diff --git a/VERSION b/VERSION
index a93bd69e..f34c30cf 100644
--- a/VERSION
+++ b/VERSION
@@ -1 +1 @@
-Version 2.1.1 (2012-10-19 12:33:48) dev
+Version 2.1.1 (2012-10-19 12:59:55) dev
diff --git a/gluon/contrib/feedparser.py b/gluon/contrib/feedparser.py
index a028d794..f8075c90 100755
--- a/gluon/contrib/feedparser.py
+++ b/gluon/contrib/feedparser.py
@@ -1,17 +1,19 @@
-#!/usr/bin/env python
"""Universal feed parser
Handles RSS 0.9x, RSS 1.0, RSS 2.0, CDF, Atom 0.3, and Atom 1.0 feeds
-Visit http://feedparser.org/ for the latest version
-Visit http://feedparser.org/docs/ for the latest documentation
+Visit https://code.google.com/p/feedparser/ for the latest version
+Visit http://packages.python.org/feedparser/ for the latest documentation
Required: Python 2.4 or later
-Recommended: CJKCodecs and iconv_codec
+Recommended: iconv_codec
"""
-__version__ = "5.0.1"
-__license__ = """Copyright (c) 2002-2008, Mark Pilgrim, All rights reserved.
+__version__ = "5.1.2"
+__license__ = """
+Copyright (c) 2010-2012 Kurt McKee
+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:
@@ -47,7 +49,7 @@ __contributors__ = ["Jason Diamond ",
# 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__
+USER_AGENT = "UniversalFeedParser/%s +https://code.google.com/p/feedparser/" % __version__
# HTTP "Accept" header to send to servers when downloading feeds. If you don't
# want to send an Accept header, set this to None.
@@ -75,6 +77,10 @@ RESOLVE_RELATIVE_URIS = 1
# HTML content, set this to 1.
SANITIZE_HTML = 1
+# If you want feedparser to automatically parse microformat content embedded
+# in entry contents, set this to 1
+PARSE_MICROFORMATS = 1
+
# ---------- Python 3 modules (make it work if possible) ----------
try:
import rfc822
@@ -98,25 +104,25 @@ else:
# Python 3.1 deprecates decodestring in favor of decodebytes
_base64decode = getattr(base64, 'decodebytes', base64.decodestring)
-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))
+# _s2bytes: convert a UTF-8 str to bytes if the interpreter is Python 3
+# _l2bytes: convert a list of ints to bytes if the interpreter is Python 3
+try:
+ if bytes is str:
+ # In Python 2.5 and below, bytes doesn't exist (NameError)
+ # In Python 2.6 and above, bytes and str are the same type
+ raise NameError
+except NameError:
+ # Python 2
+ def _s2bytes(s):
+ return s
+ def _l2bytes(l):
+ return ''.join(map(chr, l))
+else:
+ # Python 3
+ def _s2bytes(s):
+ return bytes(s, 'utf8')
+ def _l2bytes(l):
+ return bytes(l)
# If you want feedparser to allow all URL schemes, set this to ()
# List culled from Python's urlparse documentation at:
@@ -125,9 +131,10 @@ def _l2bytes(l):
# 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',
+ 'file', 'ftp', 'gopher', 'h323', 'hdl', 'http', 'https', 'imap', 'magnet',
+ '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',
@@ -136,16 +143,17 @@ ACCEPTABLE_URI_SCHEMES = (
# ---------- required modules (should come with any Python distribution) ----------
import cgi
+import codecs
import copy
import datetime
import re
import struct
-import sys
import time
import types
import urllib
import urllib2
import urlparse
+import warnings
from htmlentitydefs import name2codepoint, codepoint2name, entitydefs
@@ -170,7 +178,7 @@ except ImportError:
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
+# been tested with the built-in SAX parser 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:
@@ -220,9 +228,12 @@ else:
# feedparser's scope instead of sgmllib's scope.
charref = re.compile('&#(\d+|[xX][0-9a-fA-F]+);')
tagfind = re.compile('[a-zA-Z][-_.:a-zA-Z0-9]*')
+ attrfind = re.compile(
+ r'\s*([a-zA-Z_][-:.a-zA-Z_0-9]*)[$]?(\s*=\s*'
+ r'(\'[^\']*\'|"[^"]*"|[][\-a-zA-Z0-9./,:;+*%?!&$\(\)_#=~\'"@]*))?'
+ )
# Unfortunately, these must be copied over to prevent NameError exceptions
- attrfind = sgmllib.attrfind
entityref = sgmllib.entityref
incomplete = sgmllib.incomplete
interesting = sgmllib.interesting
@@ -250,12 +261,8 @@ else:
endbracket = _EndBracketRegEx()
-# cjkcodecs and iconv_codec provide support for more character encodings.
-# Both are available from http://cjkpython.i18n.org/
-try:
- import cjkcodecs.aliases
-except ImportError:
- pass
+# iconv_codec provides support for more character encodings.
+# It's available from http://cjkpython.i18n.org/
try:
import iconv_codec
except ImportError:
@@ -268,15 +275,14 @@ try:
except ImportError:
chardet = None
-# BeautifulSoup parser used for parsing microformats from embedded HTML content
+# BeautifulSoup is used to extract microformat content from HTML
+# feedparser is tested using BeautifulSoup 3.2.0
# 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 ImportError:
BeautifulSoup = None
+ PARSE_MICROFORMATS = False
# ---------- don't touch these ----------
class ThingsNobodyCaresAboutButMe(Exception): pass
@@ -310,6 +316,7 @@ class FeedParserDict(dict):
'date': 'updated',
'date_parsed': 'updated_parsed',
'description': ['summary', 'subtitle'],
+ 'description_detail': ['summary_detail', 'subtitle_detail'],
'url': ['href'],
'modified': 'updated',
'modified_parsed': 'updated_parsed',
@@ -330,10 +337,32 @@ class FeedParserDict(dict):
return [norel(link) for link in dict.__getitem__(self, 'links') if link['rel']==u'enclosure']
elif key == 'license':
for link in dict.__getitem__(self, 'links'):
- if link['rel']==u'license' and link.has_key('href'):
+ if link['rel']==u'license' and 'href' in link:
return link['href']
- elif key == 'categories':
- return [(tag['scheme'], tag['term']) for tag in dict.__getitem__(self, 'tags')]
+ elif key == 'updated':
+ # Temporarily help developers out by keeping the old
+ # broken behavior that was reported in issue 310.
+ # This fix was proposed in issue 328.
+ if not dict.__contains__(self, 'updated') and \
+ dict.__contains__(self, 'published'):
+ warnings.warn("To avoid breaking existing software while "
+ "fixing issue 310, a temporary mapping has been created "
+ "from `updated` to `published` if `updated` doesn't "
+ "exist. This fallback will be removed in a future version "
+ "of feedparser.", DeprecationWarning)
+ return dict.__getitem__(self, 'published')
+ return dict.__getitem__(self, 'updated')
+ elif key == 'updated_parsed':
+ if not dict.__contains__(self, 'updated_parsed') and \
+ dict.__contains__(self, 'published_parsed'):
+ warnings.warn("To avoid breaking existing software while "
+ "fixing issue 310, a temporary mapping has been created "
+ "from `updated_parsed` to `published_parsed` if "
+ "`updated_parsed` doesn't exist. This fallback will be "
+ "removed in a future version of feedparser.",
+ DeprecationWarning)
+ return dict.__getitem__(self, 'published_parsed')
+ return dict.__getitem__(self, 'updated_parsed')
else:
realkey = self.keymap.get(key, key)
if isinstance(realkey, list):
@@ -345,6 +374,11 @@ class FeedParserDict(dict):
return dict.__getitem__(self, key)
def __contains__(self, key):
+ if key in ('updated', 'updated_parsed'):
+ # Temporarily help developers out by keeping the old
+ # broken behavior that was reported in issue 310.
+ # This fix was proposed in issue 328.
