':
depth = 0
for piece in pieces[:-1]:
if piece.startswith(''):
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
if element == 'title' and self.hasTitle:
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'',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' -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.hasTitle = 0
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.hasTitle = 0
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('<>', '')
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
self.hasTitle = 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')))
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):
self.pop('link')
_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
self.hasTitle = 1
_end_dc_title = _end_title
def _end_media_title(self):
hasTitle = self.hasTitle
self._end_title()
self.hasTitle = hasTitle
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:
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
self.hasTitle = 0
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 + '>'
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'', self._shorttag_replace, data) # bug [ 1399464 ] Bad regexp for _shorttag_replace
data = re.sub(r'<([^<\s]+?)\s*/>', self._shorttag_replace, data)
data = data.replace(''', "'")
data = data.replace('"', '"')
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
, 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('>','>').replace('<','<').replace('"','"')
value = self.bare_ampersand.sub("&", 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 , tag will be 'pre'
# Reconstruct the original end tag.
if tag not in self.elements_no_end_tag:
self.pieces.append("" % locals())
def handle_charref(self, ref):
# called for each character reference, e.g. for ' ', 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 '©', ref will be 'copy'
# Reconstruct the original entity reference.
if name2codepoint.has_key(ref):
self.pieces.append('&%(ref)s;' % locals())
else:
self.pieces.append('&%(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.
# Reconstruct the original comment.
self.pieces.append('' % locals())
def handle_pi(self, text):
# called for each processing instruction, e.g.
# Reconstruct original processing instruction.
self.pieces.append('' % locals())
def handle_decl(self, text):
# called for the DOCTYPE, if present, e.g.
#
# Reconstruct original DOCTYPE
self.pieces.append('' % 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('<', '<')
data = data.replace('<', '<')
data = data.replace('<', '<')
data = data.replace('>', '>')
data = data.replace('>', '>')
data = data.replace('>', '>')
data = data.replace('&', '&')
data = data.replace('&', '&')
data = data.replace('"', '"')
data = data.replace('"', '"')
data = data.replace(''', ''')
data = data.replace(''', ''')
if self.contentparams.has_key('type') and not self.contentparams.get('type', 'xml').endswith('xml'):
data = data.replace('<', '<')
data = data.replace('>', '>')
data = data.replace('&', '&')
data = data.replace('"', '"')
data = data.replace(''', "'")
return data
def strattrs(self, attrs):
return ''.join([' %s="%s"' % (n,v.replace('"','"')) 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 = {'class': 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 None
elif iPropertyType == self.URI: return ''
elif iPropertyType == self.NODE: return None
else: return None
arValues = []
for elmResult in arResults:
sValue = None
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 = []
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 = self.getPropertyValue(elmGeo, 'latitude', self.STRING, 0, 1)
sLongitude = self.getPropertyValue(elmGeo, 'longitude', self.STRING, 0, 1)
arLines.append(self.vcardFold('GEO:' + sLatitude + ';' + sLongitude))
# TITLE
processSingleString('title')
# ROLE
processSingleString('role')
# LOGO
processSingleURI('logo')
# ORG (organization)
elmOrg = self.getPropertyValue(elmCard, 'org')
if elmOrg:
sOrganizationName = self.getPropertyValue(elmOrg, 'organization-name', self.STRING, 0, 1)
if not sOrganizationName:
# implied "organization-name" optimization
# http://microformats.org/wiki/hcard#Implied_.22organization-name.22_Optimization
sOrganizationName = self.getPropertyValue(elmCard, 'org', self.STRING, 0, 1)
if sOrganizationName:
arLines.append(self.vcardFold('ORG:' + sOrganizationName))
else:
arOrganizationUnit = self.getPropertyValue(elmOrg, 'organization-unit', self.STRING, 1, 1)
arLines.append(self.vcardFold('ORG:' + sOrganizationName + ';' + ';'.join(arOrganizationUnit)))
# CATEGORY
arCategory = self.getPropertyValue(elmCard, 'category', self.STRING, 1, 1) + self.getPropertyValue(elmCard, 'categories', self.STRING, 1, 1)
if arCategory:
arLines.append(self.vcardFold('CATEGORIES:' + ','.join(arCategory)))
# NOTE
processSingleString('note')
# REV
processSingleString('rev')
# SOUND
processSingleURI('sound')
# UID
processSingleString('uid')
# URL
processSingleURI('url')
# CLASS
processSingleString('class')
# KEY
processSingleURI('key')
if arLines:
arLines = ['BEGIN:vCard','VERSION:3.0'] + arLines + ['END:vCard']
sVCards += '\n'.join(arLines) + '\n'
return sVCards.strip()
def isProbablyDownloadable(self, elm):
attrsD = elm.attrMap
if not attrsD.has_key('href'): return 0
linktype = attrsD.get('type', '').strip()
if linktype.startswith('audio/') or \
linktype.startswith('video/') or \
(linktype.startswith('application/') and not linktype.endswith('xml')):
return 1
path = urlparse.urlparse(attrsD['href'])[2]
if path.find('.') == -1: return 0
fileext = path.split('.').pop().lower()
return fileext in self.known_binary_extensions
def findTags(self):
all = lambda x: 1
for elm in self.document(all, {'rel': re.compile(r'\btag\b')}):
href = elm.get('href')
if not href: continue
urlscheme, domain, path, params, query, fragment = \
urlparse.urlparse(_urljoin(self.baseuri, href))
segments = path.split('/')
tag = segments.pop()
if not tag:
tag = segments.pop()
tagscheme = urlparse.urlunparse((urlscheme, domain, '/'.join(segments), '', '', ''))
if not tagscheme.endswith('/'):
tagscheme += '/'
self.tags.append(FeedParserDict({"term": tag, "scheme": tagscheme, "label": elm.string or ''}))
def findEnclosures(self):
all = lambda x: 1
enclosure_match = re.compile(r'\benclosure\b')
for elm in self.document(all, {'href': re.compile(r'.+')}):
if not enclosure_match.search(elm.get('rel', '')) and not self.isProbablyDownloadable(elm): continue
