X-Git-Url: https://git.ucc.asn.au/?a=blobdiff_plain;ds=sidebyside;f=extra%2Ffeedparser.py;h=b9144a9e7138d8906d981a167035bc409b142b67;hb=d1797b9c5a2ab569399bf21361826bbfceed0375;hp=024194e724ee7ea38fe69cd168fae362a58af5b4;hpb=783caef083f5dcafaed71dfa3f3a8035109717e9;p=planet-ucc.git diff --git a/extra/feedparser.py b/extra/feedparser.py index 024194e..b9144a9 100644 --- a/extra/feedparser.py +++ b/extra/feedparser.py @@ -1,60 +1,148 @@ #!/usr/bin/env python """Universal feed parser -Visit http://diveintomark.org/projects/feed_parser/ for the latest version - -Handles RSS 0.9x, RSS 1.0, RSS 2.0, Atom feeds - -Things it handles that choke other parsers: -- bastard combinations of RSS 0.9x and RSS 1.0 -- illegal 8-bit XML characters -- naked and/or invalid HTML in description -- content:encoded, xhtml:body, fullitem -- guid -- elements in non-standard namespaces or non-default namespaces -- multiple content items per entry (Atom) -- multiple links per entry (Atom) - -Other features: -- resolves relative URIs in some elements - - uses xml:base to define base URI - - uses URI of feed if no xml:base is given - - to control which elements are resolved, set _FeedParserMixin.can_be_relative_uri -- resolves relative URIs within embedded markup - - to control which elements are resolved, set _FeedParserMixin.can_contain_relative_uris -- sanitizes embedded markup in some elements - - to allow/disallow HTML elements, set _HTMLSanitizer.acceptable_elements - - to allow/disallow HTML attributes, set _HTMLSanitizer.acceptable_attributes - - to control which feed elements are sanitized, set _FeedParserMixin.can_contain_dangerous_markup - - to disable entirely (NOT RECOMMENDED), set _FeedParserMixin.can_contain_dangerous_markup = [] -- optionally tidies embedded markup - - fixes malformed HTML - - converts to XHTML - - converts character entities to numeric entities - - requires mxTidy - -Required: Python 2.1 or later -Recommended: Python 2.3 or later +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 2.4 or later +Recommended: CJKCodecs and iconv_codec """ -__version__ = "3.0-beta-14" +__version__ = "5.0.1" +__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 " -__copyright__ = "Copyright 2002-4, Mark Pilgrim" __contributors__ = ["Jason Diamond ", "John Beimler ", - "Fazal Majid "] -__license__ = "Python" + "Fazal Majid ", + "Aaron Swartz ", + "Kevin Marks ", + "Sam Ruby ", + "Ade Oshineye ", + "Martin Pool ", + "Kurt McKee "] _debug = 0 -# if you are embedding feedparser in a larger application, you should change this to your application name and URL -USER_AGENT = "UniversalFeedParser/%s%s +http://diveintomark.org/projects/feed_parser/" % (__version__, _debug and "-debug" or "") +# 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__ -# ---------- required modules (should come with any Python distribution) ---------- -import sgmllib, re, sys, copy, urlparse, time, rfc822 +# 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 + +# ---------- Python 3 modules (make it work if possible) ---------- +try: + import rfc822 +except ImportError: + from email import _parseaddr as rfc822 + +try: + # Python 3.1 introduces bytes.maketrans and simultaneously + # deprecates string.maketrans; use bytes.maketrans if possible + _maketrans = bytes.maketrans +except (NameError, AttributeError): + import string + _maketrans = string.maketrans + +# base64 support for Atom feeds that contain embedded binary data try: - from cStringIO import StringIO as _StringIO + import base64, binascii + # Python 3.1 deprecates decodestring in favor of decodebytes + _base64decode = getattr(base64, 'decodebytes', base64.decodestring) except: - from StringIO import StringIO as _StringIO + base64 = binascii = None + +def _s2bytes(s): + # Convert a UTF-8 str to bytes if the interpreter is Python 3 + try: + return bytes(s, 'utf8') + except (NameError, TypeError): + # In Python 2.5 and below, bytes doesn't exist (NameError) + # In Python 2.6 and above, bytes and str are the same (TypeError) + return s + +def _l2bytes(l): + # Convert a list of ints to bytes if the interpreter is Python 3 + try: + if bytes is not str: + # In Python 2.6 and above, this call won't raise an exception + # but it will return bytes([65]) as '[65]' instead of 'A' + return bytes(l) + raise NameError + except NameError: + return ''.join(map(chr, l)) + +# If you want feedparser to allow all URL schemes, set this to () +# List culled from Python's urlparse documentation at: +# http://docs.python.org/library/urlparse.html +# as well as from "URI scheme" at Wikipedia: +# https://secure.wikimedia.org/wikipedia/en/wiki/URI_scheme +# Many more will likely need to be added! +ACCEPTABLE_URI_SCHEMES = ( + 'file', 'ftp', 'gopher', 'h323', 'hdl', 'http', 'https', 'imap', 'mailto', + 'mms', 'news', 'nntp', 'prospero', 'rsync', 'rtsp', 'rtspu', 'sftp', + 'shttp', 'sip', 'sips', 'snews', 'svn', 'svn+ssh', 'telnet', 'wais', + # Additional common-but-unofficial schemes + 'aim', 'callto', 'cvs', 'facetime', 'feed', 'git', 'gtalk', 'irc', 'ircs', + 'irc6', 'itms', 'mms', 'msnim', 'skype', 'ssh', 'smb', 'svn', 'ymsg', +) +#ACCEPTABLE_URI_SCHEMES = () + +# ---------- required modules (should come with any Python distribution) ---------- +import sgmllib, re, sys, copy, urlparse, time, types, cgi, urllib, urllib2, datetime +try: + from io import BytesIO as _StringIO +except ImportError: + 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) ---------- @@ -63,58 +151,103 @@ try: import gzip except: gzip = None - -# timeoutsocket allows feedparser to time out rather than hang forever on ultra-slow servers. -# Python 2.3 now has this functionality available in the standard socket library, so under -# 2.3 you don't need to install anything. -import socket -if hasattr(socket, 'setdefaulttimeout'): - socket.setdefaulttimeout(10) -else: - try: - import timeoutsocket # http://www.timo-tasi.org/python/timeoutsocket.py - timeoutsocket.setDefaultSocketTimeout(10) - except ImportError: - pass -import urllib2 - -# mxtidy allows feedparser to tidy malformed embedded HTML markup in description, content, etc. -# this does not affect HTML sanitizing, which is self-contained in the _HTMLSanitizer class try: - from mx.Tidy import Tidy as _mxtidy # http://www.lemburg.com/files/python/mxTidy.html + import zlib except: - _mxtidy = None - -# If a real XML parser is available, feedparser will attempt to use it. feedparser works -# with both the built-in SAX parser and PyXML SAX parser. 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 just fall back on regex-based parsing. If XML libraries are -# available but the feed turns out not to be well-formed XML, feedparser will fall back -# on regex-based parsing and set the "bozo" bit in the results to indicate that the feed -# author is a bozo who can't generate well-formed XML. The two advantages of using a real -# XML parser are (1) Unicode support, and (2) to get people to stop yelling at me for not -# using one. + 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 - from xml.sax.saxutils import escape as xmlescape + 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): - data = data.replace("&", "&") - data = data.replace(">", ">") - data = data.replace("<", "<") + 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 +# cjkcodecs and iconv_codec provide support for more character encodings. +# Both are available from http://cjkpython.i18n.org/ try: - import base64, binascii + import cjkcodecs.aliases except: - base64 = binascii = None - + 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): + match = self.endbracket.match(string,index) + if match is not None: + # Returning a new object in the calling thread's context + # resolves a thread-safety. + return EndBracketMatch(match) + return None + class EndBracketMatch: + def __init__(self, match): + self.match = match + def start(self, n): + return self.match.end(n) + sgmllib.endbracket = EndBracketRegEx() SUPPORTED_VERSIONS = {'': 'unknown', 'rss090': 'RSS 0.90', @@ -129,118 +262,325 @@ SUPPORTED_VERSIONS = {'': 'unknown', 'atom01': 'Atom 0.1', 'atom02': 'Atom 0.2', 'atom03': 'Atom 0.3', - 'atom': 'Atom (unknown version)' + 'atom10': 'Atom 1.0', + 'atom': 'Atom (unknown version)', + 'cdf': 'CDF', + 'hotrss': 'Hot RSS' } try: - dict + UserDict = dict except NameError: - # Python 2.1 does not have a built-in dict() function + # 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': ['summary', 'subtitle'], + '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.__contains__(self, k): + return UserDict.__getitem__(self, k) + if UserDict.__contains__(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.