestack and self.basestack[-1]:
-                self.baseuri = self.basestack[-1]
-        if self.langstack:
-            self.langstack.pop()
-            if self.langstack: # and (self.langstack[-1] is not None):
-                self.lang = self.langstack[-1]
-
-    def handle_charref(self, ref):
-        # called for each character reference, e.g. for '&#160;', ref will be '160'
-        if not self.elementstack: return
-        ref = ref.lower()
-        if ref in ('34', '38', '39', '60', '62', 'x22', 'x26', 'x27', 'x3c', 'x3e'):
-            text = '&#%s;' % ref
-        else:
-            if ref[0] == 'x':
-                c = int(ref[1:], 16)
-            else:
-                c = int(ref)
-            text = unichr(c).encode('utf-8')
-        self.elementstack[-1][2].append(text)
-
-    def handle_entityref(self, ref):
-        # called for each entity reference, e.g. for '&copy;', ref will be 'copy'
-        if not self.elementstack: return
-        if _debug: sys.stderr.write('entering handle_entityref with %s\n' % ref)
-        if ref in ('lt', 'gt', 'quot', 'amp', 'apos'):
-            text = '&%s;' % ref
-        elif ref in self.entities.keys():
-            text = self.entities[ref]
-            if text.startswith('&#') and text.endswith(';'):
-                return self.handle_entityref(text)
-        else:
-            try: name2codepoint[ref]
-            except KeyError: text = '&%s;' % ref
-            else: text = unichr(name2codepoint[ref]).encode('utf-8')
-        self.elementstack[-1][2].append(text)
-
-    def handle_data(self, text, escape=1):
-        # called for each block of plain text, i.e. outside of any tag and
-        # not containing any character or entity references
-        if not self.elementstack: return
-        if escape and self.contentparams.get('type') == 'application/xhtml+xml':
-            text = _xmlescape(text)
-        self.elementstack[-1][2].append(text)
-
-    def handle_comment(self, text):
-        # called for each comment, e.g. <!-- insert message here -->
-        pass
-
-    def handle_pi(self, text):
-        # called for each processing instruction, e.g. <?instruction>
-        pass
-
-    def handle_decl(self, text):
-        pass
-
-    def parse_declaration(self, i):
-        # override internal declaration handler to handle CDATA blocks
-        if _debug: sys.stderr.write('entering parse_declaration\n')
-        if self.rawdata[i:i+9] == '<![CDATA[':
-            k = self.rawdata.find(']]>', i)
-            if k == -1: k = len(self.rawdata)
-            self.handle_data(_xmlescape(self.rawdata[i+9:k]), 0)
-            return k+3
-        else:
-            k = self.rawdata.find('>', i)
-            return k+1
-
-    def mapContentType(self, contentType):
-        contentType = contentType.lower()
-        if contentType == 'text':
-            contentType = 'text/plain'
-        elif contentType == 'html':
-            contentType = 'text/html'
-        elif contentType == 'xhtml':
-            contentType = 'application/xhtml+xml'
-        return contentType
-    
-    def trackNamespace(self, prefix, uri):
-        loweruri = uri.lower()
-        if (prefix, loweruri) == (None, 'http://my.netscape.com/rdf/simple/0.9/') and not self.version:
-            self.version = 'rss090'
-        if loweruri == 'http://purl.org/rss/1.0/' and not self.version:
-            self.version = 'rss10'
-        if loweruri == 'http://www.w3.org/2005/atom' and not self.version:
-            self.version = 'atom10'
-        if loweruri.find('backend.userland.com/rss') <> -1:
-            # match any backend.userland.com namespace
-            uri = 'http://backend.userland.com/rss'
-            loweruri = uri
-        if self._matchnamespaces.has_key(loweruri):
-            self.namespacemap[prefix] = self._matchnamespaces[loweruri]
-            self.namespacesInUse[self._matchnamespaces[loweruri]] = uri
-        else:
-            self.namespacesInUse[prefix or ''] = uri
-
-    def resolveURI(self, uri):
-        return _urljoin(self.baseuri or '', uri)
-    
-    def decodeEntities(self, element, data):
-        return data
-
-    def strattrs(self, attrs):
-        return ''.join([' %s="%s"' % (t[0],_xmlescape(t[1],{'"':'&quot;'})) for t in attrs])
-
-    def push(self, element, expectingText):
-        self.elementstack.append([element, expectingText, []])
-
-    def pop(self, element, stripWhitespace=1):
-        if not self.elementstack: return
-        if self.elementstack[-1][0] != element: return
-        
-        element, expectingText, pieces = self.elementstack.pop()
-
-        if self.version == 'atom10' and self.contentparams.get('type','text') == 'application/xhtml+xml':
-            # remove enclosing child element, but only if it is a <div> and
-            # only if all the remaining content is nested underneath it.
-            # This means that the divs would be retained in the following:
-            #    <div>foo</div><div>bar</div>
-            while pieces and len(pieces)>1 and not pieces[-1].strip():
-                del pieces[-1]
-            while pieces and len(pieces)>1 and not pieces[0].strip():
-                del pieces[0]
-            if pieces and (pieces[0] == '<div>' or pieces[0].startswith('<div ')) and pieces[-1]=='</div>':
-                depth = 0
-                for piece in pieces[:-1]:
-                    if piece.startswith('</'):
-                        depth -= 1
-                        if depth == 0: break
-                    elif piece.startswith('<') and not piece.endswith('/>'):
-                        depth += 1
-                else:
-                    pieces = pieces[1:-1]
-
-        output = ''.join(pieces)
-        if stripWhitespace:
-            output = output.strip()
-        if not expectingText: return output
-
-        # decode base64 content
-        if base64 and self.contentparams.get('base64', 0):
-            try:
-                output = base64.decodestring(output)
-            except binascii.Error:
-                pass
-            except binascii.Incomplete:
-                pass
-                
-        # resolve relative URIs
-        if (element in self.can_be_relative_uri) and output:
-            output = self.resolveURI(output)
-        
-        # decode entities within embedded markup
-        if not self.contentparams.get('base64', 0):
-            output = self.decodeEntities(element, output)
-
-        if self.lookslikehtml(output):
-            self.contentparams['type']='text/html'
-
-        # remove temporary cruft from contentparams
-        try:
-            del self.contentparams['mode']
-        except KeyError:
-            pass
-        try:
-            del self.contentparams['base64']
-        except KeyError:
-            pass
-
-        is_htmlish = self.mapContentType(self.contentparams.get('type', 'text/html')) in self.html_types
-        # resolve relative URIs within embedded markup
-        if is_htmlish and RESOLVE_RELATIVE_URIS:
-            if element in self.can_contain_relative_uris:
-                output = _resolveRelativeURIs(output, self.baseuri, self.encoding, self.contentparams.get('type', 'text/html'))
-                
-        # parse microformats
-        # (must do this before sanitizing because some microformats
-        # rely on elements that we sanitize)
-        if is_htmlish and element in ['content', 'description', 'summary']:
-            mfresults = _parseMicroformats(output, self.baseuri, self.encoding)
-            if mfresults:
-                for tag in mfresults.get('tags', []):
-                    self._addTag(tag['term'], tag['scheme'], tag['label'])
-                for enclosure in mfresults.get('enclosures', []):
-                    self._start_enclosure(enclosure)
-                for xfn in mfresults.get('xfn', []):
-                    self._addXFN(xfn['relationships'], xfn['href'], xfn['name'])
-                vcard = mfresults.get('vcard')
-                if vcard:
-                    self._getContext()['vcard'] = vcard
-        
-        # sanitize embedded markup
-        if is_htmlish and SANITIZE_HTML:
-            if element in self.can_contain_dangerous_markup:
-                output = _sanitizeHTML(output, self.encoding, self.contentparams.get('type', 'text/html'))
-
-        if self.encoding and type(output) != type(u''):
-            try:
-                output = unicode(output, self.encoding)
-            except:
-                pass
-
-        # address common error where people take data that is already
-        # utf-8, presume that it is iso-8859-1, and re-encode it.