if elm.attrMap not in self.enclosures:
self.enclosures.append(elm.attrMap)
if elm.string and not elm.get('title'):
self.enclosures[-1]['title'] = elm.string
def findXFN(self):
all = lambda x: 1
for elm in self.document(all, {'rel': re.compile('.+'), 'href': re.compile('.+')}):
rels = elm.get('rel', '').split()
xfn_rels = []
for rel in rels:
if rel in self.known_xfn_relationships:
xfn_rels.append(rel)
if xfn_rels:
self.xfn.append({"relationships": xfn_rels, "href": elm.get('href', ''), "name": elm.string})
def _parseMicroformats(htmlSource, baseURI, encoding):
if not BeautifulSoup: return
if _debug: sys.stderr.write('entering _parseMicroformats\n')
p = _MicroformatsParser(htmlSource, baseURI, encoding)
p.vcard = p.findVCards(p.document)
p.findTags()
p.findEnclosures()
p.findXFN()
return {"tags": p.tags, "enclosures": p.enclosures, "xfn": p.xfn, "vcard": p.vcard}
class _RelativeURIResolver(_BaseHTMLProcessor):
relative_uris = [('a', 'href'),
('applet', 'codebase'),
('area', 'href'),
('blockquote', 'cite'),
('body', 'background'),
('del', 'cite'),
('form', 'action'),
('frame', 'longdesc'),
('frame', 'src'),
('iframe', 'longdesc'),
('iframe', 'src'),
('head', 'profile'),
('img', 'longdesc'),
('img', 'src'),
('img', 'usemap'),
('input', 'src'),
('input', 'usemap'),
('ins', 'cite'),
('link', 'href'),
('object', 'classid'),
('object', 'codebase'),
('object', 'data'),
('object', 'usemap'),
('q', 'cite'),
('script', 'src')]
def __init__(self, baseuri, encoding, type):
_BaseHTMLProcessor.__init__(self, encoding, type)
self.baseuri = baseuri
def resolveURI(self, uri):
return _urljoin(self.baseuri, uri.strip())
def unknown_starttag(self, tag, attrs):
attrs = self.normalize_attrs(attrs)
attrs = [(key, ((tag, key) in self.relative_uris) and self.resolveURI(value) or value) for key, value in attrs]
_BaseHTMLProcessor.unknown_starttag(self, tag, attrs)
def _resolveRelativeURIs(htmlSource, baseURI, encoding, type):
if _debug: sys.stderr.write('entering _resolveRelativeURIs\n')
p = _RelativeURIResolver(baseURI, encoding, type)
p.feed(htmlSource)
return p.output()
class _HTMLSanitizer(_BaseHTMLProcessor):
acceptable_elements = ['a', 'abbr', 'acronym', 'address', 'area', 'article',
'aside', 'audio', 'b', 'big', 'blockquote', 'br', 'button', 'canvas',
'caption', 'center', 'cite', 'code', 'col', 'colgroup', 'command',
'datagrid', 'datalist', 'dd', 'del', 'details', 'dfn', 'dialog', 'dir',
'div', 'dl', 'dt', 'em', 'event-source', 'fieldset', 'figure', 'footer',
'font', 'form', 'header', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'hr', 'i',
'img', 'input', 'ins', 'keygen', 'kbd', 'label', 'legend', 'li', 'm', 'map',
'menu', 'meter', 'multicol', 'nav', 'nextid', 'ol', 'output', 'optgroup',
'option', 'p', 'pre', 'progress', 'q', 's', 'samp', 'section', 'select',
'small', 'sound', 'source', 'spacer', 'span', 'strike', 'strong', 'sub',
'sup', 'table', 'tbody', 'td', 'textarea', 'time', 'tfoot', 'th', 'thead',
'tr', 'tt', 'u', 'ul', 'var', 'video', 'noscript']
acceptable_attributes = ['abbr', 'accept', 'accept-charset', 'accesskey',
'action', 'align', 'alt', 'autoplay', 'autocomplete', 'autofocus', 'axis',
'background', 'balance', 'bgcolor', 'bgproperties', 'border',
'bordercolor', 'bordercolordark', 'bordercolorlight', 'bottompadding',
'cellpadding', 'cellspacing', 'ch', 'challenge', 'char', 'charoff',
'choff', 'charset', 'checked', 'cite', 'class', 'clear', 'color', 'cols',
'colspan', 'compact', 'contenteditable', 'coords', 'data', 'datafld',
'datapagesize', 'datasrc', 'datetime', 'default', 'delay', 'dir',
'disabled', 'draggable', 'dynsrc', 'enctype', 'end', 'face', 'for',
'form', 'frame', 'galleryimg', 'gutter', 'headers', 'height', 'hidefocus',
'hidden', 'high', 'href', 'hreflang', 'hspace', 'icon', 'id', 'inputmode',
'ismap', 'keytype', 'label', 'leftspacing', 'lang', 'list', 'longdesc',
'loop', 'loopcount', 'loopend', 'loopstart', 'low', 'lowsrc', 'max',
'maxlength', 'media', 'method', 'min', 'multiple', 'name', 'nohref',
'noshade', 'nowrap', 'open', 'optimum', 'pattern', 'ping', 'point-size',
'prompt', 'pqg', 'radiogroup', 'readonly', 'rel', 'repeat-max',
'repeat-min', 'replace', 'required', 'rev', 'rightspacing', 'rows',
'rowspan', 'rules', 'scope', 'selected', 'shape', 'size', 'span', 'src',
'start', 'step', 'summary', 'suppress', 'tabindex', 'target', 'template',
'title', 'toppadding', 'type', 'unselectable', 'usemap', 'urn', 'valign',
'value', 'variable', 'volume', 'vspace', 'vrml', 'width', 'wrap',
'xml:lang']
unacceptable_elements_with_end_tag = ['script', 'applet', 'style']
acceptable_css_properties = ['azimuth', 'background-color',
'border-bottom-color', 'border-collapse', 'border-color',
'border-left-color', 'border-right-color', 'border-top-color', 'clear',
'color', 'cursor', 'direction', 'display', 'elevation', 'float', 'font',
'font-family', 'font-size', 'font-style', 'font-variant', 'font-weight',
'height', 'letter-spacing', 'line-height', 'overflow', 'pause',
'pause-after', 'pause-before', 'pitch', 'pitch-range', 'richness',
'speak', 'speak-header', 'speak-numeral', 'speak-punctuation',
'speech-rate', 'stress', 'text-align', 'text-decoration', 'text-indent',
'unicode-bidi', 'vertical-align', 'voice-family', 'volume',
'white-space', 'width']
# survey of common keywords found in feeds
acceptable_css_keywords = ['auto', 'aqua', 'black', 'block', 'blue',
'bold', 'both', 'bottom', 'brown', 'center', 'collapse', 'dashed',
'dotted', 'fuchsia', 'gray', 'green', '!important', 'italic', 'left',
'lime', 'maroon', 'medium', 'none', 'navy', 'normal', 'nowrap', 'olive',
'pointer', 'purple', 'red', 'right', 'solid', 'silver', 'teal', 'top',
'transparent', 'underline', 'white', 'yellow']
valid_css_values = re.compile('^(#[0-9a-f]+|rgb\(\d+%?,\d*%?,?\d*%?\)?|' +
'\d{0,2}\.?\d{0,2}(cm|em|ex|in|mm|pc|pt|px|%|,|\))?)$')
mathml_elements = ['annotation', 'annotation-xml', 'maction', 'math',
'merror', 'mfenced', 'mfrac', 'mi', 'mmultiscripts', 'mn', 'mo', 'mover', 'mpadded',
'mphantom', 'mprescripts', 'mroot', 'mrow', 'mspace', 'msqrt', 'mstyle',
'msub', 'msubsup', 'msup', 'mtable', 'mtd', 'mtext', 'mtr', 'munder',
'munderover', 'none', 'semantics']
mathml_attributes = ['actiontype', 'align', 'columnalign', 'columnalign',
'columnalign', 'close', 'columnlines', 'columnspacing', 'columnspan', 'depth',
'display', 'displaystyle', 'encoding', 'equalcolumns', 'equalrows',
'fence', 'fontstyle', 'fontweight', 'frame', 'height', 'linethickness',