__contains__(self, key) + except AttributeError: + return False + # This alias prevents the 2to3 tool from changing the semantics of the + # __contains__ function below and exhausting the maximum recursion depth + __has_key = has_key + + 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 + ) + _ebcdic_to_ascii_map = _maketrans( \ + _l2bytes(range(256)), _l2bytes(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://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://purl.org/rss/1.0/modules/rss091#": "", + 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://postneo.com/icbm/": "icbm", - "http://purl.org/rss/1.0/modules/image/": "image", - "http://xmlns.com/foaf/0.1/": "foaf", - "http://freshmeat.net/rss/fm/": "fm", - "http://purl.org/rss/1.0/modules/link/": "l", - "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://purl.org/rss/1.0/modules/servicestatus/": "ss", - "http://hacks.benhammersley.com/rss/streaming/": "str", - "http://purl.org/rss/1.0/modules/subscription/": "sub", - "http://purl.org/rss/1.0/modules/syndication/": "sy", - "http://purl.org/rss/1.0/modules/taxonomy/": "taxo", - "http://purl.org/rss/1.0/modules/threading/": "thr", - "http://purl.org/rss/1.0/modules/textinput/": "ti", - "http://madskills.com/public/xml/rss/module/trackback/":"trackback", - "http://wellformedweb.org/CommentAPI/": "wfw", - "http://purl.org/rss/1.0/modules/wiki/": "wiki", - "http://schemas.xmlsoap.org/soap/envelope/": "soap", - "http://www.w3.org/1999/xhtml": "xhtml", - "http://www.w3.org/XML/1998/namespace": "xml" + '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', 'comments'] - can_contain_relative_uris = ['content', 'description', 'title', 'summary', 'info', 'tagline', 'copyright'] - can_contain_dangerous_markup = ['content', 'description', 'title', 'summary', 'info', 'tagline', 'copyright'] + 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): - if _debug: sys.stderr.write("initializing FeedParser\n") - self.channel = {} # channel- or feed-level data - self.items = [] # list of item- or entry-level data + 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; - # some of this is kind of out of control and should - # probably be refactored into a finite state machine - self.inchannel = 0 - self.initem = 0 + # 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.contentparams = {} + 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 = None + 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(), sgmllib.charref.sub(lambda m: unichr(int(m.groups()[0])), v).strip()) for k, v in 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] + # the sgml parser doesn't handle entities in attributes, but + # strict xml parsers do -- account for this difference + if isinstance(self, _LooseFeedParser): + attrs = [(k, v.replace('&', '&')) for k, v in attrs] # track xml:base and xml:lang attrsD = dict(attrs) - baseuri = attrsD.get('xml:base') - if baseuri: - if _debug: sys.stderr.write('self.baseuri=%s\n' % baseuri) - self.baseuri = baseuri - lang = attrsD.get('xml:lang') + 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') + # ensure that self.baseuri is always an absolute URI that + # uses a whitelisted URI scheme (e.g. not `javscript:`) + if self.baseuri: + self.baseuri = _makeSafeAbsoluteURI(self.baseuri, baseuri) or self.baseuri + else: + 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: - self.lang = lang - self.basestack.append(baseuri) + 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 @@ -251,97 +591,128 @@ class _FeedParserMixin: self.trackNamespace(None, uri) # track inline content - if self.incontent and self.contentparams.get('mode') == 'escaped': + 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['mode'] = 'xml' - if self.incontent and self.contentparams.get('mode') == 'xml': - # Note: probably shouldn't simply recreate localname here, but - # our namespace handling isn't actually 100% correct in cases where - # the feed redefines the default namespace (which is actually - # the usual case for inline content, thanks Sam), so here we - # cheat and just reconstruct the element based on localname - # because that compensates for the bugs in our namespace handling. - # This will horribly munge inline content with non-empty qnames, - # but nobody actually does that, so I'm not fixing it. - tag = tag.split(':')[-1] - return self.handle_data("<%s%s>" % (tag, "".join([' %s="%s"' % t for t in attrs])), escape=0) + 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 - try: + if tag.find(':') <> -1: prefix, suffix = tag.split(':', 1) - except ValueError: + else: prefix, suffix = '', tag prefix = self.namespacemap.get(prefix, prefix) if prefix: prefix = prefix + '_' + # special hack for better tracking of empty textinput/image elements in illformed feeds + if (not prefix) and tag not in ('title', 'link', 'description', 'name'): + self.intextinput = 0 + if (not prefix) and tag not in ('title', 'link', 'description', 'url', 'href', 'width', 'height'): + self.inimage = 0 + # call special handler (if defined) or default handler methodname = '_start_' + prefix + suffix try: method = getattr(self, methodname) return method(attrsD) except AttributeError: - return self.push(prefix + suffix, 1) + # 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 - try: + if tag.find(':') <> -1: prefix, suffix = tag.split(':', 1) - except ValueError: + 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.get('mode') == 'escaped': + if self.incontent and self.contentparams.has_key('type') and not self.contentparams.get('type', 'xml').endswith('xml'): # element declared itself as escaped markup, but it isn't really - self.contentparams['mode'] = 'xml' - if self.incontent and self.contentparams.get('mode') == 'xml': + 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) + 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]: - baseuri = self.basestack[-1] - if _debug: sys.stderr.write('self.baseuri=%s\n' % baseuri) - self.baseuri = baseuri + self.baseuri = self.basestack[-1] if self.langstack: - lang = self.langstack.pop() - if lang: - self.lang = lang + 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" - # Reconstruct the original character reference. + # called for each character reference, e.g. for ' ', ref will be '160' if not self.elementstack: return - text = "&#%s;" % ref + 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" - # Reconstruct the original entity reference. + # called for each entity reference, e.g. for '©', ref will be 'copy' if not self.elementstack: return - text = "&%s;" % ref + 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('mode') == 'xml': - text = xmlescape(text) + if escape and self.contentparams.get('type') == 'application/xhtml+xml': + text = _xmlescape(text) self.elementstack[-1][2].append(text) def handle_comment(self, text): @@ -353,144 +724,267 @@ class _FeedParserMixin: pass def handle_decl(self, text): - # called for the DOCTYPE, if present, e.g. - # - if text.count('http://my.netscape.com/publish/formats/rss-0.91.dtd'): - self.version = 'rss091n' - - _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.updatepos(declstartpos, i) - self.error("expected name token") + pass def parse_declaration(self, i): # override internal declaration handler to handle CDATA blocks - if _debug: sys.stderr.write("entering parse_declaration\n") - if re.search(r'^', self.rawdata[i:]): - if _debug: sys.stderr.write("found Netscape DOCTYPE\n") - self.version = 'rss091n' + if _debug: sys.stderr.write('entering parse_declaration\n') if self.rawdata[i:i+9] == '', i) - if k == -1: k = len(self.rawdata) - self.handle_data(xmlescape(self.rawdata[i+9:k]), 0) + 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) - return k+1 - + 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' or contentType == 'plain': + contentType = 'text/plain' + elif contentType == 'html': + contentType = 'text/html' + elif contentType == 'xhtml': + contentType = 'application/xhtml+xml' + return contentType + def trackNamespace(self, prefix, uri): - if (prefix, uri) == (None, 'http://my.netscape.com/rdf/simple/0.9/') and not self.version: + loweruri = uri.lower() + if (prefix, loweruri) == (None, 'http://my.netscape.com/rdf/simple/0.9/') and not self.version: self.version = 'rss090' - if not prefix: return - if uri.find('backend.userland.com/rss') <> -1: + 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' - if self.namespaces.has_key(uri): - self.namespacemap[prefix] = self.namespaces[uri] + 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 urlparse.urljoin(self.baseuri or '', uri) + return _urljoin(self.baseuri or '', uri) def decodeEntities(self, element, data): - if self.contentparams.get('mode') == 'escaped': - 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"' % (t[0],_xmlescape(t[1],{'"':'"'})) for t in attrs]) + def push(self, element, expectingText): -# print 'push', element, expectingText -# while self.elementstack and self.elementstack[-1][1]: -# self.pop(self.elementstack[-1][0]) self.elementstack.append([element, expectingText, []]) - def pop(self, element): -# print 'pop', element + def pop(self, element, stripWhitespace=1): if not self.elementstack: return -# while self.elementstack[-1][0] != element: self.pop(self.elementstack[-1][0]) if self.elementstack[-1][0] != element: return - + element, expectingText, pieces = self.elementstack.pop() - output = "".join(pieces) - output = output.strip() + + 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] + + # Ensure each piece is a str for Python 3 + for (i, v) in enumerate(pieces): + if not isinstance(v, basestring): + pieces[i] = v.decode('utf-8') + + output = ''.join(pieces) + if stripWhitespace: + output = output.strip() if not expectingText: return output - + # decode base64 content - if