venus/0000775000123000012300000000000011373300757011302 5ustar rubysrubysvenus/splice.py0000775000123000012300000000064711373300435013136 0ustar rubysrubys#!/usr/bin/env python """ Main program to run just the splice portion of planet """ import os.path import sys from planet import splice, config if __name__ == '__main__': if len(sys.argv) == 2 and os.path.isfile(sys.argv[1]): config.load(sys.argv[1]) doc = splice.splice() splice.apply(doc.toxml('utf-8')) else: print "Usage:" print " python %s config.ini" % sys.argv[0] venus/planet/0000775000123000012300000000000011373300435012556 5ustar rubysrubysvenus/planet/splice.py0000664000123000012300000001436711373300435014422 0ustar rubysrubys""" Splice together a planet from a cache of feed entries """ import glob, os, time, shutil from xml.dom import minidom import planet, config, feedparser, reconstitute, shell from reconstitute import createTextElement, date from spider import filename from planet import idindex def splice(): """ Splice together a planet from a cache of entries """ import planet log = planet.logger log.info("Loading cached data") cache = config.cache_directory() dir=[(os.stat(file).st_mtime,file) for file in glob.glob(cache+"/*") if not os.path.isdir(file)] dir.sort() dir.reverse() max_items=max([config.items_per_page(templ) for templ in config.template_files() or ['Planet']]) doc = minidom.parseString('') feed = doc.documentElement # insert feed information createTextElement(feed, 'title', config.name()) date(feed, 'updated', time.gmtime()) gen = createTextElement(feed, 'generator', config.generator()) gen.setAttribute('uri', config.generator_uri()) author = doc.createElement('author') createTextElement(author, 'name', config.owner_name()) createTextElement(author, 'email', config.owner_email()) feed.appendChild(author) if config.feed(): createTextElement(feed, 'id', config.feed()) link = doc.createElement('link') link.setAttribute('rel', 'self') link.setAttribute('href', config.feed()) if config.feedtype(): link.setAttribute('type', "application/%s+xml" % config.feedtype()) feed.appendChild(link) if config.link(): link = doc.createElement('link') link.setAttribute('rel', 'alternate') link.setAttribute('href', config.link()) feed.appendChild(link) # insert subscription information sub_ids = [] feed.setAttribute('xmlns:planet',planet.xmlns) sources = config.cache_sources_directory() for sub in config.subscriptions(): data=feedparser.parse(filename(sources,sub)) if data.feed.has_key('id'): sub_ids.append(data.feed.id) if not data.feed: continue xdoc=minidom.parseString('''\n''' % planet.xmlns) reconstitute.source(xdoc.documentElement, data.feed, None, None) feed.appendChild(xdoc.documentElement) index = idindex.open() # insert entry information items = 0 count = {} atomNS='http://www.w3.org/2005/Atom' new_feed_items = config.new_feed_items() for mtime,file in dir: if index != None: base = os.path.basename(file) if index.has_key(base) and index[base] not in sub_ids: continue try: entry=minidom.parse(file) # verify that this entry is currently subscribed to and that the # number of entries contributed by this feed does not exceed # config.new_feed_items entry.normalize() sources = entry.getElementsByTagNameNS(atomNS, 'source') if sources: ids = sources[0].getElementsByTagName('id') if ids: id = ids[0].childNodes[0].nodeValue count[id] = count.get(id,0) + 1 if new_feed_items and count[id] > new_feed_items: continue if id not in sub_ids: ids = sources[0].getElementsByTagName('planet:id') if not ids: continue id = ids[0].childNodes[0].nodeValue if id not in sub_ids: continue # add entry to feed feed.appendChild(entry.documentElement) items = items + 1 if items >= max_items: break except: log.error("Error parsing %s", file) if index: index.close() return doc def apply(doc): output_dir = config.output_dir() if not os.path.exists(output_dir): os.makedirs(output_dir) log = planet.logger planet_filters = config.filters('Planet') # Go-go-gadget-template for template_file in config.template_files(): output_file = shell.run(template_file, doc) # run any template specific filters if config.filters(template_file) != planet_filters: output = open(output_file).read() for filter in config.filters(template_file): if filter in planet_filters: continue if filter.find('>')>0: # tee'd output filter,dest = filter.split('>',1) tee = shell.run(filter.strip(), output, mode="filter") if tee: output_dir = planet.config.output_dir() dest_file = os.path.join(output_dir, dest.strip()) dest_file = open(dest_file,'w') dest_file.write(tee) dest_file.close() else: # pipe'd output output = shell.run(filter, output, mode="filter") if not output: os.unlink(output_file) break else: handle = open(output_file,'w') handle.write(output) handle.close() # Process bill of materials for copy_file in config.bill_of_materials(): dest = os.path.join(output_dir, copy_file) for template_dir in config.template_directories(): source = os.path.join(template_dir, copy_file) if os.path.exists(source): break else: log.error('Unable to locate %s', copy_file) log.info("Template search path:") for template_dir in config.template_directories(): log.info(" %s", os.path.realpath(template_dir)) continue mtime = os.stat(source).st_mtime if not os.path.exists(dest) or os.stat(dest).st_mtime < mtime: dest_dir = os.path.split(dest)[0] if not os.path.exists(dest_dir): os.makedirs(dest_dir) log.info("Copying %s to %s", source, dest) if os.path.exists(dest): os.chmod(dest, 0644) shutil.copyfile(source, dest) shutil.copystat(source, dest) venus/planet/opml.py0000775000123000012300000001371711373300435014113 0ustar rubysrubysfrom xml.sax import ContentHandler, make_parser, SAXParseException from xml.sax.xmlreader import InputSource from sgmllib import SGMLParser from cStringIO import StringIO from ConfigParser import ConfigParser from htmlentitydefs import entitydefs import re # input = opml, output = ConfigParser def opml2config(opml, config=None): if hasattr(opml, 'read'): opml = opml.read() if not config: config = ConfigParser() opmlParser = OpmlParser(config) try: # try SAX source = InputSource() source.setByteStream(StringIO(opml)) parser = make_parser() parser.setContentHandler(opmlParser) parser.parse(source) except SAXParseException: # try as SGML opmlParser.feed(opml) return config # Parse OPML via either SAX or SGML class OpmlParser(ContentHandler,SGMLParser): entities = re.compile('&(#?\w+);') def __init__(self, config): ContentHandler.__init__(self) SGMLParser.__init__(self) self.config = config def startElement(self, name, attrs): # we are only looking for data in 'outline' nodes. if name != 'outline': return # A type of 'rss' is meant to be used generically to indicate that # this is an entry in a subscription list, but some leave this # attribute off, and others have placed 'atom' in here if attrs.has_key('type'): if attrs['type'] == 'link' and not attrs.has_key('url'): # Auto-correct WordPress link manager OPML files attrs = dict(attrs.items()) attrs['type'] = 'rss' if attrs['type'].lower() not in['rss','atom']: return # The feed itself is supposed to be in an attribute named 'xmlUrl' # (note the camel casing), but this has proven to be problematic, # with the most common misspelling being in all lower-case if not attrs.has_key('xmlUrl') or not attrs['xmlUrl'].strip(): for attribute in attrs.keys(): if attribute.lower() == 'xmlurl' and attrs[attribute].strip(): attrs = dict(attrs.items()) attrs['xmlUrl'] = attrs[attribute] break else: return # the text attribute is nominally required in OPML, but this # data is often found in a title attribute instead if not attrs.has_key('text') or not attrs['text'].strip(): if not attrs.has_key('title') or not attrs['title'].strip(): return attrs = dict(attrs.items()) attrs['text'] = attrs['title'] # if we get this far, we either have a valid subscription list entry, # or one with a correctable error. Add it to the configuration, if # it is not already there. xmlUrl = attrs['xmlUrl'] if not self.config.has_section(xmlUrl): self.config.add_section(xmlUrl) self.config.set(xmlUrl, 'name', self.unescape(attrs['text'])) def unescape(self, text): parsed = self.entities.split(text) for i in range(1,len(parsed),2): if parsed[i] in entitydefs.keys(): # named entities codepoint=entitydefs[parsed[i]] match=self.entities.match(codepoint) if match: parsed[i]=match.group(1) else: parsed[i]=unichr(ord(codepoint)) # numeric entities if parsed[i].startswith('#'): if parsed[i].startswith('#x'): parsed[i]=unichr(int(parsed[i][2:],16)) else: parsed[i]=unichr(int(parsed[i][1:])) return u''.join(parsed).encode('utf-8') # SGML => SAX def unknown_starttag(self, name, attrs): attrs = dict(attrs) for attribute in attrs: try: attrs[attribute] = attrs[attribute].decode('utf-8') except: work = attrs[attribute].decode('iso-8859-1') work = u''.join([c in cp1252 and cp1252[c] or c for c in work]) attrs[attribute] = work self.startElement(name, attrs) # http://www.intertwingly.net/stories/2004/04/14/i18n.html#CleaningWindows cp1252 = { unichr(128): unichr(8364), # euro sign unichr(130): unichr(8218), # single low-9 quotation mark unichr(131): unichr( 402), # latin small letter f with hook unichr(132): unichr(8222), # double low-9 quotation mark unichr(133): unichr(8230), # horizontal ellipsis unichr(134): unichr(8224), # dagger unichr(135): unichr(8225), # double dagger unichr(136): unichr( 710), # modifier letter circumflex accent unichr(137): unichr(8240), # per mille sign unichr(138): unichr( 352), # latin capital letter s with caron unichr(139): unichr(8249), # single left-pointing angle quotation mark unichr(140): unichr( 338), # latin capital ligature oe unichr(142): unichr( 381), # latin capital letter z with caron unichr(145): unichr(8216), # left single quotation mark unichr(146): unichr(8217), # right single quotation mark unichr(147): unichr(8220), # left double quotation mark unichr(148): unichr(8221), # right double quotation mark unichr(149): unichr(8226), # bullet unichr(150): unichr(8211), # en dash unichr(151): unichr(8212), # em dash unichr(152): unichr( 732), # small tilde unichr(153): unichr(8482), # trade mark sign unichr(154): unichr( 353), # latin small letter s with caron unichr(155): unichr(8250), # single right-pointing angle quotation mark unichr(156): unichr( 339), # latin small ligature oe unichr(158): unichr( 382), # latin small letter z with caron unichr(159): unichr( 376)} # latin capital letter y with diaeresis if __name__ == "__main__": # small main program which converts OPML into config.ini format import sys, urllib config = ConfigParser() for opml in sys.argv[1:]: opml2config(urllib.urlopen(opml), config) config.write(sys.stdout) venus/planet/shell/0000775000123000012300000000000011373300435013665 5ustar rubysrubysvenus/planet/shell/__init__.py0000664000123000012300000000455611373300435016010 0ustar rubysrubysimport planet import os import sys logged_modes = [] def run(template_file, doc, mode='template'): """ select a template module based on file extension and execute it """ log = planet.logger if mode == 'template': dirs = planet.config.template_directories() else: dirs = planet.config.filter_directories() # parse out "extra" options if template_file.find('?') < 0: extra_options = {} else: import cgi template_file, extra_options = template_file.split('?',1) extra_options = dict(cgi.parse_qsl(extra_options)) # see if the template can be located for template_dir in dirs: template_resolved = os.path.join(template_dir, template_file) if