-        if self.encoding=='utf-8' and type(output) == type(u''):
-            try:
-                output = unicode(output.encode('iso-8859-1'), 'utf-8')
-            except:
-                pass
-
-        # map win-1252 extensions to the proper code points
-        if type(output) == type(u''):
-            output = u''.join([c in _cp1252.keys() and _cp1252[c] or c for c in output])
-
-        # categories/tags/keywords/whatever are handled in _end_category
-        if element == 'category':
-            return output
-        
-        # store output in appropriate place(s)
-        if self.inentry and not self.insource:
-            if element == 'content':
-                self.entries[-1].setdefault(element, [])
-                contentparams = copy.deepcopy(self.contentparams)
-                contentparams['value'] = output
-                self.entries[-1][element].append(contentparams)
-            elif element == 'link':
-                self.entries[-1][element] = output
-                if output:
-                    self.entries[-1]['links'][-1]['href'] = output
-            else:
-                if element == 'description':
-                    element = 'summary'
-                self.entries[-1][element] = output
-                if self.incontent:
-                    contentparams = copy.deepcopy(self.contentparams)
-                    contentparams['value'] = output
-                    self.entries[-1][element + '_detail'] = contentparams
-        elif (self.infeed or self.insource):# and (not self.intextinput) and (not self.inimage):
-            context = self._getContext()
-            if element == 'description':
-                element = 'subtitle'
-            context[element] = output
-            if element == 'link':
-                context['links'][-1]['href'] = output
-            elif self.incontent:
-                contentparams = copy.deepcopy(self.contentparams)
-                contentparams['value'] = output
-                context[element + '_detail'] = contentparams
-        return output
-
-    def pushContent(self, tag, attrsD, defaultContentType, expectingText):
-        self.incontent += 1
-        if self.lang: self.lang=self.lang.replace('_','-')
-        self.contentparams = FeedParserDict({
-            'type': self.mapContentType(attrsD.get('type', defaultContentType)),
-            'language': self.lang,
-            'base': self.baseuri})
-        self.contentparams['base64'] = self._isBase64(attrsD, self.contentparams)
-        self.push(tag, expectingText)
-
-    def popContent(self, tag):
-        value = self.pop(tag)
-        self.incontent -= 1
-        self.contentparams.clear()
-        return value
-        
-    # a number of elements in a number of RSS variants are nominally plain
-    # text, but this is routinely ignored.  This is an attempt to detect
-    # the most common cases.  As false positives often result in silent
-    # data loss, this function errs on the conservative side.
-    def lookslikehtml(self, str):
-        if self.version.startswith('atom'): return
-        if self.contentparams.get('type','text/html') != 'text/plain': return
-
-        # must have a close tag or a entity reference to qualify
-        if not (re.search(r'</(\w+)>',str) or re.search("&#?\w+;",str)): return
-
-        # all tags must be in a restricted subset of valid HTML tags
-        if filter(lambda t: t.lower() not in _HTMLSanitizer.acceptable_elements,
-            re.findall(r'</?(\w+)',str)): return
-
-        # all entities must have been defined as valid HTML entities
-        from htmlentitydefs import entitydefs
-        if filter(lambda e: e not in entitydefs.keys(),
-            re.findall(r'&(\w+);',str)): return
-
-        return 1
-
-    def _mapToStandardPrefix(self, name):
-        colonpos = name.find(':')
-        if colonpos <> -1:
-            prefix = name[:colonpos]
-            suffix = name[colonpos+1:]
-            prefix = self.namespacemap.get(prefix, prefix)
-            name = prefix + ':' + suffix
-        return name
-        
-    def _getAttribute(self, attrsD, name):
-        return attrsD.get(self._mapToStandardPrefix(name))
-
-    def _isBase64(self, attrsD, contentparams):
-        if attrsD.get('mode', '') == 'base64':
-            return 1
-        if self.contentparams['type'].startswith('text/'):
-            return 0
-        if self.contentparams['type'].endswith('+xml'):
-            return 0
-        if self.contentparams['type'].endswith('/xml'):
-            return 0
-        return 1
-
-    def _itsAnHrefDamnIt(self, attrsD):
-        href = attrsD.get('url', attrsD.get('uri', attrsD.get('href', None)))
-        if href:
-            try:
-                del attrsD['url']
-            except KeyError:
-                pass
-            try:
-                del attrsD['uri']
-            except KeyError:
-                pass
-            attrsD['href'] = href
-        return attrsD
-    
-    def _save(self, key, value):
-        context = self._getContext()
-        context.setdefault(key, value)
-
-    def _start_rss(self, attrsD):
-        versionmap = {'0.91': 'rss091u',
-                      '0.92': 'rss092',
-                      '0.93': 'rss093',
-                      '0.94': 'rss094'}
-        if not self.version:
-            attr_version = attrsD.get('version', '')
-            version = versionmap.get(attr_version)
-            if version:
-                self.version = version
-            elif attr_version.startswith('2.'):
-                self.version = 'rss20'
-            else:
-                self.version = 'rss'
-    
-    def _start_dlhottitles(self, attrsD):
-        self.version = 'hotrss'
-
-    def _start_channel(self, attrsD):
-        self.infeed = 1
-        self._cdf_common(attrsD)
-    _start_feedinfo = _start_channel
-
-    def _cdf_common(self, attrsD):
-        if attrsD.has_key('lastmod'):
-            self._start_modified({})
-            self.elementstack[-1][-1] = attrsD['lastmod']
-            self._end_modified()
-        if attrsD.has_key('href'):
-            self._start_link({})
-            self.elementstack[-1][-1] = attrsD['href']
-            self._end_link()
-    
-    def _start_feed(self, attrsD):
-        self.infeed = 1
-        versionmap = {'0.1': 'atom01',
-                      '0.2': 'atom02',
-                      '0.3': 'atom03'}
-        if not self.version:
-            attr_version = attrsD.get('version')
-            version = versionmap.get(attr_version)
-            if version:
-                self.version = version
-            else:
-                self.version = 'atom'
-
-    def _end_channel(self):
-        self.infeed = 0
-    _end_feed = _end_channel
-    
-    def _start_image(self, attrsD):
-        context = self._getContext()
-        context.setdefault('image', FeedParserDict())
-        self.inimage = 1
-        self.push('image', 0)
-            
-    def _end_image(self):
-        self.pop('image')
-        self.inimage = 0
-
-    def _start_textinput(self, attrsD):
-        context = self._getContext()
-        context.setdefault('textinput', FeedParserDict())
-        self.intextinput = 1
-        self.push('textinput', 0)
-    _start_textInput = _start_textinput
-    
-    def _end_textinput(self):
-        self.pop('textinput')
-        self.intextinput = 0
-    _end_textInput = _end_textinput
-
-    def _start_author(self, attrsD):
-        self.inauthor = 1
-        self.push('author', 1)
-    _start_managingeditor = _start_author
-    _start_dc_author = _start_author
-    _start_dc_creator = _start_author
-    _start_itunes_author = _start_author
-
-    def _end_author(self):
-        self.pop('author')
-        self.inauthor = 0
-        self._sync_author_detail()