self.contentparams.get('mode') == 'base64' and base64: + if base64 and self.contentparams.get('base64', 0): try: - output = base64.decodestring(output) + output = _base64decode(output) except binascii.Error: pass except binascii.Incomplete: pass + except TypeError: + # In Python 3, base64 takes and outputs bytes, not str + # This may not be the most correct way to accomplish this + output = _base64decode(output.encode('utf-8')).decode('utf-8') # resolve relative URIs if (element in self.can_be_relative_uri) and output: output = self.resolveURI(output) # decode entities within embedded markup - output = self.decodeEntities(element, output) + 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 element in self.can_contain_relative_uris: - output = _resolveRelativeURIs(output, self.baseuri) + 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 element in self.can_contain_dangerous_markup: - output = _sanitizeHTML(output) - + 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 in ('utf-8', 'utf-8_INVALID_PYTHON_3') 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.initem: + if self.inentry and not self.insource: if element == 'content': - self.items[-1].setdefault(element, []) + self.entries[-1].setdefault(element, []) contentparams = copy.deepcopy(self.contentparams) contentparams['value'] = output - self.items[-1][element].append(contentparams) - elif element == 'category': - self.items[-1][element] = output - domain = self.items[-1]['categories'][-1][0] - self.items[-1]['categories'][-1] = (domain, output) - elif element == 'source': - self.items[-1]['source']['value'] = output - elif element == 'link': - self.items[-1][element] = output - if output: - self.items[-1]['links'][-1]['href'] = output - else: - if self.incontent and element != 'description': - contentparams = copy.deepcopy(self.contentparams) - contentparams['value'] = output - self.items[-1][element + '_detail'] = contentparams - self.items[-1][element] = output - elif self.inchannel and (not self.intextinput) and (not self.inimage): - if element == 'category': - domain = self.channel['categories'][-1][0] - self.channel['categories'][-1] = (domain, output) + self.entries[-1][element].append(contentparams) elif element == 'link': - self.channel['links'][-1]['href'] = output + if not self.inimage: + # query variables in urls in link elements are improperly + # converted from `?a=1&b=2` to `?a=1&b;=2` as if they're + # unhandled character references. fix this special case. + output = re.sub("&([A-Za-z0-9_]+);", "&\g<1>", output) + self.entries[-1][element] = output + if output: + self.entries[-1]['links'][-1]['href'] = output else: - if self.incontent and element != 'description': + if element == 'description': + element = 'summary' + self.entries[-1][element] = output + if self.incontent: contentparams = copy.deepcopy(self.contentparams) contentparams['value'] = output - self.channel[element + '_detail'] = contentparams - self.channel[element] = 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': + # fix query variables; see above for the explanation + output = re.sub("&([A-Za-z0-9_]+);", "&\g<1>", output) + context[element] = output + 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, s): + 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'',s) or re.search("&#?\w+;",s)): 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: @@ -503,19 +997,47 @@ class _FeedParserMixin: def _getAttribute(self, attrsD, name): return attrsD.get(self._mapToStandardPrefix(name)) - def _save(self, key, value): - if value: - if self.initem: - self.items[-1].setdefault(key, value) - elif self.channel: - self.channel.setdefault(key, value) + 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, overwrite=False): + context = self._getContext() + if overwrite: + context[key] = value + else: + context.setdefault(key, value) def _start_rss(self, attrsD): versionmap = {'0.91': 'rss091u', '0.92': 'rss092', '0.93': 'rss093', '0.94': 'rss094'} - if not self.version: + #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: @@ -524,12 +1046,27 @@ class _FeedParserMixin: self.version = 'rss20' else: self.version = 'rss' + + def _start_dlhottitles(self, attrsD): + self.version = 'hotrss' def _start_channel(self, attrsD): - self.inchannel = 1 - + 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.inchannel = 1 + self.infeed = 1 versionmap = {'0.1': 'atom01', '0.2': 'atom02', '0.3': 'atom03'} @@ -542,29 +1079,45 @@ class _FeedParserMixin: self.version = 'atom' def _end_channel(self): - self.inchannel = 0 + self.infeed = 0 _end_feed = _end_channel def _start_image(self, attrsD): + context = self._getContext() + if not self.inentry: + 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) + # Append a new FeedParserDict when expecting an author + context = self._getContext() + context.setdefault('authors', []) + context['authors'].append(FeedParserDict()) _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') @@ -573,153 +1126,209 @@ class _FeedParserMixin: _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({}) + 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.inauthor: + 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) - pass elif self.intextinput: - # TODO - pass + 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('url', 0) + self.push('href', 1) _start_homepage = _start_url _start_uri = _start_url def _end_url(self): - value = self.pop('url') + value = self.pop('href') if self.inauthor: - self._save_author('url', value) + self._save_author('href', value) elif self.incontributor: - self._save_contributor('url', value) - elif self.inimage: - # TODO - pass - elif self.intextinput: - # TODO - pass + 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.inauthor: + 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) - pass - elif self.inimage: - # TODO - pass - elif self.intextinput: - # TODO - pass + _end_itunes_email = _end_email def _getContext(self): - if self.initem: - context = self.items[-1] + if self.insource: + context = self.sourcedata + elif self.inimage and self.feeddata.has_key('image'): + context = self.feeddata['image'] + elif self.intextinput: + context = self.feeddata['textinput'] + elif self.inentry: + context = self.entries[-1] else: - context = self.channel + context = self.feeddata return context - def _save_author(self, key, value): + def _save_author(self, key, value, prefix='author'): context = self._getContext() - context.setdefault('author_detail', {}) - context['author_detail'][key] = value + context.setdefault(prefix + '_detail', FeedParserDict()) + context[prefix + '_detail'][key] = value self._sync_author_detail() + context.setdefault('authors', [FeedParserDict()]) + context['authors'][-1][key] = value def _save_contributor(self, key, value): context = self._getContext() - context.setdefault('contributors', [{}]) + context.setdefault('contributors', [FeedParserDict()]) context['contributors'][-1][key] = value - def _sync_author_detail(self): + def _sync_author_detail(self, key='author'): context = self._getContext() - detail = context.get('author_detail') + detail = context.get('%s_detail' % key) if detail: name = detail.get('name') email = detail.get('email') if name and email: - context['author'] = "%s (%s)" % (name, email) + context[key] = '%s (%s)' % (name, email) elif name: - context['author'] = name + context[key] = name elif email: - context['author'] = email + context[key] = email else: - author = context.get('author') + author, email = context.get(key), None if not author: return - emailmatch = re.search(r"""(([a-zA-Z0-9\_\-\.\+]+)@((\[[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.)|(([a-zA-Z0-9\-]+\.)+))([a-zA-Z]{2,4}|[0-9]{1,3})(\]?))""", author) - if not emailmatch: return - email = emailmatch.group(0) - author = author.replace(email, '') - author = author.replace('()', '') - author = author.strip() - context.setdefault('author_detail', {}) - context['author_detail']['name'] = author - context['author_detail']['email'] = email - - def _start_tagline(self, attrsD): - self.incontent += 1 - self.contentparams = {'mode': attrsD.get('mode', 'escaped'), - 'type': attrsD.get('type', 'text/plain'), - 'language': attrsD.get('xml:lang', self.lang), - 'base': attrsD.get('xml:base', self.baseuri)} - self.push('tagline', 1) - _start_subtitle = _start_tagline - - def _end_tagline(self): - value = self.pop('tagline') - self.incontent -= 1 - self.contentparams.clear() - if self.inchannel: - self.channel['description'] = value - _end_subtitle = _end_tagline + 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_copyright(self, attrsD): - self.incontent += 1 - self.contentparams = {'mode': attrsD.get('mode', 'escaped'), - 'type': attrsD.get('type', 'text/plain'), - 'language': attrsD.get('xml:lang', self.lang), - 'base': attrsD.get('xml:base', self.baseuri)} - self.push('copyright', 1) - _start_dc_rights = _start_copyright - - def _end_copyright(self): - self.pop('copyright') - self.incontent -= 1 - self.contentparams.clear() - _end_dc_rights = _end_copyright + 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.items.append({}) + self.entries.append(FeedParserDict()) self.push('item', 0) - self.initem = 1 + 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.initem = 0 + self.inentry = 0 _end_entry = _end_item def _start_dc_language(self, attrsD): @@ -736,174 +1345,229 @@ class _FeedParserMixin: def _end_dc_publisher(self): self.pop('publisher') + self._sync_author_detail('publisher') _end_webmaster = _end_dc_publisher - - def _start_dcterms_issued(self, attrsD): - self.push('issued', 1) - _start_issued = _start_dcterms_issued - def _end_dcterms_issued(self): - value = self.pop('issued') - self._save('issued_parsed', _parse_date(value)) - _end_issued = _end_dcterms_issued - - def _start_dcterms_created(self, attrsD): + 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), overwrite=True) + _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 + _start_lastbuilddate = _start_updated + + def _end_updated(self): + value = self.pop('updated') + parsed_value = _parse_date(value) + self._save('updated_parsed', parsed_value, overwrite=True) + _end_modified = _end_updated + _end_dcterms_modified = _end_updated + _end_pubdate = _end_updated + _end_dc_date = _end_updated + _end_lastbuilddate = _end_updated + + def _start_created(self, attrsD): self.push('created', 1) - _start_created = _start_dcterms_created + _start_dcterms_created = _start_created - def _end_dcterms_created(self): + def _end_created(self): value = self.pop('created') - self._save('created_parsed', _parse_date(value)) - _end_created = _end_dcterms_created - - def _start_dcterms_modified(self, attrsD): - self.push('modified', 1) - _start_modified = _start_dcterms_modified - _start_dc_date = _start_dcterms_modified - _start_pubdate = _start_dcterms_modified - - def _end_dcterms_modified(self): - value = self.pop('modified') - parsed_value = _parse_date(value) - self._save('date', value) - self._save('date_parsed', parsed_value) - self._save('modified_parsed', parsed_value) - _end_modified = _end_dcterms_modified - _end_dc_date = _end_dcterms_modified - _end_pubdate = _end_dcterms_modified + self._save('created_parsed', _parse_date(value), overwrite=True) + _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'))) + self._save('expired_parsed', _parse_date(self.pop('expired')), overwrite=True) def _start_cc_license(self, attrsD): - self.push('license', 1) + context = self._getContext() value = self._getAttribute(attrsD, 'rdf:resource') - if value: - self.elementstack[-1][2].append(value) - self.pop('license') + 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): - self.pop('license') + 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) - domain = self._getAttribute(attrsD, 'domain') - cats = [] - if self.initem: - cats = self.items[-1].setdefault('categories', []) - elif self.inchannel: - cats = self.channel.setdefault('categories', []) - cats.append((domain, None)) _start_dc_subject = _start_category + _start_keywords = _start_category + + def _start_media_category(self, attrsD): + attrsD.setdefault('scheme', 'http://search.yahoo.com/mrss/category_schema') + self._start_category(attrsD) + + 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): - self.pop('category') + 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 + _end_media_category = _end_category + def _start_cloud(self, attrsD): - self.channel['cloud'] = attrsD + self._getContext()['cloud'] = FeedParserDict(attrsD) def _start_link(self, attrsD): attrsD.setdefault('rel', 'alternate') - attrsD.setdefault('type', 'text/html') + 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']) - expectingText = self.inchannel or self.initem - if self.initem: - self.items[-1].setdefault('links', []) - self.items[-1]['links'].append(attrsD) - elif self.inchannel: - self.channel.setdefault('links', []) - self.channel['links'].append(attrsD) + expectingText = self.infeed or self.inentry or self.insource + context.setdefault('links', []) + if not (self.inentry and self.inimage): + context['links'].append(FeedParserDict(attrsD)) if attrsD.has_key('href'): expectingText = 0 - if attrsD.get('type', '') in self.html_types: - if self.initem: - self.items[-1]['link'] = attrsD['href'] - elif self.inchannel: - self.channel['link'] = attrsD['href'] + 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('guid', 1) + self.push('id', 1) def _end_guid(self): - value = self.pop('guid') - self._save('id', value) + 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", + # 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_id(self, attrsD): - self.push('id', 1) - - def _end_id(self): - value = self.pop('id') - self._save('guid', value) - def _start_title(self, attrsD): - self.incontent += 1 - self.contentparams = {'mode': attrsD.get('mode', 'escaped'), - 'type': attrsD.get('type', 'text/plain'), - 'language': attrsD.get('xml:lang', self.lang), - 'base': attrsD.get('xml:base', self.baseuri)} - self.push('title', self.inchannel or self.initem) + 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): - self.pop('title') - self.incontent -= 1 - self.contentparams.clear() + 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): - self.incontent += 1 - self.contentparams = {'mode': attrsD.get('mode', 'escaped'), - 'type': attrsD.get('type', 'text/html'), - 'language': attrsD.get('xml:lang', self.lang), - 'base': attrsD.get('xml:base', self.baseuri)} - self.push('description', self.inchannel or self.initem) + 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): - value = self.pop('description') - if self.initem: - self.items[-1]['summary'] = value - elif self.inchannel: - self.channel['tagline'] = value - self.incontent -= 1 - self.contentparams.clear() - - def _start_info(self, attrsD): - self.incontent += 1 - self.contentparams = {'mode': attrsD.get('mode', 'escaped'), - 'type': attrsD.get('type', 'text/plain'), - 'language': attrsD.get('xml:lang', self.lang), - 'base': attrsD.get('xml:base', self.baseuri)} - self.push('info', 1) + 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.pop('info') - self.incontent -= 1 - self.contentparams.clear() + self.popContent('info') + _end_feedburner_browserfriendly = _end_info def _start_generator(self, attrsD): if attrsD: - self.channel['generator_detail'] = 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') - if self.channel.has_key('generator_detail'): - self.channel['generator_detail']['name'] = value + context = self._getContext() + if context.has_key('generator_detail'): + context['generator_detail']['name'] = value def _start_admin_generatoragent(self, attrsD): self.push('generator', 1) @@ -911,6 +1575,7 @@ class _FeedParserMixin: 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) @@ -920,88 +1585,159 @@ class _FeedParserMixin: self.pop('errorreportsto') def _start_summary(self, attrsD): - self.incontent += 1 - self.contentparams = {'mode': attrsD.get('mode', 'escaped'), - 'type': attrsD.get('type', 'text/plain'), - 'language': attrsD.get('xml:lang', self.lang), - 'base': attrsD.get('xml:base', self.baseuri)} - self.push('summary', 1) + 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): - value = self.pop('summary') - if self.items: - self.items[-1]['description'] = value - self.incontent -= 1 - self.contentparams.clear() + 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): - if self.initem: - self.items[-1].setdefault('enclosures', []) - self.items[-1]['enclosures'].append(attrsD) + attrsD = self._itsAnHrefDamnIt(attrsD) + context = self._getContext() + attrsD['rel']='enclosure' + context.setdefault('links', []).append(FeedParserDict(attrsD)) def _start_source(self, attrsD): - if self.initem: - self.items[-1]['source'] = 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.pop('source') + 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.incontent += 1 - self.contentparams = {'mode': attrsD.get('mode', 'xml'), - 'type': attrsD.get('type', 'text/plain'), - 'language': attrsD.get('xml:lang', self.lang), - 'base': attrsD.get('xml:base', self.baseuri)} + 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.incontent += 1 - self.contentparams = {'mode': 'xml', - 'type': 'application/xhtml+xml', - 'language': attrsD.get('xml:lang', self.lang), - 'base': attrsD.get('xml:base', self.baseuri)} - self.push('content', 1) + self.pushContent('content', attrsD, 'application/xhtml+xml', 1) _start_xhtml_body = _start_body def _start_content_encoded(self, attrsD): - self.incontent += 1 - self.contentparams = {'mode': 'escaped', - 'type': 'text/html', - 'language': attrsD.get('xml:lang', self.lang), - 'base': attrsD.get('xml:base', self.baseuri)} - self.push('content', 1) + self.pushContent('content', attrsD, 'text/html', 1) _start_fullitem = _start_content_encoded def _end_content(self): - value = self.pop('content') - if self.contentparams.get('type') in (['text/plain'] + self.html_types): - self._save('description', value) - self.incontent -= 1 - self.contentparams.clear() + copyToSummary = self.mapContentType(self.contentparams.get('type')) in (['text/plain'] + self.html_types) + value = self.popContent('content') + if copyToSummary: + self._save('summary', 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) + if attrsD.get('href'): + 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) + # Convert 'yes' -> True, 'clean' to False, and any other value to None + # False and None both evaluate as False, so the difference