+ return dict.__contains__(self, key)
try:
self.__getitem__(key)
except KeyError:
@@ -380,66 +414,45 @@ class FeedParserDict(dict):
except KeyError:
raise AttributeError, "object has no attribute '%s'" % key
-
-_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)
+ def __hash__(self):
+ return id(self)
_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
+ 128: unichr(8364), # euro sign
+ 130: unichr(8218), # single low-9 quotation mark
+ 131: unichr( 402), # latin small letter f with hook
+ 132: unichr(8222), # double low-9 quotation mark
+ 133: unichr(8230), # horizontal ellipsis
+ 134: unichr(8224), # dagger
+ 135: unichr(8225), # double dagger
+ 136: unichr( 710), # modifier letter circumflex accent
+ 137: unichr(8240), # per mille sign
+ 138: unichr( 352), # latin capital letter s with caron
+ 139: unichr(8249), # single left-pointing angle quotation mark
+ 140: unichr( 338), # latin capital ligature oe
+ 142: unichr( 381), # latin capital letter z with caron
+ 145: unichr(8216), # left single quotation mark
+ 146: unichr(8217), # right single quotation mark
+ 147: unichr(8220), # left double quotation mark
+ 148: unichr(8221), # right double quotation mark
+ 149: unichr(8226), # bullet
+ 150: unichr(8211), # en dash
+ 151: unichr(8212), # em dash
+ 152: unichr( 732), # small tilde
+ 153: unichr(8482), # trade mark sign
+ 154: unichr( 353), # latin small letter s with caron
+ 155: unichr(8250), # single right-pointing angle quotation mark
+ 156: unichr( 339), # latin small ligature oe
+ 158: unichr( 382), # latin small letter z with caron
+ 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:
+ if not isinstance(uri, unicode):
+ uri = uri.decode('utf-8', 'ignore')
uri = urlparse.urljoin(base, uri)
if not isinstance(uri, unicode):
return uri.decode('utf-8', 'ignore')
@@ -449,75 +462,76 @@ def _urljoin(base, uri):
# return urlparse.urljoin(base, uri)
class _FeedParserMixin:
- namespaces = {'': '',
- 'http://backend.userland.com/rss': '',
- 'http://blogs.law.harvard.edu/tech/rss': '',
- 'http://purl.org/rss/1.0/': '',
- 'http://my.netscape.com/rdf/simple/0.9/': '',
- 'http://example.com/newformat#': '',
- 'http://example.com/necho': '',
- 'http://purl.org/echo/': '',
- 'uri/of/echo/namespace#': '',
- 'http://purl.org/pie/': '',
- 'http://purl.org/atom/ns#': '',
- 'http://www.w3.org/2005/Atom': '',
- 'http://purl.org/rss/1.0/modules/rss091#': '',
+ namespaces = {
+ '': '',
+ 'http://backend.userland.com/rss': '',
+ 'http://blogs.law.harvard.edu/tech/rss': '',
+ 'http://purl.org/rss/1.0/': '',
+ 'http://my.netscape.com/rdf/simple/0.9/': '',
+ 'http://example.com/newformat#': '',
+ 'http://example.com/necho': '',
+ 'http://purl.org/echo/': '',
+ 'uri/of/echo/namespace#': '',
+ 'http://purl.org/pie/': '',
+ 'http://purl.org/atom/ns#': '',
+ 'http://www.w3.org/2005/Atom': '',
+ 'http://purl.org/rss/1.0/modules/rss091#': '',
- 'http://webns.net/mvcb/': 'admin',
- 'http://purl.org/rss/1.0/modules/aggregation/': 'ag',
- 'http://purl.org/rss/1.0/modules/annotate/': 'annotate',
- 'http://media.tangent.org/rss/1.0/': 'audio',
- 'http://backend.userland.com/blogChannelModule': 'blogChannel',
- 'http://web.resource.org/cc/': 'cc',
- 'http://backend.userland.com/creativeCommonsRssModule': 'creativeCommons',
- 'http://purl.org/rss/1.0/modules/company': 'co',
- 'http://purl.org/rss/1.0/modules/content/': 'content',
- 'http://my.theinfo.org/changed/1.0/rss/': 'cp',
- 'http://purl.org/dc/elements/1.1/': 'dc',
- 'http://purl.org/dc/terms/': 'dcterms',
- 'http://purl.org/rss/1.0/modules/email/': 'email',
- 'http://purl.org/rss/1.0/modules/event/': 'ev',
- 'http://rssnamespace.org/feedburner/ext/1.0': 'feedburner',
- 'http://freshmeat.net/rss/fm/': 'fm',
- 'http://xmlns.com/foaf/0.1/': 'foaf',
- 'http://www.w3.org/2003/01/geo/wgs84_pos#': 'geo',
- 'http://postneo.com/icbm/': 'icbm',
- 'http://purl.org/rss/1.0/modules/image/': 'image',
- 'http://www.itunes.com/DTDs/PodCast-1.0.dtd': 'itunes',
- 'http://example.com/DTDs/PodCast-1.0.dtd': 'itunes',
- 'http://purl.org/rss/1.0/modules/link/': 'l',
- 'http://search.yahoo.com/mrss': 'media',
- #Version 1.1.2 of the Media RSS spec added the trailing slash on the namespace
- 'http://search.yahoo.com/mrss/': 'media',
- 'http://madskills.com/public/xml/rss/module/pingback/': 'pingback',
- 'http://prismstandard.org/namespaces/1.2/basic/': 'prism',
- 'http://www.w3.org/1999/02/22-rdf-syntax-ns#': 'rdf',
- 'http://www.w3.org/2000/01/rdf-schema#': 'rdfs',
- 'http://purl.org/rss/1.0/modules/reference/': 'ref',
- 'http://purl.org/rss/1.0/modules/richequiv/': 'reqv',
- 'http://purl.org/rss/1.0/modules/search/': 'search',
- 'http://purl.org/rss/1.0/modules/slash/': 'slash',
- 'http://schemas.xmlsoap.org/soap/envelope/': 'soap',
- 'http://purl.org/rss/1.0/modules/servicestatus/': 'ss',
- 'http://hacks.benhammersley.com/rss/streaming/': 'str',
- 'http://purl.org/rss/1.0/modules/subscription/': 'sub',
- 'http://purl.org/rss/1.0/modules/syndication/': 'sy',
- 'http://schemas.pocketsoap.com/rss/myDescModule/': 'szf',
- 'http://purl.org/rss/1.0/modules/taxonomy/': 'taxo',
- 'http://purl.org/rss/1.0/modules/threading/': 'thr',
- 'http://purl.org/rss/1.0/modules/textinput/': 'ti',
- 'http://madskills.com/public/xml/rss/module/trackback/':'trackback',
- 'http://wellformedweb.org/commentAPI/': 'wfw',
- 'http://purl.org/rss/1.0/modules/wiki/': 'wiki',
- 'http://www.w3.org/1999/xhtml': 'xhtml',
- 'http://www.w3.org/1999/xlink': 'xlink',
- 'http://www.w3.org/XML/1998/namespace': 'xml'
-}
+ 'http://webns.net/mvcb/': 'admin',
+ 'http://purl.org/rss/1.0/modules/aggregation/': 'ag',
+ 'http://purl.org/rss/1.0/modules/annotate/': 'annotate',
+ 'http://media.tangent.org/rss/1.0/': 'audio',
+ 'http://backend.userland.com/blogChannelModule': 'blogChannel',
+ 'http://web.resource.org/cc/': 'cc',
+ 'http://backend.userland.com/creativeCommonsRssModule': 'creativeCommons',
+ 'http://purl.org/rss/1.0/modules/company': 'co',
+ 'http://purl.org/rss/1.0/modules/content/': 'content',
+ 'http://my.theinfo.org/changed/1.0/rss/': 'cp',
+ 'http://purl.org/dc/elements/1.1/': 'dc',
+ 'http://purl.org/dc/terms/': 'dcterms',
+ 'http://purl.org/rss/1.0/modules/email/': 'email',
+ 'http://purl.org/rss/1.0/modules/event/': 'ev',
+ 'http://rssnamespace.org/feedburner/ext/1.0': 'feedburner',
+ 'http://freshmeat.net/rss/fm/': 'fm',
+ 'http://xmlns.com/foaf/0.1/': 'foaf',
+ 'http://www.w3.org/2003/01/geo/wgs84_pos#': 'geo',
+ 'http://postneo.com/icbm/': 'icbm',
+ 'http://purl.org/rss/1.0/modules/image/': 'image',
+ 'http://www.itunes.com/DTDs/PodCast-1.0.dtd': 'itunes',
+ 'http://example.com/DTDs/PodCast-1.0.dtd': 'itunes',
+ 'http://purl.org/rss/1.0/modules/link/': 'l',
+ 'http://search.yahoo.com/mrss': 'media',
+ # Version 1.1.2 of the Media RSS spec added the trailing slash on the namespace
+ 'http://search.yahoo.com/mrss/': 'media',
+ 'http://madskills.com/public/xml/rss/module/pingback/': 'pingback',
+ 'http://prismstandard.org/namespaces/1.2/basic/': 'prism',
+ 'http://www.w3.org/1999/02/22-rdf-syntax-ns#': 'rdf',
+ 'http://www.w3.org/2000/01/rdf-schema#': 'rdfs',
+ 'http://purl.org/rss/1.0/modules/reference/': 'ref',
+ 'http://purl.org/rss/1.0/modules/richequiv/': 'reqv',
+ 'http://purl.org/rss/1.0/modules/search/': 'search',
+ 'http://purl.org/rss/1.0/modules/slash/': 'slash',
+ 'http://schemas.xmlsoap.org/soap/envelope/': 'soap',
+ 'http://purl.org/rss/1.0/modules/servicestatus/': 'ss',
+ 'http://hacks.benhammersley.com/rss/streaming/': 'str',
+ 'http://purl.org/rss/1.0/modules/subscription/': 'sub',
+ 'http://purl.org/rss/1.0/modules/syndication/': 'sy',
+ 'http://schemas.pocketsoap.com/rss/myDescModule/': 'szf',
+ 'http://purl.org/rss/1.0/modules/taxonomy/': 'taxo',
+ 'http://purl.org/rss/1.0/modules/threading/': 'thr',
+ 'http://purl.org/rss/1.0/modules/textinput/': 'ti',
+ 'http://madskills.com/public/xml/rss/module/trackback/': 'trackback',
+ 'http://wellformedweb.org/commentAPI/': 'wfw',
+ 'http://purl.org/rss/1.0/modules/wiki/': 'wiki',
+ 'http://www.w3.org/1999/xhtml': 'xhtml',
+ 'http://www.w3.org/1999/xlink': 'xlink',
+ 'http://www.w3.org/XML/1998/namespace': 'xml',
+ }
_matchnamespaces = {}
- can_be_relative_uri = ['link', 'id', 'wfw_comment', 'wfw_commentrss', 'docs', 'url', 'href', 'comments', 'icon', 'logo']
- can_contain_relative_uris = ['content', 'title', 'summary', 'info', 'tagline', 'subtitle', 'copyright', 'rights', 'description']
- can_contain_dangerous_markup = ['content', 'title', 'summary', 'info', 'tagline', 'subtitle', 'copyright', 'rights', 'description']
+ can_be_relative_uri = set(['link', 'id', 'wfw_comment', 'wfw_commentrss', 'docs', 'url', 'href', 'comments', 'icon', 'logo'])
+ can_contain_relative_uris = set(['content', 'title', 'summary', 'info', 'tagline', 'subtitle', 'copyright', 'rights', 'description'])
+ can_contain_dangerous_markup = set(['content', 'title', 'summary', 'info', 'tagline', 'subtitle', 'copyright', 'rights', 'description'])
html_types = [u'text/html', u'application/xhtml+xml']
def __init__(self, baseuri=None, baselang=None, encoding=u'utf-8'):
@@ -551,10 +565,20 @@ class _FeedParserMixin:
self.baseuri = baseuri or u''
self.lang = baselang or None
self.svgOK = 0
- self.hasTitle = 0
+ self.title_depth = -1
+ self.depth = 0
if baselang:
self.feeddata['language'] = baselang.replace('_','-')
+ # A map of the following form:
+ # {
+ # object_that_value_is_set_on: {
+ # property_name: depth_of_node_property_was_extracted_from,
+ # other_property: depth_of_node_property_was_extracted_from,
+ # },
+ # }
+ self.property_depth_map = {}
+
def _normalize_attributes(self, kv):
k = kv[0].lower()
v = k in ('rel', 'type') and kv[1].lower() or kv[1]
@@ -568,6 +592,9 @@ class _FeedParserMixin:
return (k, v)
def unknown_starttag(self, tag, attrs):
+ # increment depth counter
+ self.depth += 1
+
# normalize attrs
attrs = map(self._normalize_attributes, attrs)
@@ -604,8 +631,8 @@ class _FeedParserMixin:
self.trackNamespace(None, uri)
# track inline content
- if self.incontent and self.contentparams.has_key('type') and not self.contentparams.get('type', u'xml').endswith(u'xml'):
- if tag in ['xhtml:div', 'div']:
+ if self.incontent and not self.contentparams.get('type', u'xml').endswith(u'xml'):
+ if tag in ('xhtml:div', 'div'):
return # typepad does this 10/2007
# element declared itself as escaped markup, but it isn't really
self.contentparams['type'] = u'application/xhtml+xml'
@@ -675,9 +702,9 @@ class _FeedParserMixin:
self.pop(prefix + suffix)
# track inline content
- if self.incontent and self.contentparams.has_key('type') and not self.contentparams.get('type', u'xml').endswith(u'xml'):
+ if self.incontent and not self.contentparams.get('type', u'xml').endswith(u'xml'):
# element declared itself as escaped markup, but it isn't really
- if tag in ['xhtml:div', 'div']:
+ if tag in ('xhtml:div', 'div'):
return # typepad does this 10/2007
self.contentparams['type'] = u'application/xhtml+xml'
if self.incontent and self.contentparams.get('type') == u'application/xhtml+xml':
@@ -694,6 +721,8 @@ class _FeedParserMixin:
if self.langstack: # and (self.langstack[-1] is not None):
self.lang = self.langstack[-1]
+ self.depth -= 1
+
