From: Lindsey Smith Date: Thu, 17 Feb 2011 21:06:03 +0000 (-0800) Subject: Initial revision from v2.70 and feedparser 5.0 X-Git-Tag: v3.0~74^2~10 X-Git-Url: http://git.tremily.us/?a=commitdiff_plain;h=ea347eb190d4e74f03e2d951f380e9325df60da0;p=rss2email.git Initial revision from v2.70 and feedparser 5.0 --- ea347eb190d4e74f03e2d951f380e9325df60da0 diff --git a/.gitignore b/.gitignore new file mode 100644 index 0000000..1088d61 --- /dev/null +++ b/.gitignore @@ -0,0 +1,3 @@ +config.py +temp/* +Old/* diff --git a/CHANGELOG b/CHANGELOG new file mode 100644 index 0000000..9920349 --- /dev/null +++ b/CHANGELOG @@ -0,0 +1,122 @@ +v2.70 (2010-12-21) +* Improved handling of given feed email addresses to prevent mail servers rejecting poorly formed Froms +* Added X-RSS-TAGS header that lists any tags provided by an entry, which will be helpful in filtering incoming messages + +v2.69 (2010-11-12) + * Added support for connecting to SMTP server via SSL, see SMTP_SSL option + * Improved backwards compatibility by fixing issue with listing feeds when run with older Python versions + * Added selective feed email overrides through OVERRIDE_EMAIL and DEFAULT_EMAIL options + * Added NO_FRIENDLY_NAME to from from address only without the friendly name + * Added X-RSS-URL header in each message with the link to the original item + +v2.68 (2010-10-01) + * Added ability to pause/resume checking of individual feeds through pause and unpause commands + * Added ability to import and export OPML feed lists through importopml and exportopml commands + +v2.67 (2010-09-21) + * Fixed entries that include an id which is blank (i.e., an empty string) were being resent + * Fixed some entries not being sent by email because they had bad From headers + * Fixed From headers with HTML entities encoded twice + * Compatibility changes to support most recent development versions of feedparser + * Compatibility changes to support Google Reader feeds + +v2.66 (2009-12-21) + + * Complete packaging of all necessary source files (rss2email, html2text, feedparser, r2e, etc.) into one bundle + o Included a more complete config.py with all options + o Default to HTML mail and CSS results + * Added 'reset' command to erase history of already seen entries + * Changed project email to 'lindsey@allthingsrss.com' and project homepage to 'http://www.allthingsrss.com/rss2email/' + * Made exception and error output text more useful + * Added X-RSS-Feed and X-RSS-ID headers to each email for easier filtering + * Improved enclosure handling + * Fixed MacOS compatibility issues + +v2.65 (2009-01-05) + + * Fixed warnings caused by Python v2.6 (using hashlib, removing mimify, etc.) + * Deprecated QP_REQUIRED option as this is more than likely no longer needed and part of what triggered Python warnings + * Fixed unicode errors in certain post headers + * Attempted to incorporate Debian/Ubuntu patches into the mainstream release + * Support img type enclosures + * No file locking for SunOS + +v2.64 (2008-10-21) + + * Bug-fix version + o Gracefully handle missing charsets + o Friendlier and more useful message if sendmail isn't installed + o SunOS locking fix + +v2.63 (2008-06-13) + + * Bug-fix version and license change: + o Licensed under GPL 2 & 3 now + o Display feed number in warning and error message lines + o Fix for unicode handling problem with certain entry titles + +v2.62 (2008-01-14) + + * Bug-fix version: + o Simplified SunOS fix + o Local feeds (/home/user/file.xml) should work + +v2.61 (2007-12-07) + + * Bug-fix version: + o Now really compatible with SunOS + o Don't wrap long subject headers + o New parameter CHARSET_LIST to override or supplement the order in which charsets are tried against an entry + o Don't use blank content to generate id + o Using GMail as mail server should work + +v2.60 (2006-08-25) + + * Small bug-fix version: + o Now compatible with SunOS + o Correctly handle international character sets in email From + +v2.59 (2006-06-09) + + * Finally added oft-requested support for enclosures. Any enclosures, such as a podcast MP3, will be listed under the entry URL + * Made feed timeout compatible with Python versions 2.2 and higher, instead of v2.4 only + * Added optional, configurable CSS styling to HTML mail. Set USE_CSS_STYLING=1 in your config.py to enable this. If you want to tweak the look, modify STYLE_SHEET. + * Improved empty feed checking + * Improved invalid feed messages + * Unfortunately, rss2email is no longer compatible with Python v2.1. Two of the most serious lingering issues with rss2email were waiting forever for non-responsive feeds and its inablility to properly handle feeds with international characters. To properly fix these once and for all, rss2email now depends on functionality that was not available until Python v2.2. Hopefully this does not unduly inconvenience anyone that has not yet upgraded to a more current version of Python. + +v2.58 (2006-05-11) + + * Total rewrite of email code that should fix encoding problems + * Added configurable timeout for nonresponsive feeds + * Fixed incorrectly using text summary_detail instead of html content + * Fixed bug with deleting feed 0 if no default email was set + * Print name of feed that is being deleted + +v2.57 (2006-04-07) + + * Integrated Joey Hess's patches + o First, a patch that makes delete more reliable, so it no longer allows you to remove the default email address ('feed' 0) and thereby hose your feed file, or 'remove' entries that don't exist without warning; and so it only says IDs have changed when they really have. Originally from http://bugs.debian.org/313101 + o Next a patch that avoids a backtrace if there's no email address defined, and outputs a less scary error message. + o Next, a simple change to the usage; since the "email" subcommand always needs a parameter, don't mark it as optional. + o And, avoid a backtrace if the email subcommand does get run w/o a parameter. + o And also avoid backtraces if delete is run w/o a parameter. Also adds support for --help. + o Simple change, make a comment match reality (/usr/sbin/sendmail) + o This avoids another backtrace, this time if there's no feed file yet. [load()] + o Add a handler for the AttributeError exception, which feedparser can throw. Beats crashing.. + o Next, four hunks that make it more robust if no default email address is set and feeds are added w/o an email address. This patch originally comes from http://bugs.debian.org/310485 which has some examples. + o Finally, this works around a bug in mimify that causes it to add a newline to the subject header if it contains very long words. Details at http://bugs.debian.org/320185. Note that Tatsuya Kinoshita has a larger patch torard the end of that bug report that deals with some other problems in this area, Aaron has seen that patch before and said it "looks pretty reasonable". + * add() catches error case on first feed add and no email address is set + * Made "emailaddress" consistent param label throughout + * Error message improvements + * Deleted problematic "if title" line + * Deleted space in front of SMTP_USER + * Only logs into SMTP server once + * Added exception handling around SMTP server connect and login attempt + * Broke contributors across multiple lines + +v2.56 (2006-04-04) + + * SMTP AUTH support added + * Windows support + * Fixed bug with HTML in titles diff --git a/config.py.example b/config.py.example new file mode 100644 index 0000000..586ac5d --- /dev/null +++ b/config.py.example @@ -0,0 +1,94 @@ +### Options for configuring rss2email ### + +# The email address messages are from by default: +DEFAULT_FROM = "bozo@dev.null.invalid" + +# 1: Send text/html messages when possible. +# 0: Convert HTML to plain text. +HTML_MAIL = 1 + +# 1: Only use the DEFAULT_FROM address. +# 0: Use the email address specified by the feed, when possible. +FORCE_FROM = 0 + +# 1: Receive one email per post. +# 0: Receive an email every time a post changes. +TRUST_GUID = 1 + +# 1: Generate Date header based on item's date, when possible. +# 0: Generate Date header based on time sent. +DATE_HEADER = 1 + +# A tuple consisting of some combination of +# ('issued', 'created', 'modified', 'expired') +# expressing ordered list of preference in dates +# to use for the Date header of the email. +DATE_HEADER_ORDER = ('modified', 'issued', 'created') + +# 1: Apply Q-P conversion (required for some MUAs). +# 0: Send message in 8-bits. +# http://cr.yp.to/smtp/8bitmime.html +#DEPRECATED +QP_REQUIRED = 0 +#DEPRECATED + +# 1: Name feeds as they're being processed. +# 0: Keep quiet. +VERBOSE = 0 + +# 1: Use the publisher's email if you can't find the author's. +# 0: Just use the DEFAULT_FROM email instead. +USE_PUBLISHER_EMAIL = 0 + +# 1: Use SMTP_SERVER to send mail. +# 0: Call /usr/sbin/sendmail to send mail. +SMTP_SEND = 1 + +SMTP_SERVER = "smtp.yourisp.net:25" +AUTHREQUIRED = 0 # if you need to use SMTP AUTH set to 1 +SMTP_USER = 'username' # for SMTP AUTH, set SMTP username here +SMTP_PASS = 'password' # for SMTP AUTH, set SMTP password here + +# Connect to the SMTP server using SSL + +SMTP_SSL = 0 + + + +# Set this to add a bonus header to all emails (start with '\n'). +BONUS_HEADER = '' +# Example: BONUS_HEADER = '\nApproved: joe@bob.org' + +# Set this to override From addresses. Keys are feed URLs, values are new titles. +OVERRIDE_FROM = {} + +# Set this to override From email addresses. Keys are feed URLs, values are new emails. + +OVERRIDE_EMAIL = {} + + + +# Set this to default From email addresses. Keys are feed URLs, values are new email addresses. + +DEFAULT_EMAIL = {} + + +# Only use the email from address rather than friendly name plus email address + +NO_FRIENDLY_NAME = 0 + + + +# Set this to override the timeout (in seconds) for feed server response +FEED_TIMEOUT = 60 + +# Optional CSS styling +USE_CSS_STYLING = 1 +STYLE_SHEET='h1 {font: 18pt Georgia, "Times New Roman";} body {font: 12pt Arial;} a:link {font: 12pt Arial; font-weight: bold; color: #0000cc} blockquote {font-family: monospace; } .header { background: #e0ecff; border-bottom: solid 4px #c3d9ff; padding: 5px; margin-top: 0px; color: red;} .header a { font-size: 20px; text-decoration: none; } .footer { background: #c3d9ff; border-top: solid 4px #c3d9ff; padding: 5px; margin-bottom: 0px; } #entry {border: solid 4px #c3d9ff; } #body { margin-left: 5px; margin-right: 5px; }' + +# If you have an HTTP Proxy set this in the format 'http://your.proxy.here:8080/' +PROXY="" + +# To most correctly encode emails with international characters, we iterate through the list below and use the first character set that works +# Eventually (and theoretically) ISO-8859-1 and UTF-8 are our catch-all failsafes +CHARSET_LIST='US-ASCII', 'BIG5', 'ISO-2022-JP', 'ISO-8859-1', 'UTF-8' diff --git a/feedparser.py b/feedparser.py new file mode 100644 index 0000000..0760cc2 --- /dev/null +++ b/feedparser.py @@ -0,0 +1,3688 @@ +#!