can be ignored + # by applications that only need to know if the content is explicit. + self._getContext()['itunes_explicit'] = (None, False, True)[(value == 'yes' and 2) or value == 'clean' 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 + + def _start_newlocation(self, attrsD): + self.push('newlocation', 1) + + def _end_newlocation(self): + url = self.pop('newlocation') + context = self._getContext() + # don't set newlocation if the context isn't right + if context is not self.feeddata: + return + context['newlocation'] = _makeSafeAbsoluteURI(self.baseuri, url.strip()) if _XML_AVAILABLE: - class _StrictFeedParser(_FeedParserMixin, xml.sax.handler.ContentHandler):#, xml.sax.handler.DTDHandler): - def __init__(self, baseuri): + 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) + _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 - namespace = str(namespace) - prefix = self.namespaces.get(namespace, '') - if prefix: - localname = prefix + ':' + localname + 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 @@ -1011,111 +1747,686 @@ if _XML_AVAILABLE: # 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 = {} + 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(): - prefix = self.namespaces.get(namespace, '') + lowernamespace = (namespace or '').lower() + prefix = self._matchnamespaces.get(lowernamespace, '') if prefix: - attrlocalname = prefix + ":" + attrlocalname + 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 resolveEntity(self, publicId, systemId): - return _StringIO() - def characters(self, text): self.handle_data(text) def endElementNS(self, name, qname): namespace, localname = name - namespace = str(namespace) - prefix = self.namespaces.get(namespace, '') + 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 fatalError(self, exc): + def error(self, exc): self.bozo = 1 self.exc = exc - error = fatalError -class _LooseFeedParser(_FeedParserMixin, sgmllib.SGMLParser): - def __init__(self, baseuri): - sgmllib.SGMLParser.__init__(self) - _FeedParserMixin.__init__(self, baseuri) + def fatalError(self, exc): + self.error(exc) + raise exc class _BaseHTMLProcessor(sgmllib.SGMLParser): - elements_no_end_tag = ['area', 'base', 'basefont', 'br', 'col', 'frame', 'hr', - 'img', 'input', 'isindex', 'link', 'meta', 'param'] - - def __init__(self): + 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): - # extend (called by sgmllib.SGMLParser.__init__) 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('"', '"') + try: + bytes + if bytes is str: + raise NameError + self.encoding = self.encoding + '_INVALID_PYTHON_3' + except NameError: + 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 = [(k.lower(), sgmllib.charref.sub(lambda m: unichr(int(m.groups()[0])), v).strip()) for k, v in attrs] + 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")]
-        strattrs = "".join([' %s="%s"' % (key, value) for key, value in attrs])
+        # 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')
+                try:
+                    # Currently, in Python 3 the key is already a str, and cannot be decoded again
+                    uattrs.append((unicode(key, self.encoding), value))
+                except TypeError:
+                    uattrs.append((key, 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())
+            self.pieces.append('<%(tag)s%(strattrs)s />' % locals())
         else:
-            self.pieces.append("<%(tag)s%(strattrs)s>" % locals())
-        
+            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"
+        # 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" + # called for each character reference, e.g. for ' ', ref will be '160' # Reconstruct the original character reference. - self.pieces.append("&#%(ref)s;" % locals()) + 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" + # called for each entity reference, e.g. for '©', ref will be 'copy' # Reconstruct the original entity reference. - self.pieces.append("&%(ref)s;" % locals()) + 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()) + self.pieces.append('' % locals()) def handle_pi(self, text): # called for each processing instruction, e.g. # Reconstruct original processing instruction. - self.pieces.append("" % locals()) + self.pieces.append('' % locals()) def handle_decl(self, text): # called for the DOCTYPE, if present, e.g. # # Reconstruct original DOCTYPE - self.pieces.append("" % locals()) + 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(self.pieces) + '''Return processed HTML as a single string''' + return ''.join([str(p) for p in self.pieces]) + + def parse_declaration(self, i): + try: + return sgmllib.SGMLParser.parse_declaration(self, i) + except sgmllib.SGMLParseError: + # escape the doctype declaration and continue parsing + self.handle_data('<') + return i+1 + +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).decode(self.encoding) + if sValue: + arLines.append(self.vcardFold(sProperty.upper() + ':' + sValue)) + return sValue or u'' + + 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)) + # Completely remove the agent element from the parse tree + elmAgent.extract() + 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 = [u'BEGIN:vCard',u'VERSION:3.0'] + arLines + [u'END:vCard'] + sVCards += u'\n'.join(arLines) + u'\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') + try: + p = _MicroformatsParser(htmlSource, baseURI, encoding) + except UnicodeEncodeError: + # sgmllib throws this exception when performing lookups of tags + # with non-ASCII characters in them. + return + 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'), @@ -1144,64 +2455,246 @@ class _RelativeURIResolver(_BaseHTMLProcessor): ('q', 'cite'), ('script', 'src')] - def __init__(self, baseuri): - _BaseHTMLProcessor.__init__(self) + def __init__(self, baseuri, encoding, _type): + _BaseHTMLProcessor.__init__(self, encoding, _type) self.baseuri = baseuri def resolveURI(self, uri): - return urlparse.urljoin(self.baseuri, uri) + return _makeSafeAbsoluteURI(_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): - p = _RelativeURIResolver(baseURI) + +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() +def _makeSafeAbsoluteURI(base, rel=None): + # bail if ACCEPTABLE_URI_SCHEMES is empty + if not ACCEPTABLE_URI_SCHEMES: + return _urljoin(base, rel or u'') + if not base: + return rel or u'' + if not rel: + scheme = urlparse.urlparse(base)[0] + if not scheme or scheme in ACCEPTABLE_URI_SCHEMES: + return base + return u'' + uri = _urljoin(base, rel) + if uri.strip().split(':', 1)[0] not in ACCEPTABLE_URI_SCHEMES: + return u'' + return uri + class _HTMLSanitizer(_BaseHTMLProcessor): - acceptable_elements = ['a', 'abbr', 'acronym', 'address', 'area', 'b', 'big', - 'blockquote', 'br', 'button', 'caption', 'center', 'cite', 'code', 'col', - 'colgroup', 'dd', 'del', 'dfn', 'dir', 'div', 'dl', 'dt', 'em', 'fieldset', - 'font', 'form', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'hr', 'i', 'img', 'input', - 'ins', 'kbd', 'label', 'legend', 'li', 'map', 'menu', 'ol', 'optgroup', - 'option', 'p', 'pre', 'q', 's', 'samp', 'select', 'small', 'span', 'strike', - 'strong', 'sub', 'sup', 'table', 'tbody', 'td', 'textarea', 'tfoot', 'th', - 'thead', 'tr', 'tt', 'u', 'ul', 'var'] + 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', + 'figcaption', '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', 'axis', 'border', 'cellpadding', 'cellspacing', - 'char', 'charoff', 'charset', 'checked', 'cite', 'class', 'clear', 'cols', - 'colspan', 'color', 'compact', 'coords', 'datetime', 'dir', 'disabled', - 'enctype', 'for', 'frame', 'headers', 'height', 'href', 'hreflang', 'hspace', - 'id', 'ismap', 'label', 'lang', 'longdesc', 'maxlength', 'media', 'method', - 'multiple', 'name', 'nohref', 'noshade', 'nowrap', 'prompt', 'readonly', - 'rel', 'rev', 'rows', 'rowspan', 'rules', 'scope', 'selected', 'shape', 'size', - 'span', 'src', 'start', 'summary', 'tabindex', 'target', 'title', 'type', - 'usemap', 'valign', 'value', 'vspace', 'width'] - - unacceptable_elements_with_end_tag = ['script', 'applet'] + '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): - if not tag in self.acceptable_elements: + 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 - return - attrs = self.normalize_attrs(attrs) - attrs = [(key, value) for key, value in attrs if key in self.acceptable_attributes] - _BaseHTMLProcessor.unknown_starttag(self, tag, attrs) + + # add implicit namespaces to html5 inline svg/mathml + if self._type.endswith('html'): + if not dict(attrs).get('xmlns'): + if tag=='svg': + attrs.append( ('xmlns','http://www.w3.org/2000/svg') ) + if tag=='math': + attrs.append( ('xmlns','http://www.w3.org/1998/Math/MathML') ) + + # 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) + # make sure the uri uses an acceptable uri scheme + if key == u'href': + value = _makeSafeAbsoluteURI(value) + 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 - return + 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): @@ -1214,44 +2707,140 @@ class _HTMLSanitizer(_BaseHTMLProcessor): if not self.unacceptablestack: _BaseHTMLProcessor.handle_data(self, text) -def _sanitizeHTML(htmlSource): - p = _HTMLSanitizer() + 