'lspace', 'mathbackground', 'mathcolor', 'mathvariant', 'mathvariant',
'maxsize', 'minsize', 'open', 'other', 'rowalign', 'rowalign', 'rowalign',
'rowlines', 'rowspacing', 'rowspan', 'rspace', 'scriptlevel', 'selection',
'separator', 'separators', 'stretchy', 'width', 'width', 'xlink:href',
'xlink:show', 'xlink:type', 'xmlns', 'xmlns:xlink']
# svgtiny - foreignObject + linearGradient + radialGradient + stop
svg_elements = ['a', 'animate', 'animateColor', 'animateMotion',
'animateTransform', 'circle', 'defs', 'desc', 'ellipse', 'foreignObject',
'font-face', 'font-face-name', 'font-face-src', 'g', 'glyph', 'hkern',
'linearGradient', 'line', 'marker', 'metadata', 'missing-glyph', 'mpath',
'path', 'polygon', 'polyline', 'radialGradient', 'rect', 'set', 'stop',
'svg', 'switch', 'text', 'title', 'tspan', 'use']
# svgtiny + class + opacity + offset + xmlns + xmlns:xlink
svg_attributes = ['accent-height', 'accumulate', 'additive', 'alphabetic',
'arabic-form', 'ascent', 'attributeName', 'attributeType',
'baseProfile', 'bbox', 'begin', 'by', 'calcMode', 'cap-height',
'class', 'color', 'color-rendering', 'content', 'cx', 'cy', 'd', 'dx',
'dy', 'descent', 'display', 'dur', 'end', 'fill', 'fill-opacity',
'fill-rule', 'font-family', 'font-size', 'font-stretch', 'font-style',
'font-variant', 'font-weight', 'from', 'fx', 'fy', 'g1', 'g2',
'glyph-name', 'gradientUnits', 'hanging', 'height', 'horiz-adv-x',
'horiz-origin-x', 'id', 'ideographic', 'k', 'keyPoints', 'keySplines',
'keyTimes', 'lang', 'mathematical', 'marker-end', 'marker-mid',
'marker-start', 'markerHeight', 'markerUnits', 'markerWidth', 'max',
'min', 'name', 'offset', 'opacity', 'orient', 'origin',
'overline-position', 'overline-thickness', 'panose-1', 'path',
'pathLength', 'points', 'preserveAspectRatio', 'r', 'refX', 'refY',
'repeatCount', 'repeatDur', 'requiredExtensions', 'requiredFeatures',
'restart', 'rotate', 'rx', 'ry', 'slope', 'stemh', 'stemv',
'stop-color', 'stop-opacity', 'strikethrough-position',
'strikethrough-thickness', 'stroke', 'stroke-dasharray',
'stroke-dashoffset', 'stroke-linecap', 'stroke-linejoin',
'stroke-miterlimit', 'stroke-opacity', 'stroke-width', 'systemLanguage',
'target', 'text-anchor', 'to', 'transform', 'type', 'u1', 'u2',
'underline-position', 'underline-thickness', 'unicode', 'unicode-range',
'units-per-em', 'values', 'version', 'viewBox', 'visibility', 'width',
'widths', 'x', 'x-height', 'x1', 'x2', 'xlink:actuate', 'xlink:arcrole',
'xlink:href', 'xlink:role', 'xlink:show', 'xlink:title', 'xlink:type',
'xml:base', 'xml:lang', 'xml:space', 'xmlns', 'xmlns:xlink', 'y', 'y1',
'y2', 'zoomAndPan']
svg_attr_map = None
svg_elem_map = None
acceptable_svg_properties = [ 'fill', 'fill-opacity', 'fill-rule',
'stroke', 'stroke-width', 'stroke-linecap', 'stroke-linejoin',
'stroke-opacity']
def reset(self):
_BaseHTMLProcessor.reset(self)
self.unacceptablestack = 0
self.mathmlOK = 0
self.svgOK = 0
def unknown_starttag(self, tag, attrs):
acceptable_attributes = self.acceptable_attributes
keymap = {}
if not tag in self.acceptable_elements or self.svgOK:
if tag in self.unacceptable_elements_with_end_tag:
self.unacceptablestack += 1
# not otherwise acceptable, perhaps it is MathML or SVG?
if tag=='math' and ('xmlns','http://www.w3.org/1998/Math/MathML') in attrs:
self.mathmlOK = 1
if tag=='svg' and ('xmlns','http://www.w3.org/2000/svg') in attrs:
self.svgOK = 1
# chose acceptable attributes based on tag class, else bail
if self.mathmlOK and tag in self.mathml_elements:
acceptable_attributes = self.mathml_attributes
elif self.svgOK and tag in self.svg_elements:
# for most vocabularies, lowercasing is a good idea. Many
# svg elements, however, are camel case
if not self.svg_attr_map:
lower=[attr.lower() for attr in self.svg_attributes]
mix=[a for a in self.svg_attributes if a not in lower]
self.svg_attributes = lower
self.svg_attr_map = dict([(a.lower(),a) for a in mix])
lower=[attr.lower() for attr in self.svg_elements]
mix=[a for a in self.svg_elements if a not in lower]
self.svg_elements = lower
self.svg_elem_map = dict([(a.lower(),a) for a in mix])
acceptable_attributes = self.svg_attributes
tag = self.svg_elem_map.get(tag,tag)
keymap = self.svg_attr_map
elif not tag in self.acceptable_elements:
return
# declare xlink namespace, if needed
if self.mathmlOK or self.svgOK:
if filter(lambda (n,v): n.startswith('xlink:'),attrs):
if not ('xmlns:xlink','http://www.w3.org/1999/xlink') in attrs:
attrs.append(('xmlns:xlink','http://www.w3.org/1999/xlink'))
clean_attrs = []
for key, value in self.normalize_attrs(attrs):
if key in acceptable_attributes:
key=keymap.get(key,key)
clean_attrs.append((key,value))
elif key=='style':
clean_value = self.sanitize_style(value)
if clean_value: clean_attrs.append((key,clean_value))
_BaseHTMLProcessor.unknown_starttag(self, tag, clean_attrs)
def unknown_endtag(self, tag):
if not tag in self.acceptable_elements:
if tag in self.unacceptable_elements_with_end_tag:
self.unacceptablestack -= 1
if self.mathmlOK and tag in self.mathml_elements:
if tag == 'math': self.mathmlOK = 0
elif self.svgOK and tag in self.svg_elements:
tag = self.svg_elem_map.get(tag,tag)
if tag == 'svg': self.svgOK = 0
else:
return
_BaseHTMLProcessor.unknown_endtag(self, tag)
def handle_pi(self, text):
pass
def handle_decl(self, text):
pass
def handle_data(self, text):
if not self.unacceptablestack:
_BaseHTMLProcessor.handle_data(self, text)
def sanitize_style(self, style):
# disallow urls
style=re.compile('url\s*\(\s*[^\s)]+?\s*\)\s*').sub(' ',style)
# gauntlet
if not re.match("""^([:,;#%.\sa-zA-Z0-9!]|\w-\w|'[\s\w]+'|"[\s\w]+"|\([\d,\s]+\))*$""", style): return ''
if not re.match("^(\s*[-\w]+\s*:\s*[^:;]*(;|$))*$", style): return ''
clean = []
for prop,value in re.findall("([-\w]+)\s*:\s*([^:;]*)",style):
if not value: continue
if prop.lower() in self.acceptable_css_properties:
clean.append(prop + ': ' + value + ';')
elif prop.split('-')[0].lower() in ['background','border','margin','padding']:
for keyword in value.split():
if not keyword in self.acceptable_css_keywords and \
not self.valid_css_values.match(keyword):
break
else:
clean.append(prop + ': ' + value + ';')
elif self.svgOK and prop.lower() in self.acceptable_svg_properties:
clean.append(prop + ': ' + value + ';')
return ' '.join(clean)
def _sanitizeHTML(htmlSource, encoding, type):
p = _HTMLSanitizer(encoding, type)
p.feed(htmlSource)
data = p.output()
if TIDY_MARKUP:
# loop through list of preferred Tidy interfaces looking for one that's installed,
# then set up a common _tidy function to wrap the interface-specific API.
_tidy = None
for tidy_interface in PREFERRED_TIDY_INTERFACES:
try:
if tidy_interface == "uTidy":
from tidy import parseString as _utidy
def _tidy(data, **kwargs):
return str(_utidy(data, **kwargs))
break
elif tidy_interface == "mxTidy":
from mx.Tidy import Tidy as _mxtidy
def _tidy(data, **kwargs):
nerrors, nwarnings, data, errordata = _mxtidy.tidy(data, **kwargs)
return data
break
except:
pass
if _tidy:
utf8 = type(data) == type(u'')
if utf8:
data = data.encode('utf-8')
data = _tidy(data, output_xhtml=1, numeric_entities=1, wrap=0, char_encoding="utf8")
if utf8:
data = unicode(data, 'utf-8')
if data.count('
'):
data = data.split('>', 1)[1]
if data.count('= '2.3.3'
assert base64 != None
user, passw = base64.decodestring(req.headers['Authorization'].split(' ')[1]).split(':')
realm = re.findall('realm="([^"]*)"', headers['WWW-Authenticate'])[0]
self.add_password(realm, host, user, passw)
retry = self.http_error_auth_reqed('www-authenticate', host, req, headers)
self.reset_retry_count()
return retry
except:
return self.http_error_default(req, fp, code, msg, headers)
def _open_resource(url_file_stream_or_string, etag, modified, agent, referrer, handlers):
"""URL, filename, or string --> stream
This function lets you define parsers that take any input source
(URL, pathname to local or network file, or actual data as a string)
and deal with it in a uniform manner. Returned object is guaranteed
to have all the basic stdio read methods (read, readline, readlines).
Just .close() the object when you're done with it.
If the etag argument is supplied, it will be used as the value of an
If-None-Match request header.
If the modified argument is supplied, it can be a tuple of 9 integers
(as returned by gmtime() in the standard Python time module) or a date
string in any format supported by feedparser. Regardless, it MUST
be in GMT (Greenwich Mean Time). It will be reformatted into an
RFC 1123-compliant date and used as the value of an If-Modified-Since
request header.
If the agent argument is supplied, it will be used as the value of a
User-Agent request header.
If the referrer argument is supplied, it will be used as the value of a
Referer[sic] request header.
If handlers is supplied, it is a list of handlers used to build a
urllib2 opener.
"""
if hasattr(url_file_stream_or_string, 'read'):
return url_file_stream_or_string
if url_file_stream_or_string == '-':
return sys.stdin
if urlparse.urlparse(url_file_stream_or_string)[0] in ('http', 'https', 'ftp'):
if not agent:
agent = USER_AGENT
# test for inline user:password for basic auth
auth = None
if base64:
urltype, rest = urllib.splittype(url_file_stream_or_string)
realhost, rest = urllib.splithost(rest)
if realhost:
user_passwd, realhost = urllib.splituser(realhost)
if user_passwd:
url_file_stream_or_string = '%s://%s%s' % (urltype, realhost, rest)
auth = base64.encodestring(user_passwd).strip()
# iri support
try:
if isinstance(url_file_stream_or_string,unicode):
url_file_stream_or_string = url_file_stream_or_string.encode('idna')
else:
url_file_stream_or_string = url_file_stream_or_string.decode('utf-8').encode('idna')
except:
pass
# try to open with urllib2 (to use optional headers)
request = urllib2.Request(url_file_stream_or_string)
request.add_header('User-Agent', agent)
if etag:
request.add_header('If-None-Match', etag)
if type(modified) == type(''):
modified = _parse_date(modified)
if modified:
# format into an RFC 1123-compliant timestamp. We can't use
# time.strftime() since the %a and %b directives can be affected
# by the current locale, but RFC 2616 states that dates must be
# in English.
short_weekdays = ['Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat', 'Sun']
months = ['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec']
request.add_header('If-Modified-Since', '%s, %02d %s %04d %02d:%02d:%02d GMT' % (short_weekdays[modified[6]], modified[2], months[modified[1] - 1], modified[0], modified[3], modified[4], modified[5]))
if referrer:
request.add_header('Referer', referrer)
if gzip and zlib:
request.add_header('Accept-encoding', 'gzip, deflate')
elif gzip:
request.add_header('Accept-encoding', 'gzip')
elif zlib:
request.add_header('Accept-encoding', 'deflate')
else:
request.add_header('Accept-encoding', '')
if auth:
request.add_header('Authorization', 'Basic %s' % auth)
if ACCEPT_HEADER:
request.add_header('Accept', ACCEPT_HEADER)
request.add_header('A-IM', 'feed') # RFC 3229 support
opener = apply(urllib2.build_opener, tuple([_FeedURLHandler()] + handlers))
opener.addheaders = [] # RMK - must clear so we only send our custom User-Agent
try:
return opener.open(request)
finally:
opener.close() # JohnD
# try to open with native open function (if url_file_stream_or_string is a filename)
try:
return open(url_file_stream_or_string)
except:
pass
# treat url_file_stream_or_string as string
return _StringIO(str(url_file_stream_or_string))
_date_handlers = []
def registerDateHandler(func):
'''Register a date handler function (takes string, returns 9-tuple date in GMT)'''
_date_handlers.insert(0, func)
# ISO-8601 date parsing routines written by Fazal Majid.
# The ISO 8601 standard is very convoluted and irregular - a full ISO 8601
# parser is beyond the scope of feedparser and would be a worthwhile addition
# to the Python library.
# A single regular expression cannot parse ISO 8601 date formats into groups
# as the standard is highly irregular (for instance is 030104 2003-01-04 or
# 0301-04-01), so we use templates instead.
# Please note the order in templates is significant because we need a
# greedy match.
_iso8601_tmpl = ['YYYY-?MM-?DD', 'YYYY-0MM?-?DD', 'YYYY-MM', 'YYYY-?OOO',
'YY-?MM-?DD', 'YY-?OOO', 'YYYY',
'-YY-?MM', '-OOO', '-YY',
'--MM-?DD', '--MM',
'---DD',
'CC', '']
_iso8601_re = [
tmpl.replace(
'YYYY', r'(?P
\d{4})').replace(
'YY', r'(?P\d\d)').replace(
'MM', r'(?P[01]\d)').replace(
'DD', r'(?P[0123]\d)').replace(
'OOO', r'(?P[0123]\d\d)').replace(
'CC', r'(?P\d\d$)')
+ r'(T?(?P\d{2}):(?P\d{2})'
+ r'(:(?P\d{2}(\.\d*)?))?'