-    _end_managingeditor = _end_author
-    _end_dc_author = _end_author
-    _end_dc_creator = _end_author
-    _end_itunes_author = _end_author
-
-    def _start_itunes_owner(self, attrsD):
-        self.inpublisher = 1
-        self.push('publisher', 0)
-
-    def _end_itunes_owner(self):
-        self.pop('publisher')
-        self.inpublisher = 0
-        self._sync_author_detail('publisher')
-
-    def _start_contributor(self, attrsD):
-        self.incontributor = 1
-        context = self._getContext()
-        context.setdefault('contributors', [])
-        context['contributors'].append(FeedParserDict())
-        self.push('contributor', 0)
-
-    def _end_contributor(self):
-        self.pop('contributor')
-        self.incontributor = 0
-
-    def _start_dc_contributor(self, attrsD):
-        self.incontributor = 1
-        context = self._getContext()
-        context.setdefault('contributors', [])
-        context['contributors'].append(FeedParserDict())
-        self.push('name', 0)
-
-    def _end_dc_contributor(self):
-        self._end_name()
-        self.incontributor = 0
-
-    def _start_name(self, attrsD):
-        self.push('name', 0)
-    _start_itunes_name = _start_name
-
-    def _end_name(self):
-        value = self.pop('name')
-        if self.inpublisher:
-            self._save_author('name', value, 'publisher')
-        elif self.inauthor:
-            self._save_author('name', value)
-        elif self.incontributor:
-            self._save_contributor('name', value)
-        elif self.intextinput:
-            context = self._getContext()
-            context['name'] = value
-    _end_itunes_name = _end_name
-
-    def _start_width(self, attrsD):
-        self.push('width', 0)
-
-    def _end_width(self):
-        value = self.pop('width')
-        try:
-            value = int(value)
-        except:
-            value = 0
-        if self.inimage:
-            context = self._getContext()
-            context['width'] = value
-
-    def _start_height(self, attrsD):
-        self.push('height', 0)
-
-    def _end_height(self):
-        value = self.pop('height')
-        try:
-            value = int(value)
-        except:
-            value = 0
-        if self.inimage:
-            context = self._getContext()
-            context['height'] = value
-
-    def _start_url(self, attrsD):
-        self.push('href', 1)
-    _start_homepage = _start_url
-    _start_uri = _start_url
-
-    def _end_url(self):
-        value = self.pop('href')
-        if self.inauthor:
-            self._save_author('href', value)
-        elif self.incontributor:
-            self._save_contributor('href', value)
-    _end_homepage = _end_url
-    _end_uri = _end_url
-
-    def _start_email(self, attrsD):
-        self.push('email', 0)
-    _start_itunes_email = _start_email
-
-    def _end_email(self):
-        value = self.pop('email')
-        if self.inpublisher:
-            self._save_author('email', value, 'publisher')
-        elif self.inauthor:
-            self._save_author('email', value)
-        elif self.incontributor:
-            self._save_contributor('email', value)
-    _end_itunes_email = _end_email
-
-    def _getContext(self):
-        if self.insource:
-            context = self.sourcedata
-        elif self.inimage:
-            context = self.feeddata['image']
-        elif self.intextinput:
-            context = self.feeddata['textinput']
-        elif self.inentry:
-            context = self.entries[-1]
-        else:
-            context = self.feeddata
-        return context
-
-    def _save_author(self, key, value, prefix='author'):
-        context = self._getContext()
-        context.setdefault(prefix + '_detail', FeedParserDict())
-        context[prefix + '_detail'][key] = value
-        self._sync_author_detail()
-
-    def _save_contributor(self, key, value):
-        context = self._getContext()
-        context.setdefault('contributors', [FeedParserDict()])
-        context['contributors'][-1][key] = value
-
-    def _sync_author_detail(self, key='author'):
-        context = self._getContext()
-        detail = context.get('%s_detail' % key)
-        if detail:
-            name = detail.get('name')
-            email = detail.get('email')
-            if name and email:
-                context[key] = '%s (%s)' % (name, email)
-            elif name:
-                context[key] = name
-            elif email:
-                context[key] = email
-        else:
-            author, email = context.get(key), None
-            if not author: return
-            emailmatch = re.search(r'''(([a-zA-Z0-9\_\-\.\+]+)@((\[[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.)|(([a-zA-Z0-9\-]+\.)+))([a-zA-Z]{2,4}|[0-9]{1,3})(\]?))(\?subject=\S+)?''', author)
-            if emailmatch:
-                email = emailmatch.group(0)
-                # probably a better way to do the following, but it passes all the tests
-                author = author.replace(email, '')
-                author = author.replace('()', '')
-                author = author.replace('<>', '')
-                author = author.replace('&lt;&gt;', '')
-                author = author.strip()
-                if author and (author[0] == '('):
-                    author = author[1:]
-                if author and (author[-1] == ')'):
-                    author = author[:-1]
-                author = author.strip()
-            if author or email:
-                context.setdefault('%s_detail' % key, FeedParserDict())
-            if author:
-                context['%s_detail' % key]['name'] = author
-            if email:
-                context['%s_detail' % key]['email'] = email
-
-    def _start_subtitle(self, attrsD):
-        self.pushContent('subtitle', attrsD, 'text/plain', 1)
-    _start_tagline = _start_subtitle
-    _start_itunes_subtitle = _start_subtitle
-
-    def _end_subtitle(self):
-        self.popContent('subtitle')
-    _end_tagline = _end_subtitle
-    _end_itunes_subtitle = _end_subtitle
-            
-    def _start_rights(self, attrsD):
-        self.pushContent('rights', attrsD, 'text/plain', 1)
-    _start_dc_rights = _start_rights
-    _start_copyright = _start_rights
-
-    def _end_rights(self):
-        self.popContent('rights')
-    _end_dc_rights = _end_rights
-    _end_copyright = _end_rights
-
-    def _start_item(self, attrsD):
-        self.entries.append(FeedParserDict())
-        self.push('item', 0)
-        self.inentry = 1
-        self.guidislink = 0
-        id = self._getAttribute(attrsD, 'rdf:about')
-        if id:
-            context = self._getContext()
-            context['id'] = id
-        self._cdf_common(attrsD)
-    _start_entry = _start_item
-    _start_product = _start_item
-
-    def _end_item(self):
-        self.pop('item')
-        self.inentry = 0
-    _end_entry = _end_item
-
-    def _start_dc_language(self, attrsD):
-        self.push('language', 1)
-    _start_language = _start_dc_language
-
-    def _end_dc_language(self):
-        self.lang = self.pop('language')
-    _end_language = _end_dc_language
-
-    def _start_dc_publisher(self, attrsD):
-        self.push('publisher', 1)
-    _start_webmaster = _start_dc_publisher
-
-    def _end_dc_publisher(self):
-        self.pop('publisher')
-        self._sync_author_detail('publisher')
-    _end_webmaster = _end_dc_publisher
-
-    def _start_published(self, attrsD):
-        self.push('published', 1)
-    _start_dcterms_issued = _start_published
-    _start_issued = _start_published
-
-    def _end_published(self):
-        value = self.pop('published')
-        self._save('published_parsed', _parse_date(value))
-    _end_dcterms_issued = _end_published
-    _end_issued = _end_published
-