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 parse_comment(self, i, report=1): + ret = _BaseHTMLProcessor.parse_comment(self, i, report) + if ret >= 0: + return ret + # if ret == -1, this may be a malicious attempt to circumvent + # sanitization, or a page-destroying unclosed comment + match = re.compile(r'--[^>]*>').search(self.rawdata, i+4) + if match: + return match.end() + # unclosed comment; deliberately fail to handle_data() + return len(self.rawdata) + + +def _sanitizeHTML(htmlSource, encoding, _type): + p = _HTMLSanitizer(encoding, _type) + htmlSource = htmlSource.replace(''): - data = data.split('>', 1)[1] - if data.count(''): + data = data.split('>', 1)[1] + if data.count('= '2.3.3' + assert base64 != None + user, passw = _base64decode(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, request_headers): """URL, filename, or string --> stream This function lets you define parsers that take any input source @@ -1263,76 +2852,401 @@ def _open_resource(url_file_stream_or_string, etag=None, modified=None, agent=No 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 must be a tuple of 9 integers - as returned by gmtime() in the standard Python time module. This MUST - be in GMT (Greenwich Mean Time). The formatted date/time will be used - as the value of an If-Modified-Since 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 request_headers is supplied it is a dictionary of HTTP request headers + that will override the values generated by FeedParser. """ - if hasattr(url_file_stream_or_string, "read"): + if hasattr(url_file_stream_or_string, 'read'): return url_file_stream_or_string - if url_file_stream_or_string == "-": + if url_file_stream_or_string == '-': return sys.stdin - if not agent: - agent = USER_AGENT - - # try to open with urllib2 (to use optional headers) - request = urllib2.Request(url_file_stream_or_string) - if etag: - request.add_header("If-None-Match", etag) - 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])) - request.add_header("User-Agent", agent) - if referrer: - request.add_header("Referer", referrer) - if gzip: - request.add_header("Accept-encoding", "gzip") - opener = urllib2.build_opener(_FeedURLHandler()) - opener.addheaders = [] # RMK - must clear so we only send our custom User-Agent - try: + if urlparse.urlparse(url_file_stream_or_string)[0] in ('http', 'https', 'ftp', 'file', 'feed'): + # Deal with the feed URI scheme + if url_file_stream_or_string.startswith('feed:http'): + url_file_stream_or_string = url_file_stream_or_string[5:] + elif url_file_stream_or_string.startswith('feed:'): + url_file_stream_or_string = 'http:' + url_file_stream_or_string[5:] + 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.standard_b64encode(user_passwd).strip() + + # iri support try: - return opener.open(request) + if isinstance(url_file_stream_or_string,unicode): + url_file_stream_or_string = url_file_stream_or_string.encode('idna').decode('utf-8') + else: + url_file_stream_or_string = url_file_stream_or_string.decode('utf-8').encode('idna').decode('utf-8') except: - # url_file_stream_or_string is not a valid URL, but it might be a valid filename pass - finally: - opener.close() # JohnD + + # try to open with urllib2 (to use optional headers) + request = _build_urllib2_request(url_file_stream_or_string, agent, etag, modified, referrer, auth, request_headers) + opener = apply(urllib2.build_opener, tuple(handlers + [_FeedURLHandler()])) + 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) + return open(url_file_stream_or_string, 'rb') except: pass # treat url_file_stream_or_string as string return _StringIO(str(url_file_stream_or_string)) +def _build_urllib2_request(url, agent, etag, modified, referrer, auth, request_headers): + request = urllib2.Request(url) + request.add_header('User-Agent', agent) + if etag: + request.add_header('If-None-Match', etag) + if type(modified) == type(''): + modified = _parse_date(modified) + elif isinstance(modified, datetime.datetime): + modified = modified.utctimetuple() + 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) + # use this for whatever -- cookies, special headers, etc + # [('Cookie','Something'),('x-special-header','Another Value')] + for header_name, header_value in request_headers.items(): + request.add_header(header_name, header_value) + request.add_header('A-IM', 'feed') # RFC 3229 support + return request + +_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] +try: + del tmpl +except NameError: + pass +_iso8601_matches = [re.compile(regex).match for regex in _iso8601_re] +try: + del regex +except NameError: + pass +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(tuple(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 _w3dtf_parse(s): +def _parse_date_w3dtf(dateString): def __extract_date(m): - year = int(m.group("year")) + 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") + julian = m.group('julian') if julian: julian = int(julian) month = julian / 30 + 1 @@ -1354,13 +3268,13 @@ def _w3dtf_parse(s): else: month = month + 1 return year, month, day - month = m.group("month") + month = m.group('month') day = 1 if month is None: month = 1 else: month = int(month) - day = m.group("day") + day = m.group('day') if day: day = int(day) else: @@ -1370,12 +3284,12 @@ def _w3dtf_parse(s): def __extract_time(m): if not m: return 0, 0, 0 - hours = m.group("hours") + hours = m.group('hours') if not hours: return 0, 0, 0 hours = int(hours) - minutes = int(m.group("minutes")) - seconds = m.group("seconds") + minutes = int(m.group('minutes')) + seconds = m.group('seconds') if seconds: seconds = int(seconds) else: @@ -1383,231 +3297,511 @@ def _w3dtf_parse(s): return hours, minutes, seconds def __extract_tzd(m): - """Return the Time Zone Designator as an offset in seconds from UTC.""" + '''Return the Time Zone Designator as an offset in seconds from UTC.''' if not m: return 0 - tzd = m.group("tzd") + tzd = m.group('tzd') if not tzd: return 0 - if tzd == "Z": + if tzd == 'Z': return 0 - hours = int(m.group("tzdhours")) - minutes = m.group("tzdminutes") + 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] == "+": + 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)" + __date_re = ('(?P\d\d\d\d)' + '(?:(?P-|)' + '(?:(?P\d\d)(?:(?P=dsep)(?P\d\d))?' + '|(?P\d\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+)?))?" + __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_re = '%s(?:T%s)?' % (__date_re, __time_re) __datetime_rx = re.compile(__datetime_re) - m = __datetime_rx.match(s) - if m is None or m.group() != s: - return None + 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.mktime(gmt) + __extract_tzd(m) - time.timezone - -# Additional 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-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[+-](?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 - + 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) + # Account for the Etc/GMT timezone by stripping 'Etc/' + elif len(data) == 5 and data[4].lower().startswith('etc/'): + data[4] = data[4][4:] + 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. +# '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 -def _parse_date(date): - """Parses a variety of date formats into a tuple of 9 integers""" - date = str(date) + 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', 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: - # try the standard rfc822 library, which handles - # RFC822, RFC1123, RFC2822, and asctime - tm = rfc822.parsedate_tz(date) - if tm: - return time.gmtime(rfc822.mktime_tz(tm)) - # not a RFC2822 date, try W3DTF profile of ISO-8601 - try: - tm = _w3dtf_parse(date) - except ValueError: - tm = None - if tm: - return time.gmtime(tm) - # try various non-W3DTF ISO-8601-compatible formats like 20040105 - m = None - for _iso8601_match in _iso8601_matches: - m = _iso8601_match(date) - if m: break - if not m: return - # catch truly malformed strings - if m.span() == (0, 0): return - params = m.groupdict() - ordinal = params.get("ordinal", 0) - if ordinal: - ordinal = int(ordinal) + if xml_data[:4] == _l2bytes([0x4c, 0x6f, 0xa7, 0x94]): + # EBCDIC + xml_data = _ebcdic_to_ascii(xml_data) + elif xml_data[:4] == _l2bytes([0x00, 0x3c, 0x00, 0x3f]): + # 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] == _l2bytes([0xfe, 0xff])) and (xml_data[2:4] != _l2bytes([0x00, 0x00])): + # UTF-16BE with BOM + sniffed_xml_encoding = 'utf-16be' + xml_data = unicode(xml_data[2:], 'utf-16be').encode('utf-8') + elif xml_data[:4] == _l2bytes([0x3c, 0x00, 0x3f, 0x00]): + # 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] == _l2bytes([0xff, 0xfe])) and (xml_data[2:4] != _l2bytes([0x00, 0x00])): + # UTF-16LE with BOM + sniffed_xml_encoding = 'utf-16le' + xml_data = unicode(xml_data[2:], 'utf-16le').encode('utf-8') + elif xml_data[:4] == _l2bytes([0x00, 0x00, 0x00, 0x3c]): + # UTF-32BE + sniffed_xml_encoding = 'utf-32be' + xml_data = unicode(xml_data, 'utf-32be').encode('utf-8') + elif xml_data[:4] == _l2bytes([0x3c, 0x00, 0x00, 0x00]): + # UTF-32LE + sniffed_xml_encoding = 'utf-32le' + xml_data = unicode(xml_data, 'utf-32le').encode('utf-8') + elif xml_data[:4] == _l2bytes([0x00, 0x00, 0xfe, 0xff]): + # UTF-32BE with BOM + sniffed_xml_encoding = 'utf-32be' + xml_data = unicode(xml_data[4:], 'utf-32be').encode('utf-8') + elif xml_data[:4] == _l2bytes([0xff, 0xfe, 0x00, 0x00]): + # UTF-32LE with BOM + sniffed_xml_encoding = 'utf-32le' + xml_data = unicode(xml_data[4:], 'utf-32le').encode('utf-8') + elif xml_data[:3] == _l2bytes([0xef, 0xbb, 0xbf]): + # UTF-8 with BOM + sniffed_xml_encoding = 'utf-8' + xml_data = unicode(xml_data[3:], 'utf-8').encode('utf-8') 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(params.get("second", 0)) - # weekday is normalized by mktime(), we can ignore it - weekday = 0 - # daylight savings is complex, but not needed for feedparser's purposes - # as time zones, if specified, include mention of whether it is active - # (e.g. PST vs. PDT, CET). Using -1 is implementation-dependent and - # and most implementations have DST bugs - daylight_savings_flag = 0 - 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)) + # ASCII-compatible + pass + xml_encoding_match = re.compile(_s2bytes('^<\?.