os.path.exists(template_resolved): break else: log.error("Unable to locate %s %s", mode, template_file) if not mode in logged_modes: log.info("%s search path:", mode) for template_dir in dirs: log.info(" %s", os.path.realpath(template_dir)) logged_modes.append(mode) return template_resolved = os.path.realpath(template_resolved) # Add shell directory to the path, if not already there shellpath = os.path.join(sys.path[0],'planet','shell') if shellpath not in sys.path: sys.path.append(shellpath) # Try loading module for processing this template, based on the extension base,ext = os.path.splitext(os.path.basename(template_resolved)) module_name = ext[1:] try: try: module = __import__("_" + module_name) except: module = __import__(module_name) except Exception, inst: return log.error("Skipping %s '%s' after failing to load '%s': %s", mode, template_resolved, module_name, inst) # Execute the shell module options = planet.config.template_options(template_file) if module_name == 'plugin': options['__file__'] = template_file options.update(extra_options) log.debug("Processing %s %s using %s", mode, os.path.realpath(template_resolved), module_name) if mode == 'filter': return module.run(template_resolved, doc, None, options) else: output_dir = planet.config.output_dir() output_file = os.path.join(output_dir, base) module.run(template_resolved, doc, output_file, options) return output_file venus/planet/shell/py.py0000664000123000012300000000103011373300435014661 0ustar rubysrubysfrom subprocess import Popen, PIPE import sys def run(script, doc, output_file=None, options={}): """ process an Python script """ if output_file: out = open(output_file, 'w') else: out = PIPE options = sum([['--'+key, value] for key,value in options.items()], []) proc = Popen([sys.executable, script] + options, stdin=PIPE, stdout=out, stderr=PIPE) stdout, stderr = proc.communicate(doc) if stderr: import planet planet.logger.error(stderr) return stdout venus/planet/shell/plugin.py0000664000123000012300000000367111373300435015544 0ustar rubysrubysimport os, sys, imp from StringIO import StringIO def run(script, doc, output_file=None, options={}): """ process an Python script using imp """ save_sys = (sys.stdin, sys.stdout, sys.stderr, sys.argv) plugin_stdout = StringIO() plugin_stderr = StringIO() try: # redirect stdin sys.stdin = StringIO(doc) # redirect stdout if output_file: sys.stdout = open(output_file, 'w') else: sys.stdout = plugin_stdout # redirect stderr sys.stderr = plugin_stderr # determine __file__ value if options.has_key("__file__"): plugin_file = options["__file__"] del options["__file__"] else: plugin_file = script # set sys.argv options = sum([['--'+key, value] for key,value in options.items()], []) sys.argv = [plugin_file] + options # import script handle = open(script, 'r') cwd = os.getcwd() try: try: try: description=('.plugin', 'rb', imp.PY_SOURCE) imp.load_module('__main__',handle,plugin_file,description) except SystemExit,e: if e.code: log.error('%s exit rc=%d',(plugin_file,e.code)) except Exception, e: import traceback type, value, tb = sys.exc_info() plugin_stderr.write(''.join( traceback.format_exception_only(type,value) + traceback.format_tb(tb))) finally: handle.close() if cwd != os.getcwd(): os.chdir(cwd) finally: # restore system state sys.stdin, sys.stdout, sys.stderr, sys.argv = save_sys # log anything sent to stderr if plugin_stderr.getvalue(): import planet planet.logger.error(plugin_stderr.getvalue()) # return stdout return plugin_stdout.getvalue() venus/planet/shell/tmpl.py0000664000123000012300000002317411373300435015222 0ustar rubysrubysfrom xml.sax.saxutils import escape import sgmllib, time, os, sys, new, urlparse, re from planet import config, feedparser import htmltmpl voids=feedparser._BaseHTMLProcessor.elements_no_end_tag empty=re.compile(r"<((%s)[^>]*)>" % '|'.join(voids)) class stripHtml(sgmllib.SGMLParser): "remove all tags from the data" def __init__(self, data): sgmllib.SGMLParser.__init__(self) self.result='' if isinstance(data, str): try: self.feed(data.decode('utf-8')) except: self.feed(data) else: self.feed(data) self.close() def __str__(self): if isinstance(self.result, unicode): return self.result.encode('utf-8') return self.result def handle_entityref(self, ref): import htmlentitydefs if ref in htmlentitydefs.entitydefs: ref=htmlentitydefs.entitydefs[ref] if len(ref)==1: self.result+=unichr(ord(ref)) elif ref.startswith('&#') and ref.endswith(';'): self.handle_charref(ref[2:-1]) else: self.result+='&%s;' % ref else: self.result+='&%s;' % ref def handle_charref(self, ref): try: if ref.startswith('x'): self.result+=unichr(int(ref[1:],16)) else: self.result+=unichr(int(ref)) except: self.result+='&#%s;' % ref def handle_data(self, data): if data: self.result+=data # Data format mappers def String(value): if isinstance(value, unicode): return value.encode('utf-8') return value def Plain(value): return str(stripHtml(value)) def PlanetDate(value): return time.strftime(config.date_format(), value) def NewDate(value): return time.strftime(config.new_date_format(), value) def Rfc822(value): return time.strftime("%a, %d %b %Y %H:%M:%S +0000", value) def Rfc3399(value): return time.strftime("%Y-%m-%dT%H:%M:%S+00:00", value) # Map from FeedParser path to Planet tmpl names Base = [ ['author', String, 'author'], ['author_name', String, 'author_detail', 'name'], ['generator', String, 'generator'], ['id', String, 'id'], ['icon', String, 'icon'], ['last_updated_822', Rfc822, 'updated_parsed'], ['last_updated_iso', Rfc3399, 'updated_parsed'], ['last_updated', PlanetDate, 'updated_parsed'], ['link', String, 'link'], ['logo', String, 'logo'], ['rights', String, 'rights_detail', 'value'], ['subtitle', String, 'subtitle_detail', 'value'], ['title', String, 'title_detail', 'value'], ['title_plain', Plain, 'title_detail', 'value'], ['url', String, 'links', {'rel':'self'}, 'href'], ['url', String, 'planet_http_location'], ] Items = [ ['author', String, 'author'], ['author_email', String, 'author_detail', 'email'], ['author_name', String, 'author_detail', 'name'], ['author_uri', String, 'author_detail', 'href'], ['content_language', String, 'content', 0, 'language'], ['content', String, 'summary_detail', 'value'], ['content', String, 'content', 0, 'value'], ['date', PlanetDate, 'published_parsed'], ['date', PlanetDate, 'updated_parsed'], ['date_822', Rfc822, 'published_parsed'], ['date_822', Rfc822, 'updated_parsed'], ['date_iso', Rfc3399, 'published_parsed'], ['date_iso', Rfc3399, 'updated_parsed'], ['enclosure_href', String, 'links', {'rel': 'enclosure'}, 'href'], ['enclosure_length', String, 'links', {'rel': 'enclosure'}, 'length'], ['enclosure_type', String, 'links', {'rel': 'enclosure'}, 'type'], ['id', String, 'id'], ['link', String, 'links', {'rel': 'alternate'}, 'href'], ['new_channel', String, 'source', 'id'], ['new_date', NewDate, 'published_parsed'], ['new_date', NewDate, 'updated_parsed'], ['rights', String, 'rights_detail', 'value'], ['title_language', String, 'title_detail', 'language'], ['title_plain', Plain, 'title_detail', 'value'], ['title', String, 'title_detail', 'value'], ['summary_language', String, 'summary_detail', 'language'], ['updated', PlanetDate, 'updated_parsed'], ['updated_822', Rfc822, 'updated_parsed'], ['updated_iso', Rfc3399, 'updated_parsed'], ['published', PlanetDate, 'published_parsed'], ['published_822', Rfc822, 'published_parsed'], ['published_iso', Rfc3399, 'published_parsed'], ] # Add additional rules for source information for rule in Base: Items.append(['channel_'+rule[0], rule[1], 'source'] + rule[2:]) def tmpl_mapper(source, rules): "Apply specified rules to the source, and return a template dictionary" output = {} for rule in rules: node = source for path in rule[2:]: if isinstance(path, str) and path in node: if path == 'value': if node.get('type','')=='text/plain': node['value'] = escape(node['value']) node['type'] = 'text/html' elif node.get('type','')=='application/xhtml+xml': node['value'] = empty.sub(r"<\1 />", node['value']) node = node[path] elif isinstance(path, int): node = node[path] elif isinstance(path, dict): for test in node: for key, value in path.items(): if test.get(key,None) != value: break else: node = test break else: break else: break else: if node: output[rule[0]] = rule[1](node) # copy over all planet namespaced elements from parent source for name,value in source.items(): if name.startswith('planet_'): output[name[7:]] = String(value) if not output.get('name') and source.has_key('title_detail'): output['name'] = Plain(source.title_detail.value) # copy over all planet namespaced elements from child source element if 'source' in source: for name,value in source.source.items(): if name.startswith('planet_'): output['channel_' + name[7:]] = String(value) if not output.get('channel_name') and \ source.source.has_key('title_detail'): output['channel_name'] = Plain(source.source.title_detail.value) return output def _end_planet_source(self): self._end_source() context = self._getContext() if not context.has_key('sources'): context['sources'] = [] context.sources.append(context.source) del context['source'] def template_info(source): """ get template information from a feedparser output """ # wire in support for planet:source, call feedparser, unplug planet:source mixin=feedparser._FeedParserMixin mixin._start_planet_source = mixin._start_source mixin._end_planet_source = \ new.instancemethod(_end_planet_source, None, mixin) data=feedparser.parse(source) del mixin._start_planet_source del mixin._end_planet_source # apply rules to convert feed parser output to htmltmpl input output = {'Channels': [], 'Items': []} output.update(tmpl_mapper(data.feed, Base)) sources = [] for feed in data.feed.get('sources',[]): source = tmpl_mapper(feed, Base) sources.append([source.get('name'), source]) sources.sort() output['Channels'] = [source for name,source in sources] for entry in data.entries: output['Items'].append(tmpl_mapper(entry, Items)) # synthesize isPermaLink attribute for item in output['Items']: if item.get('id') == item.get('link'): item['guid_isPermaLink']='true' else: item['guid_isPermaLink']='false' # feed level information output['generator'] = config.generator_uri() output['name'] = config.name() output['link'] = config.link() output['owner_name'] = config.owner_name() output['owner_email'] = config.owner_email() if config.feed(): output['feed'] = config.feed() output['feedtype'] = config.feed().find('rss')>=0 and 'rss' or 'atom' # date/time information date = time.gmtime() output['date'] = PlanetDate(date) output['date_iso'] = Rfc3399(date) output['date_822'] = Rfc822(date) # remove new_dates and new_channels that aren't "new" date = channel = None for item in output['Items']: if item.has_key('new_date'): if item['new_date'] == date: del item['new_date'] else: date = item['new_date'] if item.has_key('new_channel'): if item['new_channel'] == channel and not item.has_key('new_date'): del item['new_channel'] else: channel = item['new_channel'] return output def run(script, doc, output_file=None, options={}): """ process an HTMLTMPL file """ manager = htmltmpl.TemplateManager() template = manager.prepare(script) tp = htmltmpl.TemplateProcessor(html_escape=0) for key,value in template_info(doc).items(): tp.set(key, value) if output_file: reluri = os.path.splitext(os.path.basename(output_file))[0] tp.set('url', urlparse.urljoin(config.link(),reluri)) output = open(output_file, "w") output.write(tp.process(template)) output.close() else: return tp.process(template) if __name__ == '__main__': sys.path.insert(0, os.path.split(sys.path[0])[0]) for test in sys.argv[1:]: from pprint import pprint pprint(template_info('/home/rubys/bzr/venus/tests/data/filter/tmpl/'+test)) venus/planet/shell/dj.py0000664000123000012300000000306611373300435014641 0ustar rubysrubysimport os.path import urlparse import datetime import tmpl from planet import config def DjangoPlanetDate(value): return datetime.datetime(*value[:6]) # remap PlanetDate to be a datetime, so Django template authors can use # the "date" filter on these values tmpl.PlanetDate = DjangoPlanetDate def run(script, doc, output_file=None, options={}): """process a Django template file""" # this is needed to use the Django template system as standalone # I need to re-import the settings at every call because I have to # set the TEMPLATE_DIRS variable programmatically from django.conf import settings try: settings.configure( DEBUG=True, TEMPLATE_DEBUG=True, TEMPLATE_DIRS=(os.path.dirname(script),) ) except EnvironmentError: pass from django.template import Context from django.template.loader import get_template # set up the Django context by using the default htmltmpl # datatype converters context = Context() context.update(tmpl.template_info(doc)) context['Config'] = config.planet_options() t = get_template(script) if output_file: reluri = os.path.splitext(os.path.basename(output_file))[0] context['url'] = urlparse.urljoin(config.link(),reluri) f = open(output_file, 'w') ss = t.render(context) if isinstance(ss,unicode): ss=ss.encode('utf-8') f.write(ss) f.close() else: # @@this is useful for testing purposes, but does it # belong here? return t.render(context) venus/planet/shell/xslt.py0000664000123000012300000000455311373300435015240 0ustar rubysrubysimport os def quote(string, apos): """ quote a string so that it can be passed as a parameter """ if type(string) == unicode: string=string.encode('utf-8') if apos.startswith("\\"): string.replace('\\','\\\\') if string.find("'")<0: return "'" + string + "'" elif string.find("'")<0: return '"' + string + '"' else: # unclear how to quote strings with both types of quotes for libxslt return "'" + string.replace("'",apos) + "'" def run(script, doc, output_file=None, options={}): """ process an XSLT stylesheet """ try: # if available, use the python interface to libxslt import libxml2 import libxslt dom = libxml2.parseDoc(doc) docfile = None except: # otherwise, use the command line interface dom = None # do it result = None if dom: styledoc = libxml2.parseFile(script) style = libxslt.parseStylesheetDoc(styledoc) for key in options.keys(): options[key] = quote(options[key], apos="\xe2\x80\x99") output = style.applyStylesheet(dom, options) if output_file: style.saveResultToFilename(output_file, output, 0) else: result = output.serialize('utf-8') style.freeStylesheet() output.freeDoc() elif output_file: import warnings if hasattr(warnings, 'simplefilter'): warnings.simplefilter('ignore', RuntimeWarning) docfile = os.tmpnam() file = open(docfile,'w') file.write(doc) file.close() cmdopts = [] for key,value in options.items(): if value.find("'")>=0 and value.find('"')>=0: continue cmdopts += ['--stringparam', key, quote(value, apos=r"\'")] os.system('xsltproc %s %s %s > %s' % (' '.join(cmdopts), script, docfile, output_file)) os.unlink(docfile) else: import sys from subprocess import Popen, PIPE options = sum([['--stringparam', key, value] for key,value in options.items()], []) proc = Popen(['xsltproc'] + options + [script, '-'], stdin=PIPE, stdout=PIPE, stderr=PIPE) result, stderr = proc.communicate(doc) if stderr: import planet planet.logger.error(stderr) if dom: dom.freeDoc() return result venus/planet/shell/_genshi.py0000664000123000012300000001150611373300435015656 0ustar rubysrubysfrom StringIO import StringIO from xml.sax.saxutils import escape from genshi.input import HTMLParser, XMLParser from genshi.template import Context, MarkupTemplate subscriptions = [] feed_types = [ 'application/atom+xml', 'application/rss+xml', 'application/rdf+xml' ] def norm(value): """ Convert to Unicode """ if hasattr(value,'items'): return dict([(norm(n),norm(v)) for n,v in value.items()]) try: return value.decode('utf-8') except: return value.decode('iso-8859-1') def find_config(config, feed): # match based on self link for link in feed.links: if link.has_key('rel') and link.rel=='self': if link.has_key('type') and link.type in feed_types: if link.has_key('href') and link.href in subscriptions: return norm(dict(config.parser.items(link.href))) # match based on name for sub in subscriptions: if config.parser.has_option(sub, 'name') and \ norm(config.parser.get(sub, 'name')) == feed.planet_name: return norm(dict(config.parser.items(sub))) return {} class XHTMLParser(object): """ parse an XHTML fragment """ def __init__(self, text): self.parser = XMLParser(StringIO("
%s
" % text)) self.depth = 0 def __iter__(self): self.iter = self.parser.__iter__() return self def next(self): object = self.iter.next() if object[0] == 'END': self.depth = self.depth - 1 predepth = self.depth if object[0] == 'START': self.depth = self.depth + 1 if predepth: return object return self.next() def streamify(text,bozo): """ add a .stream to a _detail textConstruct """ if text.type == 'text/plain': text.stream = HTMLParser(StringIO(escape(text.value))) elif text.type == 'text/html' or bozo != 'false': text.stream = HTMLParser(StringIO(text.value)) else: text.stream = XHTMLParser(text.value) def run(script, doc, output_file=None, options={}): """ process an Genshi template """ context = Context(**options) tmpl_fileobj = open(script) tmpl = MarkupTemplate(tmpl_fileobj, script) tmpl_fileobj.close() if not output_file: # filter context.push({'input':XMLParser(StringIO(doc))}) else: # template import time from planet import config,feedparser from planet.spider import filename # gather a list of subscriptions, feeds global subscriptions feeds = [] sources = config.cache_sources_directory() for sub in config.subscriptions(): data=feedparser.parse(filename(sources,sub)) data.feed.config = norm(dict(config.parser.items(sub))) if data.feed.has_key('link'): feeds.append((data.feed.config.get('name',''),data.feed)) subscriptions.append(norm(sub)) feeds.sort() # annotate each entry new_date_format = config.new_date_format() vars = feedparser.parse(StringIO(doc)) vars.feeds = [value for name,value in feeds] last_feed = None last_date = None for entry in vars.entries: entry.source.config = find_config(config, entry.source) # add new_feed and new_date fields entry.new_feed = entry.source.id entry.new_date = date = None if entry.has_key('published_parsed'): date=entry.published_parsed if entry.has_key('updated_parsed'): date=entry.updated_parsed if date: entry.new_date = time.strftime(new_date_format, date) # remove new_feed and new_date fields if not "new" if entry.new_date == last_date: entry.new_date = None if entry.new_feed == last_feed: entry.new_feed = None else: last_feed = entry.new_feed elif entry.new_date: last_date = entry.new_date last_feed = None # add streams for all text constructs for key in entry.keys(): if key.endswith("_detail") and entry[key].has_key('type') and \ entry[key].has_key('value'): streamify(entry[key],entry.source.planet_bozo) if entry.has_key('content'): for content in entry.content: streamify(content,entry.source.planet_bozo) # add cumulative feed information to the Genshi context vars.feed.config = dict(config.parser.items('Planet',True)) context.push(vars) # apply template output=tmpl.generate(context).render('xml') if output_file: out_file = open(output_file,'w') out_file.write(output) out_file.close() else: return output venus/planet/shell/sed.py0000664000123000012300000000066211373300435015016 0ustar rubysrubysfrom subprocess import Popen, PIPE def run(script, doc, output_file=None, options={}): """ process an Python script """ if output_file: out = open(output_file, 'w') else: out = PIPE proc = Popen(['sed', '-f', script], stdin=PIPE, stdout=out, stderr=PIPE) stdout, stderr = proc.communicate(doc) if stderr: import planet planet.logger.error(stderr) return stdout venus/planet/__init__.py0000664000123000012300000000167111373300435014674 0ustar rubysrubysxmlns = 'http://planet.intertwingly.net/' logger = None loggerParms = None import os, sys, re import config config.__init__() from ConfigParser import ConfigParser from urlparse import urljoin def getLogger(level, format): """ get a logger with the specified log level """ global logger, loggerParms if logger and loggerParms == (level,format): return logger try: import logging logging.basicConfig(format=format) except: import compat_logging as logging logging.basicConfig(format=format) logger = logging.getLogger("planet.runner") logger.setLevel(logging.getLevelName(level)) try: logger.warning except: logger.warning = logger.warn loggerParms = (level,format) return logger sys.path.insert(1, os.path.join(os.path.dirname(__file__),'vendor')) # Configure feed parser import feedparser feedparser.SANITIZE_HTML=1 feedparser.RESOLVE_RELATIVE_URIS=0 venus/planet/idindex.py0000664000123000012300000000605111373300435014556 0ustar rubysrubysfrom glob import glob import os, sys if __name__ == '__main__': rootdir = os.path.dirname(os.path.dirname(os.path.abspath(__file__))) sys.path.insert(0, rootdir) from planet.spider import filename from planet import config def open(): try: cache = config.cache_directory() index=os.path.join(cache,'index') if not os.path.exists(index): return None import dbhash return dbhash.open(filename(index, 'id'),'w') except Exception, e: if e.__class__.__name__ == 'DBError': e = e.args[-1] from planet import logger as log log.error(str(e)) def destroy(): from planet import logger as log cache = config.cache_directory() index=os.path.join(cache,'index') if not os.path.exists(index): return None idindex = filename(index, 'id') if os.path.exists(idindex): os.unlink(idindex) os.removedirs(index) log.info(idindex + " deleted") def create(): from planet import logger as log cache = config.cache_directory() index=os.path.join(cache,'index') if not os.path.exists(index): os.makedirs(index) import dbhash index = dbhash.open(filename(index, 'id'),'c') try: import libxml2 except: libxml2 = False from xml.dom import minidom for file in glob(cache+"/*"): if os.path.isdir(file): continue elif libxml2: try: doc = libxml2.parseFile(file) ctxt = doc.xpathNewContext() ctxt.xpathRegisterNs('atom','http://www.w3.org/2005/Atom') entry = ctxt.xpathEval('/atom:entry/atom:id') source = ctxt.xpathEval('/atom:entry/atom:source/atom:id') if entry and source: index[filename('',entry[0].content)] = source[0].content doc.freeDoc() except: log.error(file) else: try: doc = minidom.parse(file) doc.normalize() ids = doc.getElementsByTagName('id') entry = [e for e in ids if e.parentNode.nodeName == 'entry'] source = [e for e in ids if e.parentNode.nodeName == 'source'] if entry and source: index[filename('',entry[0].childNodes[0].nodeValue)] = \ source[0].childNodes[0].nodeValue doc.freeDoc() except: log.error(file) log.info(str(len(index.keys())) + " entries indexed") index.close() return open() if __name__ == '__main__': if len(sys.argv) < 2: print 'Usage: %s [-c|-d]' % sys.argv[0] sys.exit(1) config.load(sys.argv[1]) if len(sys.argv) > 2 and sys.argv[2] == '-c': create() elif len(sys.argv) > 2 and sys.argv[2] == '-d': destroy() else: from planet import logger as log index = open() if index: log.info(str(len(index.keys())) + " entries indexed") index.close() else: log.info("no entries indexed") venus/planet/vendor/0000775000123000012300000000000011373300435014053 5ustar rubysrubysvenus/planet/vendor/feedparser.py0000775000123000012300000046270211373300435016563 0ustar rubysrubys#!