and
- # only if all the remaining content is nested underneath it.
- # This means that the divs would be retained in the following:
- #
foo
bar
- 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] == '
' or pieces[0].startswith('
':
- 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 we're here then this is an RSS feed.
- #If we don't have a version or have a version that starts with something
- #other than RSS then there's been a mistake. Correct it.
- if not self.version or not self.version.startswith('rss'):
- 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 and self.feeddata.has_key('image'):
- 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):
- 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()
- 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:
- 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):
- if 'url' in attrsD:
- # This means that we're processing a source element from an RSS 2.0 feed
- self.sourcedata['href'] = attrsD[u'url']
- self.push('source', 1)
- self.insource = 1
- self.hasTitle = 0
-
- def _end_source(self):
- self.insource = 0
- value = self.pop('source')
- if value:
- self.sourcedata['title'] = value
- 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
-
- def _start_media_content(self, attrsD):
- context = self._getContext()
- context.setdefault('media_content', [])
- context['media_content'].append(attrsD)
-
- def _start_media_thumbnail(self, attrsD):
- context = self._getContext()
- context.setdefault('media_thumbnail', [])
- self.push('url', 1) # new
- context['media_thumbnail'].append(attrsD)
-
- def _end_media_thumbnail(self):
- url = self.pop('url')
- context = self._getContext()
- if url != None and len(url.strip()) != 0:
- if not context['media_thumbnail'][-1].has_key('url'):
- context['media_thumbnail'][-1]['url'] = url
-
- def _start_media_player(self, attrsD):
- self.push('media_player', 0)
- self._getContext()['media_player'] = FeedParserDict(attrsD)
-
- def _end_media_player(self):
- value = self.pop('media_player')
- context = self._getContext()
- context['media_player']['content'] = value
-
-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
- self.decls = {}
-
- def startPrefixMapping(self, prefix, uri):
- self.trackNamespace(prefix, uri)
- if uri == 'http://www.w3.org/1999/xlink':
- self.decls['xmlns:'+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, self.decls = self.decls, {}
- 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', 'command', 'embed', 'frame',
- 'hr', 'img', 'input', 'isindex', 'keygen', 'link', 'meta', 'param',
- 'source', 'track', 'wbr'
- ]
-
- 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