+ r'(?P[+-](?P\d{2})(:(?P\d{2}))?|Z)?)?'
for tmpl in _iso8601_tmpl]
del tmpl
_iso8601_matches = [re.compile(regex).match for regex in _iso8601_re]
del regex
def _parse_date_iso8601(dateString):
'''Parse a variety of ISO-8601-compatible formats like 20040105'''
m = None
for _iso8601_match in _iso8601_matches:
m = _iso8601_match(dateString)
if m: break
if not m: return
if m.span() == (0, 0): return
params = m.groupdict()
ordinal = params.get('ordinal', 0)
if ordinal:
ordinal = int(ordinal)
else:
ordinal = 0
year = params.get('year', '--')
if not year or year == '--':
year = time.gmtime()[0]
elif len(year) == 2:
# ISO 8601 assumes current century, i.e. 93 -> 2093, NOT 1993
year = 100 * int(time.gmtime()[0] / 100) + int(year)
else:
year = int(year)
month = params.get('month', '-')
if not month or month == '-':
# ordinals are NOT normalized by mktime, we simulate them
# by setting month=1, day=ordinal
if ordinal:
month = 1
else:
month = time.gmtime()[1]
month = int(month)
day = params.get('day', 0)
if not day:
# see above
if ordinal:
day = ordinal
elif params.get('century', 0) or \
params.get('year', 0) or params.get('month', 0):
day = 1
else:
day = time.gmtime()[2]
else:
day = int(day)
# special case of the century - is the first year of the 21st century
# 2000 or 2001 ? The debate goes on...
if 'century' in params.keys():
year = (int(params['century']) - 1) * 100 + 1
# in ISO 8601 most fields are optional
for field in ['hour', 'minute', 'second', 'tzhour', 'tzmin']:
if not params.get(field, None):
params[field] = 0
hour = int(params.get('hour', 0))
minute = int(params.get('minute', 0))
second = int(float(params.get('second', 0)))
# weekday is normalized by mktime(), we can ignore it
weekday = 0
daylight_savings_flag = -1
tm = [year, month, day, hour, minute, second, weekday,
ordinal, daylight_savings_flag]
# ISO 8601 time zone adjustments
tz = params.get('tz')
if tz and tz != 'Z':
if tz[0] == '-':
tm[3] += int(params.get('tzhour', 0))
tm[4] += int(params.get('tzmin', 0))
elif tz[0] == '+':
tm[3] -= int(params.get('tzhour', 0))
tm[4] -= int(params.get('tzmin', 0))
else:
return None
# Python's time.mktime() is a wrapper around the ANSI C mktime(3c)
# which is guaranteed to normalize d/m/y/h/m/s.
# Many implementations have bugs, but we'll pretend they don't.
return time.localtime(time.mktime(tm))
registerDateHandler(_parse_date_iso8601)
# 8-bit date handling routines written by ytrewq1.
_korean_year = u'\ub144' # b3e2 in euc-kr
_korean_month = u'\uc6d4' # bff9 in euc-kr
_korean_day = u'\uc77c' # c0cf in euc-kr
_korean_am = u'\uc624\uc804' # bfc0 c0fc in euc-kr
_korean_pm = u'\uc624\ud6c4' # bfc0 c8c4 in euc-kr
_korean_onblog_date_re = \
re.compile('(\d{4})%s\s+(\d{2})%s\s+(\d{2})%s\s+(\d{2}):(\d{2}):(\d{2})' % \
(_korean_year, _korean_month, _korean_day))
_korean_nate_date_re = \
re.compile(u'(\d{4})-(\d{2})-(\d{2})\s+(%s|%s)\s+(\d{,2}):(\d{,2}):(\d{,2})' % \
(_korean_am, _korean_pm))
def _parse_date_onblog(dateString):
'''Parse a string according to the OnBlog 8-bit date format'''
m = _korean_onblog_date_re.match(dateString)
if not m: return
w3dtfdate = '%(year)s-%(month)s-%(day)sT%(hour)s:%(minute)s:%(second)s%(zonediff)s' % \
{'year': m.group(1), 'month': m.group(2), 'day': m.group(3),\
'hour': m.group(4), 'minute': m.group(5), 'second': m.group(6),\
'zonediff': '+09:00'}
if _debug: sys.stderr.write('OnBlog date parsed as: %s\n' % w3dtfdate)
return _parse_date_w3dtf(w3dtfdate)
registerDateHandler(_parse_date_onblog)
def _parse_date_nate(dateString):
'''Parse a string according to the Nate 8-bit date format'''
m = _korean_nate_date_re.match(dateString)
if not m: return
hour = int(m.group(5))
ampm = m.group(4)
if (ampm == _korean_pm):
hour += 12
hour = str(hour)
if len(hour) == 1:
hour = '0' + hour
w3dtfdate = '%(year)s-%(month)s-%(day)sT%(hour)s:%(minute)s:%(second)s%(zonediff)s' % \
{'year': m.group(1), 'month': m.group(2), 'day': m.group(3),\
'hour': hour, 'minute': m.group(6), 'second': m.group(7),\
'zonediff': '+09:00'}
if _debug: sys.stderr.write('Nate date parsed as: %s\n' % w3dtfdate)
return _parse_date_w3dtf(w3dtfdate)
registerDateHandler(_parse_date_nate)
_mssql_date_re = \
re.compile('(\d{4})-(\d{2})-(\d{2})\s+(\d{2}):(\d{2}):(\d{2})(\.\d+)?')