/usr/bin/env python """Universal feed parser Handles RSS 0.9x, RSS 1.0, RSS 2.0, CDF, Atom 0.3, and Atom 1.0 feeds Visit http://feedparser.org/ for the latest version Visit http://feedparser.org/docs/ for the latest documentation Required: Python 2.1 or later Recommended: Python 2.3 or later Recommended: CJKCodecs and iconv_codec """ __version__ = "4.2-pre-" + "$Revision: 308 $"[11:14] + "-svn" __license__ = """Copyright (c) 2002-2008, Mark Pilgrim, All rights reserved. Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: * Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. * Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS 'AS IS' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.""" __author__ = "Mark Pilgrim " __contributors__ = ["Jason Diamond ", "John Beimler ", "Fazal Majid ", "Aaron Swartz ", "Kevin Marks ", "Sam Ruby ", "Ade Oshineye "] _debug = 0 # HTTP "User-Agent" header to send to servers when downloading feeds. # If you are embedding feedparser in a larger application, you should # change this to your application name and URL. USER_AGENT = "UniversalFeedParser/%s +http://feedparser.org/" % __version__ # HTTP "Accept" header to send to servers when downloading feeds. If you don't # want to send an Accept header, set this to None. ACCEPT_HEADER = "application/atom+xml,application/rdf+xml,application/rss+xml,application/x-netcdf,application/xml;q=0.9,text/xml;q=0.2,*/*;q=0.1" # List of preferred XML parsers, by SAX driver name. These will be tried first, # but if they're not installed, Python will keep searching through its own list # of pre-installed parsers until it finds one that supports everything we need. PREFERRED_XML_PARSERS = ["drv_libxml2"] # If you want feedparser to automatically run HTML markup through HTML Tidy, set # this to 1. Requires mxTidy # or utidylib . TIDY_MARKUP = 0 # List of Python interfaces for HTML Tidy, in order of preference. Only useful # if TIDY_MARKUP = 1 PREFERRED_TIDY_INTERFACES = ["uTidy", "mxTidy"] # If you want feedparser to automatically resolve all relative URIs, set this # to 1. RESOLVE_RELATIVE_URIS = 1 # If you want feedparser to automatically sanitize all potentially unsafe # HTML content, set this to 1. SANITIZE_HTML = 1 # ---------- required modules (should come with any Python distribution) ---------- import sgmllib, re, sys, copy, urlparse, time, rfc822, types, cgi, urllib, urllib2 try: from cStringIO import StringIO as _StringIO except: from StringIO import StringIO as _StringIO # ---------- optional modules (feedparser will work without these, but with reduced functionality) ---------- # gzip is included with most Python distributions, but may not be available if you compiled your own try: import gzip except: gzip = None try: import zlib except: zlib = None # If a real XML parser is available, feedparser will attempt to use it. feedparser has # been tested with the built-in SAX parser, PyXML, and libxml2. On platforms where the # Python distribution does not come with an XML parser (such as Mac OS X 10.2 and some # versions of FreeBSD), feedparser will quietly fall back on regex-based parsing. try: import xml.sax xml.sax.make_parser(PREFERRED_XML_PARSERS) # test for valid parsers from xml.sax.saxutils import escape as _xmlescape _XML_AVAILABLE = 1 except: _XML_AVAILABLE = 0 def _xmlescape(data,entities={}): data = data.replace('&', '&') data = data.replace('>', '>') data = data.replace('<', '<') for char, entity in entities: data = data.replace(char, entity) return data # base64 support for Atom feeds that contain embedded binary data try: import base64, binascii except: base64 = binascii = None # cjkcodecs and iconv_codec provide support for more character encodings. # Both are available from http://cjkpython.i18n.org/ try: import cjkcodecs.aliases except: pass try: import iconv_codec except: pass # chardet library auto-detects character encodings # Download from http://chardet.feedparser.org/ try: import chardet if _debug: import chardet.constants chardet.constants._debug = 1 except: chardet = None # reversable htmlentitydefs mappings for Python 2.2 try: from htmlentitydefs import name2codepoint, codepoint2name except: import htmlentitydefs name2codepoint={} codepoint2name={} for (name,codepoint) in htmlentitydefs.entitydefs.iteritems(): if codepoint.startswith('&#'): codepoint=unichr(int(codepoint[2:-1])) name2codepoint[name]=ord(codepoint) codepoint2name[ord(codepoint)]=name # BeautifulSoup parser used for parsing microformats from embedded HTML content # http://www.crummy.com/software/BeautifulSoup/ # feedparser is tested with BeautifulSoup 3.0.x, but it might work with the # older 2.x series. If it doesn't, and you can figure out why, I'll accept a # patch and modify the compatibility statement accordingly. try: import BeautifulSoup except: BeautifulSoup = None # ---------- don't touch these ---------- class ThingsNobodyCaresAboutButMe(Exception): pass class CharacterEncodingOverride(ThingsNobodyCaresAboutButMe): pass class CharacterEncodingUnknown(ThingsNobodyCaresAboutButMe): pass class NonXMLContentType(ThingsNobodyCaresAboutButMe): pass class UndeclaredNamespace(Exception): pass sgmllib.tagfind = re.compile('[a-zA-Z][-_.:a-zA-Z0-9]*') sgmllib.special = re.compile(']|"[^"]*"(?=>|/|\s|\w+=)|'[^']*'(?=>|/|\s|\w+=))*(?=[<>])|.*?(?=[<>])''') def search(self,string,index=0): self.match = self.endbracket.match(string,index) if self.match: return self def start(self,n): return self.match.end(n) sgmllib.endbracket = EndBracketMatch() SUPPORTED_VERSIONS = {'': 'unknown', 'rss090': 'RSS 0.90', 'rss091n': 'RSS 0.91 (Netscape)', 'rss091u': 'RSS 0.91 (Userland)', 'rss092': 'RSS 0.92', 'rss093': 'RSS 0.93', 'rss094': 'RSS 0.94', 'rss20': 'RSS 2.0', 'rss10': 'RSS 1.0', 'rss': 'RSS (unknown version)', 'atom01': 'Atom 0.1', 'atom02': 'Atom 0.2', 'atom03': 'Atom 0.3', 'atom10': 'Atom 1.0', 'atom': 'Atom (unknown version)', 'cdf': 'CDF', 'hotrss': 'Hot RSS' } try: UserDict = dict except NameError: # Python 2.1 does not have dict from UserDict import UserDict def dict(aList): rc = {} for k, v in aList: rc[k] = v return rc class FeedParserDict(UserDict): keymap = {'channel': 'feed', 'items': 'entries', 'guid': 'id', 'date': 'updated', 'date_parsed': 'updated_parsed', 'description': ['subtitle', 'summary'], 'url': ['href'], 'modified': 'updated', 'modified_parsed': 'updated_parsed', 'issued': 'published', 'issued_parsed': 'published_parsed', 'copyright': 'rights', 'copyright_detail': 'rights_detail', 'tagline': 'subtitle', 'tagline_detail': 'subtitle_detail'} def __getitem__(self, key): if key == 'category': return UserDict.__getitem__(self, 'tags')[0]['term'] if key == 'enclosures': norel = lambda link: FeedParserDict([(name,value) for (name,value) in link.items() if name!='rel']) return [norel(link) for link in UserDict.__getitem__(self, 'links') if link['rel']=='enclosure'] if key == 'license': for link in UserDict.__getitem__(self, 'links'): if link['rel']=='license' and link.has_key('href'): return link['href'] if key == 'categories': return [(tag['scheme'], tag['term']) for tag in UserDict.__getitem__(self, 'tags')] realkey = self.keymap.get(key, key) if type(realkey) == types.ListType: for k in realkey: if UserDict.has_key(self, k): return UserDict.__getitem__(self, k) if UserDict.has_key(self, key): return UserDict.__getitem__(self, key) return UserDict.__getitem__(self, realkey) def __setitem__(self, key, value): for k in self.keymap.keys(): if key == k: key = self.keymap[k] if type(key) == types.ListType: key = key[0] return UserDict.__setitem__(self, key, value) def get(self, key, default=None): if self.has_key(key): return self[key] else: return default def setdefault(self, key, value): if not self.has_key(key): self[key] = value return self[key] def has_key(self, key): try: return hasattr(self, key) or UserDict.has_key(self, key) except AttributeError: return False def __getattr__(self, key): try: return self.__dict__[key] except KeyError: pass try: assert not key.startswith('_') return self.__getitem__(key) except: raise AttributeError, "object has no attribute '%s'" % key def __setattr__(self, key, value): if key.startswith('_') or key == 'data': self.__dict__[key] = value else: return self.__setitem__(key, value) def __contains__(self, key): return self.has_key(key) def zopeCompatibilityHack(): global FeedParserDict del FeedParserDict def FeedParserDict(aDict=None): rc = {} if aDict: rc.update(aDict) return rc _ebcdic_to_ascii_map = None def _ebcdic_to_ascii(s): global _ebcdic_to_ascii_map if not _ebcdic_to_ascii_map: emap = ( 0,1,2,3,156,9,134,127,151,141,142,11,12,13,14,15, 16,17,18,19,157,133,8,135,24,25,146,143,28,29,30,31, 128,129,130,131,132,10,23,27,136,137,138,139,140,5,6,7, 144,145,22,147,148,149,150,4,152,153,154,155,20,21,158,26, 32,160,161,162,163,164,165,166,167,168,91,46,60,40,43,33, 38,169,170,171,172,173,174,175,176,177,93,36,42,41,59,94, 45,47,178,179,180,181,182,183,184,185,124,44,37,95,62,63, 186,187,188,189,190,191,192,193,194,96,58,35,64,39,61,34, 195,97,98,99,100,101,102,103,104,105,196,197,198,199,200,201, 202,106,107,108,109,110,111,112,113,114,203,204,205,206,207,208, 209,126,115,116,117,118,119,120,121,122,210,211,212,213,214,215, 216,217,218,219,220,221,222,223,224,225,226,227,228,229,230,231, 123,65,66,67,68,69,70,71,72,73,232,233,234,235,236,237, 125,74,75,76,77,78,79,80,81,82,238,239,240,241,242,243, 92,159,83,84,85,86,87,88,89,90,244,245,246,247,248,249, 48,49,50,51,52,53,54,55,56,57,250,251,252,253,254,255 ) import string _ebcdic_to_ascii_map = string.maketrans( \ ''.join(map(chr, range(256))), ''.join(map(chr, emap))) return s.translate(_ebcdic_to_ascii_map) _cp1252 = { unichr(128): unichr(8364), # euro sign unichr(130): unichr(8218), # single low-9 quotation mark unichr(131): unichr( 402), # latin small letter f with hook unichr(132): unichr(8222), # double low-9 quotation mark unichr(133): unichr(8230), # horizontal ellipsis unichr(134): unichr(8224), # dagger unichr(135): unichr(8225), # double dagger unichr(136): unichr( 710), # modifier letter circumflex accent unichr(137): unichr(8240), # per mille sign unichr(138): unichr( 352), # latin capital letter s with caron unichr(139): unichr(8249), # single left-pointing angle quotation mark unichr(140): unichr( 338), # latin capital ligature oe unichr(142): unichr( 381), # latin capital letter z with caron unichr(145): unichr(8216), # left single quotation mark unichr(146): unichr(8217), # right single quotation mark unichr(147): unichr(8220), # left double quotation mark unichr(148): unichr(8221), # right double quotation mark unichr(149): unichr(8226), # bullet unichr(150): unichr(8211), # en dash unichr(151): unichr(8212), # em dash unichr(152): unichr( 732), # small tilde unichr(153): unichr(8482), # trade mark sign unichr(154): unichr( 353), # latin small letter s with caron unichr(155): unichr(8250), # single right-pointing angle quotation mark unichr(156): unichr( 339), # latin small ligature oe unichr(158): unichr( 382), # latin small letter z with caron unichr(159): unichr( 376)} # latin capital letter y with diaeresis _urifixer = re.compile('^([A-Za-z][A-Za-z0-9+-.]*://)(/*)(.