-    def _start_updated(self, attrsD):
-        self.push('updated', 1)
-    _start_modified = _start_updated
-    _start_dcterms_modified = _start_updated
-    _start_pubdate = _start_updated
-    _start_dc_date = _start_updated
-
-    def _end_updated(self):
-        value = self.pop('updated')
-        parsed_value = _parse_date(value)
-        self._save('updated_parsed', parsed_value)
-    _end_modified = _end_updated
-    _end_dcterms_modified = _end_updated
-    _end_pubdate = _end_updated
-    _end_dc_date = _end_updated
-
-    def _start_created(self, attrsD):
-        self.push('created', 1)
-    _start_dcterms_created = _start_created
-
-    def _end_created(self):
-        value = self.pop('created')
-        self._save('created_parsed', _parse_date(value))
-    _end_dcterms_created = _end_created
-
-    def _start_expirationdate(self, attrsD):
-        self.push('expired', 1)
-
-    def _end_expirationdate(self):
-        self._save('expired_parsed', _parse_date(self.pop('expired')))
-
-    # geospatial location, or "where", from georss.org
-
-    def _start_georssgeom(self, attrsD):
-        self.push('geometry', 0)
-    _start_georss_point = _start_georssgeom
-    _start_georss_line = _start_georssgeom
-    _start_georss_polygon = _start_georssgeom
-    _start_georss_box = _start_georssgeom
-
-    def _save_where(self, geometry):
-        context = self._getContext()
-        context.setdefault('where', {})
-        context['where'] = FeedParserDict(geometry)
-
-    def _end_georss_point(self):
-        geometry = _parse_georss_point(self.pop('geometry'))
-        self._save_where(geometry)
-
-    def _end_georss_line(self):
-        geometry = _parse_georss_line(self.pop('geometry'))
-        self._save_where(geometry)
-    
-    def _end_georss_polygon(self):
-        this = self.pop('geometry')
-        geometry = _parse_georss_polygon(this)
-        self._save_where(geometry)
-
-    def _end_georss_box(self):
-        geometry = _parse_georss_box(self.pop('geometry'))
-        self._save_where(geometry)
-
-    def _start_where(self, attrsD):
-        self.push('where', 0)
-    _start_georss_where = _start_where
-
-    def _start_gml_point(self, attrsD):
-        self.ingeometry = 'point'
-        self.push('geometry', 0)
-
-    def _start_gml_linestring(self, attrsD):
-        self.ingeometry = 'linestring'
-        self.push('geometry', 0)
-
-    def _start_gml_polygon(self, attrsD):
-        self.push('geometry', 0)
-
-    def _start_gml_exterior(self, attrsD):
-        self.push('geometry', 0)
-
-    def _start_gml_linearring(self, attrsD):
-        self.ingeometry = 'polygon'
-        self.push('geometry', 0)
-
-    def _start_gml_pos(self, attrsD):
-        self.push('pos', 0)
-
-    def _end_gml_pos(self):
-        this = self.pop('pos')
-        geometry = _parse_georss_point(this)
-        self._save_where(geometry)
-
-    def _start_gml_poslist(self, attrsD):
-        self.push('pos', 0)
-
-    def _end_gml_poslist(self):
-        geometry = _parse_poslist(self.pop('pos'), self.ingeometry)
-        self._save_where(geometry)
-
-    def _end_geom(self):
-        self.ingeometry = 0
-        self.pop('geometry')
-    _end_gml_point = _end_geom
-    _end_gml_linestring = _end_geom
-    _end_gml_linearring = _end_geom
-    _end_gml_exterior = _end_geom
-    _end_gml_polygon = _end_geom
-
-    def _end_where(self):
-        self.pop('where')
-    _end_georss_where = _end_where
-
-    # end geospatial
-
-    def _start_cc_license(self, attrsD):
-        context = self._getContext()
-        value = self._getAttribute(attrsD, 'rdf:resource')
-        attrsD = FeedParserDict()
-        attrsD['rel']='license'
-        if value: attrsD['href']=value
-        context.setdefault('links', []).append(attrsD)
-        
-    def _start_creativecommons_license(self, attrsD):
-        self.push('license', 1)
-    _start_creativeCommons_license = _start_creativecommons_license
-
-    def _end_creativecommons_license(self):
-        value = self.pop('license')
-        context = self._getContext()
-        attrsD = FeedParserDict()
-        attrsD['rel']='license'
-        if value: attrsD['href']=value
-        context.setdefault('links', []).append(attrsD)
-        del context['license']
-    _end_creativeCommons_license = _end_creativecommons_license
-
-    def _addXFN(self, relationships, href, name):
-        context = self._getContext()
-        xfn = context.setdefault('xfn', [])
-        value = FeedParserDict({'relationships': relationships, 'href': href, 'name': name})
-        if value not in xfn:
-            xfn.append(value)
-        
-    def _addTag(self, term, scheme, label):
-        context = self._getContext()
-        tags = context.setdefault('tags', [])
-        if (not term) and (not scheme) and (not label): return
-        value = FeedParserDict({'term': term, 'scheme': scheme, 'label': label})
-        if value not in tags:
-            tags.append(value)
-
-    def _start_category(self, attrsD):
-        if _debug: sys.stderr.write('entering _start_category with %s\n' % repr(attrsD))
-        term = attrsD.get('term')
-        scheme = attrsD.get('scheme', attrsD.get('domain'))
-        label = attrsD.get('label')
-        self._addTag(term, scheme, label)
-        self.push('category', 1)
-    _start_dc_subject = _start_category
-    _start_keywords = _start_category
-        
-    def _end_itunes_keywords(self):
-        for term in self.pop('itunes_keywords').split():
-            self._addTag(term, 'http://www.itunes.com/', None)
-        
-    def _start_itunes_category(self, attrsD):
-        self._addTag(attrsD.get('text'), 'http://www.itunes.com/', None)
-        self.push('category', 1)
-        
-    def _end_category(self):
-        value = self.pop('category')
-        if not value: return
-        context = self._getContext()
-        tags = context['tags']
-        if value and len(tags) and not tags[-1]['term']:
-            tags[-1]['term'] = value
-        else:
-            self._addTag(value, None, None)
-    _end_dc_subject = _end_category
-    _end_keywords = _end_category
-    _end_itunes_category = _end_category
-
-    def _start_cloud(self, attrsD):
-        self._getContext()['cloud'] = FeedParserDict(attrsD)
-        
-    def _start_link(self, attrsD):
-        attrsD.setdefault('rel', 'alternate')
-        if attrsD['rel'] == 'self':
-            attrsD.setdefault('type', 'application/atom+xml')
-        else:
-            attrsD.setdefault('type', 'text/html')
-        context = self._getContext()
-        attrsD = self._itsAnHrefDamnIt(attrsD)
-        if attrsD.has_key('href'):
-            attrsD['href'] = self.resolveURI(attrsD['href'])
-            if attrsD.get('rel')=='enclosure' and not context.get('id'):
-                context['id'] = attrsD.get('href')
-        expectingText = self.infeed or self.inentry or self.insource
-        context.setdefault('links', [])
-        context['links'].append(FeedParserDict(attrsD))
-        if attrsD.has_key('href'):
-            expectingText = 0
-            if (attrsD.get('rel') == 'alternate') and (self.mapContentType(attrsD.get('type')) in self.html_types):
-                context['link'] = attrsD['href']
-        else:
-            self.push('link', expectingText)
-    _start_producturl = _start_link
-
-    def _end_link(self):
-        value = self.pop('link')
-        context = self._getContext()
-    _end_producturl = _end_link
-
-    def _start_guid(self, attrsD):
-        self.guidislink = (attrsD.get('ispermalink', 'true') == 'true')
-        self.push('id', 1)
-
-    def _end_guid(self):
-        value = self.pop('id')
-        self._save('guidislink', self.guidislink and not self._getContext().has_key('link'))
-        if self.guidislink:
-            # guid acts as link, but only if 'ispermalink' is not present or is 'true',
-            # and only if the item doesn't already have a link element
-            self._save('link', value)
-
-    def _start_title(self, attrsD):
-        if self.svgOK: return self.unknown_starttag('title', attrsD.items())
-        self.pushContent('title', attrsD, 'text/plain', self.infeed or self.inentry or self.insource)
-    _start_dc_title = _start_title
-    _start_media_title = _start_title
-
-    def _end_title(self):
-        if self.svgOK: return
-        value = self.popContent('title')
-        if not value: return
-        context = self._getContext()
-    _end_dc_title = _end_title
-    _end_media_title = _end_title
-
-    def _start_description(self, attrsD):
-        context = self._getContext()
-        if context.has_key('summary'):
-            self._summaryKey = 'content'
-            self._start_content(attrsD)
-        else:
-            self.pushContent('description', attrsD, 'text/html', self.infeed or self.inentry or self.insource)
-    _start_dc_description = _start_description
-
-    def _start_abstract(self, attrsD):
-        self.pushContent('description', attrsD, 'text/plain', self.infeed or self.inentry or self.insource)
-
-    def _end_description(self):
-        if self._summaryKey == 'content':
-            self._end_content()
-        else:
-            value = self.popContent('description')
-        self._summaryKey = None
-    _end_abstract = _end_description
-    _end_dc_description = _end_description
-
-    def _start_info(self, attrsD):
-        self.pushContent('info', attrsD, 'text/plain', 1)
-    _start_feedburner_browserfriendly = _start_info
-
-    def _end_info(self):
-        self.popContent('info')
-    _end_feedburner_browserfriendly = _end_info
-
-    def _start_generator(self, attrsD):
-        if attrsD:
-            attrsD = self._itsAnHrefDamnIt(attrsD)
-            if attrsD.has_key('href'):
-                attrsD['href'] = self.resolveURI(attrsD['href'])
-        self._getContext()['generator_detail'] = FeedParserDict(attrsD)
-        self.push('generator', 1)
-
-    def _end_generator(self):
-        value = self.pop('generator')
-        context = self._getContext()
-        if context.has_key('generator_detail'):
-            context['generator_detail']['name'] = value
-            
-    def _start_admin_generatoragent(self, attrsD):
-        self.push('generator', 1)
-        value = self._getAttribute(attrsD, 'rdf:resource')
-        if value:
-            self.elementstack[-1][2].append(value)
-        self.pop('generator')
-        self._getContext()['generator_detail'] = FeedParserDict({'href': value})
-
-    def _start_admin_errorreportsto(self, attrsD):
-        self.push('errorreportsto', 1)
-        value = self._getAttribute(attrsD, 'rdf:resource')
-        if value:
-            self.elementstack[-1][2].append(value)
-        self.pop('errorreportsto')
-        
-    def _start_summary(self, attrsD):
-        context = self._getContext()
-        if context.has_key('summary'):
-            self._summaryKey = 'content'
-            self._start_content(attrsD)
-        else:
-            self._summaryKey = 'summary'
-            self.pushContent(self._summaryKey, attrsD, 'text/plain', 1)
-    _start_itunes_summary = _start_summary
-
-    def _end_summary(self):
-        if self._summaryKey == 'content':
-            self._end_content()
-        else:
-            self.popContent(self._summaryKey or 'summary')
-        self._summaryKey = None
-    _end_itunes_summary = _end_summary
-        
-    def _start_enclosure(self, attrsD):
-        attrsD = self._itsAnHrefDamnIt(attrsD)
-        context = self._getContext()
-        attrsD['rel']='enclosure'
-        context.setdefault('links', []).append(FeedParserDict(attrsD))
-        href = attrsD.get('href')
-        if href and not context.get('id'):
-            context['id'] = href
-            
-    def _start_source(self, attrsD):
-        self.insource = 1
-
-    def _end_source(self):
-        self.insource = 0
-        self._getContext()['source'] = copy.deepcopy(self.sourcedata)
-        self.sourcedata.clear()
-
-    def _start_content(self, attrsD):
-        self.pushContent('content', attrsD, 'text/plain', 1)
-        src = attrsD.get('src')
-        if src:
-            self.contentparams['src'] = src
-        self.push('content', 1)
-
-    def _start_prodlink(self, attrsD):
-        self.pushContent('content', attrsD, 'text/html', 1)
-
-    def _start_body(self, attrsD):
-        self.pushContent('content', attrsD, 'application/xhtml+xml', 1)
-    _start_xhtml_body = _start_body
-
-    def _start_content_encoded(self, attrsD):
-        self.pushContent('content', attrsD, 'text/html', 1)
-    _start_fullitem = _start_content_encoded
-
-    def _end_content(self):
-        copyToDescription = self.mapContentType(self.contentparams.get('type')) in (['text/plain'] + self.html_types)
-        value = self.popContent('content')
-        if copyToDescription:
-            self._save('description', value)
-
-    _end_body = _end_content
-    _end_xhtml_body = _end_content
-    _end_content_encoded = _end_content
-    _end_fullitem = _end_content
-    _end_prodlink = _end_content
-
-    def _start_itunes_image(self, attrsD):
-        self.push('itunes_image', 0)
-        self._getContext()['image'] = FeedParserDict({'href': attrsD.get('href')})
-    _start_itunes_link = _start_itunes_image
-        
-    def _end_itunes_block(self):
-        value = self.pop('itunes_block', 0)
-        self._getContext()['itunes_block'] = (value == 'yes') and 1 or 0
-
-    def _end_itunes_explicit(self):
-        value = self.pop('itunes_explicit', 0)
-        self._getContext()['itunes_explicit'] = (value == 'yes') and 1 or 0
-
-if _XML_AVAILABLE:
-    class _StrictFeedParser(_FeedParserMixin, xml.sax.handler.ContentHandler):
-        def __init__(self, baseuri, baselang, encoding):
-            if _debug: sys.stderr.write('trying StrictFeedParser\n')
-            xml.sax.handler.ContentHandler.__init__(self)
-            _FeedParserMixin.__init__(self, baseuri, baselang, encoding)
-            self.bozo = 0
-            self.exc = None
-        
-        def startPrefixMapping(self, prefix, uri):
-            self.trackNamespace(prefix, uri)
-        
-        def startElementNS(self, name, qname, attrs):
-            namespace, localname = name
-            lowernamespace = str(namespace or '').lower()
-            if lowernamespace.find('backend.userland.com/rss') <> -1:
-                # match any backend.userland.com namespace
-                namespace = 'http://backend.userland.com/rss'
-                lowernamespace = namespace
-            if qname and qname.find(':') > 0:
-                givenprefix = qname.split(':')[0]
-            else:
-                givenprefix = None
-            prefix = self._matchnamespaces.get(lowernamespace, givenprefix)
-            if givenprefix and (prefix == None or (prefix == '' and lowernamespace == '')) and not self.namespacesInUse.has_key(givenprefix):
-                    raise UndeclaredNamespace, "'%s' is not associated with a namespace" % givenprefix
-            localname = str(localname).lower()
-
-            # qname implementation is horribly broken in Python 2.1 (it
-            # doesn't report any), and slightly broken in Python 2.2 (it
-            # doesn't report the xml: namespace). So we match up namespaces
-            # with a known list first, and then possibly override them with
-            # the qnames the SAX parser gives us (if indeed it gives us any
-            # at all).  Thanks to MatejC for helping me test this and
-            # tirelessly telling me that it didn't work yet.