def _parse_date_mssql(dateString):
'''Parse a string according to the MS SQL date format'''
m = _mssql_date_re.match(dateString)
if not m: return
w3dtfdate = '%(year)s-%(month)s-%(day)sT%(hour)s:%(minute)s:%(second)s%(zonediff)s' % \
{'year': m.group(1), 'month': m.group(2), 'day': m.group(3),\
'hour': m.group(4), 'minute': m.group(5), 'second': m.group(6),\
'zonediff': '+09:00'}
if _debug: sys.stderr.write('MS SQL date parsed as: %s\n' % w3dtfdate)
return _parse_date_w3dtf(w3dtfdate)
registerDateHandler(_parse_date_mssql)
# Unicode strings for Greek date strings
_greek_months = \
{ \
u'\u0399\u03b1\u03bd': u'Jan', # c9e1ed in iso-8859-7
u'\u03a6\u03b5\u03b2': u'Feb', # d6e5e2 in iso-8859-7
u'\u039c\u03ac\u03ce': u'Mar', # ccdcfe in iso-8859-7
u'\u039c\u03b1\u03ce': u'Mar', # cce1fe in iso-8859-7
u'\u0391\u03c0\u03c1': u'Apr', # c1f0f1 in iso-8859-7
u'\u039c\u03ac\u03b9': u'May', # ccdce9 in iso-8859-7
u'\u039c\u03b1\u03ca': u'May', # cce1fa in iso-8859-7
u'\u039c\u03b1\u03b9': u'May', # cce1e9 in iso-8859-7
u'\u0399\u03bf\u03cd\u03bd': u'Jun', # c9effded in iso-8859-7
u'\u0399\u03bf\u03bd': u'Jun', # c9efed in iso-8859-7
u'\u0399\u03bf\u03cd\u03bb': u'Jul', # c9effdeb in iso-8859-7
u'\u0399\u03bf\u03bb': u'Jul', # c9f9eb in iso-8859-7
u'\u0391\u03cd\u03b3': u'Aug', # c1fde3 in iso-8859-7
u'\u0391\u03c5\u03b3': u'Aug', # c1f5e3 in iso-8859-7
u'\u03a3\u03b5\u03c0': u'Sep', # d3e5f0 in iso-8859-7
u'\u039f\u03ba\u03c4': u'Oct', # cfeaf4 in iso-8859-7
u'\u039d\u03bf\u03ad': u'Nov', # cdefdd in iso-8859-7
u'\u039d\u03bf\u03b5': u'Nov', # cdefe5 in iso-8859-7
u'\u0394\u03b5\u03ba': u'Dec', # c4e5ea in iso-8859-7
}
_greek_wdays = \
{ \
u'\u039a\u03c5\u03c1': u'Sun', # caf5f1 in iso-8859-7
u'\u0394\u03b5\u03c5': u'Mon', # c4e5f5 in iso-8859-7
u'\u03a4\u03c1\u03b9': u'Tue', # d4f1e9 in iso-8859-7
u'\u03a4\u03b5\u03c4': u'Wed', # d4e5f4 in iso-8859-7
u'\u03a0\u03b5\u03bc': u'Thu', # d0e5ec in iso-8859-7
u'\u03a0\u03b1\u03c1': u'Fri', # d0e1f1 in iso-8859-7
u'\u03a3\u03b1\u03b2': u'Sat', # d3e1e2 in iso-8859-7
}
_greek_date_format_re = \
re.compile(u'([^,]+),\s+(\d{2})\s+([^\s]+)\s+(\d{4})\s+(\d{2}):(\d{2}):(\d{2})\s+([^\s]+)')
def _parse_date_greek(dateString):
'''Parse a string according to a Greek 8-bit date format.'''
m = _greek_date_format_re.match(dateString)
if not m: return
try:
wday = _greek_wdays[m.group(1)]
month = _greek_months[m.group(3)]
except:
return
rfc822date = '%(wday)s, %(day)s %(month)s %(year)s %(hour)s:%(minute)s:%(second)s %(zonediff)s' % \
{'wday': wday, 'day': m.group(2), 'month': month, 'year': m.group(4),\
'hour': m.group(5), 'minute': m.group(6), 'second': m.group(7),\
'zonediff': m.group(8)}
if _debug: sys.stderr.write('Greek date parsed as: %s\n' % rfc822date)
return _parse_date_rfc822(rfc822date)
registerDateHandler(_parse_date_greek)
# Unicode strings for Hungarian date strings
_hungarian_months = \
{ \
u'janu\u00e1r': u'01', # e1 in iso-8859-2
u'febru\u00e1ri': u'02', # e1 in iso-8859-2
u'm\u00e1rcius': u'03', # e1 in iso-8859-2
u'\u00e1prilis': u'04', # e1 in iso-8859-2
u'm\u00e1ujus': u'05', # e1 in iso-8859-2
u'j\u00fanius': u'06', # fa in iso-8859-2
u'j\u00falius': u'07', # fa in iso-8859-2
u'augusztus': u'08',
u'szeptember': u'09',
u'okt\u00f3ber': u'10', # f3 in iso-8859-2
u'november': u'11',
u'december': u'12',
}
_hungarian_date_format_re = \
re.compile(u'(\d{4})-([^-]+)-(\d{,2})T(\d{,2}):(\d{2})((\+|-)(\d{,2}:\d{2}))')
def _parse_date_hungarian(dateString):
'''Parse a string according to a Hungarian 8-bit date format.'''
m = _hungarian_date_format_re.match(dateString)
if not m: return
try:
month = _hungarian_months[m.group(2)]
day = m.group(3)
if len(day) == 1:
day = '0' + day
hour = m.group(4)
if len(hour) == 1:
hour = '0' + hour
except:
return
w3dtfdate = '%(year)s-%(month)s-%(day)sT%(hour)s:%(minute)s%(zonediff)s' % \
{'year': m.group(1), 'month': month, 'day': day,\
'hour': hour, 'minute': m.group(5),\
'zonediff': m.group(6)}
if _debug: sys.stderr.write('Hungarian date parsed as: %s\n' % w3dtfdate)
return _parse_date_w3dtf(w3dtfdate)
registerDateHandler(_parse_date_hungarian)
# W3DTF-style date parsing adapted from PyXML xml.utils.iso8601, written by
# Drake and licensed under the Python license. Removed all range checking
# for month, day, hour, minute, and second, since mktime will normalize
# these later
def _parse_date_w3dtf(dateString):
def __extract_date(m):
year = int(m.group('year'))
if year < 100:
year = 100 * int(time.gmtime()[0] / 100) + int(year)
if year < 1000:
return 0, 0, 0
julian = m.group('julian')
if julian:
julian = int(julian)
month = julian / 30 + 1
day = julian % 30 + 1
jday = None
while jday != julian:
t = time.mktime((year, month, day, 0, 0, 0, 0, 0, 0))
jday = time.gmtime(t)[-2]
diff = abs(jday - julian)
if jday > julian:
if diff < day:
day = day - diff
else:
month = month - 1
day = 31
elif jday < julian:
if day + diff < 28:
day = day + diff
else:
month = month + 1
return year, month, day
month = m.group('month')
day = 1
if month is None:
month = 1
else:
month = int(month)
day = m.group('day')
if day:
day = int(day)
else:
day = 1
return year, month, day
def __extract_time(m):
if not m:
return 0, 0, 0
hours = m.group('hours')
if not hours:
return 0, 0, 0
hours = int(hours)
minutes = int(m.group('minutes'))
seconds = m.group('seconds')
if seconds:
seconds = int(seconds)
else:
seconds = 0
return hours, minutes, seconds
def __extract_tzd(m):
'''Return the Time Zone Designator as an offset in seconds from UTC.'''
if not m:
return 0
tzd = m.group('tzd')
if not tzd:
return 0
if tzd == 'Z':
return 0
hours = int(m.group('tzdhours'))
minutes = m.group('tzdminutes')
if minutes:
minutes = int(minutes)
else:
minutes = 0
offset = (hours*60 + minutes) * 60
if tzd[0] == '+':
return -offset
return offset
__date_re = ('(?P\d\d\d\d)'
'(?:(?P-|)'
'(?:(?P\d\d\d)'
'|(?P\d\d)(?:(?P=dsep)(?P\d\d))?))?')