*?)') def _urljoin(base, uri): uri = _urifixer.sub(r'\1\3', uri) try: return urlparse.urljoin(base, uri) except: uri = urlparse.urlunparse([urllib.quote(part) for part in urlparse.urlparse(uri)]) return urlparse.urljoin(base, uri) class _FeedParserMixin: namespaces = {'': '', 'http://backend.userland.com/rss': '', 'http://blogs.law.harvard.edu/tech/rss': '', 'http://purl.org/rss/1.0/': '', 'http://my.netscape.com/rdf/simple/0.9/': '', 'http://example.com/newformat#': '', 'http://example.com/necho': '', 'http://purl.org/echo/': '', 'uri/of/echo/namespace#': '', 'http://purl.org/pie/': '', 'http://purl.org/atom/ns#': '', 'http://www.w3.org/2005/Atom': '', 'http://purl.org/rss/1.0/modules/rss091#': '', 'http://webns.net/mvcb/': 'admin', 'http://purl.org/rss/1.0/modules/aggregation/': 'ag', 'http://purl.org/rss/1.0/modules/annotate/': 'annotate', 'http://media.tangent.org/rss/1.0/': 'audio', 'http://backend.userland.com/blogChannelModule': 'blogChannel', 'http://web.resource.org/cc/': 'cc', 'http://backend.userland.com/creativeCommonsRssModule': 'creativeCommons', 'http://purl.org/rss/1.0/modules/company': 'co', 'http://purl.org/rss/1.0/modules/content/': 'content', 'http://my.theinfo.org/changed/1.0/rss/': 'cp', 'http://purl.org/dc/elements/1.1/': 'dc', 'http://purl.org/dc/terms/': 'dcterms', 'http://purl.org/rss/1.0/modules/email/': 'email', 'http://purl.org/rss/1.0/modules/event/': 'ev', 'http://rssnamespace.org/feedburner/ext/1.0': 'feedburner', 'http://freshmeat.net/rss/fm/': 'fm', 'http://xmlns.com/foaf/0.1/': 'foaf', 'http://www.w3.org/2003/01/geo/wgs84_pos#': 'geo', 'http://postneo.com/icbm/': 'icbm', 'http://purl.org/rss/1.0/modules/image/': 'image', 'http://www.itunes.com/DTDs/PodCast-1.0.dtd': 'itunes', 'http://example.com/DTDs/PodCast-1.0.dtd': 'itunes', 'http://purl.org/rss/1.0/modules/link/': 'l', 'http://search.yahoo.com/mrss': 'media', #Version 1.1.2 of the Media RSS spec added the trailing slash on the namespace 'http://search.yahoo.com/mrss/': 'media', 'http://madskills.com/public/xml/rss/module/pingback/': 'pingback', 'http://prismstandard.org/namespaces/1.2/basic/': 'prism', 'http://www.w3.org/1999/02/22-rdf-syntax-ns#': 'rdf', 'http://www.w3.org/2000/01/rdf-schema#': 'rdfs', 'http://purl.org/rss/1.0/modules/reference/': 'ref', 'http://purl.org/rss/1.0/modules/richequiv/': 'reqv', 'http://purl.org/rss/1.0/modules/search/': 'search', 'http://purl.org/rss/1.0/modules/slash/': 'slash', 'http://schemas.xmlsoap.org/soap/envelope/': 'soap', 'http://purl.org/rss/1.0/modules/servicestatus/': 'ss', 'http://hacks.benhammersley.com/rss/streaming/': 'str', 'http://purl.org/rss/1.0/modules/subscription/': 'sub', 'http://purl.org/rss/1.0/modules/syndication/': 'sy', 'http://schemas.pocketsoap.com/rss/myDescModule/': 'szf', 'http://purl.org/rss/1.0/modules/taxonomy/': 'taxo', 'http://purl.org/rss/1.0/modules/threading/': 'thr', 'http://purl.org/rss/1.0/modules/textinput/': 'ti', 'http://madskills.com/public/xml/rss/module/trackback/':'trackback', 'http://wellformedweb.org/commentAPI/': 'wfw', 'http://purl.org/rss/1.0/modules/wiki/': 'wiki', 'http://www.w3.org/1999/xhtml': 'xhtml', 'http://www.w3.org/1999/xlink': 'xlink', 'http://www.w3.org/XML/1998/namespace': 'xml' } _matchnamespaces = {} can_be_relative_uri = ['link', 'id', 'wfw_comment', 'wfw_commentrss', 'docs', 'url', 'href', 'comments', 'icon', 'logo'] can_contain_relative_uris = ['content', 'title', 'summary', 'info', 'tagline', 'subtitle', 'copyright', 'rights', 'description'] can_contain_dangerous_markup = ['content', 'title', 'summary', 'info', 'tagline', 'subtitle', 'copyright', 'rights', 'description'] html_types = ['text/html', 'application/xhtml+xml'] def __init__(self, baseuri=None, baselang=None, encoding='utf-8'): if _debug: sys.stderr.write('initializing FeedParser\n') if not self._matchnamespaces: for k, v in self.namespaces.items(): self._matchnamespaces[k.lower()] = v self.feeddata = FeedParserDict() # feed-level data self.encoding = encoding # character encoding self.entries = [] # list of entry-level data self.version = '' # feed type/version, see SUPPORTED_VERSIONS self.namespacesInUse = {} # dictionary of namespaces defined by the feed # the following are used internally to track state; # this is really out of control and should be refactored self.infeed = 0 self.inentry = 0 self.incontent = 0 self.intextinput = 0 self.inimage = 0 self.inauthor = 0 self.incontributor = 0 self.inpublisher = 0 self.insource = 0 self.sourcedata = FeedParserDict() self.contentparams = FeedParserDict() self._summaryKey = None self.namespacemap = {} self.elementstack = [] self.basestack = [] self.langstack = [] self.baseuri = baseuri or '' self.lang = baselang or None self.svgOK = 0 self.hasTitle = 0 if baselang: self.feeddata['language'] = baselang.replace('_','-') def unknown_starttag(self, tag, attrs): if _debug: sys.stderr.write('start %s with %s\n' % (tag, attrs)) # normalize attrs attrs = [(k.lower(), v) for k, v in attrs] attrs = [(k, k in ('rel', 'type') and v.lower() or v) for k, v in attrs] # track xml:base and xml:lang attrsD = dict(attrs) baseuri = attrsD.get('xml:base', attrsD.get('base')) or self.baseuri if type(baseuri) != type(u''): try: baseuri = unicode(baseuri, self.encoding) except: baseuri = unicode(baseuri, 'iso-8859-1') self.baseuri = _urljoin(self.baseuri, baseuri) lang = attrsD.get('xml:lang', attrsD.get('lang')) if lang == '': # xml:lang could be explicitly set to '', we need to capture that lang = None elif lang is None: # if no xml:lang is specified, use parent lang lang = self.lang if lang: if tag in ('feed', 'rss', 'rdf:RDF'): self.feeddata['language'] = lang.replace('_','-') self.lang = lang self.basestack.append(self.baseuri) self.langstack.append(lang) # track namespaces for prefix, uri in attrs: if prefix.startswith('xmlns:'): self.trackNamespace(prefix[6:], uri) elif prefix == 'xmlns': self.trackNamespace(None, uri) # track inline content if self.incontent and self.contentparams.has_key('type') and not self.contentparams.get('type', 'xml').endswith('xml'): if tag in ['xhtml:div', 'div']: return # typepad does this 10/2007 # element declared itself as escaped markup, but it isn't really self.contentparams['type'] = 'application/xhtml+xml' if self.incontent and self.contentparams.get('type') == 'application/xhtml+xml': if tag.find(':') <> -1: prefix, tag = tag.split(':', 1) namespace = self.namespacesInUse.get(prefix, '') if tag=='math' and namespace=='http://www.w3.org/1998/Math/MathML': attrs.append(('xmlns',namespace)) if tag=='svg' and namespace=='http://www.w3.org/2000/svg': attrs.append(('xmlns',namespace)) if tag == 'svg': self.svgOK += 1 return self.handle_data('<%s%s>' % (tag, self.strattrs(attrs)), escape=0) # match namespaces if tag.find(':') <> -1: prefix, suffix = tag.split(':', 1) else: prefix, suffix = '', tag prefix = self.namespacemap.get(prefix, prefix) if prefix: prefix = prefix + '_' # special hack for better tracking of empty textinput/image elements in illformed feeds if (not prefix) and tag not in ('title', 'link', 'description', 'name'): self.intextinput = 0 if (not prefix) and tag not in ('title', 'link', 'description', 'url', 'href', 'width', 'height'): self.inimage = 0 # call special handler (if defined) or default handler methodname = '_start_' + prefix + suffix try: method = getattr(self, methodname) return method(attrsD) except AttributeError: # Since there's no handler or something has gone wrong we explicitly add the element and its attributes unknown_tag = prefix + suffix if len(attrsD) == 0: # No attributes so merge it into the encosing dictionary return self.push(unknown_tag, 1) else: # Has attributes so create it in its own dictionary context = self._getContext() context[unknown_tag] = attrsD def unknown_endtag(self, tag): if _debug: sys.stderr.write('end %s\n' % tag) # match namespaces if tag.find(':') <> -1: prefix, suffix = tag.split(':', 1) else: prefix, suffix = '', tag prefix = self.namespacemap.get(prefix, prefix) if prefix: prefix = prefix + '_' if suffix == 'svg' and self.svgOK: self.svgOK -= 1 # call special handler (if defined) or default handler methodname = '_end_' + prefix + suffix try: if self.svgOK: raise AttributeError() method = getattr(self, methodname) method() except AttributeError: self.pop(prefix + suffix) # track inline content if self.incontent and self.contentparams.has_key('type') and not self.contentparams.get('type', 'xml').endswith('xml'): # element declared itself as escaped markup, but it isn't really if tag in ['xhtml:div', 'div']: return # typepad does this 10/2007 self.contentparams['type'] = 'application/xhtml+xml' if self.incontent and self.contentparams.get('type') == 'application/xhtml+xml': tag = tag.split(':')[-1] self.handle_data('' % tag, escape=0) # track xml:base and xml:lang going out of scope if self.basestack: self.basestack.pop() if self.basestack and self.basestack[-1]: self.baseuri = self.basestack[-1] if self.langstack: self.langstack.pop() if self.langstack: # and (self.langstack[-1] is not None): self.lang = self.langstack[-1] def handle_charref(self, ref): # called for each character reference, e.g. for ' ', ref will be '160' if not self.elementstack: return ref = ref.lower() if ref in ('34', '38', '39', '60', '62', 'x22', 'x26', 'x27', 'x3c', 'x3e'): text = '&#%s;' % ref else: if ref[0] == 'x': c = int(ref[1:], 16) else: c = int(ref) text = unichr(c).encode('utf-8') self.elementstack[-1][2].append(text) def handle_entityref(self, ref): # called for each entity reference, e.g. for '©', ref will be 'copy' if not self.elementstack: return if _debug: sys.stderr.write('entering handle_entityref with %s\n' % ref) if ref in ('lt', 'gt', 'quot', 'amp', 'apos'): text = '&%s;' % ref elif ref in self.entities.keys(): text = self.entities[ref] if text.startswith('&#') and text.endswith(';'): return self.handle_entityref(text) else: try: name2codepoint[ref] except KeyError: text = '&%s;' % ref else: text = unichr(name2codepoint[ref]).encode('utf-8') self.elementstack[-1][2].append(text) def handle_data(self, text, escape=1): # called for each block of plain text, i.e. outside of any tag and # not containing any character or entity references if not self.elementstack: return if escape and self.contentparams.get('type') == 'application/xhtml+xml': text = _xmlescape(text) self.elementstack[-1][2].append(text) def handle_comment(self, text): # called for each comment, e.g. pass def handle_pi(self, text): # called for each processing instruction, e.g. pass def handle_decl(self, text): pass def parse_declaration(self, i): # override internal declaration handler to handle CDATA blocks if _debug: sys.stderr.write('entering parse_declaration\n') if self.rawdata[i:i+9] == '', i) if k == -1: # CDATA block began but didn't finish k = len(self.rawdata) return k self.handle_data(_xmlescape(self.rawdata[i+9:k]), 0) return k+3 else: k = self.rawdata.find('>', i) if k >= 0: return k+1 else: # We have an incomplete CDATA block. return k def mapContentType(self, contentType): contentType = contentType.lower() if contentType == 'text': contentType = 'text/plain' elif contentType == 'html': contentType = 'text/html' elif contentType == 'xhtml': contentType = 'application/xhtml+xml' return contentType def trackNamespace(self, prefix, uri): loweruri = uri.lower() if (prefix, loweruri) == (None, 'http://my.netscape.com/rdf/simple/0.9/') and not self.version: self.version = 'rss090' if loweruri == 'http://purl.org/rss/1.0/' and not self.version: self.version = 'rss10' if loweruri == 'http://www.w3.org/2005/atom' and not self.version: self.version = 'atom10' if loweruri.find('backend.userland.com/rss') <> -1: # match any backend.userland.com namespace uri = 'http://backend.userland.com/rss' loweruri = uri if self._matchnamespaces.has_key(loweruri): self.namespacemap[prefix] = self._matchnamespaces[loweruri] self.namespacesInUse[self._matchnamespaces[loweruri]] = uri else: self.namespacesInUse[prefix or ''] = uri def resolveURI(self, uri): return _urljoin(self.baseuri or '', uri) def decodeEntities(self, element, data): return data def strattrs(self, attrs): return ''.join([' %s="%s"' % (t[0],_xmlescape(t[1],{'"':'"'})) for t in attrs]) def push(self, element, expectingText): self.elementstack.append([element, expectingText, []]) def pop(self, element, stripWhitespace=1): if not self.elementstack: return if self.elementstack[-1][0] != element: return element, expectingText, pieces = self.elementstack.pop() if self.version == 'atom10' and self.contentparams.get('type','text') == 'application/xhtml+xml': # remove enclosing child element, but only if it is a