def handle_charref(self, ref):
# called for each character reference, e.g. for ' ', ref will be '160'
if not self.elementstack:
@@ -715,7 +744,7 @@ class _FeedParserMixin:
return
if ref in ('lt', 'gt', 'quot', 'amp', 'apos'):
text = '&%s;' % ref
- elif ref in self.entities.keys():
+ elif ref in self.entities:
text = self.entities[ref]
if text.startswith('&#') and text.endswith(';'):
return self.handle_entityref(text)
@@ -778,17 +807,18 @@ class _FeedParserMixin:
def trackNamespace(self, prefix, uri):
loweruri = uri.lower()
- if (prefix, loweruri) == (None, 'http://my.netscape.com/rdf/simple/0.9/') and not self.version:
- self.version = u'rss090'
- if loweruri == 'http://purl.org/rss/1.0/' and not self.version:
- self.version = u'rss10'
- if loweruri == 'http://www.w3.org/2005/atom' and not self.version:
- self.version = u'atom10'
+ if not self.version:
+ if (prefix, loweruri) == (None, 'http://my.netscape.com/rdf/simple/0.9/'):
+ self.version = u'rss090'
+ elif loweruri == 'http://purl.org/rss/1.0/':
+ self.version = u'rss10'
+ elif loweruri == 'http://www.w3.org/2005/atom':
+ self.version = u'atom10'
if loweruri.find(u'backend.userland.com/rss') <> -1:
# match any backend.userland.com namespace
uri = u'http://backend.userland.com/rss'
loweruri = uri
- if self._matchnamespaces.has_key(loweruri):
+ if loweruri in self._matchnamespaces:
self.namespacemap[prefix] = self._matchnamespaces[loweruri]
self.namespacesInUse[self._matchnamespaces[loweruri]] = uri
else:
@@ -892,7 +922,7 @@ class _FeedParserMixin:
# 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']:
+ if PARSE_MICROFORMATS and is_htmlish and element in ['content', 'description', 'summary']:
mfresults = _parseMicroformats(output, self.baseuri, self.encoding)
if mfresults:
for tag in mfresults.get('tags', []):
@@ -923,13 +953,13 @@ class _FeedParserMixin:
# map win-1252 extensions to the proper code points
if isinstance(output, unicode):
- output = u''.join([c in _cp1252.keys() and _cp1252[c] or c for c in output])
+ output = output.translate(_cp1252)
# categories/tags/keywords/whatever are handled in _end_category
if element == 'category':
return output
- if element == 'title' and self.hasTitle:
+ if element == 'title' and -1 < self.title_depth <= self.depth:
return output
# store output in appropriate place(s)
@@ -951,7 +981,10 @@ class _FeedParserMixin:
else:
if element == 'description':
element = 'summary'
- self.entries[-1][element] = output
+ old_value_depth = self.property_depth_map.setdefault(self.entries[-1], {}).get(element)
+ if old_value_depth is None or self.depth <= old_value_depth:
+ self.property_depth_map[self.entries[-1]][element] = self.depth
+ self.entries[-1][element] = output
if self.incontent:
contentparams = copy.deepcopy(self.contentparams)
contentparams['value'] = output
@@ -995,7 +1028,7 @@ class _FeedParserMixin:
# data loss, this function errs on the conservative side.
@staticmethod
def lookslikehtml(s):
- # must have a close tag or a entity reference to qualify
+ # must have a close tag or an entity reference to qualify
if not (re.search(r'',s) or re.search("&#?\w+;",s)):
return
@@ -1077,11 +1110,11 @@ class _FeedParserMixin:
self._cdf_common(attrsD)
def _cdf_common(self, attrsD):
- if attrsD.has_key('lastmod'):
+ if 'lastmod' in attrsD:
self._start_modified({})
self.elementstack[-1][-1] = attrsD['lastmod']
self._end_modified()
- if attrsD.has_key('href'):
+ if 'href' in attrsD:
self._start_link({})
self.elementstack[-1][-1] = attrsD['href']
self._end_link()
@@ -1108,7 +1141,7 @@ class _FeedParserMixin:
if not self.inentry:
context.setdefault('image', FeedParserDict())
self.inimage = 1
- self.hasTitle = 0
+ self.title_depth = -1
self.push('image', 0)
def _end_image(self):
@@ -1119,7 +1152,7 @@ class _FeedParserMixin:
context = self._getContext()
context.setdefault('textinput', FeedParserDict())
self.intextinput = 1
- self.hasTitle = 0
+ self.title_depth = -1
self.push('textinput', 0)
_start_textInput = _start_textinput
@@ -1254,7 +1287,7 @@ class _FeedParserMixin:
def _getContext(self):
if self.insource:
context = self.sourcedata
- elif self.inimage and self.feeddata.has_key('image'):
+ elif self.inimage and 'image' in self.feeddata:
context = self.feeddata['image']
elif self.intextinput:
context = self.feeddata['textinput']
@@ -1339,7 +1372,7 @@ class _FeedParserMixin:
self.push('item', 0)
self.inentry = 1
self.guidislink = 0
- self.hasTitle = 0
+ self.title_depth = -1
id = self._getAttribute(attrsD, 'rdf:about')
if id:
context = self._getContext()
@@ -1373,18 +1406,19 @@ class _FeedParserMixin:
self.push('published', 1)
_start_dcterms_issued = _start_published
_start_issued = _start_published
+ _start_pubdate = _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
+ _end_pubdate = _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
@@ -1394,7 +1428,6 @@ class _FeedParserMixin:
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
@@ -1467,8 +1500,9 @@ class _FeedParserMixin:
self._start_category(attrsD)
def _end_itunes_keywords(self):
- for term in self.pop('itunes_keywords').split():
- self._addTag(term, u'http://www.itunes.com/', None)
+ for term in self.pop('itunes_keywords').split(','):
+ if term.strip():
+ self._addTag(term.strip(), u'http://www.itunes.com/', None)
def _start_itunes_category(self, attrsD):
self._addTag(attrsD.get('text'), u'http://www.itunes.com/', None)
@@ -1500,13 +1534,13 @@ class _FeedParserMixin:
attrsD.setdefault('type', u'text/html')
context = self._getContext()
attrsD = self._itsAnHrefDamnIt(attrsD)
- if attrsD.has_key('href'):
+ if 'href' in attrsD:
attrsD['href'] = self.resolveURI(attrsD['href'])
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'):
+ if 'href' in attrsD:
expectingText = 0
if (attrsD.get('rel') == u'alternate') and (self.mapContentType(attrsD.get('type')) in self.html_types):
context['link'] = attrsD['href']
@@ -1515,19 +1549,20 @@ class _FeedParserMixin:
def _end_link(self):
value = self.pop('link')
- context = self._getContext()
def _start_guid(self, attrsD):
self.guidislink = (attrsD.get('ispermalink', 'true') == 'true')
self.push('id', 1)
+ _start_id = _start_guid
def _end_guid(self):
value = self.pop('id')
- self._save('guidislink', self.guidislink and not self._getContext().has_key('link'))
+ self._save('guidislink', self.guidislink and 'link' not in self._getContext())
if self.guidislink:
# guid acts as link, but only if 'ispermalink' is not present or is 'true',
# and only if the item doesn't already have a link element
self._save('link', value)
+ _end_id = _end_guid
def _start_title(self, attrsD):
if self.svgOK:
@@ -1542,18 +1577,17 @@ class _FeedParserMixin:
value = self.popContent('title')
if not value:
return
- context = self._getContext()
- self.hasTitle = 1
+ self.title_depth = self.depth
_end_dc_title = _end_title
def _end_media_title(self):
- hasTitle = self.hasTitle
+ title_depth = self.title_depth
self._end_title()
- self.hasTitle = hasTitle
+ self.title_depth = title_depth
def _start_description(self, attrsD):
context = self._getContext()
- if context.has_key('summary'):
+ if 'summary' in context:
self._summaryKey = 'content'
self._start_content(attrsD)
else:
@@ -1583,7 +1617,7 @@ class _FeedParserMixin:
def _start_generator(self, attrsD):
if attrsD:
attrsD = self._itsAnHrefDamnIt(attrsD)
- if attrsD.has_key('href'):
+ if 'href' in attrsD:
attrsD['href'] = self.resolveURI(attrsD['href'])
self._getContext()['generator_detail'] = FeedParserDict(attrsD)
self.push('generator', 1)
@@ -1591,7 +1625,7 @@ class _FeedParserMixin:
def _end_generator(self):
value = self.pop('generator')
context = self._getContext()
- if context.has_key('generator_detail'):
+ if 'generator_detail' in context:
context['generator_detail']['name'] = value
def _start_admin_generatoragent(self, attrsD):
@@ -1611,7 +1645,7 @@ class _FeedParserMixin:
def _start_summary(self, attrsD):
context = self._getContext()
- if context.has_key('summary'):
+ if 'summary' in context:
self._summaryKey = 'content'
self._start_content(attrsD)
else:
@@ -1635,17 +1669,17 @@ class _FeedParserMixin:
def _start_source(self, attrsD):
if 'url' in attrsD:
- # This means that we're processing a source element from an RSS 2.0 feed
- self.sourcedata['href'] = attrsD[u'url']
+ # 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
+ self.title_depth = -1
def _end_source(self):
self.insource = 0
value = self.pop('source')
if value:
- self.sourcedata['title'] = value
+ self.sourcedata['title'] = value
self._getContext()['source'] = copy.deepcopy(self.sourcedata)
self.sourcedata.clear()
@@ -1679,6 +1713,8 @@ class _FeedParserMixin:
self.push('itunes_image', 0)
if attrsD.get('href'):
self._getContext()['image'] = FeedParserDict({'href': attrsD.get('href')})
+ elif attrsD.get('url'):
+ self._getContext()['image'] = FeedParserDict({'href': attrsD.get('url')})
_start_itunes_link = _start_itunes_image
def _end_itunes_block(self):
@@ -1707,7 +1743,7 @@ class _FeedParserMixin:
url = self.pop('url')
context = self._getContext()
if url != None and len(url.strip()) != 0:
- if not context['media_thumbnail'][-1].has_key('url'):
+ if 'url' not in context['media_thumbnail'][-1]:
context['media_thumbnail'][-1]['url'] = url
def _start_media_player(self, attrsD):
@@ -1760,8 +1796,8 @@ if _XML_AVAILABLE:
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
+ if givenprefix and (prefix == None or (prefix == '' and lowernamespace == '')) and givenprefix not in self.namespacesInUse:
+ raise UndeclaredNamespace, "'%s' is not associated with a namespace" % givenprefix
localname = str(localname).lower()
# qname implementation is horribly broken in Python 2.1 (it
@@ -1781,9 +1817,9 @@ if _XML_AVAILABLE:
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 name and value == namespace:
+ localname = name + ':' + localname
+ break
for (namespace, attrlocalname), attrvalue in attrs.items():
lowernamespace = (namespace or '').lower()
@@ -1810,9 +1846,9 @@ if _XML_AVAILABLE:
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
+ if name and value == namespace:
+ localname = name + ':' + localname
+ break
localname = str(localname).lower()
self.unknown_endtag(localname)
@@ -1820,6 +1856,9 @@ if _XML_AVAILABLE:
self.bozo = 1
self.exc = exc
+ # drv_libxml2 calls warning() in some cases
+ warning = error
+
def fatalError(self, exc):
self.error(exc)
raise exc
@@ -1827,11 +1866,11 @@ if _XML_AVAILABLE:
class _BaseHTMLProcessor(sgmllib.SGMLParser):
special = re.compile('''[<>'"]''')
bare_ampersand = re.compile("&(?!#\d+;|#x[0-9a-fA-F]+;|\w+;)")
- elements_no_end_tag = [
+ elements_no_end_tag = set([
'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
@@ -1871,7 +1910,6 @@ class _BaseHTMLProcessor(sgmllib.SGMLParser):
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('"', '"')
@@ -1920,36 +1958,36 @@ class _BaseHTMLProcessor(sgmllib.SGMLParser):
except (UnicodeEncodeError, LookupError):
pass
if tag in self.elements_no_end_tag:
- self.pieces.append('<%(tag)s%(strattrs)s />' % locals())
+ self.pieces.append('<%s%s />' % (tag, strattrs))
else:
- self.pieces.append('<%(tag)s%(strattrs)s>' % locals())
+ self.pieces.append('<%s%s>' % (tag, strattrs))
def unknown_endtag(self, tag):