/usr/bin/env python +"""Universal feed parser + +Handles RSS 0.9x, RSS 1.0, RSS 2.0, CDF, Atom 0.3, and Atom 1.0 feeds + +Visit http://feedparser.org/ for the latest version +Visit http://feedparser.org/docs/ for the latest documentation + +Required: Python 2.1 or later +Recommended: Python 2.3 or later +Recommended: CJKCodecs and iconv_codec +""" + +__version__ = "4.2-pre-" + "$Revision$"[11:14] + "-svn" +__license__ = """Copyright (c) 2002-2008, Mark Pilgrim, All rights reserved. + +Redistribution and use in source and binary forms, with or without modification, +are permitted provided that the following conditions are met: + +* Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS 'AS IS' +AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE +IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE +ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE +LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR +CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF +SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS +INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN +CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) +ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE +POSSIBILITY OF SUCH DAMAGE.""" +__author__ = "Mark Pilgrim " +__contributors__ = ["Jason Diamond ", + "John Beimler ", + "Fazal Majid ", + "Aaron Swartz ", + "Kevin Marks ", + "Sam Ruby ", + "Ade Oshineye "] +_debug = 0 + +# HTTP "User-Agent" header to send to servers when downloading feeds. +# If you are embedding feedparser in a larger application, you should +# change this to your application name and URL. +USER_AGENT = "UniversalFeedParser/%s +http://feedparser.org/" % __version__ + +# HTTP "Accept" header to send to servers when downloading feeds. If you don't +# want to send an Accept header, set this to None. +ACCEPT_HEADER = "application/atom+xml,application/rdf+xml,application/rss+xml,application/x-netcdf,application/xml;q=0.9,text/xml;q=0.2,*/*;q=0.1" + +# List of preferred XML parsers, by SAX driver name. These will be tried first, +# but if they're not installed, Python will keep searching through its own list +# of pre-installed parsers until it finds one that supports everything we need. +PREFERRED_XML_PARSERS = ["drv_libxml2"] + +# If you want feedparser to automatically run HTML markup through HTML Tidy, set +# this to 1. Requires mxTidy +# or utidylib . +TIDY_MARKUP = 0 + +# List of Python interfaces for HTML Tidy, in order of preference. Only useful +# if TIDY_MARKUP = 1 +PREFERRED_TIDY_INTERFACES = ["uTidy", "mxTidy"] + +# If you want feedparser to automatically resolve all relative URIs, set this +# to 1. +RESOLVE_RELATIVE_URIS = 1 + +# If you want feedparser to automatically sanitize all potentially unsafe +# HTML content, set this to 1. +SANITIZE_HTML = 1 + +# ---------- required modules (should come with any Python distribution) ---------- +import sgmllib, re, sys, copy, urlparse, time, rfc822, types, cgi, urllib, urllib2 +try: + from cStringIO import StringIO as _StringIO +except: + from StringIO import StringIO as _StringIO + +# ---------- optional modules (feedparser will work without these, but with reduced functionality) ---------- + +# gzip is included with most Python distributions, but may not be available if you compiled your own +try: + import gzip +except: + gzip = None +try: + import zlib +except: + zlib = None + +# If a real XML parser is available, feedparser will attempt to use it. feedparser has +# been tested with the built-in SAX parser, PyXML, and libxml2. On platforms where the +# Python distribution does not come with an XML parser (such as Mac OS X 10.2 and some +# versions of FreeBSD), feedparser will quietly fall back on regex-based parsing. +try: + import xml.sax + xml.sax.make_parser(PREFERRED_XML_PARSERS) # test for valid parsers + from xml.sax.saxutils import escape as _xmlescape + _XML_AVAILABLE = 1 +except: + _XML_AVAILABLE = 0 + def _xmlescape(data,entities={}): + data = data.replace('&', '&') + data = data.replace('>', '>') + data = data.replace('<', '<') + for char, entity in entities: + data = data.replace(char, entity) + return data + +# base64 support for Atom feeds that contain embedded binary data +try: + import base64, binascii +except: + base64 = binascii = None + +# cjkcodecs and iconv_codec provide support for more character encodings. +# Both are available from http://cjkpython.i18n.org/ +try: + import cjkcodecs.aliases +except: + pass +try: + import iconv_codec +except: + pass + +# chardet library auto-detects character encodings +# Download from http://chardet.feedparser.org/ +try: + import chardet + if _debug: + import chardet.constants + chardet.constants._debug = 1 +except: + chardet = None + +# reversable htmlentitydefs mappings for Python 2.2 +try: + from htmlentitydefs import name2codepoint, codepoint2name +except: + import htmlentitydefs + name2codepoint={} + codepoint2name={} + for (name,codepoint) in htmlentitydefs.entitydefs.iteritems(): + if codepoint.startswith('&#'): codepoint=unichr(int(codepoint[2:-1])) + name2codepoint[name]=ord(codepoint) + codepoint2name[ord(codepoint)]=name + +# BeautifulSoup parser used for parsing microformats from embedded HTML content +# http://www.crummy.com/software/BeautifulSoup/ +# feedparser is tested with BeautifulSoup 3.0.x, but it might work with the +# older 2.x series. If it doesn't, and you can figure out why, I'll accept a +# patch and modify the compatibility statement accordingly. +try: + import BeautifulSoup +except: + BeautifulSoup = None + +# ---------- don't touch these ---------- +class ThingsNobodyCaresAboutButMe(Exception): pass +class CharacterEncodingOverride(ThingsNobodyCaresAboutButMe): pass +class CharacterEncodingUnknown(ThingsNobodyCaresAboutButMe): pass +class NonXMLContentType(ThingsNobodyCaresAboutButMe): pass +class UndeclaredNamespace(Exception): pass + +sgmllib.tagfind = re.compile('[a-zA-Z][-_.:a-zA-Z0-9]*') +sgmllib.special = re.compile(']|"[^"]*"(?=>|/|\s|\w+=)|'[^']*'(?=>|/|\s|\w+=))*(?=[<>])|.*?(?=[<>])''') + def search(self,string,index=0): + self.match = self.endbracket.match(string,index) + if self.match: return self + def start(self,n): + return self.match.end(n) + sgmllib.endbracket = EndBracketMatch() + +SUPPORTED_VERSIONS = {'': 'unknown', + 'rss090': 'RSS 0.90', + 'rss091n': 'RSS 0.91 (Netscape)', + 'rss091u': 'RSS 0.91 (Userland)', + 'rss092': 'RSS 0.92', + 'rss093': 'RSS 0.93', + 'rss094': 'RSS 0.94', + 'rss20': 'RSS 2.0', + 'rss10': 'RSS 1.0', + 'rss': 'RSS (unknown version)', + 'atom01': 'Atom 0.1', + 'atom02': 'Atom 0.2', + 'atom03': 'Atom 0.3', + 'atom10': 'Atom 1.0', + 'atom': 'Atom (unknown version)', + 'cdf': 'CDF', + 'hotrss': 'Hot RSS' + } + +try: + UserDict = dict +except NameError: + # Python 2.1 does not have dict + from UserDict import UserDict + def dict(aList): + rc = {} + for k, v in aList: + rc[k] = v + return rc + +class FeedParserDict(UserDict): + keymap = {'channel': 'feed', + 'items': 'entries', + 'guid': 'id', + 'date': 'updated', + 'date_parsed': 'updated_parsed', + 'description': ['subtitle', 'summary'], + 'url': ['href'], + 'modified': 'updated', + 'modified_parsed': 'updated_parsed', + 'issued': 'published', + 'issued_parsed': 'published_parsed', + 'copyright': 'rights', + 'copyright_detail': 'rights_detail', + 'tagline': 'subtitle', + 'tagline_detail': 'subtitle_detail'} + def __getitem__(self, key): + if key == 'category': + return UserDict.__getitem__(self, 'tags')[0]['term'] + if key == 'enclosures': + norel = lambda link: FeedParserDict([(name,value) for (name,value) in link.items() if name!='rel']) + return [norel(link) for link in UserDict.__getitem__(self, 'links') if link['rel']=='enclosure'] + if key == 'license': + for link in UserDict.__getitem__(self, 'links'): + if link['rel']=='license' and link.has_key('href'): + return link['href'] + if key == 'categories': + return [(tag['scheme'], tag['term']) for tag in UserDict.__getitem__(self, 'tags')] + realkey = self.keymap.get(key, key) + if type(realkey) == types.ListType: + for k in realkey: + if UserDict.has_key(self, k): + return UserDict.__getitem__(self, k) + if UserDict.has_key(self, key): + return UserDict.__getitem__(self, key) + return UserDict.__getitem__(self, realkey) + + def __setitem__(self, key, value): + for k in self.keymap.keys(): + if key == k: + key = self.keymap[k] + if type(key) == types.ListType: + key = key[0] + return UserDict.__setitem__(self, key, value) + + def get(self, key, default=None): + if self.has_key(key): + return self[key] + else: + return default + + def setdefault(self, key, value): + if not self.has_key(key): + self[key] = value + return self[key] + + def has_key(self, key): + try: + return hasattr(self, key) or UserDict.has_key(self, key) + except AttributeError: + return False + + def __getattr__(self, key): + try: + return self.__dict__[key] + except KeyError: + pass + try: + assert not key.startswith('_') + return self.__getitem__(key) + except: + raise AttributeError, "object has no attribute '%s'" % key + + def __setattr__(self, key, value): + if key.startswith('_') or key == 'data': + self.__dict__[key] = value + else: + return self.__setitem__(key, value) + + def __contains__(self, key): + return self.has_key(key) + +def zopeCompatibilityHack(): + global FeedParserDict + del FeedParserDict + def FeedParserDict(aDict=None): + rc = {} + if aDict: + rc.update(aDict) + return rc + +_ebcdic_to_ascii_map = None +def _ebcdic_to_ascii(s): + global _ebcdic_to_ascii_map + if not _ebcdic_to_ascii_map: + emap = ( + 0,1,2,3,156,9,134,127,151,141,142,11,12,13,14,15, + 16,17,18,19,157,133,8,135,24,25,146,143,28,29,30,31, + 128,129,130,131,132,10,23,27,136,137,138,139,140,5,6,7, + 144,145,22,147,148,149,150,4,152,153,154,155,20,21,158,26, + 32,160,161,162,163,164,165,166,167,168,91,46,60,40,43,33, + 38,169,170,171,172,173,174,175,176,177,93,36,42,41,59,94, + 45,47,178,179,180,181,182,183,184,185,124,44,37,95,62,63, + 186,187,188,189,190,191,192,193,194,96,58,35,64,39,61,34, + 195,97,98,99,100,101,102,103,104,105,196,197,198,199,200,201, + 202,106,107,108,109,110,111,112,113,114,203,204,205,206,207,208, + 209,126,115,116,117,118,119,120,121,122,210,211,212,213,214,215, + 216,217,218,219,220,221,222,223,224,225,226,227,228,229,230,231, + 123,65,66,67,68,69,70,71,72,73,232,233,234,235,236,237, + 125,74,75,76,77,78,79,80,81,82,238,239,240,241,242,243, + 92,159,83,84,85,86,87,88,89,90,244,245,246,247,248,249, + 48,49,50,51,52,53,54,55,56,57,250,251,252,253,254,255 + ) + import string + _ebcdic_to_ascii_map = string.maketrans( \ + ''.join(map(chr, range(256))), ''.join(map(chr, emap))) + return s.translate(_ebcdic_to_ascii_map) + +_cp1252 = { + unichr(128): unichr(8364), # euro sign + unichr(130): unichr(8218), # single low-9 quotation mark + unichr(131): unichr( 402), # latin small letter f with hook + unichr(132): unichr(8222), # double low-9 quotation mark + unichr(133): unichr(8230), # horizontal ellipsis + unichr(134): unichr(8224), # dagger + unichr(135): unichr(8225), # double dagger + unichr(136): unichr( 710), # modifier letter circumflex accent + unichr(137): unichr(8240), # per mille sign + unichr(138): unichr( 352), # latin capital letter s with caron + unichr(139): unichr(8249), # single left-pointing angle quotation mark + unichr(140): unichr( 338), # latin capital ligature oe + unichr(142): unichr( 381), # latin capital letter z with caron + unichr(145): unichr(8216), # left single quotation mark + unichr(146): unichr(8217), # right single quotation mark + unichr(147): unichr(8220), # left double quotation mark + unichr(148): unichr(8221), # right double quotation mark + unichr(149): unichr(8226), # bullet + unichr(150): unichr(8211), # en dash + unichr(151): unichr(8212), # em dash + unichr(152): unichr( 732), # small tilde + unichr(153): unichr(8482), # trade mark sign + unichr(154): unichr( 353), # latin small letter s with caron + unichr(155): unichr(8250), # single right-pointing angle quotation mark + unichr(156): unichr( 339), # latin small ligature oe + unichr(158): unichr( 382), # latin small letter z with caron + unichr(159): unichr( 376)} # latin capital letter y with diaeresis + +_urifixer = re.compile('^([A-Za-z][A-Za-z0-9+-.]