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: 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
, 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_data, 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): if _debug: sys.stderr.write('tag: [%s] with attributes: [%s]\n' % (tag, str(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', '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', 'controls', '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' and self.mathmlOK: self.mathmlOK -= 1 elif self.svgOK and tag in self.svg_elements: tag = self.svg_elem_map.get(tag,tag) if tag == 'svg' and self.svgOK: self.svgOK -= 1 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 '' # This replaced a regexp that used re.match and was prone to pathological back-tracking. if re.sub("\s*[-\w]+\s*:\s*[^:;]*;?", '', style).strip(): 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}))?' + r'(\.(?P\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)' __tzd_rx = re.compile(__tzd_re) __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 = 0: if token["systemId"].find(u"'") >= 0: self.serializeError(_("System identifer contains both single and double quote characters")) quote_char = u"'" else: quote_char = u'"' doctype += u" %s%s%s" % (quote_char, token["systemId"], quote_char) doctype += u">" if encoding: yield doctype.encode(encoding) else: yield doctype elif type in ("Characters", "SpaceCharacters"): if type == "SpaceCharacters" or in_cdata: if in_cdata and token["data"].find("= 0: self.serializeError(_("Unexpected \"'=", False) v = v.replace("&", "&") if self.escape_lt_in_attrs: v = v.replace("<", "<") if encoding: v = encode(v, encoding) if quote_attr: quote_char = self.quote_char if self.use_best_quote_char: if "'" in v and '"' not in v: quote_char = '"' elif '"' in v and "'" not in v: quote_char = "'" if quote_char == "'": v = v.replace("'", "'") else: v = v.replace('"', """) attributes.append(quote_char) attributes.append(v) attributes.append(quote_char) else: attributes.append(v) if name in voidElements and self.use_trailing_solidus: if self.space_before_trailing_solidus: attributes.append(" /") else: attributes.append("/") if encoding: yield "<%s%s>" % (name.encode(encoding, "strict"), "".join(attributes)) else: yield u"<%s%s>" % (name, u"".join(attributes)) elif type == "EndTag": name = token["name"] if name in rcdataElements: in_cdata = False elif in_cdata: self.serializeError(_("Unexpected child element of a CDATA element")) end_tag = u"" % name if encoding: end_tag = end_tag.encode(encoding, "strict") yield end_tag elif type == "Comment": data = token["data"] if data.find("--") >= 0: self.serializeError(_("Comment contains --")) comment = u"" % token["data"] if encoding: comment = comment.encode(encoding, unicode_encode_errors) yield comment elif type == "Entity": name = token["name"] key = name + ";" if not key in entities: self.serializeError(_("Entity %s not recognized" % name)) if self.resolve_entities and key not in xmlEntities: data = entities[key] else: data = u"&%s;" % name if encoding: data = data.encode(encoding, unicode_encode_errors) yield data else: self.serializeError(token["data"]) def render(self, treewalker, encoding=None): if encoding: return "".join(list(self.serialize(treewalker, encoding))) else: return u"".join(list(self.serialize(treewalker))) def serializeError(self, data="XXX ERROR MESSAGE NEEDED"): # XXX The idea is to make data mandatory. self.errors.append(data) if self.strict: raise SerializeError def SerializeError(Exception): """Error in serialized tree""" pass venus/planet/vendor/html5lib/html5parser.py0000664000123000012300000031631411373300435020423 0ustar rubysrubystry: frozenset except NameError: # Import from the sets module for python 2.3 from sets import Set as set from sets import ImmutableSet as frozenset try: any except: # Implement 'any' for python 2.4 and previous def any(iterable): for element in iterable: if element: return True return False import sys import inputstream import tokenizer import treebuilders from treebuilders._base import Marker from treebuilders import simpletree import utils from constants import spaceCharacters, asciiUpper2Lower from constants import scopingElements, formattingElements, specialElements from constants import headingElements, tableInsertModeElements from constants import cdataElements, rcdataElements, voidElements from constants import tokenTypes, ReparseException, namespaces def parse(doc, treebuilder="simpletree", encoding=None, namespaceHTMLElements=True): tb = treebuilders.getTreeBuilder(treebuilder) p = HTMLParser(tb, namespaceHTMLElements=namespaceHTMLElements) return p.parse(doc, encoding=encoding) def parseFragment(doc, container="div", treebuilder="simpletree", encoding=None, namespaceHTMLElements=True): tb = treebuilders.getTreeBuilder(treebuilder) p = HTMLParser(tb, namespaceHTMLElements=namespaceHTMLElements) return p.parseFragment(doc, container=container, encoding=encoding) class HTMLParser(object): """HTML parser. Generates a tree structure from a stream of (possibly malformed) HTML""" def __init__(self, tree = simpletree.TreeBuilder, tokenizer = tokenizer.HTMLTokenizer, strict = False, namespaceHTMLElements = True): """ strict - raise an exception when a parse error is encountered tree - a treebuilder class controlling the type of tree that will be returned. Built in treebuilders can be accessed through html5lib.treebuilders.getTreeBuilder(treeType) tokenizer - a class that provides a stream of tokens to the treebuilder. This may be replaced for e.g. a sanitizer which converts some tags to text """ # Raise an exception on the first error encountered self.strict = strict self.tree = tree(namespaceHTMLElements) self.tokenizer_class = tokenizer self.errors = [] self.phases = { "initial": InitialPhase(self, self.tree), "beforeHtml": BeforeHtmlPhase(self, self.tree), "beforeHead": BeforeHeadPhase(self, self.tree), "inHead": InHeadPhase(self, self.tree), # XXX "inHeadNoscript": InHeadNoScriptPhase(self, self.tree), "afterHead": AfterHeadPhase(self, self.tree), "inBody": InBodyPhase(self, self.tree), "text": TextPhase(self, self.tree), "inTable": InTablePhase(self, self.tree), "inTableText": InTableTextPhase(self, self.tree), "inCaption": InCaptionPhase(self, self.tree), "inColumnGroup": InColumnGroupPhase(self, self.tree), "inTableBody": InTableBodyPhase(self, self.tree), "inRow": InRowPhase(self, self.tree), "inCell": InCellPhase(self, self.tree), "inSelect": InSelectPhase(self, self.tree), "inSelectInTable": InSelectInTablePhase(self, self.tree), "inForeignContent": InForeignContentPhase(self, self.tree), "afterBody": AfterBodyPhase(self, self.tree), "inFrameset": InFramesetPhase(self, self.tree), "afterFrameset": AfterFramesetPhase(self, self.tree), "afterAfterBody": AfterAfterBodyPhase(self, self.tree), "afterAfterFrameset": AfterAfterFramesetPhase(self, self.tree), # XXX after after frameset } def _parse(self, stream, innerHTML=False, container="div", encoding=None, parseMeta=True, useChardet=True, **kwargs): self.innerHTMLMode = innerHTML self.container = container self.tokenizer = self.tokenizer_class(stream, encoding=encoding, parseMeta=parseMeta, useChardet=useChardet, **kwargs) self.reset() while True: try: self.mainLoop() break except ReparseException, e: self.reset() def reset(self): self.tree.reset() self.firstStartTag = False self.errors = [] # "quirks" / "limited quirks" / "no quirks" self.compatMode = "no quirks" if self.innerHTMLMode: self.innerHTML = self.container.lower() if self.innerHTML in cdataElements: self.tokenizer.state = self.tokenizer.rcdataState elif self.innerHTML in rcdataElements: self.tokenizer.state = self.tokenizer.rawtextState elif self.innerHTML == 'plaintext': self.tokenizer.state = self.tokenizer.plaintextState else: # state already is data state # self.tokenizer.state = self.tokenizer.dataState pass self.phase = self.phases["beforeHtml"] self.phase.insertHtmlElement() self.resetInsertionMode() else: self.innerHTML = False self.phase = self.phases["initial"] self.lastPhase = None self.secondaryPhase = None self.beforeRCDataPhase = None self.framesetOK = True def mainLoop(self): (CharactersToken, SpaceCharactersToken, StartTagToken, EndTagToken, CommentToken, DoctypeToken) = (tokenTypes["Characters"], tokenTypes["SpaceCharacters"], tokenTypes["StartTag"], tokenTypes["EndTag"], tokenTypes["Comment"], tokenTypes["Doctype"]) CharactersToken = tokenTypes["Characters"] SpaceCharactersToken = tokenTypes["SpaceCharacters"] StartTagToken = tokenTypes["StartTag"] EndTagToken = tokenTypes["EndTag"] CommentToken = tokenTypes["Comment"] DoctypeToken = tokenTypes["Doctype"] for token in self.normalizedTokens(): type = token["type"] if type == CharactersToken: self.phase.processCharacters(token) elif type == SpaceCharactersToken: self.phase.processSpaceCharacters(token) elif type == StartTagToken: self.selfClosingAcknowledged = False self.phase.processStartTag(token) if (token["selfClosing"] and not self.selfClosingAcknowledged): self.parseError("non-void-element-with-trailing-solidus", {"name":token["name"]}) elif type == EndTagToken: self.phase.processEndTag(token) elif type == CommentToken: self.phase.processComment(token) elif type == DoctypeToken: self.phase.processDoctype(token) else: self.parseError(token["data"], token.get("datavars", {})) # When the loop finishes it's EOF self.phase.processEOF() def normalizedTokens(self): for token in self.tokenizer: yield self.normalizeToken(token) def parse(self, stream, encoding=None, parseMeta=True, useChardet=True): """Parse a HTML document into a well-formed tree stream - a filelike object or string containing the HTML to be parsed The optional encoding parameter must be a string that indicates the encoding. If specified, that encoding will be used, regardless of any BOM or later declaration (such as in a meta element) """ self._parse(stream, innerHTML=False, encoding=encoding, parseMeta=parseMeta, useChardet=useChardet) return self.tree.getDocument() def parseFragment(self, stream, container="div", encoding=None, parseMeta=False, useChardet=True): """Parse a HTML fragment into a well-formed tree fragment container - name of the element we're setting the innerHTML property if set to None, default to 'div' stream - a filelike object or string containing the HTML to be parsed The optional encoding parameter must be a string that indicates the encoding. If