# called for each end tag, e.g. for , tag will be 'pre'
# Reconstruct the original end tag.
if tag not in self.elements_no_end_tag:
- self.pieces.append("" % locals())
+ self.pieces.append("" % tag)
def handle_charref(self, ref):
# called for each character reference, e.g. for ' ', ref will be '160'
# Reconstruct the original character reference.
if ref.startswith('x'):
- value = unichr(int(ref[1:],16))
+ value = int(ref[1:], 16)
else:
- value = unichr(int(ref))
+ value = int(ref)
- if value in _cp1252.keys():
+ if value in _cp1252:
self.pieces.append('&#%s;' % hex(ord(_cp1252[value]))[1:])
else:
- self.pieces.append('&#%(ref)s;' % locals())
+ self.pieces.append('&#%s;' % ref)
def handle_entityref(self, ref):
# called for each entity reference, e.g. for '©', ref will be 'copy'
# Reconstruct the original entity reference.
- if name2codepoint.has_key(ref):
- self.pieces.append('&%(ref)s;' % locals())
+ if ref in name2codepoint or ref == 'apos':
+ self.pieces.append('&%s;' % ref)
else:
- self.pieces.append('&%(ref)s' % locals())
+ self.pieces.append('&%s' % ref)
def handle_data(self, text):
# called for each block of plain text, i.e. outside of any tag and
@@ -1960,19 +1998,19 @@ class _BaseHTMLProcessor(sgmllib.SGMLParser):
def handle_comment(self, text):
# called for each HTML comment, e.g.
# Reconstruct the original comment.
- self.pieces.append('' % locals())
+ self.pieces.append('' % text)
def handle_pi(self, text):
# called for each processing instruction, e.g.
# Reconstruct original processing instruction.
- self.pieces.append('' % locals())
+ self.pieces.append('' % text)
def handle_decl(self, text):
# called for the DOCTYPE, if present, e.g.
#
# Reconstruct original DOCTYPE
- self.pieces.append('' % locals())
+ self.pieces.append('' % text)
_new_declname_match = re.compile(r'[a-zA-Z][-_.a-zA-Z0-9:]*\s*').match
def _scan_name(self, i, declstartpos):
@@ -2030,7 +2068,7 @@ class _LooseFeedParser(_FeedParserMixin, _BaseHTMLProcessor):
data = data.replace('"', '"')
data = data.replace(''', ''')
data = data.replace(''', ''')
- if self.contentparams.has_key('type') and not self.contentparams.get('type', u'xml').endswith(u'xml'):
+ if not self.contentparams.get('type', u'xml').endswith(u'xml'):
data = data.replace('<', '<')
data = data.replace('>', '>')
data = data.replace('&', '&')
@@ -2048,8 +2086,8 @@ class _MicroformatsParser:
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']
+ known_xfn_relationships = set(['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 = set(['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)
@@ -2410,7 +2448,7 @@ class _MicroformatsParser:
def isProbablyDownloadable(self, elm):
attrsD = elm.attrMap
- if not attrsD.has_key('href'):
+ if 'href' not in attrsD:
return 0
linktype = attrsD.get('type', '').strip()
if linktype.startswith('audio/') or \
@@ -2459,10 +2497,7 @@ class _MicroformatsParser:
all = lambda x: 1
for elm in self.document(all, {'rel': re.compile('.+'), 'href': re.compile('.+')}):
rels = elm.get('rel', u'').split()
- xfn_rels = []
- for rel in rels:
- if rel in self.known_xfn_relationships:
- xfn_rels.append(rel)
+ xfn_rels = [r for r in rels if r in self.known_xfn_relationships]
if xfn_rels:
self.xfn.append({"relationships": xfn_rels, "href": elm.get('href', ''), "name": elm.string})
@@ -2482,7 +2517,7 @@ def _parseMicroformats(htmlSource, baseURI, encoding):
return {"tags": p.tags, "enclosures": p.enclosures, "xfn": p.xfn, "vcard": p.vcard}
class _RelativeURIResolver(_BaseHTMLProcessor):
- relative_uris = [('a', 'href'),
+ relative_uris = set([('a', 'href'),
('applet', 'codebase'),
('area', 'href'),
('blockquote', 'cite'),
@@ -2506,14 +2541,14 @@ class _RelativeURIResolver(_BaseHTMLProcessor):
('object', 'data'),
('object', 'usemap'),
('q', 'cite'),
- ('script', 'src')]
+ ('script', 'src')])
def __init__(self, baseuri, encoding, _type):
_BaseHTMLProcessor.__init__(self, encoding, _type)
self.baseuri = baseuri
def resolveURI(self, uri):
- return _makeSafeAbsoluteURI(_urljoin(self.baseuri, uri.strip()))
+ return _makeSafeAbsoluteURI(self.baseuri, uri.strip())
def unknown_starttag(self, tag, attrs):
attrs = self.normalize_attrs(attrs)
@@ -2531,21 +2566,30 @@ def _resolveRelativeURIs(htmlSource, baseURI, encoding, _type):
def _makeSafeAbsoluteURI(base, rel=None):
# bail if ACCEPTABLE_URI_SCHEMES is empty
if not ACCEPTABLE_URI_SCHEMES:
- return _urljoin(base, rel or u'')
+ try:
+ return _urljoin(base, rel or u'')
+ except ValueError:
+ return u''
if not base:
return rel or u''
if not rel:
- scheme = urlparse.urlparse(base)[0]
+ try:
+ scheme = urlparse.urlparse(base)[0]
+ except ValueError:
+ return u''
if not scheme or scheme in ACCEPTABLE_URI_SCHEMES:
return base
return u''
- uri = _urljoin(base, rel)
+ try:
+ uri = _urljoin(base, rel)
+ except ValueError:
+ return u''
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',
+ acceptable_elements = set(['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',
@@ -2557,9 +2601,9 @@ class _HTMLSanitizer(_BaseHTMLProcessor):
'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']
+ 'th', 'thead', 'tr', 'tt', 'u', 'ul', 'var', 'video', 'noscript'])
- acceptable_attributes = ['abbr', 'accept', 'accept-charset', 'accesskey',
+ acceptable_attributes = set(['abbr', 'accept', 'accept-charset', 'accesskey',
'action', 'align', 'alt', 'autocomplete', 'autofocus', 'axis',
'background', 'balance', 'bgcolor', 'bgproperties', 'border',
'bordercolor', 'bordercolordark', 'bordercolorlight', 'bottompadding',
@@ -2580,11 +2624,11 @@ class _HTMLSanitizer(_BaseHTMLProcessor):
'start', 'step', 'summary', 'suppress', 'tabindex', 'target', 'template',
'title', 'toppadding', 'type', 'unselectable', 'usemap', 'urn', 'valign',
'value', 'variable', 'volume', 'vspace', 'vrml', 'width', 'wrap',
- 'xml:lang']
+ 'xml:lang'])
- unacceptable_elements_with_end_tag = ['script', 'applet', 'style']
+ unacceptable_elements_with_end_tag = set(['script', 'applet', 'style'])
- acceptable_css_properties = ['azimuth', 'background-color',
+ acceptable_css_properties = set(['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',
@@ -2594,26 +2638,26 @@ class _HTMLSanitizer(_BaseHTMLProcessor):
'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']
+ 'white-space', 'width'])
# survey of common keywords found in feeds
- acceptable_css_keywords = ['auto', 'aqua', 'black', 'block', 'blue',
+ acceptable_css_keywords = set(['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']
+ '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',
+ mathml_elements = set(['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']
+ 'munderover', 'none', 'semantics'])
- mathml_attributes = ['actiontype', 'align', 'columnalign', 'columnalign',
+ mathml_attributes = set(['actiontype', 'align', 'columnalign', 'columnalign',
'columnalign', 'close', 'columnlines', 'columnspacing', 'columnspan', 'depth',
'display', 'displaystyle', 'encoding', 'equalcolumns', 'equalrows',
'fence', 'fontstyle', 'fontweight', 'frame', 'height', 'linethickness',
@@ -2621,18 +2665,18 @@ class _HTMLSanitizer(_BaseHTMLProcessor):
'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']
+ 'xlink:show', 'xlink:type', 'xmlns', 'xmlns:xlink'])
# svgtiny - foreignObject + linearGradient + radialGradient + stop
- svg_elements = ['a', 'animate', 'animateColor', 'animateMotion',
+ svg_elements = set(['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']
+ 'svg', 'switch', 'text', 'title', 'tspan', 'use'])
# svgtiny + class + opacity + offset + xmlns + xmlns:xlink
- svg_attributes = ['accent-height', 'accumulate', 'additive', 'alphabetic',
+ svg_attributes = set(['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',
@@ -2658,14 +2702,14 @@ class _HTMLSanitizer(_BaseHTMLProcessor):
'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']
+ 'y2', 'zoomAndPan'])
svg_attr_map = None
svg_elem_map = None
- acceptable_svg_properties = [ 'fill', 'fill-opacity', 'fill-rule',
+ acceptable_svg_properties = set([ 'fill', 'fill-opacity', 'fill-rule',
'stroke', 'stroke-width', 'stroke-linecap', 'stroke-linejoin',
- 'stroke-opacity']
+ 'stroke-opacity'])
def reset(self):
_BaseHTMLProcessor.reset(self)
@@ -2718,7 +2762,7 @@ class _HTMLSanitizer(_BaseHTMLProcessor):
# declare xlink namespace, if needed
if self.mathmlOK or self.svgOK:
- if filter((lambda n,v: n.startswith('xlink:')),attrs):
+ 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'))
@@ -2923,9 +2967,6 @@ def _open_resource(url_file_stream_or_string, etag, modified, agent, referrer, h
if hasattr(url_file_stream_or_string, 'read'):
return url_file_stream_or_string
- if url_file_stream_or_string == '-':
- return sys.stdin
-
if isinstance(url_file_stream_or_string, basestring) \
and urlparse.urlparse(url_file_stream_or_string)[0] in ('http', 'https', 'ftp', 'file', 'feed'):
# Deal with the feed URI scheme
@@ -2952,7 +2993,7 @@ def _open_resource(url_file_stream_or_string, etag, modified, agent, referrer, h
# 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 = urllib2.build_opener(*tuple(handlers + [_FeedURLHandler()]))
opener.addheaders = [] # RMK - must clear so we only send our custom User-Agent
try:
return opener.open(request)
@@ -2962,7 +3003,14 @@ def _open_resource(url_file_stream_or_string, etag, modified, agent, referrer, h
# try to open with native open function (if url_file_stream_or_string is a filename)
try:
return open(url_file_stream_or_string, 'rb')
- except IOError:
+ except (IOError, UnicodeEncodeError, TypeError):
+ # if url_file_stream_or_string is a unicode object that
+ # cannot be converted to the encoding returned by
+ # sys.getfilesystemencoding(), a UnicodeEncodeError
+ # will be thrown
+ # If url_file_stream_or_string is a string that contains NULL
+ # (such as an XML document encoded in UTF-32), TypeError will
+ # be thrown.