-            attrsD = {}
-            if localname=='math' and namespace=='http://www.w3.org/1998/Math/MathML':
-                attrsD['xmlns']=namespace
-            if localname=='svg' and namespace=='http://www.w3.org/2000/svg':
-                attrsD['xmlns']=namespace
-
-            if prefix:
-                localname = prefix.lower() + ':' + localname
-            elif namespace and not qname: #Expat
-                for name,value in self.namespacesInUse.items():
-                     if name and value == namespace:
-                         localname = name + ':' + localname
-                         break
-            if _debug: sys.stderr.write('startElementNS: qname = %s, namespace = %s, givenprefix = %s, prefix = %s, attrs = %s, localname = %s\n' % (qname, namespace, givenprefix, prefix, attrs.items(), localname))
-
-            for (namespace, attrlocalname), attrvalue in attrs._attrs.items():
-                lowernamespace = (namespace or '').lower()
-                prefix = self._matchnamespaces.get(lowernamespace, '')
-                if prefix:
-                    attrlocalname = prefix + ':' + attrlocalname
-                attrsD[str(attrlocalname).lower()] = attrvalue
-            for qname in attrs.getQNames():
-                attrsD[str(qname).lower()] = attrs.getValueByQName(qname)
-            self.unknown_starttag(localname, attrsD.items())
-
-        def characters(self, text):
-            self.handle_data(text)
-
-        def endElementNS(self, name, qname):
-            namespace, localname = name
-            lowernamespace = str(namespace or '').lower()
-            if qname and qname.find(':') > 0:
-                givenprefix = qname.split(':')[0]
-            else:
-                givenprefix = ''
-            prefix = self._matchnamespaces.get(lowernamespace, givenprefix)
-            if prefix:
-                localname = prefix + ':' + localname
-            elif namespace and not qname: #Expat
-                for name,value in self.namespacesInUse.items():
-                     if name and value == namespace:
-                         localname = name + ':' + localname
-                         break
-            localname = str(localname).lower()
-            self.unknown_endtag(localname)
-
-        def error(self, exc):
-            self.bozo = 1
-            self.exc = exc
-            
-        def fatalError(self, exc):
-            self.error(exc)
-            raise exc
-
-class _BaseHTMLProcessor(sgmllib.SGMLParser):
-    special = re.compile('''[<>'"]''')
-    bare_ampersand = re.compile("&(?!#\d+;|#x[0-9a-fA-F]+;|\w+;)")
-    elements_no_end_tag = ['area', 'base', 'basefont', 'br', 'col', 'frame', 'hr',
-      'img', 'input', 'isindex', 'link', 'meta', 'param']
-    
-    def __init__(self, encoding, type):
-        self.encoding = encoding
-        self.type = type
-        if _debug: sys.stderr.write('entering BaseHTMLProcessor, encoding=%s\n' % self.encoding)
-        sgmllib.SGMLParser.__init__(self)
-        
-    def reset(self):
-        self.pieces = []
-        sgmllib.SGMLParser.reset(self)
-
-    def _shorttag_replace(self, match):
-        tag = match.group(1)
-        if tag in self.elements_no_end_tag:
-            return '<' + tag + ' />'
-        else:
-            return '<' + tag + '></' + tag + '>'
-
-    def parse_starttag(self,i):
-        j=sgmllib.SGMLParser.parse_starttag(self, i)
-        if self.type == 'application/xhtml+xml':
-            if j>2 and self.rawdata[j-2:j]=='/>':
-                self.unknown_endtag(self.lasttag)
-        return j
-
-    def feed(self, data):
-        data = re.compile(r'<!((?!DOCTYPE|--|\[))', re.IGNORECASE).sub(r'&lt;!\1', data)
-        #data = re.sub(r'<(\S+?)\s*?/>', self._shorttag_replace, data) # bug [ 1399464 ] Bad regexp for _shorttag_replace
-        data = re.sub(r'<([^<\s]+?)\s*/>', self._shorttag_replace, data) 
-        data = data.replace('&#39;', "'")
-        data = data.replace('&#34;', '"')
-        if self.encoding and type(data) == type(u''):
-            data = data.encode(self.encoding)
-        sgmllib.SGMLParser.feed(self, data)
-        sgmllib.SGMLParser.close(self)
-
-    def normalize_attrs(self, attrs):
-        if not attrs: return attrs
-        # utility method to be called by descendants
-        attrs = dict([(k.lower(), v) for k, v in attrs]).items()
-        attrs = [(k, k in ('rel', 'type') and v.lower() or v) for k, v in attrs]
-        attrs.sort()
-        return attrs
-
-    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())
-
-    def unknown_endtag(self, tag):
-        # called for each end tag, e.g. for </pre>, tag will be 'pre'
-        # Reconstruct the original end tag.
-        if tag not in self.elements_no_end_tag:
-            self.pieces.append("</%(tag)s>" % locals())
-
-    def handle_charref(self, ref):
-        # called for each character reference, e.g. for '&#160;', ref will be '160'
-        # Reconstruct the original character reference.
-        if ref.startswith('x'):
-            value = unichr(int(ref[1:],16))
-        else:
-            value = unichr(int(ref))
-
-        if value in _cp1252.keys():
-            self.pieces.append('&#%s;' % hex(ord(_cp1252[value]))[1:])
-        else:
-            self.pieces.append('&#%(ref)s;' % locals())
-        
-    def handle_entityref(self, ref):
-        # called for each entity reference, e.g. for '&copy;', ref will be 'copy'
-        # Reconstruct the original entity reference.
-        if name2codepoint.has_key(ref):
-            self.pieces.append('&%(ref)s;' % locals())
-        else:
-            self.pieces.append('&amp;%(ref)s' % locals())
-
-    def handle_data(self, text):
-        # called for each block of plain text, i.e. outside of any tag and
-        # not containing any character or entity references
-        # Store the original text verbatim.
-        if _debug: sys.stderr.write('_BaseHTMLProcessor, handle_text, text=%s\n' % text)
-        self.pieces.append(text)
-        
-    def handle_comment(self, text):
-        # called for each HTML comment, e.g. <!-- insert Javascript code here -->
-        # Reconstruct the original comment.
-        self.pieces.append('<!--%(text)s-->' % locals())
-        
-    def handle_pi(self, text):
-        # called for each processing instruction, e.g. <?instruction>
-        # Reconstruct original processing instruction.
-        self.pieces.append('<?%(text)s>' % locals())
-
-    def handle_decl(self, text):
-        # called for the DOCTYPE, if present, e.g.