__tzd_re = '(?P[-+](?P\d\d)(?::?(?P\d\d))|Z)'
__time_re = ('(?P\d\d)(?P:|)(?P\d\d)'
'(?:(?P=tsep)(?P\d\d(?:[.,]\d+)?))?'
+ __tzd_re)
__datetime_re = '%s(?:T%s)?' % (__date_re, __time_re)
__datetime_rx = re.compile(__datetime_re)
m = __datetime_rx.match(dateString)
if (m is None) or (m.group() != dateString): return
gmt = __extract_date(m) + __extract_time(m) + (0, 0, 0)
if gmt[0] == 0: return
return time.gmtime(time.mktime(gmt) + __extract_tzd(m) - time.timezone)
registerDateHandler(_parse_date_w3dtf)
def _parse_date_rfc822(dateString):
'''Parse an RFC822, RFC1123, RFC2822, or asctime-style date'''
data = dateString.split()
if data[0][-1] in (',', '.') or data[0].lower() in rfc822._daynames:
del data[0]
if len(data) == 4:
s = data[3]
i = s.find('+')
if i > 0:
data[3:] = [s[:i], s[i+1:]]
else:
data.append('')
dateString = " ".join(data)
if len(data) < 5:
dateString += ' 00:00:00 GMT'
tm = rfc822.parsedate_tz(dateString)
if tm:
return time.gmtime(rfc822.mktime_tz(tm))
# rfc822.py defines several time zones, but we define some extra ones.
# 'ET' is equivalent to 'EST', etc.
_additional_timezones = {'AT': -400, 'ET': -500, 'CT': -600, 'MT': -700, 'PT': -800}
rfc822._timezones.update(_additional_timezones)
registerDateHandler(_parse_date_rfc822)
def _parse_date_perforce(aDateString):
"""parse a date in yyyy/mm/dd hh:mm:ss TTT format"""
# Fri, 2006/09/15 08:19:53 EDT
_my_date_pattern = re.compile( \
r'(\w{,3}), (\d{,4})/(\d{,2})/(\d{2}) (\d{,2}):(\d{2}):(\d{2}) (\w{,3})')
dow, year, month, day, hour, minute, second, tz = \
_my_date_pattern.search(aDateString).groups()
months = ['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec']
dateString = "%s, %s %s %s %s:%s:%s %s" % (dow, day, months[int(month) - 1], year, hour, minute, second, tz)
tm = rfc822.parsedate_tz(dateString)
if tm:
return time.gmtime(rfc822.mktime_tz(tm))
registerDateHandler(_parse_date_perforce)
def _parse_date(dateString):
'''Parses a variety of date formats into a 9-tuple in GMT'''
for handler in _date_handlers:
try:
date9tuple = handler(dateString)
if not date9tuple: continue
if len(date9tuple) != 9:
if _debug: sys.stderr.write('date handler function must return 9-tuple\n')
raise ValueError
map(int, date9tuple)
return date9tuple
except Exception, e:
if _debug: sys.stderr.write('%s raised %s\n' % (handler.__name__, repr(e)))
pass
return None
def _getCharacterEncoding(http_headers, xml_data):
'''Get the character encoding of the XML document
http_headers is a dictionary
xml_data is a raw string (not Unicode)
This is so much trickier than it sounds, it's not even funny.
According to RFC 3023 ('XML Media Types'), if the HTTP Content-Type
is application/xml, application/*+xml,
application/xml-external-parsed-entity, or application/xml-dtd,
the encoding given in the charset parameter of the HTTP Content-Type
takes precedence over the encoding given in the XML prefix within the
document, and defaults to 'utf-8' if neither are specified. But, if
the HTTP Content-Type is text/xml, text/*+xml, or
text/xml-external-parsed-entity, the encoding given in the XML prefix
within the document is ALWAYS IGNORED and only the encoding given in
the charset parameter of the HTTP Content-Type header should be
respected, and it defaults to 'us-ascii' if not specified.
Furthermore, discussion on the atom-syntax mailing list with the
author of RFC 3023 leads me to the conclusion that any document
served with a Content-Type of text/* and no charset parameter
must be treated as us-ascii. (We now do this.) And also that it
must always be flagged as non-well-formed. (We now do this too.)
If Content-Type is unspecified (input was local file or non-HTTP source)
or unrecognized (server just got it totally wrong), then go by the
encoding given in the XML prefix of the document and default to
'iso-8859-1' as per the HTTP specification (RFC 2616).
Then, assuming we didn't find a character encoding in the HTTP headers
(and the HTTP Content-type allowed us to look in the body), we need
to sniff the first few bytes of the XML data and try to determine
whether the encoding is ASCII-compatible. Section F of the XML
specification shows the way here:
http://www.w3.org/TR/REC-xml/#sec-guessing-no-ext-info
If the sniffed encoding is not ASCII-compatible, we need to make it
ASCII compatible so that we can sniff further into the XML declaration
to find the encoding attribute, which will tell us the true encoding.
Of course, none of this guarantees that we will be able to parse the
feed in the declared character encoding (assuming it was declared
correctly, which many are not). CJKCodecs and iconv_codec help a lot;
you should definitely install them if you can.