pass
# treat url_file_stream_or_string as string
@@ -3122,7 +3170,7 @@ def _parse_date_iso8601(dateString):
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():
+ if 'century' in params:
year = (int(params['century']) - 1) * 100 + 1
# in ISO 8601 most fields are optional
for field in ['hour', 'minute', 'second', 'tzhour', 'tzmin']:
@@ -3197,20 +3245,6 @@ def _parse_date_nate(dateString):
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'}
- return _parse_date_w3dtf(w3dtfdate)
-registerDateHandler(_parse_date_mssql)
-
# Unicode strings for Greek date strings
_greek_months = \
{ \
@@ -3306,6 +3340,9 @@ registerDateHandler(_parse_date_hungarian)
# Drake and licensed under the Python license. Removed all range checking
# for month, day, hour, minute, and second, since mktime will normalize
# these later
+# Modified to also support MSSQL-style datetimes as defined at:
+# http://msdn.microsoft.com/en-us/library/ms186724.aspx
+# (which basically means allowing a space as a date/time/timezone separator)
def _parse_date_w3dtf(dateString):
def __extract_date(m):
year = int(m.group('year'))
@@ -3331,7 +3368,7 @@ def _parse_date_w3dtf(dateString):
day = 31
elif jday < julian:
if day + diff < 28:
- day = day + diff
+ day = day + diff
else:
month = month + 1
return year, month, day
@@ -3387,12 +3424,11 @@ def _parse_date_w3dtf(dateString):
'(?:(?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)
+ __tzd_re = ' ?(?P[-+](?P\d\d)(?::?(?P\d\d))|Z)?'
__time_re = ('(?P\d\d)(?P:|)(?P\d\d)'
'(?:(?P=tsep)(?P\d\d)(?:[.,]\d+)?)?'
+ __tzd_re)
- __datetime_re = '%s(?:T%s)?' % (__date_re, __time_re)
+ __datetime_re = '%s(?:[T ]%s)?' % (__date_re, __time_re)
__datetime_rx = re.compile(__datetime_re)
m = __datetime_rx.match(dateString)
if (m is None) or (m.group() != dateString):
@@ -3403,41 +3439,91 @@ def _parse_date_w3dtf(dateString):
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 not data:
+# Define the strings used by the RFC822 datetime parser
+_rfc822_months = ['jan', 'feb', 'mar', 'apr', 'may', 'jun',
+ 'jul', 'aug', 'sep', 'oct', 'nov', 'dec']
+_rfc822_daynames = ['mon', 'tue', 'wed', 'thu', 'fri', 'sat', 'sun']
+
+# Only the first three letters of the month name matter
+_rfc822_month = "(?P%s)(?:[a-z]*,?)" % ('|'.join(_rfc822_months))
+# The year may be 2 or 4 digits; capture the century if it exists
+_rfc822_year = "(?P(?:\d{2})?\d{2})"
+_rfc822_day = "(?P *\d{1,2})"
+_rfc822_date = "%s %s %s" % (_rfc822_day, _rfc822_month, _rfc822_year)
+
+_rfc822_hour = "(?P\d{2}):(?P\d{2})(?::(?P\d{2}))?"
+_rfc822_tz = "(?Put|gmt(?:[+-]\d{2}:\d{2})?|[aecmp][sd]?t|[zamny]|[+-]\d{4})"
+_rfc822_tznames = {
+ 'ut': 0, 'gmt': 0, 'z': 0,
+ 'adt': -3, 'ast': -4, 'at': -4,
+ 'edt': -4, 'est': -5, 'et': -5,
+ 'cdt': -5, 'cst': -6, 'ct': -6,
+ 'mdt': -6, 'mst': -7, 'mt': -7,
+ 'pdt': -7, 'pst': -8, 'pt': -8,
+ 'a': -1, 'n': 1,
+ 'm': -12, 'y': 12,
+ }
+# The timezone may be prefixed by 'Etc/'
+_rfc822_time = "%s (?:etc/)?%s" % (_rfc822_hour, _rfc822_tz)
+
+_rfc822_dayname = "(?P%s)" % ('|'.join(_rfc822_daynames))
+_rfc822_match = re.compile(
+ "(?:%s, )?%s(?: %s)?" % (_rfc822_dayname, _rfc822_date, _rfc822_time)
+).match
+
+def _parse_date_rfc822(dt):
+ """Parse RFC 822 dates and times, with one minor
+ difference: years may be 4DIGIT or 2DIGIT.
+ http://tools.ietf.org/html/rfc822#section-5"""
+ try:
+ m = _rfc822_match(dt.lower()).groupdict(0)
+ except AttributeError:
return None
- 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:
- # Jython doesn't adjust for 2-digit years like CPython does,
- # so account for it by shifting the year so that it's in the
- # range 1970-2069 (1970 being the year of the Unix epoch).
- if tm[0] < 100:
- tm = (tm[0] + (1900, 2000)[tm[0] < 70],) + tm[1:]
- return time.gmtime(rfc822.mktime_tz(tm))
-# rfc822.py defines several time zones, but we define some extra ones.
-# 'ET' is equivalent to 'EST', etc.
-_additional_timezones = {'AT': -400, 'ET': -500, 'CT': -600, 'MT': -700, 'PT': -800}
-rfc822._timezones.update(_additional_timezones)
+
+ # Calculate a date and timestamp
+ for k in ('year', 'day', 'hour', 'minute', 'second'):
+ m[k] = int(m[k])
+ m['month'] = _rfc822_months.index(m['month']) + 1
+ # If the year is 2 digits, assume everything in the 90's is the 1990's
+ if m['year'] < 100:
+ m['year'] += (1900, 2000)[m['year'] < 90]
+ stamp = datetime.datetime(*[m[i] for i in
+ ('year', 'month', 'day', 'hour', 'minute', 'second')])
+
+ # Use the timezone information to calculate the difference between
+ # the given date and timestamp and Universal Coordinated Time
+ tzhour = 0
+ tzmin = 0
+ if m['tz'] and m['tz'].startswith('gmt'):
+ # Handle GMT and GMT+hh:mm timezone syntax (the trailing
+ # timezone info will be handled by the next `if` block)
+ m['tz'] = ''.join(m['tz'][3:].split(':')) or 'gmt'
+ if not m['tz']:
+ pass
+ elif m['tz'].startswith('+'):
+ tzhour = int(m['tz'][1:3])
+ tzmin = int(m['tz'][3:])
+ elif m['tz'].startswith('-'):
+ tzhour = int(m['tz'][1:3]) * -1
+ tzmin = int(m['tz'][3:]) * -1
+ else:
+ tzhour = _rfc822_tznames[m['tz']]
+ delta = datetime.timedelta(0, 0, 0, 0, tzmin, tzhour)
+
+ # Return the date and timestamp in UTC
+ return (stamp - delta).utctimetuple()
registerDateHandler(_parse_date_rfc822)
+def _parse_date_asctime(dt):
+ """Parse asctime-style dates"""
+ dayname, month, day, remainder = dt.split(None, 3)
+ # Convert month and day into zero-padded integers
+ month = '%02i ' % (_rfc822_months.index(month.lower()) + 1)
+ day = '%02i ' % (int(day),)
+ dt = month + day + remainder
+ return time.strptime(dt, '%m %d %H:%M:%S %Y')[:-1] + (0, )
+registerDateHandler(_parse_date_asctime)
+
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
@@ -3471,211 +3557,283 @@ def _parse_date(dateString):
return date9tuple
return None
-def _getCharacterEncoding(http_headers, xml_data):
- '''Get the character encoding of the XML document
+# Each marker represents some of the characters of the opening XML
+# processing instruction ('
+RE_XML_DECLARATION = re.compile('^<\?xml[^>]*?>')
+
+# Capture the value of the XML processing instruction's encoding attribute.
+# Example:
+RE_XML_PI_ENCODING = re.compile(_s2bytes('^<\?.*encoding=[\'"](.*?)[\'"].*\?>'))
+
+def convert_to_utf8(http_headers, data):
+ '''Detect and convert the character encoding to UTF-8.
http_headers is a dictionary
- xml_data is a raw string (not Unicode)
+ 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.
+ # 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.)
+ # 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).
+ # 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)
- charset = params.get('charset', '').replace("'", "")
- if not isinstance(charset, unicode):
- charset = charset.decode('utf-8', 'ignore')
- return content_type, charset
-
- sniffed_xml_encoding = u''
- xml_encoding = u''
- true_encoding = u''
- 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:
+ # 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). iconv_codec can help a lot;
+ # you should definitely install it if you can.