-        # <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
-        #     "http://www.w3.org/TR/html4/loose.dtd">
-        # Reconstruct original DOCTYPE
-        self.pieces.append('<!%(text)s>' % locals())
-        
-    _new_declname_match = re.compile(r'[a-zA-Z][-_.a-zA-Z0-9:]*\s*').match
-    def _scan_name(self, i, declstartpos):
-        rawdata = self.rawdata
-        n = len(rawdata)
-        if i == n:
-            return None, -1
-        m = self._new_declname_match(rawdata, i)
-        if m:
-            s = m.group()
-            name = s.strip()
-            if (i + len(s)) == n:
-                return None, -1  # end of buffer
-            return name.lower(), m.end()
-        else:
-            self.handle_data(rawdata)
-#            self.updatepos(declstartpos, i)
-            return None, -1
-
-    def convert_charref(self, name):
-        return '&#%s;' % name
-
-    def convert_entityref(self, name):
-        return '&%s;' % name
-
-    def output(self):
-        '''Return processed HTML as a single string'''
-        return ''.join([str(p) for p in self.pieces])
-
-class _LooseFeedParser(_FeedParserMixin, _BaseHTMLProcessor):
-    def __init__(self, baseuri, baselang, encoding, entities):
-        sgmllib.SGMLParser.__init__(self)
-        _FeedParserMixin.__init__(self, baseuri, baselang, encoding)
-        _BaseHTMLProcessor.__init__(self, encoding, 'application/xhtml+xml')
-        self.entities=entities
-
-    def decodeEntities(self, element, data):
-        data = data.replace('&#60;', '&lt;')
-        data = data.replace('&#x3c;', '&lt;')
-        data = data.replace('&#x3C;', '&lt;')
-        data = data.replace('&#62;', '&gt;')
-        data = data.replace('&#x3e;', '&gt;')
-        data = data.replace('&#x3E;', '&gt;')
-        data = data.replace('&#38;', '&amp;')
-        data = data.replace('&#x26;', '&amp;')
-        data = data.replace('&#34;', '&quot;')
-        data = data.replace('&#x22;', '&quot;')
-        data = data.replace('&#39;', '&apos;')
-        data = data.replace('&#x27;', '&apos;')
-        if self.contentparams.has_key('type') and not self.contentparams.get('type', 'xml').endswith('xml'):
-            data = data.replace('&lt;', '<')
-            data = data.replace('&gt;', '>')
-            data = data.replace('&amp;', '&')
-            data = data.replace('&quot;', '"')
-            data = data.replace('&apos;', "'")
-        return data
-        
-    def strattrs(self, attrs):
-        return ''.join([' %s="%s"' % (n,v.replace('"','&quot;')) for n,v in attrs])
-
-class _MicroformatsParser:
-    STRING = 1
-    DATE = 2
-    URI = 3
-    NODE = 4
-    EMAIL = 5
-
-    known_xfn_relationships = ['contact', 'acquaintance', 'friend', 'met', 'co-worker', 'coworker', 'colleague', 'co-resident', 'coresident', 'neighbor', 'child', 'parent', 'sibling', 'brother', 'sister', 'spouse', 'wife', 'husband', 'kin', 'relative', 'muse', 'crush', 'date', 'sweetheart', 'me']
-    known_binary_extensions =  ['zip','rar','exe','gz','tar','tgz','tbz2','bz2','z','7z','dmg','img','sit','sitx','hqx','deb','rpm','bz2','jar','rar','iso','bin','msi','mp2','mp3','ogg','ogm','mp4','m4v','m4a','avi','wma','wmv']
-
-    def __init__(self, data, baseuri, encoding):
-        self.document = BeautifulSoup.BeautifulSoup(data)
-        self.baseuri = baseuri
-        self.encoding = encoding
-        if type(data) == type(u''):
-            data = data.encode(encoding)
-        self.tags = []
-        self.enclosures = []
-        self.xfn = []
-        self.vcard = None
-    
-    def vcardEscape(self, s):
-        if type(s) in (type(''), type(u'')):
-            s = s.replace(',', '\\,').replace(';', '\\;').replace('\n', '\\n')
-        return s
-    
-    def vcardFold(self, s):
-        s = re.sub(';+$', '', s)
-        sFolded = ''
-        iMax = 75
-        sPrefix = ''
-        while len(s) > iMax:
-            sFolded += sPrefix + s[:iMax] + '\n'
-            s = s[iMax:]
-            sPrefix = ' '
-            iMax = 74
-        sFolded += sPrefix + s
-        return sFolded
-
-    def normalize(self, s):
-        return re.sub(r'\s+', ' ', s).strip()
-    
-    def unique(self, aList):
-        results = []
-        for element in aList:
-            if element not in results:
-                results.append(element)
-        return results
-    
-    def toISO8601(self, dt):
-        return time.strftime('%Y-%m-%dT%H:%M:%SZ', dt)
-
-    def getPropertyValue(self, elmRoot, sProperty, iPropertyType=4, bAllowMultiple=0, bAutoEscape=0):
-        all = lambda x: 1
-        sProperty = sProperty.lower()
-        bFound = 0
-        bNormalize = 1
-        propertyMatch = re.compile(r'\b%s\b' % sProperty)
-        if bAllowMultiple and (iPropertyType != self.NODE):
-            snapResults = []
-            containers = elmRoot(['ul', 'ol'], propertyMatch)
-            for container in containers:
-                snapResults.extend(container('li'))
-            bFound = (len(snapResults) != 0)
-        if not bFound:
-            snapResults = elmRoot(all, propertyMatch)
-            bFound = (len(snapResults) != 0)
-        if (not bFound) and (sProperty == 'value'):
-            snapResults = elmRoot('pre')
-            bFound = (len(snapResults) != 0)
-            bNormalize = not bFound
-            if not bFound:
-                snapResults = [elmRoot]
-                bFound = (len(snapResults) != 0)
-        arFilter = []
-        if sProperty == 'vcard':
-            snapFilter = elmRoot(all, propertyMatch)
-            for node in snapFilter:
-                if node.findParent(all, propertyMatch):
-                    arFilter.append(node)
-        arResults = []
-        for node in snapResults:
-            if node not in arFilter:
-                arResults.append(node)
-        bFound = (len(arResults) != 0)
-        if not bFound:
-            if bAllowMultiple: return []
-            elif iPropertyType == self.STRING: return ''
-            elif iPropertyType == self.DATE: return BeautifulSoup.Null
-            elif iPropertyType == self.URI: return ''
-            elif iPropertyType == self.NODE: return BeautifulSoup.Null
-            else: return BeautifulSoup.Null
-        arValues = []
-        for elmResult in arResults:
-            sValue = BeautifulSoup.Null
-            if iPropertyType == self.NODE:
-                if bAllowMultiple:
-                    arValues.append(elmResult)
-                    continue
-                else:
-                    return elmResult
-            sNodeName = elmResult.name.lower()
-            if (iPropertyType == self.EMAIL) and (sNodeName == 'a'):
-                sValue = (elmResult.get('href') or '').split('mailto:').pop().split('?')[0]
-            if sValue:
-                sValue = bNormalize and self.normalize(sValue) or sValue.strip()
-            if (not sValue) and (sNodeName == 'abbr'):
-                sValue = elmResult.get('title')
-            if sValue:
-                sValue = bNormalize and self.normalize(sValue) or sValue.strip()
-            if (not sValue) and (iPropertyType == self.URI):
-                if sNodeName == 'a': sValue = elmResult.get('href')
-                elif sNodeName == 'img': sValue = elmResult.get('src')
-                elif sNodeName == 'object': sValue = elmResult.get('data')
-            if sValue:
-                sValue = bNormalize and self.normalize(sValue) or sValue.strip()
-            if (not sValue) and (sNodeName == 'img'):
-                sValue = elmResult.get('alt')
-            if sValue:
-                sValue = bNormalize and self.normalize(sValue) or sValue.strip()
-            if not sValue:
-                sValue = elmResult.renderContents()
-                sValue = re.sub(r'<\S[^>]*>', '', sValue)
-                sValue = sValue.replace('\r\n', '\n')
-                sValue = sValue.replace('\r', '\n')
-            if sValue:
-                sValue = bNormalize and self.normalize(sValue) or sValue.strip()
-            if not sValue: continue
-            if iPropertyType == self.DATE:
-                sValue = _parse_date_iso8601(sValue)
-            if bAllowMultiple:
-                arValues.append(bAutoEscape and self.vcardEscape(sValue) or sValue)
-            else:
-                return bAutoEscape and self.vcardEscape(sValue) or sValue
-        return arValues
-
-    def findVCards(self, elmRoot, bAgentParsing=0):
-        sVCards = ''
-        
-        if not bAgentParsing:
-            arCards = self.getPropertyValue(elmRoot, 'vcard', bAllowMultiple=1)