specified, that encoding will be used, regardless of any BOM or later declaration (such as in a meta element) """ self._parse(stream, True, container=container, encoding=encoding) return self.tree.getFragment() def parseError(self, errorcode="XXX-undefined-error", datavars={}): # XXX The idea is to make errorcode mandatory. self.errors.append((self.tokenizer.stream.position(), errorcode, datavars)) if self.strict: raise ParseError def normalizeToken(self, token): """ HTML5 specific normalizations to the token stream """ if token["type"] == tokenTypes["StartTag"]: token["data"] = dict(token["data"][::-1]) return token def adjustMathMLAttributes(self, token): replacements = {"definitionurl":"definitionURL"} for k,v in replacements.iteritems(): if k in token["data"]: token["data"][v] = token["data"][k] del token["data"][k] def adjustSVGAttributes(self, token): replacements = { "attributename" : "attributeName", "attributetype" : "attributeType", "basefrequency" : "baseFrequency", "baseprofile" : "baseProfile", "calcmode" : "calcMode", "clippathunits" : "clipPathUnits", "contentscripttype" : "contentScriptType", "contentstyletype" : "contentStyleType", "diffuseconstant" : "diffuseConstant", "edgemode" : "edgeMode", "externalresourcesrequired" : "externalResourcesRequired", "filterres" : "filterRes", "filterunits" : "filterUnits", "glyphref" : "glyphRef", "gradienttransform" : "gradientTransform", "gradientunits" : "gradientUnits", "kernelmatrix" : "kernelMatrix", "kernelunitlength" : "kernelUnitLength", "keypoints" : "keyPoints", "keysplines" : "keySplines", "keytimes" : "keyTimes", "lengthadjust" : "lengthAdjust", "limitingconeangle" : "limitingConeAngle", "markerheight" : "markerHeight", "markerunits" : "markerUnits", "markerwidth" : "markerWidth", "maskcontentunits" : "maskContentUnits", "maskunits" : "maskUnits", "numoctaves" : "numOctaves", "pathlength" : "pathLength", "patterncontentunits" : "patternContentUnits", "patterntransform" : "patternTransform", "patternunits" : "patternUnits", "pointsatx" : "pointsAtX", "pointsaty" : "pointsAtY", "pointsatz" : "pointsAtZ", "preservealpha" : "preserveAlpha", "preserveaspectratio" : "preserveAspectRatio", "primitiveunits" : "primitiveUnits", "refx" : "refX", "refy" : "refY", "repeatcount" : "repeatCount", "repeatdur" : "repeatDur", "requiredextensions" : "requiredExtensions", "requiredfeatures" : "requiredFeatures", "specularconstant" : "specularConstant", "specularexponent" : "specularExponent", "spreadmethod" : "spreadMethod", "startoffset" : "startOffset", "stddeviation" : "stdDeviation", "stitchtiles" : "stitchTiles", "surfacescale" : "surfaceScale", "systemlanguage" : "systemLanguage", "tablevalues" : "tableValues", "targetx" : "targetX", "targety" : "targetY", "textlength" : "textLength", "viewbox" : "viewBox", "viewtarget" : "viewTarget", "xchannelselector" : "xChannelSelector", "ychannelselector" : "yChannelSelector", "zoomandpan" : "zoomAndPan" } for originalName in token["data"].keys(): if originalName in replacements: svgName = replacements[originalName] token["data"][svgName] = token["data"][originalName] del token["data"][originalName] def adjustForeignAttributes(self, token): replacements = { "xlink:actuate":("xlink", "actuate", namespaces["xlink"]), "xlink:arcrole":("xlink", "arcrole", namespaces["xlink"]), "xlink:href":("xlink", "href", namespaces["xlink"]), "xlink:role":("xlink", "role", namespaces["xlink"]), "xlink:show":("xlink", "show", namespaces["xlink"]), "xlink:title":("xlink", "title", namespaces["xlink"]), "xlink:type":("xlink", "type", namespaces["xlink"]), "xml:base":("xml", "base", namespaces["xml"]), "xml:lang":("xml", "lang", namespaces["xml"]), "xml:space":("xml", "space", namespaces["xml"]), "xmlns":(None, "xmlns", namespaces["xmlns"]), "xmlns:xlink":("xmlns", "xlink", namespaces["xmlns"]) } for originalName in token["data"].iterkeys(): if originalName in replacements: foreignName = replacements[originalName] token["data"][foreignName] = token["data"][originalName] del token["data"][originalName] def resetInsertionMode(self): # The name of this method is mostly historical. (It's also used in the # specification.) last = False newModes = { "select":"inSelect", "td":"inCell", "th":"inCell", "tr":"inRow", "tbody":"inTableBody", "thead":"inTableBody", "tfoot":"inTableBody", "caption":"inCaption", "colgroup":"inColumnGroup", "table":"inTable", "head":"inBody", "body":"inBody", "frameset":"inFrameset" } for node in self.tree.openElements[::-1]: nodeName = node.name if node == self.tree.openElements[0]: last = True if nodeName not in ['td', 'th']: # XXX assert self.innerHTML nodeName = self.innerHTML # Check for conditions that should only happen in the innerHTML # case if nodeName in ("select", "colgroup", "head", "frameset"): # XXX assert self.innerHTML if nodeName in newModes: self.phase = self.phases[newModes[nodeName]] break elif node.namespace in (namespaces["mathml"], namespaces["svg"]): self.phase = self.phases["inForeignContent"] self.secondaryPhase = self.phases["inBody"] break elif nodeName == "html": if self.tree.headPointer is None: self.phase = self.phases["beforeHead"] else: self.phase = self.phases["afterHead"] break elif last: self.phase = self.phases["inBody"] break def parseRCDataRawtext(self, token, contentType): """Generic RCDATA/RAWTEXT Parsing algorithm contentType - RCDATA or RAWTEXT """ assert contentType in ("RAWTEXT", "RCDATA") element = self.tree.insertElement(token) if contentType == "RAWTEXT": self.tokenizer.state = self.tokenizer.rawtextState else: self.tokenizer.state = self.tokenizer.rcdataState self.originalPhase = self.phase self.phase = self.phases["text"] class Phase(object): """Base class for helper object that implements each phase of processing """ # Order should be (they can be omitted): # * EOF # * Comment # * Doctype # * SpaceCharacters # * Characters # * StartTag # - startTag* methods # * EndTag # - endTag* methods def __init__(self, parser, tree): self.parser = parser self.tree = tree def processEOF(self): raise NotImplementedError def processComment(self, token): # For most phases the following is correct. Where it's not it will be # overridden. self.tree.insertComment(token, self.tree.openElements[-1]) def processDoctype(self, token): self.parser.parseError("unexpected-doctype") def processCharacters(self, token): self.tree.insertText(token["data"]) def processSpaceCharacters(self, token): self.tree.insertText(token["data"]) def processStartTag(self, token): self.startTagHandler[token["name"]](token) def startTagHtml(self, token): if self.parser.firstStartTag == False and token["name"] == "html": self.parser.parseError("non-html-root") # XXX Need a check here to see if the first start tag token emitted is # this token... If it's not, invoke self.parser.parseError(). for attr, value in token["data"].iteritems(): if attr not in self.tree.openElements[0].attributes: self.tree.openElements[0].attributes[attr] = value self.parser.firstStartTag = False def processEndTag(self, token): self.endTagHandler[token["name"]](token) class InitialPhase(Phase): def processSpaceCharacters(self, token): pass def processComment(self, token): self.tree.insertComment(token, self.tree.document) def processDoctype(self, token): name = token["name"] publicId = token["publicId"] systemId = token["systemId"] correct = token["correct"] if (name != "html" or publicId != None or systemId != None and systemId != "about:legacy-compat"): self.parser.parseError("unknown-doctype") if publicId is None: publicId = "" self.tree.insertDoctype(token) if publicId != "": publicId = publicId.translate(asciiUpper2Lower) if (not correct or token["name"] != "html" or publicId.startswith( ("+//silmaril//dtd html pro v0r11 19970101//", "-//advasoft ltd//dtd html 3.0 aswedit + extensions//", "-//as//dtd html 3.0 aswedit + extensions//", "-//ietf//dtd html 2.0 level 1//", "-//ietf//dtd html 2.0 level 2//", "-//ietf//dtd html 2.0 strict level 1//", "-//ietf//dtd html 2.0 strict level 2//", "-//ietf//dtd html 2.0 strict//", "-//ietf//dtd html 2.0//", "-//ietf//dtd html 2.1e//", "-//ietf//dtd html 3.0//", "-//ietf//dtd html 3.2 final//", "-//ietf//dtd html 3.2//", "-//ietf//dtd html 3//", "-//ietf//dtd html level 0//", "-//ietf//dtd html level 1//", "-//ietf//dtd html level 2//", "-//ietf//dtd html level 3//", "-//ietf//dtd html strict level 0//", "-//ietf//dtd html strict level 1//", "-//ietf//dtd html strict level 2//", "-//ietf//dtd html strict level 3//", "-//ietf//dtd html strict//", "-//ietf//dtd html//", "-//metrius//dtd metrius presentational//", "-//microsoft//dtd internet explorer 2.0 html strict//", "-//microsoft//dtd internet explorer 2.0 html//", "-//microsoft//dtd internet explorer 2.0 tables//", "-//microsoft//dtd internet explorer 3.0 html strict//", "-//microsoft//dtd internet explorer 3.0 html//", "-//microsoft//dtd internet explorer 3.0 tables//", "-//netscape comm. corp.//dtd html//", "-//netscape comm. corp.//dtd strict html//", "-//o'reilly and associates//dtd html 2.0//", "-//o'reilly and associates//dtd html extended 1.0//", "-//o'reilly and associates//dtd html extended relaxed 1.0//", "-//softquad software//dtd hotmetal pro 6.0::19990601::extensions to html 4.0//", "-//softquad//dtd hotmetal pro 4.0::19971010::extensions to html 4.0//", "-//spyglass//dtd html 2.0 extended//", "-//sq//dtd html 2.0 hotmetal + extensions//", "-//sun microsystems corp.//dtd hotjava html//", "-//sun microsystems corp.//dtd hotjava strict html//", "-//w3c//dtd html 3 1995-03-24//", "-//w3c//dtd html 3.2 draft//", "-//w3c//dtd html 3.2 final//", "-//w3c//dtd html 3.2//", "-//w3c//dtd html 3.2s draft//", "-//w3c//dtd html 4.0 frameset//", "-//w3c//dtd html 4.0 transitional//", "-//w3c//dtd html experimental 19960712//", "-//w3c//dtd html experimental 970421//", "-//w3c//dtd w3 html//", "-//w3o//dtd w3 html 3.0//", "-//webtechs//dtd mozilla html 2.0//", "-//webtechs//dtd mozilla html//")) or publicId in ("-//w3o//dtd w3 html strict 3.0//en//", "-/w3c/dtd html 4.0 transitional/en", "html") or publicId.startswith( ("-//w3c//dtd html 4.01 frameset//", "-//w3c//dtd html 4.01 transitional//")) and systemId == None or systemId and systemId.lower() == "http://www.ibm.com/data/dtd/v11/ibmxhtml1-transitional.dtd"): self.parser.compatMode = "quirks" elif (publicId.startswith( ("-//w3c//dtd xhtml 1.0 frameset//", "-//w3c//dtd xhtml 1.0 transitional//")) or publicId.startswith( ("-//w3c//dtd html 4.01 frameset//", "-//w3c//dtd html 4.01 transitional//")) and systemId != None): self.parser.compatMode = "limited quirks" self.parser.phase = self.parser.phases["beforeHtml"] def anythingElse(self): self.parser.compatMode = "quirks" self.parser.phase = self.parser.phases["beforeHtml"] def processCharacters(self, token): self.parser.parseError("expected-doctype-but-got-chars") self.anythingElse() self.parser.phase.processCharacters(token) def processStartTag(self, token): self.parser.parseError("expected-doctype-but-got-start-tag", {"name": token["name"]}) self.anythingElse() self.parser.phase.processStartTag(token) def processEndTag(self, token): self.parser.parseError("expected-doctype-but-got-end-tag", {"name": token["name"]}) self.anythingElse() self.parser.phase.processEndTag(token) def processEOF(self): self.parser.parseError("expected-doctype-but-got-eof") self.anythingElse() self.parser.phase.processEOF() class BeforeHtmlPhase(Phase): # helper methods def insertHtmlElement(self): self.tree.insertRoot(impliedTagToken("html", "StartTag")) self.parser.phase = self.parser.phases["beforeHead"] # other def processEOF(self): self.insertHtmlElement() self.parser.phase.processEOF() def processComment(self, token): self.tree.insertComment(token, self.tree.document) def processSpaceCharacters(self, token): pass def processCharacters(self, token): self.insertHtmlElement() self.parser.phase.processCharacters(token) def processStartTag(self, token): if token["name"] == "html": self.parser.firstStartTag = True self.insertHtmlElement() self.parser.phase.processStartTag(token) def processEndTag(self, token): if token["name"] not in ("head", "body", "html", "br"): self.parser.parseError("unexpected-end-tag-before-html", {"name": token["name"]}) else: self.insertHtmlElement() self.parser.phase.processEndTag(token) class BeforeHeadPhase(Phase): def __init__(self, parser, tree): Phase.