+ # http://cjkpython.i18n.org/
+
+ bom_encoding = u''
+ xml_encoding = u''
+ rfc3023_encoding = u''
+
+ # Look at the first few bytes of the document to guess what
+ # its encoding may be. We only need to decode enough of the
+ # document that we can use an ASCII-compatible regular
+ # expression to search for an XML encoding declaration.
+ # The heuristic follows the XML specification, section F:
+ # http://www.w3.org/TR/REC-xml/#sec-guessing-no-ext-info
+ # Check for BOMs first.
+ if data[:4] == codecs.BOM_UTF32_BE:
+ bom_encoding = u'utf-32be'
+ data = data[4:]
+ elif data[:4] == codecs.BOM_UTF32_LE:
+ bom_encoding = u'utf-32le'
+ data = data[4:]
+ elif data[:2] == codecs.BOM_UTF16_BE and data[2:4] != ZERO_BYTES:
+ bom_encoding = u'utf-16be'
+ data = data[2:]
+ elif data[:2] == codecs.BOM_UTF16_LE and data[2:4] != ZERO_BYTES:
+ bom_encoding = u'utf-16le'
+ data = data[2:]
+ elif data[:3] == codecs.BOM_UTF8:
+ bom_encoding = u'utf-8'
+ data = data[3:]
+ # Check for the characters '= 4) and (xml_data[:2] == _l2bytes([0xfe, 0xff])) and (xml_data[2:4] != _l2bytes([0x00, 0x00])):
- # UTF-16BE with BOM
- sniffed_xml_encoding = u'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 = u'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 = u'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 = u'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 = u'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 = u'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 = u'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 = u'utf-8'
- xml_data = unicode(xml_data[3:], 'utf-8').encode('utf-8')
- else:
- # ASCII-compatible
- pass
- xml_encoding_match = re.compile(_s2bytes('^<\?.*encoding=[\'"](.*?)[\'"].*\?>')).match(xml_data)
- except UnicodeDecodeError:
+ if bom_encoding:
+ tempdata = data.decode(bom_encoding).encode('utf-8')
+ except (UnicodeDecodeError, LookupError):
+ # feedparser recognizes UTF-32 encodings that aren't
+ # available in Python 2.4 and 2.5, so it's possible to
+ # encounter a LookupError during decoding.
xml_encoding_match = None
+ else:
+ xml_encoding_match = RE_XML_PI_ENCODING.match(tempdata)
+
if xml_encoding_match:
xml_encoding = xml_encoding_match.groups()[0].decode('utf-8').lower()
- if sniffed_xml_encoding and (xml_encoding in (u'iso-10646-ucs-2', u'ucs-2', u'csunicode', u'iso-10646-ucs-4', u'ucs-4', u'csucs4', u'utf-16', u'utf-32', u'utf_16', u'utf_32', u'utf16', u'u16')):
- xml_encoding = sniffed_xml_encoding
+ # Normalize the xml_encoding if necessary.
+ if bom_encoding and (xml_encoding in (
+ u'u16', u'utf-16', u'utf16', u'utf_16',
+ u'u32', u'utf-32', u'utf32', u'utf_32',
+ u'iso-10646-ucs-2', u'iso-10646-ucs-4',
+ u'csucs4', u'csunicode', u'ucs-2', u'ucs-4'
+ )):
+ xml_encoding = bom_encoding
+
+ # Find the HTTP Content-Type and, hopefully, a character
+ # encoding provided by the server. The Content-Type is used
+ # to choose the "correct" encoding among the BOM encoding,
+ # XML declaration encoding, and HTTP encoding, following the
+ # heuristic defined in RFC 3023.
+ http_content_type = http_headers.get('content-type') or ''
+ http_content_type, params = cgi.parse_header(http_content_type)
+ http_encoding = params.get('charset', '').replace("'", "")
+ if not isinstance(http_encoding, unicode):
+ http_encoding = http_encoding.decode('utf-8', 'ignore')
+
acceptable_content_type = 0
- application_content_types = (u'application/xml', u'application/xml-dtd', u'application/xml-external-parsed-entity')
+ application_content_types = (u'application/xml', u'application/xml-dtd',
+ u'application/xml-external-parsed-entity')
text_content_types = (u'text/xml', u'text/xml-external-parsed-entity')
if (http_content_type in application_content_types) or \
- (http_content_type.startswith(u'application/') and http_content_type.endswith(u'+xml')):
+ (http_content_type.startswith(u'application/') and
+ http_content_type.endswith(u'+xml')):
acceptable_content_type = 1
- true_encoding = http_encoding or xml_encoding or u'utf-8'
+ rfc3023_encoding = http_encoding or xml_encoding or u'utf-8'
elif (http_content_type in text_content_types) or \
- (http_content_type.startswith(u'text/')) and http_content_type.endswith(u'+xml'):
+ (http_content_type.startswith(u'text/') and
+ http_content_type.endswith(u'+xml')):
acceptable_content_type = 1
- true_encoding = http_encoding or u'us-ascii'
+ rfc3023_encoding = http_encoding or u'us-ascii'
elif http_content_type.startswith(u'text/'):
- true_encoding = http_encoding or u'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 u'iso-8859-1'
+ rfc3023_encoding = http_encoding or u'us-ascii'
+ elif http_headers and 'content-type' not in http_headers:
+ rfc3023_encoding = xml_encoding or u'iso-8859-1'
else:
- true_encoding = xml_encoding or u'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() == u'gb2312':
- true_encoding = u'gb18030'
- return true_encoding, http_encoding, xml_encoding, sniffed_xml_encoding, acceptable_content_type
+ rfc3023_encoding = xml_encoding or u'utf-8'
+ # gb18030 is a superset of gb2312, so always replace gb2312
+ # with gb18030 for greater compatibility.
+ if rfc3023_encoding.lower() == u'gb2312':
+ rfc3023_encoding = u'gb18030'
+ if xml_encoding.lower() == u'gb2312':
+ xml_encoding = u'gb18030'
-def _toUTF8(data, encoding):
- '''Changes an XML data stream on the fly to specify a new encoding
+ # 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 = 4) and (data[:2] == _l2bytes([0xfe, 0xff])) and (data[2:4] != _l2bytes([0x00, 0x00])):
- encoding = 'utf-16be'
- data = data[2:]
- elif (len(data) >= 4) and (data[:2] == _l2bytes([0xff, 0xfe])) and (data[2:4] != _l2bytes([0x00, 0x00])):
- encoding = 'utf-16le'
- data = data[2:]
- elif data[:3] == _l2bytes([0xef, 0xbb, 0xbf]):
- encoding = 'utf-8'
- data = data[3:]
- elif data[:4] == _l2bytes([0x00, 0x00, 0xfe, 0xff]):
- encoding = 'utf-32be'
- data = data[4:]
- elif data[:4] == _l2bytes([0xff, 0xfe, 0x00, 0x00]):
- encoding = 'utf-32le'
- data = data[4:]
- newdata = unicode(data, 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')
+ if http_headers and (not acceptable_content_type):
+ if 'content-type' in http_headers:
+ msg = '%s is not an XML media type' % http_headers['content-type']
+ else:
+ msg = 'no Content-type specified'
+ error = NonXMLContentType(msg)
-def _stripDoctype(data):
- '''Strips DOCTYPE from XML document, returns (rss_version, stripped_data)
+ # determine character encoding
+ known_encoding = 0
+ chardet_encoding = None
+ tried_encodings = []
+ if chardet:
+ chardet_encoding = unicode(chardet.detect(data)['encoding'] or '', 'ascii', 'ignore')
+ # try: HTTP encoding, declared XML encoding, encoding sniffed from BOM
+ for proposed_encoding in (rfc3023_encoding, xml_encoding, bom_encoding,
+ chardet_encoding, u'utf-8', u'windows-1252', u'iso-8859-2'):
+ if not proposed_encoding:
+ continue
+ if proposed_encoding in tried_encodings:
+ continue
+ tried_encodings.append(proposed_encoding)
+ try:
+ data = data.decode(proposed_encoding)
+ except (UnicodeDecodeError, LookupError):
+ pass
+ else:
+ known_encoding = 1
+ # Update the encoding in the opening XML processing instruction.
+ new_declaration = ''''''
+ if RE_XML_DECLARATION.search(data):
+ data = RE_XML_DECLARATION.sub(new_declaration, data)
+ else:
+ data = new_declaration + u'\n' + data
+ data = data.encode('utf-8')
+ break
+ # if still no luck, give up
+ if not known_encoding:
+ error = CharacterEncodingUnknown(
+ 'document encoding unknown, I tried ' +
+ '%s, %s, utf-8, windows-1252, and iso-8859-2 but nothing worked' %
+ (rfc3023_encoding, xml_encoding))
+ rfc3023_encoding = u''
+ elif proposed_encoding != rfc3023_encoding:
+ error = CharacterEncodingOverride(
+ 'document declared as %s, but parsed as %s' %
+ (rfc3023_encoding, proposed_encoding))
+ rfc3023_encoding = proposed_encoding
+
+ return data, rfc3023_encoding, error
+
+# Match XML entity declarations.
+# Example:
+RE_ENTITY_PATTERN = re.compile(_s2bytes(r'^\s*]*?)>'), re.MULTILINE)
+
+# Match XML DOCTYPE declarations.
+# Example:
+RE_DOCTYPE_PATTERN = re.compile(_s2bytes(r'^\s*]*?)>'), re.MULTILINE)
+
+# Match safe entity declarations.
+# This will allow hexadecimal character references through,
+# as well as text, but not arbitrary nested entities.
+# Example: cubed "³"
+# Example: copyright "(C)"
+# Forbidden: explode1 "&explode2;&explode2;"
+RE_SAFE_ENTITY_PATTERN = re.compile(_s2bytes('\s+(\w+)\s+"(&#\w+;|[^&"]*)"'))
+
+def replace_doctype(data):
+ '''Strips and replaces the DOCTYPE, returns (rss_version, stripped_data)
rss_version may be 'rss091n' or None
- stripped_data is the same XML document, minus the DOCTYPE
+ stripped_data is the same XML document with a replaced DOCTYPE
'''
+
+ # Divide the document into two groups by finding the location
+ # of the first element that doesn't begin with ']*?)>'), 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)
+ # Save and then remove all of the ENTITY declarations.
+ entity_results = RE_ENTITY_PATTERN.findall(head)
+ head = RE_ENTITY_PATTERN.sub(_s2bytes(''), head)
+
+ # Find the DOCTYPE declaration and check the feed type.