*://)(/*)(.*?)') +def _urljoin(base, uri): + uri = _urifixer.sub(r'\1\3', uri) + try: + return urlparse.urljoin(base, uri) + except: + uri = urlparse.urlunparse([urllib.quote(part) for part in urlparse.urlparse(uri)]) + return urlparse.urljoin(base, uri) + +class _FeedParserMixin: + namespaces = {'': '', + 'http://backend.userland.com/rss': '', + 'http://blogs.law.harvard.edu/tech/rss': '', + 'http://purl.org/rss/1.0/': '', + 'http://my.netscape.com/rdf/simple/0.9/': '', + 'http://example.com/newformat#': '', + 'http://example.com/necho': '', + 'http://purl.org/echo/': '', + 'uri/of/echo/namespace#': '', + 'http://purl.org/pie/': '', + 'http://purl.org/atom/ns#': '', + 'http://www.w3.org/2005/Atom': '', + 'http://purl.org/rss/1.0/modules/rss091#': '', + + 'http://webns.net/mvcb/': 'admin', + 'http://purl.org/rss/1.0/modules/aggregation/': 'ag', + 'http://purl.org/rss/1.0/modules/annotate/': 'annotate', + 'http://media.tangent.org/rss/1.0/': 'audio', + 'http://backend.userland.com/blogChannelModule': 'blogChannel', + 'http://web.resource.org/cc/': 'cc', + 'http://backend.userland.com/creativeCommonsRssModule': 'creativeCommons', + 'http://purl.org/rss/1.0/modules/company': 'co', + 'http://purl.org/rss/1.0/modules/content/': 'content', + 'http://my.theinfo.org/changed/1.0/rss/': 'cp', + 'http://purl.org/dc/elements/1.1/': 'dc', + 'http://purl.org/dc/terms/': 'dcterms', + 'http://purl.org/rss/1.0/modules/email/': 'email', + 'http://purl.org/rss/1.0/modules/event/': 'ev', + 'http://rssnamespace.org/feedburner/ext/1.0': 'feedburner', + 'http://freshmeat.net/rss/fm/': 'fm', + 'http://xmlns.com/foaf/0.1/': 'foaf', + 'http://www.w3.org/2003/01/geo/wgs84_pos#': 'geo', + 'http://postneo.com/icbm/': 'icbm', + 'http://purl.org/rss/1.0/modules/image/': 'image', + 'http://www.itunes.com/DTDs/PodCast-1.0.dtd': 'itunes', + 'http://example.com/DTDs/PodCast-1.0.dtd': 'itunes', + 'http://purl.org/rss/1.0/modules/link/': 'l', + 'http://search.yahoo.com/mrss': 'media', + #Version 1.1.2 of the Media RSS spec added the trailing slash on the namespace + 'http://search.yahoo.com/mrss/': 'media', + 'http://madskills.com/public/xml/rss/module/pingback/': 'pingback', + 'http://prismstandard.org/namespaces/1.2/basic/': 'prism', + 'http://www.w3.org/1999/02/22-rdf-syntax-ns#': 'rdf', + 'http://www.w3.org/2000/01/rdf-schema#': 'rdfs', + 'http://purl.org/rss/1.0/modules/reference/': 'ref', + 'http://purl.org/rss/1.0/modules/richequiv/': 'reqv', + 'http://purl.org/rss/1.0/modules/search/': 'search', + 'http://purl.org/rss/1.0/modules/slash/': 'slash', + 'http://schemas.xmlsoap.org/soap/envelope/': 'soap', + 'http://purl.org/rss/1.0/modules/servicestatus/': 'ss', + 'http://hacks.benhammersley.com/rss/streaming/': 'str', + 'http://purl.org/rss/1.0/modules/subscription/': 'sub', + 'http://purl.org/rss/1.0/modules/syndication/': 'sy', + 'http://schemas.pocketsoap.com/rss/myDescModule/': 'szf', + 'http://purl.org/rss/1.0/modules/taxonomy/': 'taxo', + 'http://purl.org/rss/1.0/modules/threading/': 'thr', + 'http://purl.org/rss/1.0/modules/textinput/': 'ti', + 'http://madskills.com/public/xml/rss/module/trackback/':'trackback', + 'http://wellformedweb.org/commentAPI/': 'wfw', + 'http://purl.org/rss/1.0/modules/wiki/': 'wiki', + 'http://www.w3.org/1999/xhtml': 'xhtml', + 'http://www.w3.org/1999/xlink': 'xlink', + 'http://www.w3.org/XML/1998/namespace': 'xml' +} + _matchnamespaces = {} + + can_be_relative_uri = ['link', 'id', 'wfw_comment', 'wfw_commentrss', 'docs', 'url', 'href', 'comments', 'icon', 'logo'] + can_contain_relative_uris = ['content', 'title', 'summary', 'info', 'tagline', 'subtitle', 'copyright', 'rights', 'description'] + can_contain_dangerous_markup = ['content', 'title', 'summary', 'info', 'tagline', 'subtitle', 'copyright', 'rights', 'description'] + html_types = ['text/html', 'application/xhtml+xml'] + + def __init__(self, baseuri=None, baselang=None, encoding='utf-8'): + if _debug: sys.stderr.write('initializing FeedParser\n') + if not self._matchnamespaces: + for k, v in self.namespaces.items(): + self._matchnamespaces[k.lower()] = v + self.feeddata = FeedParserDict() # feed-level data + self.encoding = encoding # character encoding + self.entries = [] # list of entry-level data + self.version = '' # feed type/version, see SUPPORTED_VERSIONS + self.namespacesInUse = {} # dictionary of namespaces defined by the feed + + # the following are used internally to track state; + # this is really out of control and should be refactored + self.infeed = 0 + self.inentry = 0 + self.incontent = 0 + self.intextinput = 0 + self.inimage = 0 + self.inauthor = 0 + self.incontributor = 0 + self.inpublisher = 0 + self.insource = 0 + self.sourcedata = FeedParserDict() + self.contentparams = FeedParserDict() + self._summaryKey = None + self.namespacemap = {} + self.elementstack = [] + self.basestack = [] + self.langstack = [] + self.baseuri = baseuri or '' + self.lang = baselang or None + self.svgOK = 0 + self.hasTitle = 0 + if baselang: + self.feeddata['language'] = baselang.replace('_','-') + + def unknown_starttag(self, tag, attrs): + if _debug: sys.stderr.write('start %s with %s\n' % (tag, attrs)) + # normalize attrs + attrs = [(k.lower(), v) for k, v in attrs] + attrs = [(k, k in ('rel', 'type') and v.lower() or v) for k, v in attrs] + + # track xml:base and xml:lang + attrsD = dict(attrs) + baseuri = attrsD.get('xml:base', attrsD.get('base')) or self.baseuri + if type(baseuri) != type(u''): + try: + baseuri = unicode(baseuri, self.encoding) + except: + baseuri = unicode(baseuri, 'iso-8859-1') + self.baseuri = _urljoin(self.baseuri, baseuri) + lang = attrsD.get('xml:lang', attrsD.get('lang')) + if lang == '': + # xml:lang could be explicitly set to '', we need to capture that + lang = None + elif lang is None: + # if no xml:lang is specified, use parent lang + lang = self.lang + if lang: + if tag in ('feed', 'rss', 'rdf:RDF'): + self.feeddata['language'] = lang.replace('_','-') + self.lang = lang + self.basestack.append(self.baseuri) + self.langstack.append(lang) + + # track namespaces + for prefix, uri in attrs: + if prefix.startswith('xmlns:'): + self.trackNamespace(prefix[6:], uri) + elif prefix == 'xmlns': + self.trackNamespace(None, uri) + + # track inline content + if self.incontent and self.contentparams.has_key('type') and not self.contentparams.get('type', 'xml').endswith('xml'): + if tag in ['xhtml:div', 'div']: return # typepad does this 10/2007 + # element declared itself as escaped markup, but it isn't really + self.contentparams['type'] = 'application/xhtml+xml' + if self.incontent and self.contentparams.get('type') == 'application/xhtml+xml': + if tag.find(':') <> -1: + prefix, tag = tag.split(':', 1) + namespace = self.namespacesInUse.get(prefix, '') + if tag=='math' and namespace=='http://www.w3.org/1998/Math/MathML': + attrs.append(('xmlns',namespace)) + if tag=='svg' and namespace=='http://www.w3.org/2000/svg': + attrs.append(('xmlns',namespace)) + if tag == 'svg': self.svgOK += 1 + return self.handle_data('<%s%s>' % (tag, self.strattrs(attrs)), escape=0) + + # match namespaces + if tag.find(':') <> -1: + prefix, suffix = tag.split(':', 1) + else: + prefix, suffix = '', tag + prefix = self.namespacemap.get(prefix, prefix) + if prefix: + prefix = prefix + '_' + + # special hack for better tracking of empty textinput/image elements in illformed feeds + if (not prefix) and tag not in ('title', 'link', 'description', 'name'): + self.intextinput = 0 + if (not prefix) and tag not in ('title', 'link', 'description', 'url', 'href', 'width', 'height'): + self.inimage = 0 + + # call special handler (if defined) or default handler + methodname = '_start_' + prefix + suffix + try: + method = getattr(self, methodname) + return method(attrsD) + except AttributeError: + # Since there's no handler or something has gone wrong we explicitly add the element and its attributes + unknown_tag = prefix + suffix + if len(attrsD) == 0: + # No attributes so merge it into the encosing dictionary + return self.push(unknown_tag, 1) + else: + # Has attributes so create it in its own dictionary + context = self._getContext() + context[unknown_tag] = attrsD + + def unknown_endtag(self, tag): + if _debug: sys.stderr.write('end %s\n' % tag) + # match namespaces + if tag.find(':') <> -1: + prefix, suffix = tag.split(':', 1) + else: + prefix, suffix = '', tag + prefix = self.namespacemap.get(prefix, prefix) + if prefix: + prefix = prefix + '_' + if suffix == 'svg' and self.svgOK: self.svgOK -= 1 + + # call special handler (if defined) or default handler + methodname = '_end_' + prefix + suffix + try: + if self.svgOK: raise AttributeError() + method = getattr(self, methodname) + method() + except AttributeError: + self.pop(prefix + suffix) + + # track inline content + if self.incontent and self.contentparams.has_key('type') and not self.contentparams.get('type', 'xml').endswith('xml'): + # element declared itself as escaped markup, but it isn't really + if tag in ['xhtml:div', 'div']: return # typepad does this 10/2007 + self.contentparams['type'] = 'application/xhtml+xml' + if self.incontent and self.contentparams.get('type') == 'application/xhtml+xml': + tag = tag.split(':')[-1] + self.handle_data('' % tag, escape=0) + + # track xml:base and xml:lang going out of scope + if self.basestack: + self.basestack.pop() + if self.basestack and self.basestack[-1]: + self.baseuri = self.basestack[-1] + if self.langstack: + self.langstack.pop() + if self.langstack: # and (self.langstack[-1] is not None): + self.lang = self.langstack[-1] + + def handle_charref(self, ref): + # called for each character reference, e.g. for ' ', ref will be '160' + if not self.elementstack: return + ref = ref.lower() + if ref in ('34', '38', '39', '60', '62', 'x22', 'x26', 'x27', 'x3c', 'x3e'): + text = '&#%s;' % ref + else: + if ref[0] == 'x': + c = int(ref[1:], 16) + else: + c = int(ref) + text = unichr(c).encode('utf-8') + self.elementstack[-1][2].append(text) + + def handle_entityref(self, ref): + # called for each entity reference, e.g. for '©', ref will be 'copy' + if not self.elementstack: return + if _debug: sys.stderr.write('entering handle_entityref with %s\n' % ref) + if ref in ('lt', 'gt', 'quot', 'amp', 'apos'): + text = '&%s;' % ref + elif ref in self.entities.keys(): + text = self.entities[ref] + if text.startswith('&#') and text.endswith(';'): + return self.handle_entityref(text) + else: + try: name2codepoint[ref] + except KeyError: text = '&%s;' % ref + else: text = unichr(name2codepoint[ref]).encode('utf-8') + self.elementstack[-1][2].append(text) + + def handle_data(self, text, escape=1): + # called for each block of plain text, i.e. outside of any tag and + # not containing any character or entity references + if not self.elementstack: return + if escape and self.contentparams.get('type') == 'application/xhtml+xml': + text = _xmlescape(text) + self.elementstack[-1][2].append(text) + + def handle_comment(self, text): + # called for each comment, e.g. + pass + + def handle_pi(self, text): + # called for each processing instruction, e.g. + pass + + def handle_decl(self, text): + pass + + def parse_declaration(self, i): + # override internal declaration handler to handle CDATA blocks + if _debug: sys.stderr.write('entering parse_declaration\n') + if self.rawdata[i:i+9] == '', i) + if k == -1: + # CDATA block began but didn't finish + k = len(self.rawdata) + return k + self.handle_data(_xmlescape(self.rawdata[i+9:k]), 0) + return k+3 + else: + k = self.rawdata.find('>', i) + if k >= 0: + return k+1 + else: + # We have an incomplete CDATA block. + return k + + def mapContentType(self, contentType): + contentType = contentType.lower() + if contentType == 'text': + contentType = 'text/plain' + elif contentType == 'html': + contentType = 'text/html' + elif contentType == 'xhtml': + contentType = 'application/xhtml+xml' + return contentType + + def trackNamespace(self, prefix, uri): + loweruri = uri.lower() + if (prefix, loweruri) == (None, 'http://my.netscape.com/rdf/simple/0.9/') and not self.version: + self.version = 'rss090' + if loweruri == 'http://purl.org/rss/1.0/' and not self.version: + self.version = 'rss10' + if loweruri == 'http://www.w3.org/2005/atom' and not self.version: + self.version = 'atom10' + if loweruri.find('backend.userland.com/rss') <> -1: + # match any backend.userland.com namespace + uri = 'http://backend.userland.com/rss' + loweruri = uri + if self._matchnamespaces.has_key(loweruri): + self.namespacemap[prefix] = self._matchnamespaces[loweruri] + self.namespacesInUse[self._matchnamespaces[loweruri]] = uri + else: + self.namespacesInUse[prefix or ''] = uri + + def resolveURI(self, uri): + return _urljoin(self.baseuri or '', uri) + + def decodeEntities(self, element, data): + return data + + def strattrs(self, attrs): + return ''.join([' %s="%s"' % (t[0],_xmlescape(t[1],{'"':'"'})) for t in attrs]) + + def push(self, element, expectingText): + self.elementstack.append([element, expectingText, []]) + + def pop(self, element, stripWhitespace=1): + if not self.elementstack: return + if self.elementstack[-1][0] != element: return + + element, expectingText, pieces = self.elementstack.pop() + + if self.version == 'atom10' and self.contentparams.get('type','text') == 'application/xhtml+xml': + # remove enclosing child element, but only if it is a
and + # only if all the remaining content is nested underneath it. + # This means that the divs would be retained in the following: + #
foo
bar
+ while pieces and len(pieces)>1 and not pieces[-1].strip(): + del pieces[-1] + while pieces and len(pieces)>1 and not pieces[0].strip(): + del pieces[0] + if pieces and (pieces[0] == '
' or pieces[0].startswith('
': + depth = 0 + for piece in pieces[:-1]: + if piece.startswith(''): + depth += 1 + else: + pieces = pieces[1:-1] + + output = ''.join(pieces) + if stripWhitespace: + output = output.strip() + if not expectingText: return output + + # decode base64 content + if base64 and self.contentparams.get('base64', 0): + try: + output = base64.decodestring(output) + except binascii.Error: + pass + except binascii.Incomplete: + pass + + # resolve relative URIs + if (element in self.can_be_relative_uri) and output: + output = self.resolveURI(output) + + # decode entities within embedded markup + if not self.contentparams.get('base64', 0): + output = self.decodeEntities(element, output) + + if self.lookslikehtml(output): + self.contentparams['type']='text/html' + + # remove temporary cruft from contentparams + try: + del self.contentparams['mode'] + except KeyError: + pass + try: + del self.contentparams['base64'] + except KeyError: + pass + + is_htmlish = self.mapContentType(self.contentparams.get('type', 'text/html')) in self.html_types + # resolve relative URIs within embedded markup + if is_htmlish and RESOLVE_RELATIVE_URIS: + if element in self.can_contain_relative_uris: + output = _resolveRelativeURIs(output, self.baseuri, self.encoding, self.contentparams.get('type', 'text/html')) + + # parse microformats + # (must do this before sanitizing because some microformats + # rely on elements that we sanitize) + if is_htmlish and element in ['content', 'description', 'summary']: + mfresults = _parseMicroformats(output, self.baseuri, self.encoding) + if mfresults: + for tag in mfresults.get('tags', []): + self._addTag(tag['term'], tag['scheme'], tag['label']) + for enclosure in mfresults.get('enclosures', []): + self._start_enclosure(enclosure) + for xfn in mfresults.get('xfn', []): + self._addXFN(xfn['relationships'], xfn['href'], xfn['name']) + vcard = mfresults.get('vcard') + if vcard: + self._getContext()['vcard'] = vcard + + # sanitize embedded markup + if is_htmlish and SANITIZE_HTML: + if element in self.can_contain_dangerous_markup: + output = _sanitizeHTML(output, self.encoding, self.contentparams.get('type', 'text/html')) + + if self.encoding and type(output) != type(u''): + try: + output = unicode(output, self.encoding) + except: + pass + + # address common error where people take data that is already + # utf-8, presume that it is iso-8859-1, and re-encode it. + if self.encoding=='utf-8' and type(output) == type(u''): + try: + output = unicode(output.encode('iso-8859-1'), 'utf-8') + except: + pass + + # map win-1252 extensions to the proper code points + if type(output) == type(u''): + output = u''.join([c in _cp1252.keys() and _cp1252[c] or c for c in output]) + + # categories/tags/keywords/whatever are handled in _end_category + if element == 'category': + return output + + if element == 'title' and self.hasTitle: + return output + + # store output in appropriate place(s) + if self.inentry and not self.insource: + if element == 'content': + self.entries[-1].setdefault(element, []) + contentparams = copy.deepcopy(self.contentparams) + contentparams['value'] = output + self.entries[-1][element].append(contentparams) + elif element == 'link': + self.entries[-1][element] = output + if output: + self.entries[-1]['links'][-1]['href'] = output + else: + if element == 'description': + element = 'summary' + self.entries[-1][element] = output + if self.incontent: + contentparams = copy.deepcopy(self.contentparams) + contentparams['value'] = output + self.entries[-1][element + '_detail'] = contentparams + elif (self.infeed or self.insource):# and (not self.intextinput) and (not self.inimage): + context = self._getContext() + if element == 'description': + element = 'subtitle' + context[element] = output + if element == 'link': + context['links'][-1]['href'] = output + elif self.incontent: + contentparams = copy.deepcopy(self.contentparams) + contentparams['value'] = output + context[element + '_detail'] = contentparams + return output + + def pushContent(self, tag, attrsD, defaultContentType, expectingText): + self.incontent += 1 + if self.lang: self.lang=self.lang.replace('_','-') + self.contentparams = FeedParserDict({ + 'type': self.mapContentType(attrsD.get('type', defaultContentType)), + 'language': self.lang, + 'base': self.baseuri}) + self.contentparams['base64'] = self._isBase64(attrsD, self.contentparams) + self.push(tag, expectingText) + + def popContent(self, tag): + value = self.pop(tag) + self.incontent -= 1 + self.contentparams.clear() + return value + + # a number of elements in a number of RSS variants are nominally plain + # text, but this is routinely ignored. This is an attempt to detect + # the most common cases. As false positives often result in silent + # data loss, this function errs on the conservative side. + def lookslikehtml(self, str): + if self.version.startswith('atom'): return + if self.contentparams.get('type','text/html') != 'text/plain': return + + # must have a close tag or a entity reference to qualify + if not (re.search(r'',str) or re.search("&#?\w+;",str)): return + + # all tags must be in a restricted subset of valid HTML tags + if filter(lambda t: t.lower() not in _HTMLSanitizer.acceptable_elements, + re.findall(r' -1: + prefix = name[:colonpos] + suffix = name[colonpos+1:] + prefix = self.namespacemap.get(prefix, prefix) + name = prefix + ':' + suffix + return name + + def _getAttribute(self, attrsD, name): + return attrsD.get(self._mapToStandardPrefix(name)) + + def _isBase64(self, attrsD, contentparams): + if attrsD.get('mode', '') == 'base64': + return 1 + if self.contentparams['type'].startswith('text/'): + return 0 + if self.contentparams['type'].endswith('+xml'): + return 0 + if self.contentparams['type'].endswith('/xml'): + return 0 + return 1 + + def _itsAnHrefDamnIt(self, attrsD): + href = attrsD.get('url', attrsD.get('uri', attrsD.get('href', None))) + if href: + try: + del attrsD['url'] + except KeyError: + pass + try: + del attrsD['uri'] + except KeyError: + pass + attrsD['href'] = href + return attrsD + + def _save(self, key, value): + context = self._getContext() + context.setdefault(key, value) + + def _start_rss(self, attrsD): + versionmap = {'0.91': 'rss091u', + '0.92': 'rss092', + '0.93': 'rss093', + '0.94': 'rss094'} + #If we're here then this is an RSS feed. + #If we don't have a version or have a version that starts with something + #other than RSS then there's been a mistake. Correct it. + if not self.version or not self.version.startswith('rss'): + attr_version = attrsD.get('version', '') + version = versionmap.get(attr_version) + if version: + self.version = version + elif attr_version.startswith('2.'): + self.version = 'rss20' + else: + self.version = 'rss' + + def _start_dlhottitles(self, attrsD): + self.version = 'hotrss' + + def _start_channel(self, attrsD): + self.infeed = 1 + self._cdf_common(attrsD) + _start_feedinfo = _start_channel + + def _cdf_common(self, attrsD): + if attrsD.has_key('lastmod'): + self._start_modified({}) + self.elementstack[-1][-1] = attrsD['lastmod'] + self._end_modified() + if attrsD.has_key('href'): + self._start_link({}) + self.elementstack[-1][-1] = attrsD['href'] + self._end_link() + + def _start_feed(self, attrsD): + self.infeed = 1 + versionmap = {'0.1': 'atom01', + '0.2': 'atom02', + '0.3': 'atom03'} + if not self.version: + attr_version = attrsD.get('version') + version = versionmap.get(attr_version) + if version: + self.version = version + else: + self.version = 'atom' + + def _end_channel(self): + self.infeed = 0 + _end_feed = _end_channel + + def _start_image(self, attrsD): + context = self._getContext() + context.setdefault('image', FeedParserDict()) + self.inimage = 1 + self.hasTitle = 0 + self.push('image', 0) + + def _end_image(self): + self.pop('image') + self.inimage = 0 + + def _start_textinput(self, attrsD): + context = self._getContext() + context.setdefault('textinput', FeedParserDict()) + self.intextinput = 1 + self.hasTitle = 0 + self.push('textinput', 0) + _start_textInput = _start_textinput + + def _end_textinput(self): + self.pop('textinput') + self.intextinput = 0 + _end_textInput = _end_textinput + + def _start_author(self, attrsD): + self.inauthor = 1 + self.push('author', 1) + _start_managingeditor = _start_author + _start_dc_author = _start_author + _start_dc_creator = _start_author + _start_itunes_author = _start_author + + def _end_author(self): + self.pop('author') + self.inauthor = 0 + self._sync_author_detail() + _end_managingeditor = _end_author + _end_dc_author = _end_author + _end_dc_creator = _end_author + _end_itunes_author = _end_author + + def _start_itunes_owner(self, attrsD): + self.inpublisher = 1 + self.push('publisher', 0) + + def _end_itunes_owner(self): + self.pop('publisher') + self.inpublisher = 0 + self._sync_author_detail('publisher') + + def _start_contributor(self, attrsD): + self.incontributor = 1 + context = self._getContext() + context.setdefault('contributors', []) + context['contributors'].append(FeedParserDict()) + self.push('contributor', 0) + + def _end_contributor(self): + self.pop('contributor') + self.incontributor = 0 + + def _start_dc_contributor(self, attrsD): + self.incontributor = 1 + context = self._getContext() + context.setdefault('contributors', []) + context['contributors'].append(FeedParserDict()) + self.push('name', 0) + + def _end_dc_contributor(self): + self._end_name() + self.incontributor = 0 + + def _start_name(self, attrsD): + self.push('name', 0) + _start_itunes_name = _start_name + + def _end_name(self): + value = self.pop('name') + if self.inpublisher: + self._save_author('name', value, 'publisher') + elif self.inauthor: + self._save_author('name', value) + elif self.incontributor: + self._save_contributor('name', value) + elif self.intextinput: + context = self._getContext() + context['name'] = value + _end_itunes_name = _end_name + + def _start_width(self, attrsD): + self.push('width', 0) + + def _end_width(self): + value = self.pop('width') + try: + value = int(value) + except: + value = 0 + if self.inimage: + context = self._getContext() + context['width'] = value + + def _start_height(self, attrsD): + self.push('height', 0) + + def _end_height(self): + value = self.pop('height') + try: + value = int(value) + except: + value = 0 + if self.inimage: + context = self._getContext() + context['height'] = value + + def _start_url(self, attrsD): + self.push('href', 1) + _start_homepage = _start_url + _start_uri = _start_url + + def _end_url(self): + value = self.pop('href') + if self.inauthor: + self._save_author('href', value) + elif self.incontributor: + self._save_contributor('href', value) + _end_homepage = _end_url + _end_uri = _end_url + + def _start_email(self, attrsD): + self.push('email', 0) + _start_itunes_email = _start_email + + def _end_email(self): + value = self.pop('email') + if self.inpublisher: + self._save_author('email', value, 'publisher') + elif self.inauthor: + self._save_author('email', value) + elif self.incontributor: + self._save_contributor('email', value) + _end_itunes_email = _end_email + + def _getContext(self): + if self.insource: + context = self.sourcedata + elif self.inimage and self.feeddata.has_key('image'): + context = self.feeddata['image'] + elif self.intextinput: + context = self.feeddata['textinput'] + elif self.inentry: + context = self.entries[-1] + else: + context = self.feeddata + return context + + def _save_author(self, key, value, prefix='author'): + context = self._getContext() + context.setdefault(prefix + '_detail', FeedParserDict()) + context[prefix + '_detail'][key] = value + self._sync_author_detail() + + def _save_contributor(self, key, value): + context = self._getContext() + context.setdefault('contributors', [FeedParserDict()]) + context['contributors'][-1][key] = value + + def _sync_author_detail(self, key='author'): + context = self._getContext() + detail = context.get('%s_detail' % key) + if detail: + name = detail.get('name') + email = detail.get('email') + if name and email: + context[key] = '%s (%s)' % (name, email) + elif name: + context[key] = name + elif email: + context[key] = email + else: + author, email = context.get(key), None + if not author: return + emailmatch = re.search(r'''(([a-zA-Z0-9\_\-\.\+]+)@((\[[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.)|(([a-zA-Z0-9\-]+\.)+))([a-zA-Z]{2,4}|[0-9]{1,3})(\]?))(\?subject=\S+)?''', author) + if emailmatch: + email = emailmatch.group(0) + # probably a better way to do the following, but it passes all the tests + author = author.replace(email, '') + author = author.replace('()', '') + author = author.replace('<>', '') + author = author.replace('<>', '') + author = author.strip() + if author and (author[0] == '('): + author = author[1:] + if author and (author[-1] == ')'): + author = author[:-1] + author = author.strip() + if author or email: + context.setdefault('%s_detail' % key, FeedParserDict()) + if author: + context['%s_detail' % key]['name'] = author + if email: + context['%s_detail' % key]['email'] = email + + def _start_subtitle(self, attrsD): + self.pushContent('subtitle', attrsD, 'text/plain', 1) + _start_tagline = _start_subtitle + _start_itunes_subtitle = _start_subtitle + + def _end_subtitle(self): + self.popContent('subtitle') + _end_tagline = _end_subtitle + _end_itunes_subtitle = _end_subtitle + + def _start_rights(self, attrsD): + self.pushContent('rights', attrsD, 'text/plain', 1) + _start_dc_rights = _start_rights + _start_copyright = _start_rights + + def _end_rights(self): + self.popContent('rights') + _end_dc_rights = _end_rights + _end_copyright = _end_rights + + def _start_item(self, attrsD): + self.entries.append(FeedParserDict()) + self.push('item', 0) + self.inentry = 1 + self.guidislink = 0 + self.hasTitle = 0 + id = self._getAttribute(attrsD, 'rdf:about') + if id: + context = self._getContext() + context['id'] = id + self._cdf_common(attrsD) + _start_entry = _start_item + _start_product = _start_item + + def _end_item(self): + self.pop('item') + self.inentry = 0 + _end_entry = _end_item + + def _start_dc_language(self, attrsD): + self.push('language', 1) + _start_language = _start_dc_language + + def _end_dc_language(self): + self.lang = self.pop('language') + _end_language = _end_dc_language + + def _start_dc_publisher(self, attrsD): + self.push('publisher', 1) + _start_webmaster = _start_dc_publisher + + def _end_dc_publisher(self): + self.pop('publisher') + self._sync_author_detail('publisher') + _end_webmaster = _end_dc_publisher + + def _start_published(self, attrsD): + self.push('published', 1) + _start_dcterms_issued = _start_published + _start_issued = _start_published + + def _end_published(self): + value = self.pop('published') + self._save('published_parsed', _parse_date(value)) + _end_dcterms_issued = _end_published + _end_issued = _end_published + + def _start_updated(self, attrsD): + self.push('updated', 1) + _start_modified = _start_updated + _start_dcterms_modified = _start_updated + _start_pubdate = _start_updated + _start_dc_date = _start_updated + + def _end_updated(self): + value = self.pop('updated') + parsed_value = _parse_date(value) + self._save('updated_parsed', parsed_value) + _end_modified = _end_updated + _end_dcterms_modified = _end_updated + _end_pubdate = _end_updated + _end_dc_date = _end_updated + + def _start_created(self, attrsD): + self.push('created', 1) + _start_dcterms_created = _start_created + + def _end_created(self): + value = self.pop('created') + self._save('created_parsed', _parse_date(value)) + _end_dcterms_created = _end_created + + def _start_expirationdate(self, attrsD): + self.push('expired', 1) + + def _end_expirationdate(self): + self._save('expired_parsed', _parse_date(self.pop('expired'))) + + def _start_cc_license(self, attrsD): + context = self._getContext() + value = self._getAttribute(attrsD, 'rdf:resource') + attrsD = FeedParserDict() + attrsD['rel']='license' + if value: attrsD['href']=value + context.setdefault('links', []).append(attrsD) + + def _start_creativecommons_license(self, attrsD): + self.push('license', 1) + _start_creativeCommons_license = _start_creativecommons_license + + def _end_creativecommons_license(self): + value = self.pop('license') + context = self._getContext() + attrsD = FeedParserDict() + attrsD['rel']='license' + if value: attrsD['href']=value + context.setdefault('links', []).append(attrsD) + del context['license'] + _end_creativeCommons_license = _end_creativecommons_license + + def _addXFN(self, relationships, href, name): + context = self._getContext() + xfn = context.setdefault('xfn', []) + value = FeedParserDict({'relationships': relationships, 'href': href, 'name': name}) + if value not in xfn: + xfn.append(value) + + def _addTag(self, term, scheme, label): + context = self._getContext() + tags = context.setdefault('tags', []) + if (not term) and (not scheme) and (not label): return + value = FeedParserDict({'term': term, 'scheme': scheme, 'label': label}) + if value not in tags: + tags.append(value) + + def _start_category(self, attrsD): + if _debug: sys.stderr.write('entering _start_category with %s\n' % repr(attrsD)) + term = attrsD.get('term') + scheme = attrsD.get('scheme', attrsD.get('domain')) + label = attrsD.get('label') + self._addTag(term, scheme, label) + self.push('category', 1) + _start_dc_subject = _start_category + _start_keywords = _start_category + + def _end_itunes_keywords(self): + for term in self.pop('itunes_keywords').split(): + self._addTag(term, 'http://www.itunes.com/', None) + + def _start_itunes_category(self, attrsD): + self._addTag(attrsD.get('text'), 'http://www.itunes.com/', None) + self.push('category', 1) + + def _end_category(self): + value = self.pop('category') + if not value: return + context = self._getContext() + tags = context['tags'] + if value and len(tags) and not tags[-1]['term']: + tags[-1]['term'] = value + else: + self._addTag(value, None, None) + _end_dc_subject = _end_category + _end_keywords = _end_category + _end_itunes_category = _end_category + + def _start_cloud(self, attrsD): + self._getContext()['cloud'] = FeedParserDict(attrsD) + + def _start_link(self, attrsD): + attrsD.setdefault('rel', 'alternate') + if attrsD['rel'] == 'self': + attrsD.setdefault('type', 'application/atom+xml') + else: + attrsD.setdefault('type', 'text/html') + context = self._getContext() + attrsD = self._itsAnHrefDamnIt(attrsD) + if attrsD.has_key('href'): + attrsD['href'] = self.resolveURI(attrsD['href']) + if attrsD.get('rel')=='enclosure' and not context.get('id'): + context['id'] = attrsD.get('href') + expectingText = self.infeed or self.inentry or self.insource + context.setdefault('links', []) + context['links'].append(FeedParserDict(attrsD)) + if attrsD.has_key('href'): + expectingText = 0 + if (attrsD.get('rel') == 'alternate') and (self.mapContentType(attrsD.get('type')) in self.html_types): + context['link'] = attrsD['href'] + else: + self.push('link', expectingText) + _start_producturl = _start_link + + def _end_link(self): + value = self.pop('link') + context = self._getContext() + _end_producturl = _end_link + + def _start_guid(self, attrsD): + self.guidislink = (attrsD.get('ispermalink', 'true') == 'true') + self.push('id', 1) + + def _end_guid(self): + value = self.pop('id') + self._save('guidislink', self.guidislink and not self._getContext().has_key('link')) + if self.guidislink: + # guid acts as link, but only if 'ispermalink' is not present or is 'true', + # and only if the item doesn't already have a link element + self._save('link', value) + + def _start_title(self, attrsD): + if self.svgOK: return self.unknown_starttag('title', attrsD.items()) + self.pushContent('title', attrsD, 'text/plain', self.infeed or self.inentry or self.insource) + _start_dc_title = _start_title + _start_media_title = _start_title + + def _end_title(self): + if self.svgOK: return + value = self.popContent('title') + if not value: return + context = self._getContext() + self.hasTitle = 1 + _end_dc_title = _end_title + + def _end_media_title(self): + hasTitle = self.hasTitle + self._end_title() + self.hasTitle = hasTitle + + def _start_description(self, attrsD): + context = self._getContext() + if