__init__(self, parser, tree) self.startTagHandler = utils.MethodDispatcher([ ("html", self.startTagHtml), ("head", self.startTagHead) ]) self.startTagHandler.default = self.startTagOther self.endTagHandler = utils.MethodDispatcher([ (("head", "body", "html", "br"), self.endTagImplyHead) ]) self.endTagHandler.default = self.endTagOther def processEOF(self): self.startTagHead(impliedTagToken("head", "StartTag")) self.parser.phase.processEOF() def processSpaceCharacters(self, token): pass def processCharacters(self, token): self.startTagHead(impliedTagToken("head", "StartTag")) self.parser.phase.processCharacters(token) def startTagHtml(self, token): self.parser.phases["inBody"].processStartTag(token) def startTagHead(self, token): self.tree.insertElement(token) self.tree.headPointer = self.tree.openElements[-1] self.parser.phase = self.parser.phases["inHead"] def startTagOther(self, token): self.startTagHead(impliedTagToken("head", "StartTag")) self.parser.phase.processStartTag(token) def endTagImplyHead(self, token): self.startTagHead(impliedTagToken("head", "StartTag")) self.parser.phase.processEndTag(token) def endTagOther(self, token): self.parser.parseError("end-tag-after-implied-root", {"name": token["name"]}) class InHeadPhase(Phase): def __init__(self, parser, tree): Phase.__init__(self, parser, tree) self.startTagHandler = utils.MethodDispatcher([ ("html", self.startTagHtml), ("title", self.startTagTitle), (("noscript", "noframes", "style"), self.startTagNoScriptNoFramesStyle), ("script", self.startTagScript), (("base", "link", "command"), self.startTagBaseLinkCommand), ("meta", self.startTagMeta), ("head", self.startTagHead) ]) self.startTagHandler.default = self.startTagOther self. endTagHandler = utils.MethodDispatcher([ ("head", self.endTagHead), (("br", "html", "body"), self.endTagHtmlBodyBr) ]) self.endTagHandler.default = self.endTagOther # helper def appendToHead(self, element): if self.tree.headPointer is not None: self.tree.headPointer.appendChild(element) else: assert self.parser.innerHTML self.tree.openElementsw[-1].appendChild(element) # the real thing def processEOF (self): self.anythingElse() self.parser.phase.processEOF() def processCharacters(self, token): self.anythingElse() self.parser.phase.processCharacters(token) def startTagHtml(self, token): self.parser.phases["inBody"].processStartTag(token) def startTagHead(self, token): self.parser.parseError("two-heads-are-not-better-than-one") def startTagBaseLinkCommand(self, token): self.tree.insertElement(token) self.tree.openElements.pop() token["selfClosingAcknowledged"] = True def startTagMeta(self, token): self.tree.insertElement(token) self.tree.openElements.pop() token["selfClosingAcknowledged"] = True attributes = token["data"] if self.parser.tokenizer.stream.charEncoding[1] == "tentative": if "charset" in attributes: self.parser.tokenizer.stream.changeEncoding(attributes["charset"]) elif "content" in attributes: # Encoding it as UTF-8 here is a hack, as really we should pass # the abstract Unicode string, and just use the # ContentAttrParser on that, but using UTF-8 allows all chars # to be encoded and as a ASCII-superset works. data = inputstream.EncodingBytes(attributes["content"].encode("utf-8")) parser = inputstream.ContentAttrParser(data) codec = parser.parse() self.parser.tokenizer.stream.changeEncoding(codec) def startTagTitle(self, token): self.parser.parseRCDataRawtext(token, "RCDATA") def startTagNoScriptNoFramesStyle(self, token): #Need to decide whether to implement the scripting-disabled case self.parser.parseRCDataRawtext(token, "RAWTEXT") def startTagScript(self, token): self.tree.insertElement(token) self.parser.tokenizer.state = self.parser.tokenizer.scriptDataState self.parser.originalPhase = self.parser.phase self.parser.phase = self.parser.phases["text"] def startTagOther(self, token): self.anythingElse() self.parser.phase.processStartTag(token) def endTagHead(self, token): node = self.parser.tree.openElements.pop() assert node.name == "head", "Expected head got %s"%node.name self.parser.phase = self.parser.phases["afterHead"] def endTagHtmlBodyBr(self, token): self.anythingElse() self.parser.phase.processEndTag(token) def endTagOther(self, token): self.parser.parseError("unexpected-end-tag", {"name": token["name"]}) def anythingElse(self): self.endTagHead(impliedTagToken("head")) # XXX If we implement a parser for which scripting is disabled we need to # implement this phase. # # class InHeadNoScriptPhase(Phase): class AfterHeadPhase(Phase): def __init__(self, parser, tree): Phase.__init__(self, parser, tree) self.startTagHandler = utils.MethodDispatcher([ ("html", self.startTagHtml), ("body", self.startTagBody), ("frameset", self.startTagFrameset), (("base", "link", "meta", "noframes", "script", "style", "title"), self.startTagFromHead), ("head", self.startTagHead) ]) self.startTagHandler.default = self.startTagOther self.endTagHandler = utils.MethodDispatcher([(("body", "html", "br"), self.endTagHtmlBodyBr)]) self.endTagHandler.default = self.endTagOther def processEOF(self): self.anythingElse() self.parser.phase.processEOF() def processCharacters(self, token): self.anythingElse() self.parser.phase.processCharacters(token) def startTagBody(self, token): self.parser.framesetOK = False self.tree.insertElement(token) self.parser.phase = self.parser.phases["inBody"] def startTagFrameset(self, token): self.tree.insertElement(token) self.parser.phase = self.parser.phases["inFrameset"] def startTagFromHead(self, token): self.parser.parseError("unexpected-start-tag-out-of-my-head", {"name": token["name"]}) self.tree.openElements.append(self.tree.headPointer) self.parser.phases["inHead"].processStartTag(token) for node in self.tree.openElements[::-1]: if node.name == "head": self.tree.openElements.remove(node) break def startTagHead(self, token): self.parser.parseError("unexpected-start-tag", {"name":token["name"]}) def startTagOther(self, token): self.anythingElse() self.parser.phase.processStartTag(token) def endTagHtmlBodyBr(self, token): self.anythingElse() self.parser.phase.processEndTag(token) def endTagOther(self, token): self.parser.parseError("unexpected-end-tag", {"name":token["name"]}) def anythingElse(self): self.tree.insertElement(impliedTagToken("body", "StartTag")) self.parser.phase = self.parser.phases["inBody"] self.parser.framesetOK = True class InBodyPhase(Phase): # http://www.whatwg.org/specs/web-apps/current-work/#parsing-main-inbody # the really-really-really-very crazy mode def __init__(self, parser, tree): Phase.__init__(self, parser, tree) #Keep a ref to this for special handling of whitespace in 
        self.processSpaceCharactersNonPre = self.processSpaceCharacters

        self.startTagHandler = utils.MethodDispatcher([
            ("html", self.startTagHtml),
            (("base", "command", "link", "meta", "noframes", "script", "style", 
              "title"), self.startTagProcessInHead),
            ("body", self.startTagBody),
            ("frameset", self.startTagFrameset),
            (("address", "article", "aside", "blockquote", "center", "datagrid",
              "details", "dir", "div", "dl", "fieldset", "figure",
              "footer", "header", "hgroup", "menu", "nav", "ol", "p",
              "section", "ul"),
              self.startTagCloseP),
            (("pre", "listing"), self.startTagPreListing),
            ("form", self.startTagForm),
            (("li", "dd", "dt"), self.startTagListItem),
            ("plaintext",self.startTagPlaintext),
            (headingElements, self.startTagHeading),
            ("a", self.startTagA),
            (("b", "big", "code", "em", "font", "i", "s", "small", "strike", 
              "strong", "tt", "u"),self.startTagFormatting),
            ("nobr", self.startTagNobr),
            ("button", self.startTagButton),
            (("applet", "marquee", "object"), self.startTagAppletMarqueeObject),
            ("xmp", self.startTagXmp),
            ("table", self.startTagTable),
            (("area", "basefont", "bgsound", "br", "embed", "img", "input",
              "keygen", "spacer", "wbr"), self.startTagVoidFormatting),
            (("param", "source"), self.startTagParamSource),
            ("hr", self.startTagHr),
            ("image", self.startTagImage),
            ("isindex", self.startTagIsIndex),
            ("textarea", self.startTagTextarea),
            ("iframe", self.startTagIFrame),
            (("noembed", "noframes", "noscript"), self.startTagRawtext),
            ("select", self.startTagSelect),
            (("rp", "rt"), self.startTagRpRt),
            (("option", "optgroup"), self.startTagOpt),
            (("math"), self.startTagMath),
            (("svg"), self.startTagSvg),
            (("caption", "col", "colgroup", "frame", "head",
              "tbody", "td", "tfoot", "th", "thead",
              "tr"), self.startTagMisplaced)
        ])
        self.startTagHandler.default = self.startTagOther

        self.endTagHandler = utils.MethodDispatcher([
            ("body",self.endTagBody),
            ("html",self.endTagHtml),
            (("address", "article", "aside", "blockquote", "center", "datagrid",
              "details", "dir", "div", "dl", "fieldset", "figure",
              "footer", "header", "hgroup", "listing", "menu", "nav", "ol", "pre", 
              "section", "ul"), self.endTagBlock),
            ("form", self.endTagForm),
            ("p",self.endTagP),
            (("dd", "dt", "li"), self.endTagListItem),
            (headingElements, self.endTagHeading),
            (("a", "b", "big", "code", "em", "font", "i", "nobr", "s", "small",
              "strike", "strong", "tt", "u"), self.endTagFormatting),
            (("applet", "button", "marquee", "object"), self.endTagAppletButtonMarqueeObject),
            ("br", self.endTagBr),
            ])
        self.endTagHandler.default = self.endTagOther

    # helper
    def addFormattingElement(self, token):
        self.tree.insertElement(token)
        self.tree.activeFormattingElements.append(
            self.tree.openElements[-1])

    # the real deal
    def processEOF(self):
        allowed_elements = frozenset(("dd", "dt", "li", "p", "tbody", "td",
                                      "tfoot", "th", "thead", "tr", "body",
                                      "html"))
        for node in self.tree.openElements[::-1]:
            if node.name not in allowed_elements:
                self.parser.parseError("expected-closing-tag-but-got-eof")
                break
        #Stop parsing
    
    def processSpaceCharactersDropNewline(self, token):
        # Sometimes (start of 
, , and