-        else:
-            arCards = [elmRoot]
-            
-        for elmCard in arCards:
-            arLines = []
-            
-            def processSingleString(sProperty):
-                sValue = self.getPropertyValue(elmCard, sProperty, self.STRING, bAutoEscape=1)
-                if sValue:
-                    arLines.append(self.vcardFold(sProperty.upper() + ':' + sValue))
-                return sValue or ''
-            
-            def processSingleURI(sProperty):
-                sValue = self.getPropertyValue(elmCard, sProperty, self.URI)
-                if sValue:
-                    sContentType = ''
-                    sEncoding = ''
-                    sValueKey = ''
-                    if sValue.startswith('data:'):
-                        sEncoding = ';ENCODING=b'
-                        sContentType = sValue.split(';')[0].split('/').pop()
-                        sValue = sValue.split(',', 1).pop()
-                    else:
-                        elmValue = self.getPropertyValue(elmCard, sProperty)
-                        if elmValue:
-                            if sProperty != 'url':
-                                sValueKey = ';VALUE=uri'
-                            sContentType = elmValue.get('type', '').strip().split('/').pop().strip()
-                    sContentType = sContentType.upper()
-                    if sContentType == 'OCTET-STREAM':
-                        sContentType = ''
-                    if sContentType:
-                        sContentType = ';TYPE=' + sContentType.upper()
-                    arLines.append(self.vcardFold(sProperty.upper() + sEncoding + sContentType + sValueKey + ':' + sValue))
-    
-            def processTypeValue(sProperty, arDefaultType, arForceType=None):
-                arResults = self.getPropertyValue(elmCard, sProperty, bAllowMultiple=1)
-                for elmResult in arResults:
-                    arType = self.getPropertyValue(elmResult, 'type', self.STRING, 1, 1)
-                    if arForceType:
-                        arType = self.unique(arForceType + arType)
-                    if not arType:
-                        arType = arDefaultType
-                    sValue = self.getPropertyValue(elmResult, 'value', self.EMAIL, 0)
-                    if sValue:
-                        arLines.append(self.vcardFold(sProperty.upper() + ';TYPE=' + ','.join(arType) + ':' + sValue))
-            
-            # AGENT
-            # must do this before all other properties because it is destructive
-            # (removes nested class="vcard" nodes so they don't interfere with
-            # this vcard's other properties)
-            arAgent = self.getPropertyValue(elmCard, 'agent', bAllowMultiple=1)
-            for elmAgent in arAgent:
-                if re.compile(r'\bvcard\b').search(elmAgent.get('class')):
-                    sAgentValue = self.findVCards(elmAgent, 1) + '\n'
-                    sAgentValue = sAgentValue.replace('\n', '\\n')
-                    sAgentValue = sAgentValue.replace(';', '\\;')
-                    if sAgentValue:
-                        arLines.append(self.vcardFold('AGENT:' + sAgentValue))
-                    elmAgent['class'] = ''
-                    elmAgent.contents = BeautifulSoup.Null
-                else:
-                    sAgentValue = self.getPropertyValue(elmAgent, 'value', self.URI, bAutoEscape=1);
-                    if sAgentValue:
-                        arLines.append(self.vcardFold('AGENT;VALUE=uri:' + sAgentValue))
-    
-            # FN (full name)
-            sFN = processSingleString('fn')
-            
-            # N (name)
-            elmName = self.getPropertyValue(elmCard, 'n')
-            if elmName:
-                sFamilyName = self.getPropertyValue(elmName, 'family-name', self.STRING, bAutoEscape=1)
-                sGivenName = self.getPropertyValue(elmName, 'given-name', self.STRING, bAutoEscape=1)
-                arAdditionalNames = self.getPropertyValue(elmName, 'additional-name', self.STRING, 1, 1) + self.getPropertyValue(elmName, 'additional-names', self.STRING, 1, 1)
-                arHonorificPrefixes = self.getPropertyValue(elmName, 'honorific-prefix', self.STRING, 1, 1) + self.getPropertyValue(elmName, 'honorific-prefixes', self.STRING, 1, 1)
-                arHonorificSuffixes = self.getPropertyValue(elmName, 'honorific-suffix', self.STRING, 1, 1) + self.getPropertyValue(elmName, 'honorific-suffixes', self.STRING, 1, 1)
-                arLines.append(self.vcardFold('N:' + sFamilyName + ';' + 
-                                         sGivenName + ';' +
-                                         ','.join(arAdditionalNames) + ';' +
-                                         ','.join(arHonorificPrefixes) + ';' +
-                                         ','.join(arHonorificSuffixes)))
-            elif sFN:
-                # implied "N" optimization
-                # http://microformats.org/wiki/hcard#Implied_.22N.22_Optimization
-                arNames = self.normalize(sFN).split()
-                if len(arNames) == 2:
-                    bFamilyNameFirst = (arNames[0].endswith(',') or
-                                        len(arNames[1]) == 1 or
-                                        ((len(arNames[1]) == 2) and (arNames[1].endswith('.'))))
-                    if bFamilyNameFirst:
-                        arLines.append(self.vcardFold('N:' + arNames[0] + ';' + arNames[1]))
-                    else:
-                        arLines.append(self.vcardFold('N:' + arNames[1] + ';' + arNames[0]))
-    
-            # SORT-STRING
-            sSortString = self.getPropertyValue(elmCard, 'sort-string', self.STRING, bAutoEscape=1)
-            if sSortString:
-                arLines.append(self.vcardFold('SORT-STRING:' + sSortString))
-            
-            # NICKNAME
-            arNickname = self.getPropertyValue(elmCard, 'nickname', self.STRING, 1, 1)
-            if arNickname:
-                arLines.append(self.vcardFold('NICKNAME:' + ','.join(arNickname)))
-            
-            # PHOTO
-            processSingleURI('photo')
-            
-            # BDAY
-            dtBday = self.getPropertyValue(elmCard, 'bday', self.DATE)
-            if dtBday:
-                arLines.append(self.vcardFold('BDAY:' + self.toISO8601(dtBday)))
-            
-            # ADR (address)
-            arAdr = self.getPropertyValue(elmCard, 'adr', bAllowMultiple=1)
-            for elmAdr in arAdr:
-                arType = self.getPropertyValue(elmAdr, 'type', self.STRING, 1, 1)
-                if not arType:
-                    arType = ['intl','postal','parcel','work'] # default adr types, see RFC 2426 section 3.2.1
-                sPostOfficeBox = self.getPropertyValue(elmAdr, 'post-office-box', self.STRING, 0, 1)
-                sExtendedAddress = self.getPropertyValue(elmAdr, 'extended-address', self.STRING, 0, 1)
-                sStreetAddress = self.getPropertyValue(elmAdr, 'street-address', self.STRING, 0, 1)
-                sLocality = self.getPropertyValue(elmAdr, 'locality', self.STRING, 0, 1)
-                sRegion = self.getPropertyValue(elmAdr, 'region', self.STRING, 0, 1)
-                sPostalCode = self.getPropertyValue(elmAdr, 'postal-code', self.STRING, 0, 1)
-                sCountryName = self.getPropertyValue(elmAdr, 'country-name', self.STRING, 0, 1)
-                arLines.append(self.vcardFold('ADR;TYPE=' + ','.join(arType) + ':' +
-                                         sPostOfficeBox + ';' +
-                                         sExtendedAddress + ';' +
-                                         sStreetAddress + ';' +
-                                         sLocality + ';' +
-                                         sRegion + ';' +
-                                         sPostalCode + ';' +
-                                         sCountryName))
-            
-            # LABEL
-            processTypeValue('label', ['intl','postal','parcel','work'])
-            
-            # TEL (phone number)
-            processTypeValue('tel', ['voice'])
-            
-            # EMAIL
-            processTypeValue('email', ['internet'], ['internet'])
-            
-            # MAILER
-            processSingleString('mailer')
-            
-            # TZ (timezone)
-            processSingleString('tz')
-    
-            # GEO (geographical information)
-            elmGeo = self.getPropertyValue(elmCard, 'geo')
-            if elmGeo:
-                sLatitude = s