*encoding=[\'"](.*?)[\'"].*\?>')).match(xml_data) except: - return None - -def parse(url_file_stream_or_string, etag=None, modified=None, agent=None, referrer=None): - """Parse a feed from a URL, file, stream, or string""" - result = {} - f = _open_resource(url_file_stream_or_string, etag=etag, modified=modified, agent=agent, referrer=referrer) - data = f.read() - if hasattr(f, "headers"): - if gzip and f.headers.get('content-encoding', '') == 'gzip': + xml_encoding_match = None + if xml_encoding_match: + xml_encoding = xml_encoding_match.groups()[0].decode('utf-8').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') or 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] == _l2bytes([0xfe, 0xff])) and (data[2:4] != _l2bytes([0x00, 0x00])): + 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] == _l2bytes([0xff, 0xfe])) and (data[2:4] != _l2bytes([0x00, 0x00])): + 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] == _l2bytes([0xef, 0xbb, 0xbf]): + 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] == _l2bytes([0x00, 0x00, 0xfe, 0xff]): + 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] == _l2bytes([0xff, 0xfe, 0x00, 0x00]): + 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(_s2bytes('<\w'), data) + start = start and start.start() or -1 + head,data = data[:start+1], data[start+1:] + + entity_pattern = re.compile(_s2bytes(r'^\s*]*?)>'), re.MULTILINE) + entity_results=entity_pattern.findall(head) + head = entity_pattern.sub(_s2bytes(''), head) + doctype_pattern = re.compile(_s2bytes(r'^\s*]*?)>'), re.MULTILINE) + doctype_results = doctype_pattern.findall(head) + doctype = doctype_results and doctype_results[0] or _s2bytes('') + if doctype.lower().count(_s2bytes('netscape')): + version = 'rss091n' + else: + version = None + + # only allow in 'safe' inline entity definitions + replacement=_s2bytes('') + if len(doctype_results)==1 and entity_results: + safe_pattern=re.compile(_s2bytes('\s+(\w+)\s+"(&#\w+;|[^&"]*)"')) + safe_entities=filter(lambda e: safe_pattern.match(e),entity_results) + if safe_entities: + replacement=_s2bytes('\n \n]>') + data = doctype_pattern.sub(replacement, head) + data + + return version, data, dict(replacement and [(k.decode('utf-8'), v.decode('utf-8')) for k, v in safe_pattern.findall(replacement)]) + +def parse(url_file_stream_or_string, etag=None, modified=None, agent=None, referrer=None, handlers=[], request_headers={}, response_headers={}): + '''Parse a feed from a URL, file, stream, or string. + + request_headers, if given, is a dict from http header name to value to add + to the request; this overrides internally generated values. + ''' + result = FeedParserDict() + result['feed'] = FeedParserDict() + result['entries'] = [] + if _XML_AVAILABLE: + result['bozo'] = 0 + if not isinstance(handlers, list): + handlers = [handlers] + try: + f = _open_resource(url_file_stream_or_string, etag, modified, agent, referrer, handlers, request_headers) + data = f.read() + except Exception, e: + result['bozo'] = 1 + result['bozo_exception'] = e + data = None + f = None + + if hasattr(f, 'headers'): + result['headers'] = dict(f.headers) + # overwrite existing headers using response_headers + if 'headers' in result: + result['headers'].update(response_headers) + elif response_headers: + result['headers'] = copy.deepcopy(response_headers) + + # if feed is gzip-compressed, decompress it + if f and data and 'headers' in result: + if gzip and result['headers'].get('content-encoding') == 'gzip': try: data = gzip.GzipFile(fileobj=_StringIO(data)).read() - except: - # some feeds claim to be gzipped but they're not, so we get garbage + except Exception, e: + # Some feeds claim to be gzipped but they're not, so + # we get garbage. Ideally, we should re-request the + # feed without the 'Accept-encoding: gzip' header, + # but we don't. + result['bozo'] = 1 + result['bozo_exception'] = e data = '' - if hasattr(f, "info"): - info = f.info() - result["etag"] = info.getheader("ETag") - last_modified = info.getheader("Last-Modified") - if last_modified: - result["modified"] = _parse_date(last_modified) - if hasattr(f, "url"): - result["url"] = f.url - result["status"] = 200 # default, may be overridden later - if hasattr(f, "status"): - result["status"] = f.status - if hasattr(f, "headers"): - result["headers"] = f.headers.dict - # get the xml encoding - xmlheaderRe = re.compile('<\?.*encoding=[\'"](.*?)[\'"].*\?>') # Andrei's version - match = xmlheaderRe.match(data) - if match: - result["encoding"] = match.groups()[0].lower() - f.close() - result['channel'] = {} - result['items'] = {} - baseuri = result.get('headers', {}).get('content-location', result.get('url')) - # try true XML parser first - if _XML_AVAILABLE: - if _debug: sys.stderr.write('using xml library\n') - result['bozo'] = 0 - feedparser = _StrictFeedParser(baseuri) - if re.search(r'', data): - feedparser.version = 'rss091n' - source = xml.sax.xmlreader.InputSource() - source.setByteStream(_StringIO(data)) - saxparser = xml.sax.make_parser()#["drv_libxml2"]) + elif zlib and result['headers'].get('content-encoding') == 'deflate': + try: + data = zlib.decompress(data, -zlib.MAX_WBITS) + except Exception, e: + result['bozo'] = 1 + result['bozo_exception'] = e + data = '' + + # save HTTP headers + if 'headers' in result: + if 'etag' in result['headers'] or 'ETag' in result['headers']: + etag = result['headers'].get('etag', result['headers'].get('ETag')) + if etag: + result['etag'] = etag + if 'last-modified' in result['headers'] or 'Last-Modified' in result['headers']: + modified = result['headers'].get('last-modified', result['headers'].get('Last-Modified')) + if modified: + result['modified'] = _parse_date(modified) + if hasattr(f, 'url'): + result['href'] = f.url + result['status'] = 200 + if hasattr(f, 'status'): + result['status'] = f.status + if hasattr(f, 'close'): + f.close() + + # there are four encodings to keep track of: + # - http_encoding is the encoding declared in the Content-Type HTTP header + # - xml_encoding is the encoding declared in the becomes - data = re.sub(r'<(\S+)/>', r'<\1>', data) - feedparser = _LooseFeedParser(baseuri) - feedparser.feed(data) - else: - if _debug: sys.stderr.write('no xml libraries available, using regexes\n') - data = re.sub(r'<(\S+)/>', r'<\1>', data) - feedparser = _LooseFeedParser(baseuri) - feedparser.feed(data) - result['channel'] = feedparser.channel - result['items'] = feedparser.items - result['version'] = feedparser.version + result['bozo_exception'] = feedparser.exc or e + use_strict_parser = 0 + if not use_strict_parser: + feedparser = _LooseFeedParser(baseuri, baselang, 'utf-8', entities) + feedparser.feed(data.decode('utf-8', 'replace')) + result['feed'] = feedparser.feeddata + result['entries'] = feedparser.entries + result['version'] = result['version'] or feedparser.version + result['namespaces'] = feedparser.namespacesInUse return result -_TEST_SUITE = ('http://www.pocketsoap.com/rssTests/rss1.0withModules.xml', - 'http://www.pocketsoap.com/rssTests/rss1.0withModulesNoDefNS.xml', - 'http://www.pocketsoap.com/rssTests/rss1.0withModulesNoDefNSLocalNameClash.xml', - 'http://www.pocketsoap.com/rssTests/rss2.0noNSwithModules.xml', - 'http://www.pocketsoap.com/rssTests/rss2.0noNSwithModulesLocalNameClash.xml', - 'http://www.pocketsoap.com/rssTests/rss2.0NSwithModules.xml', - 'http://www.pocketsoap.com/rssTests/rss2.0NSwithModulesNoDefNS.xml', - 'http://www.pocketsoap.com/rssTests/rss2.0NSwithModulesNoDefNSLocalNameClash.xml') - +class Serializer: + def __init__(self, results): + self.results = results + +class TextSerializer(Serializer): + def write(self, stream=sys.stdout): + self._writer(stream, self.results, '') + + def _writer(self, stream, node, prefix): + if not node: return + if hasattr(node, 'keys'): + keys = node.keys() + keys.sort() + for k in keys: + if k in ('description', 'link'): continue + if node.has_key(k + '_detail'): continue + if node.has_key(k + '_parsed'): continue + self._writer(stream, node[k], prefix + k + '.') + elif type(node) == types.ListType: + index = 0 + for n in node: + self._writer(stream, n, prefix[:-1] + '[' + str(index) + '].') + index += 1 + else: + try: + s = str(node).encode('utf-8') + s = s.replace('\\', '\\\\') + s = s.replace('\r', '') + s = s.replace('\n', r'\n') + stream.write(prefix[:-1]) + stream.write('=') + stream.write(s) + stream.write('\n') + except: + pass + +class PprintSerializer(Serializer): + def write(self, stream=sys.stdout): + if self.results.has_key('href'): + stream.write(self.results['href'] + '\n\n') + from pprint import pprint + pprint(self.results, stream) + stream.write('\n') + if __name__ == '__main__': - if sys.argv[1:]: - urls = sys.argv[1:] + try: + from optparse import OptionParser + except: + OptionParser = None + + if OptionParser: + optionParser = OptionParser(version=__version__, usage="%prog [options] url_or_filename_or_-") + optionParser.set_defaults(format="pprint") + optionParser.add_option("-A", "--user-agent", dest="agent", metavar="AGENT", help="User-Agent for HTTP URLs") + optionParser.add_option("-e", "--referer", "--referrer", dest="referrer", metavar="URL", help="Referrer for HTTP URLs") + optionParser.add_option("-t", "--etag", dest="etag", metavar="TAG", help="ETag/If-None-Match for HTTP URLs") + optionParser.add_option("-m", "--last-modified", dest="modified", metavar="DATE", help="Last-modified/If-Modified-Since for HTTP URLs (any supported date format)") + optionParser.add_option("-f", "--format", dest="format", metavar="FORMAT", help="output results in FORMAT (text, pprint)") + optionParser.add_option("-v", "--verbose", action="store_true", dest="verbose", default=False, help="write debugging information to stderr") + (options, urls) = optionParser.parse_args() + if options.verbose: + _debug = 1 + if not urls: + optionParser.print_help() + sys.exit(0) else: - urls = _TEST_SUITE - from pprint import pprint + if not sys.argv[1:]: + print __doc__ + sys.exit(0) + class _Options: + etag = modified = agent = referrer = None + format = 'pprint' + options = _Options() + urls = sys.argv[1:] + + zopeCompatibilityHack() + + serializer = globals().get(options.format.capitalize() + 'Serializer', Serializer) for url in urls: - print url - print - result = parse(url) - pprint(result) - print - -#TODO -#- image -#- textinput/textInput -#- comments -# -#encoding notes: -#- RFC 3023 -#- content-type.startswith('text/') and content-type.endswith('xml') --> look for charset="(.*?)" in HTTP content-type header, else "us-ascii" -#- content-type.startswith('application/') and content-type.endswith('xml') --> look for charset="(.*?)" in HTTP content-type header, else look for encoding="(.*?)" in document, else "utf-8" -#- parsing encoding: http://www.w3.org/TR/REC-xml#NT-EncodingDecl -# -#REVISION HISTORY -#1.0 - 9/27/2002 - MAP - fixed namespace processing on prefixed RSS 2.0 elements, -# added Simon Fell's test suite -#1.1 - 9/29/2002 - MAP - fixed infinite loop on incomplete CDATA sections -#2.0 - 10/19/2002 -# JD - use inchannel to watch out for image and textinput elements which can -# also contain title, link, and description elements -# JD - check for isPermaLink="false" attribute on guid elements -# JD - replaced openAnything with open_resource supporting ETag and -# If-Modified-Since request headers -# JD - parse now accepts etag, modified, agent, and referrer optional -# arguments -# JD - modified parse to return a dictionary instead of a tuple so that any -# etag or modified information can be returned and cached by the caller -#2.0.1 - 10/21/2002 - MAP - changed parse() so that if we don't get anything -# because of etag/modified, return the old etag/modified to the caller to -# indicate why nothing is being returned -#2.0.2 - 10/21/2002 - JB - added the inchannel to the if statement, otherwise its -# useless. Fixes the problem JD was addressing by adding it. -#2.1 - 11/14/2002 - MAP - added gzip support -#2.2 - 1/27/2003 - MAP - added attribute support, admin:generatorAgent. -# start_admingeneratoragent is an example of how to handle elements with -# only attributes, no content. -#2.3 - 6/11/2003 - MAP - added USER_AGENT for default (if caller doesn't specify); -# also, make sure we send the User-Agent even if urllib2 isn't available. -# Match any variation of backend.userland.com/rss namespace. -#2.3.1 - 6/12/2003 - MAP - if item has both link and guid, return both as-is. -#2.4 - 7/9/2003 - MAP - added preliminary Pie/Atom/Echo support based on Sam Ruby's -# snapshot of July 1 ; changed -# project name -#2.5 - 7/25/2003 - MAP - changed to Python license (all contributors agree); -# removed unnecessary urllib code -- urllib2 should always be available anyway; -# return actual url, status, and full HTTP headers (as result['url'], -# result['status'], and result['headers']) if parsing a remote feed over HTTP -- -# this should pass all the HTTP tests at ; -# added the latest namespace-of-the-week for RSS 2.0 -#2.5.1 - 7/26/2003 - RMK - clear opener.addheaders so we only send our custom -# User-Agent (otherwise urllib2 sends two, which confuses some servers) -#2.5.2 - 7/28/2003 - MAP - entity-decode inline xml properly; added support for -# inline and as used in some RSS 2.0 feeds -#2.5.3 - 8/6/2003 - TvdV - patch to track whether we're inside an image or -# textInput, and also to return the character encoding (if specified) -#2.6 - 1/1/2004 - MAP - dc:author support (MarekK); fixed bug tracking -# nested divs within content (JohnD); fixed missing sys import (JohanS); -# fixed regular expression to capture XML character encoding (Andrei); -# added support for Atom 0.3-style links; fixed bug with textInput tracking; -# added support for cloud (MartijnP); added support for multiple -# category/dc:subject (MartijnP); normalize content model: "description" gets -# description (which can come from description, summary, or full content if no -# description), "content" gets dict of base/language/type/value (which can come -# from content:encoded, xhtml:body, content, or fullitem); -# fixed bug matching arbitrary Userland namespaces; added xml:base and xml:lang -# tracking; fixed bug tracking unknown tags; fixed bug tracking content when -# element is not in default namespace (like Pocketsoap feed); -# resolve relative URLs in link, guid, docs, url, comments, wfw:comment, -# wfw:commentRSS; resolve relative URLs within embedded HTML markup in -# description, xhtml:body, content, content:encoded, title, subtitle, -# summary, info, tagline, and copyright; added support for pingback and -# trackback namespaces -#2.7 - 1/5/2004 - MAP - really added support for trackback and pingback -# namespaces, as opposed to 2.6 when I said I did but didn't really; -# sanitize HTML markup within some elements; added mxTidy support (if -# installed) to tidy HTML markup within some elements; fixed indentation -# bug in _parse_date (FazalM); use socket.setdefaulttimeout if available -# (FazalM); universal date parsing and normalization (FazalM): 'created', modified', -# 'issued' are parsed into 9-tuple date format and stored in 'created_parsed', -# 'modified_parsed', and 'issued_parsed'; 'date' is duplicated in 'modified' -# and vice-versa; 'date_parsed' is duplicated in 'modified_parsed' and vice-versa -#2.7.1 - 1/9/2004 - MAP - fixed bug handling " and '. fixed memory -# leak not closing url opener (JohnD); added dc:publisher support (MarekK); -# added admin:errorReportsTo support (MarekK); Python 2.1 dict support (MarekK) -#2.7.4 - 1/14/2004 - MAP - added workaround for improperly formed
tags in -# encoded HTML (skadz); fixed unicode handling in normalize_attrs (ChrisL); -# fixed relative URI processing for guid (skadz); added ICBM support; added -# base64 support -#2.7.5 - 1/15/2004 - MAP - added workaround for malformed DOCTYPE (seen on many -# blogspot.com sites); added _debug variable -#2.7.6 - 1/16/2004 - MAP - fixed bug with StringIO importing -#3.0 - MAP - parse entire feed with real XML parser (if available); added several -# new supported namespaces; fixed bug tracking naked markup in description; -# added support for enclosure; added support for source; re-added support for -# cloud which got dropped somehow; added support for expirationDate; fixed -# xml:lang inheritance; fixed multiple bugs tracking xml:base URI, one for -# documents that don't define one explicitly and one for documents that define -# an outer and an inner xml:base that goes out of scope before the end of the -# document; fixed bug parsing multiple links at feed level; added feed type and -# version detection, results["version"] will be one of SUPPORTED_VERSIONS.keys() -# or empty string if unrecognized; added support for creativeCommons:license and -# cc:license; added support for full Atom content model in title, tagline, info, -# copyright, summary; fixed bug with gzip encoding (not always telling server -# we support it when we do); support Atom-style author element in author_detail -# (dictionary of "name", "url", "email"); map author to author_detail if author -# contains name + email address; better handling of empty HTML tags (br, hr, img, -# etc.) in embedded markup, in either HTML or XHTML form (
,
,
); -# fixed CDATA handling in non-wellformed feeds under Python 2.1 + results = parse(url, etag=options.etag, modified=options.modified, agent=options.agent, referrer=options.referrer) + serializer(results).write(sys.stdout)