+ doctype_results = RE_DOCTYPE_PATTERN.findall(head)
doctype = doctype_results and doctype_results[0] or _s2bytes('')
- if doctype.lower().count(_s2bytes('netscape')):
+ if _s2bytes('netscape') in doctype.lower():
version = u'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
+ # Re-insert the safe ENTITY declarations if a DOCTYPE was found.
+ replacement = _s2bytes('')
+ if len(doctype_results) == 1 and entity_results:
+ match_safe_entities = lambda e: RE_SAFE_ENTITY_PATTERN.match(e)
+ safe_entities = filter(match_safe_entities, entity_results)
+ if safe_entities:
+ replacement = _s2bytes('\n\n]>')
+ data = RE_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)])
+ # Precompute the safe entities for the loose parser.
+ safe_entities = dict((k.decode('utf-8'), v.decode('utf-8'))
+ for k, v in RE_SAFE_ENTITY_PATTERN.findall(replacement))
+ return version, data, safe_entities
def parse(url_file_stream_or_string, etag=None, modified=None, agent=None, referrer=None, handlers=None, request_headers=None, response_headers=None):
'''Parse a feed from a URL, file, stream, or string.
@@ -3714,41 +3872,51 @@ def parse(url_file_stream_or_string, etag=None, modified=None, agent=None, refer
elif response_headers:
result['headers'] = copy.deepcopy(response_headers)
+ # lowercase all of the HTTP headers for comparisons per RFC 2616
+ if 'headers' in result:
+ http_headers = dict((k.lower(), v) for k, v in result['headers'].items())
+ else:
+ http_headers = {}
+
# if feed is gzip-compressed, decompress it
- if f and data and 'headers' in result:
- if gzip and 'gzip' in (result['headers'].get('content-encoding'), result['headers'].get('Content-Encoding')):
+ if f and data and http_headers:
+ if gzip and 'gzip' in http_headers.get('content-encoding', ''):
try:
data = gzip.GzipFile(fileobj=_StringIO(data)).read()
except (IOError, struct.error), e:
- # IOError can occur if the gzip header is bad
- # struct.error can occur if the data is damaged
- # 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.
+ # IOError can occur if the gzip header is bad.
+ # struct.error can occur if the data is damaged.
result['bozo'] = 1
result['bozo_exception'] = e
- data = None
- elif zlib and 'deflate' in (result['headers'].get('content-encoding'), result['headers'].get('Content-Encoding')):
+ if isinstance(e, struct.error):
+ # A gzip header was found but the data is corrupt.
+ # Ideally, we should re-request the feed without the
+ # 'Accept-encoding: gzip' header, but we don't.
+ data = None
+ elif zlib and 'deflate' in http_headers.get('content-encoding', ''):
try:
data = zlib.decompress(data)
except zlib.error, e:
- result['bozo'] = 1
- result['bozo_exception'] = e
- data = None
+ try:
+ # The data may have no headers and no checksum.
+ data = zlib.decompress(data, -15)
+ except zlib.error, e:
+ result['bozo'] = 1
+ result['bozo_exception'] = e
# 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', u''))
+ if http_headers:
+ if 'etag' in http_headers:
+ etag = http_headers.get('etag', u'')
if not isinstance(etag, unicode):
etag = etag.decode('utf-8', 'ignore')
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 'last-modified' in http_headers:
+ modified = http_headers.get('last-modified', u'')
if modified:
- result['modified'] = _parse_date(modified)
+ result['modified'] = modified
+ result['modified_parsed'] = _parse_date(modified)
if hasattr(f, 'url'):
if not isinstance(f.url, unicode):
result['href'] = f.url.decode('utf-8', 'ignore')
@@ -3763,118 +3931,29 @@ def parse(url_file_stream_or_string, etag=None, modified=None, agent=None, refer
if data is None:
return result
- # 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 '
+Copyright (c) 2003, Dalke Scientific Software, LLC
-_generator_name = __name__ + '-' + '.'.join(map(str, __version__))
+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.
+
+ * Neither the name of the Dalke Scientific Softare, LLC, Andrew
+ Dalke, nor the names of its contributors may be used to endorse or
+ promote products derived from this software without specific prior
+ written permission.
+
+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.
+"""
+
+__name__ = "PyRSS2Gen"
+__version__ = (1, 1, 0)
+__author__ = "Andrew Dalke "
+
+_generator_name = __name__ + "-" + ".".join(map(str, __version__))
import datetime
-import cStringIO
+
+import sys
+
+if sys.version_info[0] == 3:
+ # Python 3
+ basestring = str
+ from io import StringIO
+else:
+ # Python 2
+ try:
+ from cStringIO import StringIO
+ except ImportError:
+ # Very old (or memory constrained) systems might
+ # have left out the compiled C version. Fall back
+ # to the pure Python one. Haven't seen this sort
+ # of system since the early 2000s.
+ from StringIO import StringIO
# Could make this the base class; will need to add 'publish'
class WriteXmlMixin:
-
- def write_xml(self, outfile, encoding='iso-8859-1'):
+ def write_xml(self, outfile, encoding="iso-8859-1"):
from xml.sax import saxutils
handler = saxutils.XMLGenerator(outfile, encoding)
handler.startDocument()
self.publish(handler)
handler.endDocument()
- def to_xml(self, encoding='iso-8859-1'):
- try:
- import cStringIO as StringIO
- except ImportError:
- import StringIO
- f = StringIO.StringIO()
+ def to_xml(self, encoding="iso-8859-1"):
+ f = StringIO()
self.write_xml(f, encoding)
return f.getvalue()
-def _element(
- handler,
- name,
- obj,
- d={},
- ):
+def _element(handler, name, obj, d={}):
if isinstance(obj, basestring) or obj is None:
-
# special-case handling to make the API easier
# to use for the common case.
-
handler.startElement(name, d)
if obj is not None:
handler.characters(obj)
handler.endElement(name)
else:
-
# It better know how to emit the correct XML.
-
obj.publish(handler)
@@ -67,7 +104,6 @@ def _format_date(dt):
Input date must be in GMT.
"""
-
# Looks like:
# Sat, 07 Sep 2002 00:00:01 GMT
# Can't use strftime because that's locale dependent
@@ -75,46 +111,18 @@ def _format_date(dt):
# Isn't there a standard way to do this for Python? The
# rfc822 and email.Utils modules assume a timestamp. The
# following is based on the rfc822 module.
-
- return '%s, %02d %s %04d %02d:%02d:%02d GMT' % (
- [
- 'Mon',
- 'Tue',
- 'Wed',
- 'Thu',
- 'Fri',
- 'Sat',
- 'Sun',
- ][dt.weekday()],
+ return "%s, %02d %s %04d %02d:%02d:%02d GMT" % (
+ ["Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun"][dt.weekday()],
dt.day,
- [
- 'Jan',
- 'Feb',
- 'Mar',
- 'Apr',
- 'May',
- 'Jun',
- 'Jul',
- 'Aug',
- 'Sep',
- 'Oct',
- 'Nov',
- 'Dec',
- ][dt.month - 1],
- dt.year,
- dt.hour,
- dt.minute,
- dt.second,
- )
+ ["Jan", "Feb", "Mar", "Apr", "May", "Jun",
+ "Jul", "Aug", "Sep", "Oct", "Nov", "Dec"][dt.month - 1],
+ dt.year, dt.hour, dt.minute, dt.second)
##
# A couple simple wrapper objects for the fields which
# take a simple value other than a string.
-
-
class IntElement:
-
"""implements the 'publish' API for integers
Takes the tag name and the integer value to publish.
@@ -122,7 +130,6 @@ class IntElement:
(Could be used for anything which uses str() to be published
to text for XML.)
"""
-
element_attrs = {}
def __init__(self, name, val):
@@ -136,29 +143,23 @@ class IntElement:
class DateElement:
-
"""implements the 'publish' API for a datetime.datetime
Takes the tag name and the datetime to publish.
Converts the datetime to RFC 2822 timestamp (4-digit year).
"""
-
def __init__(self, name, dt):
self.name = name
self.dt = dt
def publish(self, handler):
_element(handler, self.name, _format_date(self.dt))
-
-
-# ###
+####
class Category:
-
"""Publish a category element"""
-
def __init__(self, category, domain=None):
self.category = category
self.domain = domain
@@ -166,22 +167,14 @@ class Category:
def publish(self, handler):
d = {}
if self.domain is not None:
- d['domain'] = self.domain
- _element(handler, 'category', self.category, d)
+ d["domain"] = self.domain
+ _element(handler, "category", self.category, d)
class Cloud:
-
"""Publish a cloud"""
-
- def __init__(
- self,
- domain,
- port,
- path,
- registerProcedure,
- protocol,
- ):
+ def __init__(self, domain, port, path,
+ registerProcedure, protocol):
self.domain = domain
self.port = port
self.path = path
@@ -189,30 +182,20 @@ class Cloud:
self.protocol = protocol
def publish(self, handler):
- _element(handler, 'cloud', None, {
- 'domain': self.domain,
- 'port': str(self.port),
- 'path': self.path,
- 'registerProcedure': self.registerProcedure,
- 'protocol': self.protocol,
- })
+ _element(handler, "cloud", None, {
+ "domain": self.domain,
+ "port": str(self.port),
+ "path": self.path,
+ "registerProcedure": self.registerProcedure,
+ "protocol": self.protocol})
class Image:
-
"""Publish a channel Image"""
-
element_attrs = {}
- def __init__(
- self,
- url,
- title,
- link,
- width=None,
- height=None,
- description=None,
- ):
+ def __init__(self, url, title, link,
+ width=None, height=None, description=None):
self.url = url
self.title = title
self.link = link
@@ -221,35 +204,33 @@ class Image:
self.description = description
def publish(self, handler):
- handler.startElement('image', self.element_attrs)
+ handler.startElement("image", self.element_attrs)
- _element(handler, 'url', self.url)
- _element(handler, 'title', self.title)
- _element(handler, 'link', self.link)
+ _element(handler, "url", self.url)
+ _element(handler, "title", self.title)
+ _element(handler, "link", self.link)
width = self.width
if isinstance(width, int):
- width = IntElement('width', width)
- _opt_element(handler, 'width', width)
+ width = IntElement("width", width)
+ _opt_element(handler, "width", width)
height = self.height
if isinstance(height, int):
- height = IntElement('height', height)
- _opt_element(handler, 'height', height)
+ height = IntElement("height", height)
+ _opt_element(handler, "height", height)
- _opt_element(handler, 'description', self.description)
+ _opt_element(handler, "description", self.description)
- handler.endElement('image')
+ handler.endElement("image")
class Guid:
-
"""Publish a guid
Defaults to being a permalink, which is the assumption if it's
omitted. Hence strings are always permalinks.