http://cjkpython.i18n.org/
'''
def _parseHTTPContentType(content_type):
'''takes HTTP Content-Type header and returns (content type, charset)
If no charset is specified, returns (content type, '')
If no content type is specified, returns ('', '')
Both return parameters are guaranteed to be lowercase strings
'''
content_type = content_type or ''
content_type, params = cgi.parse_header(content_type)
return content_type, params.get('charset', '').replace("'", '')
sniffed_xml_encoding = ''
xml_encoding = ''
true_encoding = ''
http_content_type, http_encoding = _parseHTTPContentType(http_headers.get('content-type'))
# Must sniff for non-ASCII-compatible character encodings before
# searching for XML declaration. This heuristic is defined in
# section F of the XML specification:
# http://www.w3.org/TR/REC-xml/#sec-guessing-no-ext-info
try:
if xml_data[:4] == '\x4c\x6f\xa7\x94':
# EBCDIC
xml_data = _ebcdic_to_ascii(xml_data)
elif xml_data[:4] == '\x00\x3c\x00\x3f':
# UTF-16BE
sniffed_xml_encoding = 'utf-16be'
xml_data = unicode(xml_data, 'utf-16be').encode('utf-8')
elif (len(xml_data) >= 4) and (xml_data[:2] == '\xfe\xff') and (xml_data[2:4] != '\x00\x00'):
# UTF-16BE with BOM
sniffed_xml_encoding = 'utf-16be'
xml_data = unicode(xml_data[2:], 'utf-16be').encode('utf-8')
elif xml_data[:4] == '\x3c\x00\x3f\x00':
# UTF-16LE
sniffed_xml_encoding = 'utf-16le'
xml_data = unicode(xml_data, 'utf-16le').encode('utf-8')
elif (len(xml_data) >= 4) and (xml_data[:2] == '\xff\xfe') and (xml_data[2:4] != '\x00\x00'):
# UTF-16LE with BOM
sniffed_xml_encoding = 'utf-16le'
xml_data = unicode(xml_data[2:], 'utf-16le').encode('utf-8')
elif xml_data[:4] == '\x00\x00\x00\x3c':
# UTF-32BE
sniffed_xml_encoding = 'utf-32be'
xml_data = unicode(xml_data, 'utf-32be').encode('utf-8')
elif xml_data[:4] == '\x3c\x00\x00\x00':
# UTF-32LE
sniffed_xml_encoding = 'utf-32le'
xml_data = unicode(xml_data, 'utf-32le').encode('utf-8')
elif xml_data[:4] == '\x00\x00\xfe\xff':
# UTF-32BE with BOM
sniffed_xml_encoding = 'utf-32be'
xml_data = unicode(xml_data[4:], 'utf-32be').encode('utf-8')
elif xml_data[:4] == '\xff\xfe\x00\x00':
# UTF-32LE with BOM
sniffed_xml_encoding = 'utf-32le'
xml_data = unicode(xml_data[4:], 'utf-32le').encode('utf-8')
elif xml_data[:3] == '\xef\xbb\xbf':
# UTF-8 with BOM
sniffed_xml_encoding = 'utf-8'
xml_data = unicode(xml_data[3:], 'utf-8').encode('utf-8')
else:
# ASCII-compatible
pass
xml_encoding_match = re.compile('^<\?.*encoding=[\'"](.*?)[\'"].*\?>').match(xml_data)
except:
xml_encoding_match = None
if xml_encoding_match:
xml_encoding = xml_encoding_match.groups()[0].lower()
if sniffed_xml_encoding and (xml_encoding in ('iso-10646-ucs-2', 'ucs-2', 'csunicode', 'iso-10646-ucs-4', 'ucs-4', 'csucs4', 'utf-16', 'utf-32', 'utf_16', 'utf_32', 'utf16', 'u16')):
xml_encoding = sniffed_xml_encoding
acceptable_content_type = 0
application_content_types = ('application/xml', 'application/xml-dtd', 'application/xml-external-parsed-entity')
text_content_types = ('text/xml', 'text/xml-external-parsed-entity')
if (http_content_type in application_content_types) or \
(http_content_type.startswith('application/') and http_content_type.endswith('+xml')):
acceptable_content_type = 1
true_encoding = http_encoding or xml_encoding or 'utf-8'
elif (http_content_type in text_content_types) or \
(http_content_type.startswith('text/')) and http_content_type.endswith('+xml'):
acceptable_content_type = 1
true_encoding = http_encoding or 'us-ascii'
elif http_content_type.startswith('text/'):
true_encoding = http_encoding or 'us-ascii'
elif http_headers and (not http_headers.has_key('content-type')):
true_encoding = xml_encoding or 'iso-8859-1'
else:
true_encoding = xml_encoding or 'utf-8'
# some feeds claim to be gb2312 but are actually gb18030.
# apparently MSIE and Firefox both do the following switch:
if true_encoding.lower() == 'gb2312':
true_encoding = 'gb18030'
return true_encoding, http_encoding, xml_encoding, sniffed_xml_encoding, acceptable_content_type
def _toUTF8(data, encoding):
'''Changes an XML data stream on the fly to specify a new encoding
data is a raw sequence of bytes (not Unicode) that is presumed to be in %encoding already
encoding is a string recognized by encodings.aliases
'''
if _debug: sys.stderr.write('entering _toUTF8, trying encoding %s\n' % encoding)
# strip Byte Order Mark (if present)
if (len(data) >= 4) and (data[:2] == '\xfe\xff') and (data[2:4] != '\x00\x00'):
if _debug:
sys.stderr.write('stripping BOM\n')
if encoding != 'utf-16be':
sys.stderr.write('trying utf-16be instead\n')
encoding = 'utf-16be'
data = data[2:]
elif (len(data) >= 4) and (data[:2] == '\xff\xfe') and (data[2:4] != '\x00\x00'):
if _debug:
sys.stderr.write('stripping BOM\n')
if encoding != 'utf-16le':
sys.stderr.write('trying utf-16le instead\n')
encoding = 'utf-16le'
data = data[2:]
elif data[:3] == '\xef\xbb\xbf':
if _debug:
sys.stderr.write('stripping BOM\n')
if encoding != 'utf-8':
sys.stderr.write('trying utf-8 instead\n')
encoding = 'utf-8'
data = data[3:]
elif data[:4] == '\x00\x00\xfe\xff':
if _debug:
sys.stderr.write('stripping BOM\n')
if encoding != 'utf-32be':
sys.stderr.write('trying utf-32be instead\n')
encoding = 'utf-32be'
data = data[4:]
elif data[:4] == '\xff\xfe\x00\x00':
if _debug:
sys.stderr.write('stripping BOM\n')
if encoding != 'utf-32le':
sys.stderr.write('trying utf-32le instead\n')
encoding = 'utf-32le'
data = data[4:]
newdata = unicode(data, encoding)
if _debug: sys.stderr.write('successfully converted %s data to unicode\n' % encoding)
declmatch = re.compile('^<\?xml[^>]*?>')
newdecl = ''''''
if declmatch.search(newdata):
newdata = declmatch.sub(newdecl, newdata)
else:
newdata = newdecl + u'\n' + newdata
return newdata.encode('utf-8')
def _stripDoctype(data):
'''Strips DOCTYPE from XML document, returns (rss_version, stripped_data)
rss_version may be 'rss091n' or None
stripped_data is the same XML document, minus the DOCTYPE
'''
start = re.search('<\w',data)
start = start and start.start() or -1
head,data = data[:start+1], data[start+1:]
entity_pattern = re.compile(r'^\s*]*?)>', re.MULTILINE)
entity_results=entity_pattern.findall(head)
head = entity_pattern.sub('', head)
doctype_pattern = re.compile(r'^\s*]*?)>', re.MULTILINE)
doctype_results = doctype_pattern.findall(head)
doctype = doctype_results and doctype_results[0] or ''
if doctype.lower().count('netscape'):
version = 'rss091n'
else:
version = None
# only allow in 'safe' inline entity definitions
replacement=''
if len(doctype_results)==1 and entity_results:
safe_pattern=re.compile('\s+(\w+)\s+"(&#\w+;|[^&"]*)"')
safe_entities=filter(lambda e: safe_pattern.match(e),entity_results)
if safe_entities:
replacement='\n]>' % '>\n