"""
-
def __init__(self, guid, isPermaLink=1):
self.guid = guid
self.isPermaLink = isPermaLink
@@ -257,80 +238,64 @@ class Guid:
def publish(self, handler):
d = {}
if self.isPermaLink:
- d['isPermaLink'] = 'true'
+ d["isPermaLink"] = "true"
else:
- d['isPermaLink'] = 'false'
- _element(handler, 'guid', self.guid, d)
+ d["isPermaLink"] = "false"
+ _element(handler, "guid", self.guid, d)
class TextInput:
-
"""Publish a textInput
Apparently this is rarely used.
"""
-
element_attrs = {}
- def __init__(
- self,
- title,
- description,
- name,
- link,
- ):
+ def __init__(self, title, description, name, link):
self.title = title
self.description = description
self.name = name
self.link = link
def publish(self, handler):
- handler.startElement('textInput', self.element_attrs)
- _element(handler, 'title', self.title)
- _element(handler, 'description', self.description)
- _element(handler, 'name', self.name)
- _element(handler, 'link', self.link)
- handler.endElement('textInput')
+ handler.startElement("textInput", self.element_attrs)
+ _element(handler, "title", self.title)
+ _element(handler, "description", self.description)
+ _element(handler, "name", self.name)
+ _element(handler, "link", self.link)
+ handler.endElement("textInput")
class Enclosure:
-
"""Publish an enclosure"""
-
- def __init__(
- self,
- url,
- length,
- type,
- ):
+ def __init__(self, url, length, type):
self.url = url
self.length = length
self.type = type
def publish(self, handler):
- _element(handler, 'enclosure', None,
- {'url': self.url, 'length': str(self.length), 'type': self.type})
+ _element(handler, "enclosure", None,
+ {"url": self.url,
+ "length": str(self.length),
+ "type": self.type,
+ })
class Source:
-
"""Publish the item's original source, used by aggregators"""
-
def __init__(self, name, url):
self.name = name
self.url = url
def publish(self, handler):
- _element(handler, 'source', self.name, {'url': self.url})
+ _element(handler, "source", self.name, {"url": self.url})
class SkipHours:
-
"""Publish the skipHours
This takes a list of hours, as integers.
"""
-
element_attrs = {}
def __init__(self, hours):
@@ -338,19 +303,17 @@ class SkipHours:
def publish(self, handler):
if self.hours:
- handler.startElement('skipHours', self.element_attrs)
+ handler.startElement("skipHours", self.element_attrs)
for hour in self.hours:
- _element(handler, 'hour', str(hour))
- handler.endElement('skipHours')
+ _element(handler, "hour", str(hour))
+ handler.endElement("skipHours")
class SkipDays:
-
"""Publish the skipDays
This takes a list of days as strings.
"""
-
element_attrs = {}
def __init__(self, days):
@@ -358,47 +321,48 @@ class SkipDays:
def publish(self, handler):
if self.days:
- handler.startElement('skipDays', self.element_attrs)
+ handler.startElement("skipDays", self.element_attrs)
for day in self.days:
- _element(handler, 'day', day)
- handler.endElement('skipDays')
+ _element(handler, "day", day)
+ handler.endElement("skipDays")
class RSS2(WriteXmlMixin):
-
"""The main RSS class.
- Stores the channel attributes, with the \"category\" elements under
- \".categories\" and the RSS items under \".items\".
+ Stores the channel attributes, with the "category" elements under
+ ".categories" and the RSS items under ".items".
"""
- rss_attrs = {'version': '2.0'}
+ rss_attrs = {"version": "2.0"}
element_attrs = {}
- def __init__(
- self,
- title,
- link,
- description,
- language=None,
- copyright=None,
- managingEditor=None,
- webMaster=None,
- pubDate=None,
- lastBuildDate=None,
- categories=None,
- generator=_generator_name,
- docs='http://blogs.law.harvard.edu/tech/rss',
- cloud=None,
- ttl=None,
- image=None,
- rating=None,
- textInput=None,
- skipHours=None,
- skipDays=None,
- items=None,
- ):
+ def __init__(self,
+ title,
+ link,
+ description,
+ language=None,
+ copyright=None,
+ managingEditor=None,
+ webMaster=None,
+ pubDate=None, # a datetime, *in* *GMT*
+ lastBuildDate=None, # a datetime
+
+ categories=None, # list of strings or Category
+ generator=_generator_name,
+ docs="http://blogs.law.harvard.edu/tech/rss",
+ cloud=None, # a Cloud
+ ttl=None, # integer number of minutes
+
+ image=None, # an Image
+ rating=None, # a string; I don't know how it's used
+ textInput=None, # a TextInput
+ skipHours=None, # a SkipHours with a list of integers
+ skipDays=None, # a SkipDays with a list of strings
+
+ items=None, # list of RSSItems
+ ):
self.title = title
self.link = link
self.description = description
@@ -428,49 +392,49 @@ class RSS2(WriteXmlMixin):
self.items = items
def publish(self, handler):
- handler.startElement('rss', self.rss_attrs)
- handler.startElement('channel', self.element_attrs)
- _element(handler, 'title', self.title)
- _element(handler, 'link', self.link)
- _element(handler, 'description', self.description)
+ handler.startElement("rss", self.rss_attrs)
+ handler.startElement("channel", self.element_attrs)
+ _element(handler, "title", self.title)
+ _element(handler, "link", self.link)
+ _element(handler, "description", self.description)
self.publish_extensions(handler)
- _opt_element(handler, 'language', self.language)
- _opt_element(handler, 'copyright', self.copyright)
- _opt_element(handler, 'managingEditor', self.managingEditor)
- _opt_element(handler, 'webMaster', self.webMaster)
+ _opt_element(handler, "language", self.language)
+ _opt_element(handler, "copyright", self.copyright)
+ _opt_element(handler, "managingEditor", self.managingEditor)
+ _opt_element(handler, "webMaster", self.webMaster)
pubDate = self.pubDate
if isinstance(pubDate, datetime.datetime):
- pubDate = DateElement('pubDate', pubDate)
- _opt_element(handler, 'pubDate', pubDate)
+ pubDate = DateElement("pubDate", pubDate)
+ _opt_element(handler, "pubDate", pubDate)
lastBuildDate = self.lastBuildDate
if isinstance(lastBuildDate, datetime.datetime):
- lastBuildDate = DateElement('lastBuildDate', lastBuildDate)
- _opt_element(handler, 'lastBuildDate', lastBuildDate)
+ lastBuildDate = DateElement("lastBuildDate", lastBuildDate)
+ _opt_element(handler, "lastBuildDate", lastBuildDate)
for category in self.categories:
if isinstance(category, basestring):
category = Category(category)
category.publish(handler)
- _opt_element(handler, 'generator', self.generator)
- _opt_element(handler, 'docs', self.docs)
+ _opt_element(handler, "generator", self.generator)
+ _opt_element(handler, "docs", self.docs)
if self.cloud is not None:
self.cloud.publish(handler)
ttl = self.ttl
if isinstance(self.ttl, int):
- ttl = IntElement('ttl', ttl)
- _opt_element(handler, 'tt', ttl)
+ ttl = IntElement("ttl", ttl)
+ _opt_element(handler, "ttl", ttl)
if self.image is not None:
self.image.publish(handler)
- _opt_element(handler, 'rating', self.rating)
+ _opt_element(handler, "rating", self.rating)
if self.textInput is not None:
self.textInput.publish(handler)
if self.skipHours is not None:
@@ -481,40 +445,35 @@ class RSS2(WriteXmlMixin):
for item in self.items:
item.publish(handler)
- handler.endElement('channel')
- handler.endElement('rss')
+ handler.endElement("channel")
+ handler.endElement("rss")
def publish_extensions(self, handler):
-
# Derived classes can hook into this to insert
# output after the three required fields.
-
pass
class RSSItem(WriteXmlMixin):
-
"""Publish an RSS Item"""
-
element_attrs = {}
- def __init__(
- self,
- title=None,
- link=None,
- description=None,
- author=None,
- categories=None,
- comments=None,
- enclosure=None,
- guid=None,
- pubDate=None,
- source=None,
- ):
+ def __init__(self,
+ title=None, # string
+ link=None, # url as string
+ description=None, # string
+ author=None, # email address as string
+ categories=None, # list of string or Category
+ comments=None, # url as string
+ enclosure=None, # an Enclosure
+ guid=None, # a unique string
+ pubDate=None, # a datetime
+ source=None, # a Source
+ ):
if title is None and description is None:
raise TypeError(
- "RSSItem must define at least one of 'title' or 'description'")
+ "must define at least one of 'title' or 'description'")
self.title = title
self.link = link
self.description = description
@@ -527,62 +486,37 @@ class RSSItem(WriteXmlMixin):
self.guid = guid
self.pubDate = pubDate
self.source = source
-
# It sure does get tedious typing these names three times...
def publish(self, handler):
- handler.startElement('item', self.element_attrs)
- _opt_element(handler, 'title', self.title)
- _opt_element(handler, 'link', self.link)
+ handler.startElement("item", self.element_attrs)
+ _opt_element(handler, "title", self.title)
+ _opt_element(handler, "link", self.link)
self.publish_extensions(handler)
- _opt_element(handler, 'description', self.description)
- _opt_element(handler, 'author', self.author)
+ _opt_element(handler, "description", self.description)
+ _opt_element(handler, "author", self.author)
for category in self.categories:
if isinstance(category, basestring):
category = Category(category)
category.publish(handler)
- _opt_element(handler, 'comments', self.comments)
+ _opt_element(handler, "comments", self.comments)
if self.enclosure is not None:
self.enclosure.publish(handler)
- _opt_element(handler, 'guid', self.guid)
+ _opt_element(handler, "guid", self.guid)
pubDate = self.pubDate
if isinstance(pubDate, datetime.datetime):
- pubDate = DateElement('pubDate', pubDate)
- _opt_element(handler, 'pubDate', pubDate)
+ pubDate = DateElement("pubDate", pubDate)
+ _opt_element(handler, "pubDate", pubDate)
if self.source is not None:
self.source.publish(handler)
- handler.endElement('item')
+ handler.endElement("item")
def publish_extensions(self, handler):
-
# Derived classes can hook into this to insert
# output after the title and link elements
-
pass
-
-
-def dumps(rss, encoding='utf-8'):
- s = cStringIO.StringIO()
- rss.write_xml(s, encoding)
- return s.getvalue()
-
-
-def test():
- rss = RSS2(title='web2py feed', link='http://www.web2py.com',
- description='About web2py',
- lastBuildDate=datetime.datetime.now(),
- items=[RSSItem(title='web2py and PyRSS2Gen-0.0',
- link='http://www.web2py.com/examples/simple_examples/getrss',
- description='web2py can now make rss feeds!',
- guid=Guid('http://www.web2py.com/'),
- pubDate=datetime.datetime(2007, 11, 14, 10, 30))])
- return dumps(rss)
-
-
-if __name__ == '__main__':
- print test()