context.has_key('summary'): + self._summaryKey = 'content' + self._start_content(attrsD) + else: + self.pushContent('description', attrsD, 'text/html', self.infeed or self.inentry or self.insource) + _start_dc_description = _start_description + + def _start_abstract(self, attrsD): + self.pushContent('description', attrsD, 'text/plain', self.infeed or self.inentry or self.insource) + + def _end_description(self): + if self._summaryKey == 'content': + self._end_content() + else: + value = self.popContent('description') + self._summaryKey = None + _end_abstract = _end_description + _end_dc_description = _end_description + + def _start_info(self, attrsD): + self.pushContent('info', attrsD, 'text/plain', 1) + _start_feedburner_browserfriendly = _start_info + + def _end_info(self): + self.popContent('info') + _end_feedburner_browserfriendly = _end_info + + def _start_generator(self, attrsD): + if attrsD: + attrsD = self._itsAnHrefDamnIt(attrsD) + if attrsD.has_key('href'): + attrsD['href'] = self.resolveURI(attrsD['href']) + self._getContext()['generator_detail'] = FeedParserDict(attrsD) + self.push('generator', 1) + + def _end_generator(self): + value = self.pop('generator') + context = self._getContext() + if context.has_key('generator_detail'): + context['generator_detail']['name'] = value + + def _start_admin_generatoragent(self, attrsD): + self.push('generator', 1) + value = self._getAttribute(attrsD, 'rdf:resource') + if value: + self.elementstack[-1][2].append(value) + self.pop('generator') + self._getContext()['generator_detail'] = FeedParserDict({'href': value}) + + def _start_admin_errorreportsto(self, attrsD): + self.push('errorreportsto', 1) + value = self._getAttribute(attrsD, 'rdf:resource') + if value: + self.elementstack[-1][2].append(value) + self.pop('errorreportsto') + + def _start_summary(self, attrsD): + context = self._getContext() + if context.has_key('summary'): + self._summaryKey = 'content' + self._start_content(attrsD) + else: + self._summaryKey = 'summary' + self.pushContent(self._summaryKey, attrsD, 'text/plain', 1) + _start_itunes_summary = _start_summary + + def _end_summary(self): + if self._summaryKey == 'content': + self._end_content() + else: + self.popContent(self._summaryKey or 'summary') + self._summaryKey = None + _end_itunes_summary = _end_summary + + def _start_enclosure(self, attrsD): + attrsD = self._itsAnHrefDamnIt(attrsD) + context = self._getContext() + attrsD['rel']='enclosure' + context.setdefault('links', []).append(FeedParserDict(attrsD)) + href = attrsD.get('href') + if href and not context.get('id'): + context['id'] = href + + def _start_source(self, attrsD): + if 'url' in attrsD: + # This means that we're processing a source element from an RSS 2.0 feed + self.sourcedata['href'] = attrsD[u'url'] + self.push('source', 1) + self.insource = 1 + self.hasTitle = 0 + + def _end_source(self): + self.insource = 0 + value = self.pop('source') + if value: + self.sourcedata['title'] = value + self._getContext()['source'] = copy.deepcopy(self.sourcedata) + self.sourcedata.clear() + + def _start_content(self, attrsD): + self.pushContent('content', attrsD, 'text/plain', 1) + src = attrsD.get('src') + if src: + self.contentparams['src'] = src + self.push('content', 1) + + def _start_prodlink(self, attrsD): + self.pushContent('content', attrsD, 'text/html', 1) + + def _start_body(self, attrsD): + self.pushContent('content', attrsD, 'application/xhtml+xml', 1) + _start_xhtml_body = _start_body + + def _start_content_encoded(self, attrsD): + self.pushContent('content', attrsD, 'text/html', 1) + _start_fullitem = _start_content_encoded + + def _end_content(self): + copyToDescription = self.mapContentType(self.contentparams.get('type')) in (['text/plain'] + self.html_types) + value = self.popContent('content') + if copyToDescription: + self._save('description', value) + + _end_body = _end_content + _end_xhtml_body = _end_content + _end_content_encoded = _end_content + _end_fullitem = _end_content + _end_prodlink = _end_content + + def _start_itunes_image(self, attrsD): + self.push('itunes_image', 0) + self._getContext()['image'] = FeedParserDict({'href': attrsD.get('href')}) + _start_itunes_link = _start_itunes_image + + def _end_itunes_block(self): + value = self.pop('itunes_block', 0) + self._getContext()['itunes_block'] = (value == 'yes') and 1 or 0 + + def _end_itunes_explicit(self): + value = self.pop('itunes_explicit', 0) + self._getContext()['itunes_explicit'] = (value == 'yes') and 1 or 0 + + def _start_media_content(self, attrsD): + context = self._getContext() + context.setdefault('media_content', []) + context['media_content'].append(attrsD) + + def _start_media_thumbnail(self, attrsD): + context = self._getContext() + context.setdefault('media_thumbnail', []) + self.push('url', 1) # new + context['media_thumbnail'].append(attrsD) + + def _end_media_thumbnail(self): + url = self.pop('url') + context = self._getContext() + if url != None and len(url.strip()) != 0: + if not context['media_thumbnail'][-1].has_key('url'): + context['media_thumbnail'][-1]['url'] = url + + def _start_media_player(self, attrsD): + self.push('media_player', 0) + self._getContext()['media_player'] = FeedParserDict(attrsD) + + def _end_media_player(self): + value = self.pop('media_player') + context = self._getContext() + context['media_player']['content'] = value + +if _XML_AVAILABLE: + class _StrictFeedParser(_FeedParserMixin, xml.sax.handler.ContentHandler): + def __init__(self, baseuri, baselang, encoding): + if _debug: sys.stderr.write('trying StrictFeedParser\n') + xml.sax.handler.ContentHandler.__init__(self) + _FeedParserMixin.__init__(self, baseuri, baselang, encoding) + self.bozo = 0 + self.exc = None + self.decls = {} + + def startPrefixMapping(self, prefix, uri): + self.trackNamespace(prefix, uri) + if uri == 'http://www.w3.org/1999/xlink': + self.decls['xmlns:'+prefix] = uri + + def startElementNS(self, name, qname, attrs): + namespace, localname = name + lowernamespace = str(namespace or '').lower() + if lowernamespace.find('backend.userland.com/rss') <> -1: + # match any backend.userland.com namespace + namespace = 'http://backend.userland.com/rss' + lowernamespace = namespace + if qname and qname.find(':') > 0: + givenprefix = qname.split(':')[0] + else: + givenprefix = None + prefix = self._matchnamespaces.get(lowernamespace, givenprefix) + if givenprefix and (prefix == None or (prefix == '' and lowernamespace == '')) and not self.namespacesInUse.has_key(givenprefix): + raise UndeclaredNamespace, "'%s' is not associated with a namespace" % givenprefix + localname = str(localname).lower() + + # qname implementation is horribly broken in Python 2.1 (it + # doesn't report any), and slightly broken in Python 2.2 (it + # doesn't report the xml: namespace). So we match up namespaces + # with a known list first, and then possibly override them with + # the qnames the SAX parser gives us (if indeed it gives us any + # at all). Thanks to MatejC for helping me test this and + # tirelessly telling me that it didn't work yet. + attrsD, self.decls = self.decls, {} + if localname=='math' and namespace=='http://www.w3.org/1998/Math/MathML': + attrsD['xmlns']=namespace + if localname=='svg' and namespace=='http://www.w3.org/2000/svg': + attrsD['xmlns']=namespace + + if prefix: + localname = prefix.lower() + ':' + localname + elif namespace and not qname: #Expat + for name,value in self.namespacesInUse.items(): + if name and value == namespace: + localname = name + ':' + localname + break + if _debug: sys.stderr.write('startElementNS: qname = %s, namespace = %s, givenprefix = %s, prefix = %s, attrs = %s, localname = %s\n' % (qname, namespace, givenprefix, prefix, attrs.items(), localname)) + + for (namespace, attrlocalname), attrvalue in attrs._attrs.items(): + lowernamespace = (namespace or '').lower() + prefix = self._matchnamespaces.get(lowernamespace, '') + if prefix: + attrlocalname = prefix + ':' + attrlocalname + attrsD[str(attrlocalname).lower()] = attrvalue + for qname in attrs.getQNames(): + attrsD[str(qname).lower()] = attrs.getValueByQName(qname) + self.unknown_starttag(localname, attrsD.items()) + + def characters(self, text): + self.handle_data(text) + + def endElementNS(self, name, qname): + namespace, localname = name + lowernamespace = str(namespace or '').lower() + if qname and qname.find(':') > 0: + givenprefix = qname.split(':')[0] + else: + givenprefix = '' + prefix = self._matchnamespaces.get(lowernamespace, givenprefix) + if prefix: + localname = prefix + ':' + localname + elif namespace and not qname: #Expat + for name,value in self.namespacesInUse.items(): + if name and value == namespace: + localname = name + ':' + localname + break + localname = str(localname).lower() + self.unknown_endtag(localname) + + def error(self, exc): + self.bozo = 1 + self.exc = exc + + def fatalError(self, exc): + self.error(exc) + raise exc + +class _BaseHTMLProcessor(sgmllib.SGMLParser): + special = re.compile('''[<>'"]''') + bare_ampersand = re.compile("&(?!#\d+;|#x[0-9a-fA-F]+;|\w+;)") + elements_no_end_tag = [ + 'area', 'base', 'basefont', 'br', 'col', 'command', 'embed', 'frame', + 'hr', 'img', 'input', 'isindex', 'keygen', 'link', 'meta', 'param', + 'source', 'track', 'wbr' + ] + + def __init__(self, encoding, type): + self.encoding = encoding + self.type = type + if _debug: sys.stderr.write('entering BaseHTMLProcessor, encoding=%s\n' % self.encoding) + sgmllib.SGMLParser.__init__(self) + + def reset(self): + self.pieces = [] + sgmllib.SGMLParser.reset(self) + + def _shorttag_replace(self, match): + tag = match.group(1) + if tag in self.elements_no_end_tag: + return '<' + tag + ' />' + else: + return '<' + tag + '>' + + def parse_starttag(self,i): + j=sgmllib.SGMLParser.parse_starttag(self, i) + if self.type == 'application/xhtml+xml': + if j>2 and self.rawdata[j-2:j]=='/>': + self.unknown_endtag(self.lasttag) + return j + + def feed(self, data): + data = re.compile(r'', self._shorttag_replace, data) # bug [ 1399464 ] Bad regexp for _shorttag_replace + data = re.sub(r'<([^<>\s]+?)\s*/>', self._shorttag_replace, data) + data = data.replace(''', "'") + data = data.replace('"', '"') + if self.encoding and type(data) == type(u''): + data = data.encode(self.encoding) + sgmllib.SGMLParser.feed(self, data) + sgmllib.SGMLParser.close(self) + + def normalize_attrs(self, attrs): + if not attrs: return attrs + # utility method to be called by descendants + attrs = dict([(k.lower(), v) for k, v in attrs]).items() + attrs = [(k, k in ('rel', 'type') and v.lower() or v) for k, v in attrs] + attrs.sort() + return attrs + + def unknown_starttag(self, tag, attrs): + # called for each start tag + # attrs is a list of (attr, value) tuples + # e.g. for 
, tag='pre', attrs=[('class', 'screen')]
+        if _debug: sys.stderr.write('_BaseHTMLProcessor, unknown_starttag, tag=%s\n' % tag)
+        uattrs = []
+        strattrs=''
+        if attrs:
+            for key, value in attrs:
+                value=value.replace('>','>').replace('<','<').replace('"','"')
+                value = self.bare_ampersand.sub("&", value)
+                # thanks to Kevin Marks for this breathtaking hack to deal with (valid) high-bit attribute values in UTF-8 feeds
+                if type(value) != type(u''):
+                    try:
+                        value = unicode(value, self.encoding)
+                    except:
+                        value = unicode(value, 'iso-8859-1')
+                uattrs.append((unicode(key, self.encoding), value))
+            strattrs = u''.join([u' %s="%s"' % (key, value) for key, value in uattrs])
+            if self.encoding:
+                try:
+                    strattrs=strattrs.encode(self.encoding)
+                except:
+                    pass
+        if tag in self.elements_no_end_tag:
+            self.pieces.append('<%(tag)s%(strattrs)s />' % locals())
+        else:
+            self.pieces.append('<%(tag)s%(strattrs)s>' % locals())
+
+    def unknown_endtag(self, tag):
+        # called for each end tag, e.g. for
, tag will be 'pre' + # Reconstruct the original end tag. + if tag not in self.elements_no_end_tag: + self.pieces.append("" % locals()) + + def handle_charref(self, ref): + # called for each character reference, e.g. for ' ', ref will be '160' + # Reconstruct the original character reference. + if ref.startswith('x'): + value = unichr(int(ref[1:],16)) + else: + value = unichr(int(ref)) + + if value in _cp1252.keys(): + self.pieces.append('&#%s;' % hex(ord(_cp1252[value]))[1:]) + else: + self.pieces.append('&#%(ref)s;' % locals()) + + def handle_entityref(self, ref): + # called for each entity reference, e.g. for '©', ref will be 'copy' + # Reconstruct the original entity reference. + if name2codepoint.has_key(ref): + self.pieces.append('&%(ref)s;' % locals()) + else: + self.pieces.append('&%(ref)s' % locals()) + + def handle_data(self, text): + # called for each block of plain text, i.e. outside of any tag and + # not containing any character or entity references + # Store the original text verbatim. + if _debug: sys.stderr.write('_BaseHTMLProcessor, handle_data, text=%s\n' % text) + self.pieces.append(text) + + def handle_comment(self, text): + # called for each HTML comment, e.g. + # Reconstruct the original comment. + self.pieces.append('' % locals()) + + def handle_pi(self, text): + # called for each processing instruction, e.g. + # Reconstruct original processing instruction. + self.pieces.append('' % locals()) + + def handle_decl(self, text): + # called for the DOCTYPE, if present, e.g. + # + # Reconstruct original DOCTYPE + self.pieces.append('' % locals()) + + _new_declname_match = re.compile(r'[a-zA-Z][-_.a-zA-Z0-9:]*\s*').match + def _scan_name(self, i, declstartpos): + rawdata = self.rawdata + n = len(rawdata) + if i == n: + return None, -1 + m = self._new_declname_match(rawdata, i) + if m: + s = m.group() + name = s.strip() + if (i + len(s)) == n: + return None, -1 # end of buffer + return name.lower(), m.end() + else: + self.handle_data(rawdata) +# self.updatepos(declstartpos, i) + return None, -1 + + def convert_charref(self, name): + return '&#%s;' % name + + def convert_entityref(self, name): + return '&%s;' % name + + def output(self): + '''Return processed HTML as a single string''' + return ''.join([str(p) for p in self.pieces]) + +class _LooseFeedParser(_FeedParserMixin, _BaseHTMLProcessor): + def __init__(self, baseuri, baselang, encoding, entities): + sgmllib.SGMLParser.__init__(self) + _FeedParserMixin.__init__(self, baseuri, baselang, encoding) + _BaseHTMLProcessor.__init__(self, encoding, 'application/xhtml+xml') + self.entities=entities + + def decodeEntities(self, element, data): + data = data.replace('<', '<') + data = data.replace('<', '<') + data = data.replace('<', '<') + data = data.replace('>', '>') + data = data.replace('>', '>') + data = data.replace('>', '>') + data = data.replace('&', '&') + data = data.replace('&', '&') + data = data.replace('"', '"') + data = data.replace('"', '"') + data = data.replace(''', ''') + data = data.replace(''', ''') + if self.contentparams.has_key('type') and not self.contentparams.get('type', 'xml').endswith('xml'): + data = data.replace('<', '<') + data = data.replace('>', '>') + data = data.replace('&', '&') + data = data.replace('"', '"') + data = data.replace(''', "'") + return data + + def strattrs(self, attrs): + return ''.join([' %s="%s"' % (n,v.replace('"','"')) for n,v in attrs]) + +class _MicroformatsParser: + STRING = 1 + DATE = 2 + URI = 3 + NODE = 4 + EMAIL = 5 + + known_xfn_relationships = ['contact', 'acquaintance', 'friend', 'met', 'co-worker', 'coworker', 'colleague', 'co-resident', 'coresident', 'neighbor', 'child', 'parent', 'sibling', 'brother', 'sister', 'spouse', 'wife', 'husband', 'kin', 'relative', 'muse', 'crush', 'date', 'sweetheart', 'me'] + known_binary_extensions = ['zip','rar','exe','gz','tar','tgz','tbz2','bz2','z','7z','dmg','img','sit','sitx','hqx','deb','rpm','bz2','jar','rar','iso','bin','msi','mp2','mp3','ogg','ogm','mp4','m4v','m4a','avi','wma','wmv'] + + def __init__(self, data, baseuri, encoding): + self.document = BeautifulSoup.BeautifulSoup(data) + self.baseuri = baseuri + self.encoding = encoding + if type(data) == type(u''): + data = data.encode(encoding) + self.tags = [] + self.enclosures = [] + self.xfn = [] + self.vcard = None + + def vcardEscape(self, s): + if type(s) in (type(''), type(u'')): + s = s.replace(',', '\\,').replace(';', '\\;').replace('\n', '\\n') + return s + + def vcardFold(self, s): + s = re.sub(';+$', '', s) + sFolded = '' + iMax = 75 + sPrefix = '' + while len(s) > iMax: + sFolded += sPrefix + s[:iMax] + '\n' + s = s[iMax:] + sPrefix = ' ' + iMax = 74 + sFolded += sPrefix + s + return sFolded + + def normalize(self, s): + return re.sub(r'\s+', ' ', s).strip() + + def unique(self, aList): + results = [] + for element in aList: + if element not in results: + results.append(element) + return results + + def toISO8601(self, dt): + return time.strftime('%Y-%m-%dT%H:%M:%SZ', dt) + + def getPropertyValue(self, elmRoot, sProperty, iPropertyType=4, bAllowMultiple=0, bAutoEscape=0): + all = lambda x: 1 + sProperty = sProperty.lower() + bFound = 0 + bNormalize = 1 + propertyMatch = {'class': re.compile(r'\b%s\b' % sProperty)} + if bAllowMultiple and (iPropertyType != self.NODE): + snapResults = [] + containers = elmRoot(['ul', 'ol'], propertyMatch) + for container in containers: + snapResults.extend(container('li')) + bFound = (len(snapResults) != 0) + if not bFound: + snapResults = elmRoot(all, propertyMatch) + bFound = (len(snapResults) != 0) + if (not bFound) and (sProperty == 'value'): + snapResults = elmRoot('pre') + bFound = (len(snapResults) != 0) + bNormalize = not bFound + if not bFound: + snapResults = [elmRoot] + bFound = (len(snapResults) != 0) + arFilter = [] + if sProperty == 'vcard': + snapFilter = elmRoot(all, propertyMatch) + for node in snapFilter: + if node.findParent(all, propertyMatch): + arFilter.append(node) + arResults = [] + for node in snapResults: + if node not in arFilter: + arResults.append(node) + bFound = (len(arResults) != 0) + if not bFound: + if bAllowMultiple: return [] + elif iPropertyType == self.STRING: return '' + elif iPropertyType == self.DATE: return None + elif iPropertyType == self.URI: return '' + elif iPropertyType == self.NODE: return None + else: return None + arValues = [] + for elmResult in arResults: + sValue = None + if iPropertyType == self.NODE: + if bAllowMultiple: + arValues.append(elmResult) + continue + else: + return elmResult + sNodeName = elmResult.name.lower() + if (iPropertyType == self.EMAIL) and (sNodeName == 'a'): + sValue = (elmResult.get('href') or '').split('mailto:').pop().split('?')[0] + if sValue: + sValue = bNormalize and self.normalize(sValue) or sValue.strip() + if (not sValue) and (sNodeName == 'abbr'): + sValue = elmResult.get('title') + if sValue: + sValue = bNormalize and self.normalize(sValue) or sValue.strip() + if (not sValue) and (iPropertyType == self.URI): + if sNodeName == 'a': sValue = elmResult.get('href') + elif sNodeName == 'img': sValue = elmResult.get('src') + elif sNodeName == 'object': sValue = elmResult.get('data') + if sValue: + sValue = bNormalize and self.normalize(sValue) or sValue.strip() + if (not sValue) and (sNodeName == 'img'): + sValue = elmResult.get('alt') + if sValue: + sValue = bNormalize and self.normalize(sValue) or sValue.strip() + if not sValue: + sValue = elmResult.renderContents() + sValue = re.sub(r'<\S[^>]*>', '', sValue) + sValue = sValue.replace('\r\n', '\n') + sValue = sValue.replace('\r', '\n') + if sValue: + sValue = bNormalize and self.normalize(sValue) or sValue.strip() + if not sValue: continue + if iPropertyType == self.DATE: + sValue = _parse_date_iso8601(sValue) + if bAllowMultiple: + arValues.append(bAutoEscape and self.vcardEscape(sValue) or sValue) + else: + return bAutoEscape and self.vcardEscape(sValue) or sValue + return arValues + + def findVCards(self, elmRoot, bAgentParsing=0): + sVCards = '' + + if not bAgentParsing: + arCards = self.getPropertyValue(elmRoot, 'vcard', bAllowMultiple=1) + else: + arCards = [elmRoot] + + for elmCard in arCards: + arLines = [] + + def processSingleString(sProperty): + sValue = self.getPropertyValue(elmCard, sProperty, self.STRING, bAutoEscape=1) + if sValue: + arLines.append(self.vcardFold(sProperty.upper() + ':' + sValue)) + return sValue or '' + + def processSingleURI(sProperty): + sValue = self.getPropertyValue(elmCard, sProperty, self.URI) + if sValue: + sContentType = '' + sEncoding = '' + sValueKey = '' + if sValue.startswith('data:'): + sEncoding = ';ENCODING=b' + sContentType = sValue.split(';')[0].split('/').pop() + sValue = sValue.split(',', 1).pop() + else: + elmValue = self.getPropertyValue(elmCard, sProperty) + if elmValue: + if sProperty != 'url': + sValueKey = ';VALUE=uri' + sContentType = elmValue.get('type', '').strip().split('/').pop().strip() + sContentType = sContentType.upper() + if sContentType == 'OCTET-STREAM': + sContentType = '' + if sContentType: + sContentType = ';TYPE=' + sContentType.upper() + arLines.append(self.vcardFold(sProperty.upper() + sEncoding + sContentType + sValueKey + ':' + sValue)) + + def processTypeValue(sProperty, arDefaultType, arForceType=None): + arResults = self.getPropertyValue(elmCard, sProperty, bAllowMultiple=1) + for elmResult in arResults: + arType = self.getPropertyValue(elmResult, 'type', self.STRING, 1, 1) + if arForceType: + arType = self.unique(arForceType + arType) + if not arType: + arType = arDefaultType + sValue = self.getPropertyValue(elmResult, 'value', self.EMAIL, 0) + if sValue: + arLines.append(self.vcardFold(sProperty.upper() + ';TYPE=' + ','.join(arType) + ':' + sValue)) + + # AGENT + # must do this before all other properties because it is destructive + # (removes nested class="vcard" nodes so they don't interfere with + # this vcard's other properties) + arAgent = self.getPropertyValue(elmCard, 'agent', bAllowMultiple=1) + for elmAgent in arAgent: + if re.compile(r'\bvcard\b').search(elmAgent.get('class')): + sAgentValue = self.findVCards(elmAgent, 1) + '\n' + sAgentValue = sAgentValue.replace('\n', '\\n') + sAgentValue = sAgentValue.replace(';', '\\;') + if sAgentValue: + arLines.append(self.vcardFold('AGENT:' + sAgentValue)) + elmAgent['class'] = '' + elmAgent.contents = [] + else: + sAgentValue = self.getPropertyValue(elmAgent, 'value', self.URI, bAutoEscape=1); + if sAgentValue: + arLines.append(self.vcardFold('AGENT;VALUE=uri:' + sAgentValue)) + + # FN (full name) + sFN = processSingleString('fn') + + # N (name) + elmName = self.getPropertyValue(elmCard, 'n') + if elmName: + sFamilyName = self.getPropertyValue(elmName, 'family-name', self.STRING, bAutoEscape=1) + sGivenName = self.getPropertyValue(elmName, 'given-name', self.STRING, bAutoEscape=1) + arAdditionalNames = self.getPropertyValue(elmName, 'additional-name', self.STRING, 1, 1) + self.getPropertyValue(elmName, 'additional-names', self.STRING, 1, 1) + arHonorificPrefixes = self.getPropertyValue(elmName, 'honorific-prefix', self.STRING, 1, 1) + self.getPropertyValue(elmName, 'honorific-prefixes', self.STRING, 1, 1) + arHonorificSuffixes = self.getPropertyValue(elmName, 'honorific-suffix', self.STRING, 1, 1) + self.getPropertyValue(elmName, 'honorific-suffixes', self.STRING, 1, 1) + arLines.append(self.vcardFold('N:' + sFamilyName + ';' + + sGivenName + ';' + + ','.join(arAdditionalNames) + ';' + + ','.join(arHonorificPrefixes) + ';' + + ','.join(arHonorificSuffixes))) + elif sFN: + # implied "N" optimization + # http://microformats.org/wiki/hcard#Implied_.22N.22_Optimization + arNames = self.normalize(sFN).split() + if len(arNames) == 2: + bFamilyNameFirst = (arNames[0].endswith(',') or + len(arNames[1]) == 1 or + ((len(arNames[1]) == 2) and (arNames[1].endswith('.')))) + if bFamilyNameFirst: + arLines.append(self.vcardFold('N:' + arNames[0] + ';' + arNames[1])) + else: + arLines.append(self.vcardFold('N:' + arNames[1] + ';' + arNames[0])) + + # SORT-STRING + sSortString = self.getPropertyValue(elmCard, 'sort-string', self.STRING, bAutoEscape=1) + if sSortString: + arLines.append(self.vcardFold('SORT-STRING:' + sSortString)) + + # NICKNAME + arNickname = self.getPropertyValue(elmCard, 'nickname', self.STRING, 1, 1) + if arNickname: + arLines.append(self.vcardFold('NICKNAME:' + ','.join(arNickname))) + + # PHOTO + processSingleURI('photo') + + # BDAY + dtBday = self.getPropertyValue(elmCard, 'bday', self.DATE) + if dtBday: + arLines.append(self.vcardFold('BDAY:' + self.toISO8601(dtBday))) + + # ADR (address) + arAdr = self.getPropertyValue(elmCard, 'adr', bAllowMultiple=1) + for elmAdr in arAdr: + arType = self.getPropertyValue(elmAdr, 'type', self.STRING, 1, 1) + if not arType: + arType = ['intl','postal','parcel','work'] # default adr types, see RFC 2426 section 3.2.1 + sPostOfficeBox = self.getPropertyValue(elmAdr, 'post-office-box', self.STRING, 0, 1) + sExtendedAddress = self.getPropertyValue(elmAdr, 'extended-address', self.STRING, 0, 1) + sStreetAddress = self.getPropertyValue(elmAdr, 'street-address', self.STRING, 0, 1) + sLocality = self.getPropertyValue(elmAdr, 'locality', self.STRING, 0, 1) + sRegion = self.getPropertyValue(elmAdr, 'region', self.STRING, 0, 1) + sPostalCode = self.getPropertyValue(elmAdr, 'postal-code', self.STRING, 0, 1) + sCountryName = self.getPropertyValue(elmAdr, 'country-name', self.STRING, 0, 1) + arLines.append(self.vcardFold('ADR;TYPE=' + ','.join(arType) + ':' + + sPostOfficeBox + ';' + + sExtendedAddress + ';' + + sStreetAddress + ';' + + sLocality + ';' + + sRegion + ';' + + sPostalCode + ';' + + sCountryName)) + + # LABEL + processTypeValue('label', ['intl','postal','parcel','work']) + + # TEL (phone number) + processTypeValue('tel', ['voice']) + + # EMAIL + processTypeValue('email', ['internet'], ['internet']) + + # MAILER + processSingleString('mailer') + + # TZ (timezone) + processSingleString('tz') + + # GEO (geographical information) + elmGeo = self.getPropertyValue(elmCard, 'geo') + if elmGeo: + sLatitude = self.getPropertyValue(elmGeo, 'latitude', self.STRING, 0, 1) + sLongitude = self.getPropertyValue(elmGeo, 'longitude', self.STRING, 0, 1) + arLines.append(self.vcardFold('GEO:' + sLatitude + ';' + sLongitude)) + + # TITLE + processSingleString('title') + + # ROLE + processSingleString('role') + + # LOGO + processSingleURI('logo') + + # ORG (organization) + elmOrg = self.getPropertyValue(elmCard, 'org') + if elmOrg: + sOrganizationName = self.getPropertyValue(elmOrg, 'organization-name', self.STRING, 0, 1) + if not sOrganizationName: + # implied "organization-name" optimization + # http://microformats.org/wiki/hcard#Implied_.22organization-name.22_Optimization + sOrganizationName = self.getPropertyValue(elmCard, 'org', self.STRING, 0, 1) + if sOrganizationName: + arLines.append(self.vcardFold('ORG:' + sOrganizationName)) + else: + arOrganizationUnit = self.getPropertyValue(elmOrg, 'organization-unit', self.STRING, 1, 1) + arLines.append(self.vcardFold('ORG:' + sOrganizationName + ';' + ';'.join(arOrganizationUnit))) + + # CATEGORY + arCategory = self.getPropertyValue(elmCard, 'category', self.STRING, 1, 1) + self.getPropertyValue(elmCard, 'categories', self.STRING, 1, 1) + if arCategory: + arLines.append(self.vcardFold('CATEGORIES:' + ','.join(arCategory))) + + # NOTE + processSingleString('note') + + # REV + processSingleString('rev') + + # SOUND + processSingleURI('sound') + + # UID + processSingleString('uid') + + # URL + processSingleURI('url') + + # CLASS + processSingleString('class') + + # KEY + processSingleURI('key') + + if arLines: + arLines = ['BEGIN:vCard','VERSION:3.0'] + arLines + ['END:vCard'] + sVCards += '\n'.join(arLines) + '\n' + + return sVCards.strip() + + def isProbablyDownloadable(self, elm): + attrsD = elm.attrMap + if not attrsD.has_key('href'): return 0 + linktype = attrsD.get('type', '').strip() + if linktype.startswith('audio/') or \ + linktype.startswith('video/') or \ + (linktype.startswith('application/') and not linktype.endswith('xml')): + return 1 + path = urlparse.urlparse(attrsD['href'])[2] + if path.find('.') == -1: return 0 + fileext = path.split('.').pop().lower() + return fileext in self.known_binary_extensions + + def findTags(self): + all = lambda x: 1 + for elm in self.document(all, {'rel': re.compile(r'\btag\b')}): + href = elm.get('href') + if not href: continue + urlscheme, domain, path, params, query, fragment = \ + urlparse.urlparse(_urljoin(self.baseuri, href)) + segments = path.split('/') + tag = segments.pop() + if not tag: + tag = segments.pop() + tagscheme = urlparse.urlunparse((urlscheme, domain, '/'.join(segments), '', '', '')) + if not tagscheme.endswith('/'): + tagscheme += '/' + self.tags.append(FeedParserDict({"term": tag, "scheme": tagscheme, "label": elm.string or ''})) + + def findEnclosures(self): + all = lambda x: 1 + enclosure_match = re.compile(r'\benclosure\b') + for elm in self.document(all, {'href': re.compile(r'.+')}): + if not enclosure_match.search(elm.get('rel', '')) and not self.isProbablyDownloadable(elm): continue + if elm.attrMap not in self.enclosures: + self.enclosures.append(elm.attrMap) + if elm.string and not elm.get('title'): + self.enclosures[-1]['title'] = elm.string + + def findXFN(self): + all = lambda x: 1 + for elm in self.document(all, {'rel': re.compile('.+'), 'href': re.compile('.+')}): + rels = elm.get('rel', '').split() + xfn_rels = [] + for rel in rels: + if rel in self.known_xfn_relationships: + xfn_rels.append(rel) + if xfn_rels: + self.xfn.append({"relationships": xfn_rels, "href": elm.get('href', ''), "name": elm.string}) + +def _parseMicroformats(htmlSource, baseURI, encoding): + if not BeautifulSoup: return + if _debug: sys.stderr.write('entering _parseMicroformats\n') + p = _MicroformatsParser(htmlSource, baseURI, encoding) + p.vcard = p.findVCards(p.document) + p.findTags() + p.findEnclosures() + p.findXFN() + return {"tags": p.tags, "enclosures": p.enclosures, "xfn": p.xfn, "vcard": p.vcard} + +class _RelativeURIResolver(_BaseHTMLProcessor): + relative_uris = [('a', 'href'), + ('applet', 'codebase'), + ('area', 'href'), + ('blockquote', 'cite'), + ('body', 'background'), + ('del', 'cite'), + ('form', 'action'), + ('frame', 'longdesc'), + ('frame', 'src'), + ('iframe', 'longdesc'), + ('iframe', 'src'), + ('head', 'profile'), + ('img', 'longdesc'), + ('img', 'src'), + ('img', 'usemap'), + ('input', 'src'), + ('input', 'usemap'), + ('ins', 'cite'), + ('link', 'href'), + ('object', 'classid'), + ('object', 'codebase'), + ('object', 'data'), + ('object', 'usemap'), + ('q', 'cite'), + ('script', 'src')] + + def __init__(self, baseuri, encoding, type): + _BaseHTMLProcessor.__init__(self, encoding, type) + self.baseuri = baseuri + + def resolveURI(self, uri): + return _urljoin(self.baseuri, uri.strip()) + + def unknown_starttag(self, tag, attrs): + if _debug: + sys.stderr.write('tag: [%s] with attributes: [%s]\n' % (tag, str(attrs))) + attrs = self.normalize_attrs(attrs) + attrs = [(key, ((tag, key) in self.relative_uris) and self.resolveURI(value) or value) for key, value in attrs] + _BaseHTMLProcessor.unknown_starttag(self, tag, attrs) + +def _resolveRelativeURIs(htmlSource, baseURI, encoding, type): + if _debug: + sys.stderr.write('entering _resolveRelativeURIs\n') + + p = _RelativeURIResolver(baseURI, encoding, type) + p.feed(htmlSource) + return p.output() + +class _HTMLSanitizer(_BaseHTMLProcessor): + acceptable_elements = ['a', 'abbr', 'acronym', 'address', 'area', 'article', + 'aside', 'audio', 'b', 'big', 'blockquote', 'br', 'button', 'canvas', + 'caption', 'center', 'cite', 'code', 'col', 'colgroup', 'command', + 'datagrid', 'datalist', 'dd', 'del', 'details', 'dfn', 'dialog', 'dir', + 'div', 'dl', 'dt', 'em', 'event-source', 'fieldset', 'figure', 'footer', + 'font', 'form', 'header', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'hr', 'i', + 'img', 'input', 'ins', 'keygen', 'kbd', 'label', 'legend', 'li', 'm', 'map', + 'menu', 'meter', 'multicol', 'nav', 'nextid', 'ol', 'output', 'optgroup', + 'option', 'p', 'pre', 'progress', 'q', 's', 'samp', 'section', 'select', + 'small', 'sound', 'source', 'spacer', 'span', 'strike', 'strong', 'sub', + 'sup', 'table', 'tbody', 'td', 'textarea', 'time', 'tfoot', 'th', 'thead', + 'tr', 'tt', 'u', 'ul', 'var', 'video', 'noscript'] + + acceptable_attributes = ['abbr', 'accept', 'accept-charset', 'accesskey', + 'action', 'align', 'alt', 'autocomplete', 'autofocus', 'axis', + 'background', 'balance', 'bgcolor', 'bgproperties', 'border', + 'bordercolor', 'bordercolordark', 'bordercolorlight', 'bottompadding', + 'cellpadding', 'cellspacing', 'ch', 'challenge', 'char', 'charoff', + 'choff', 'charset', 'checked', 'cite', 'class', 'clear', 'color', 'cols', + 'colspan', 'compact', 'contenteditable', 'controls', 'coords', 'data', + 'datafld', 'datapagesize', 'datasrc', 'datetime', 'default', 'delay', + 'dir', 'disabled', 'draggable', 'dynsrc', 'enctype', 'end', 'face', 'for', + 'form', 'frame', 'galleryimg', 'gutter', 'headers', 'height', 'hidefocus', + 'hidden', 'high', 'href', 'hreflang', 'hspace', 'icon', 'id', 'inputmode', + 'ismap', 'keytype', 'label', 'leftspacing', 'lang', 'list', 'longdesc', + 'loop', 'loopcount', 'loopend', 'loopstart', 'low', 'lowsrc', 'max', + 'maxlength', 'media', 'method', 'min', 'multiple', 'name', 'nohref', + 'noshade', 'nowrap', 'open', 'optimum', 'pattern', 'ping', 'point-size', + 'prompt', 'pqg', 'radiogroup', 'readonly', 'rel', 'repeat-max', + 'repeat-min', 'replace', 'required', 'rev', 'rightspacing', 'rows', + 'rowspan', 'rules', 'scope', 'selected', 'shape', 'size', 'span', 'src', + 'start', 'step', 'summary', 'suppress', 'tabindex', 'target', 'template', + 'title', 'toppadding', 'type', 'unselectable', 'usemap', 'urn', 'valign', + 'value', 'variable', 'volume', 'vspace', 'vrml', 'width', 'wrap', + 'xml:lang'] + + unacceptable_elements_with_end_tag = ['script', 'applet', 'style'] + + acceptable_css_properties = ['azimuth', 'background-color', + 'border-bottom-color', 'border-collapse', 'border-color', + 'border-left-color', 'border-right-color', 'border-top-color', 'clear', + 'color', 'cursor', 'direction', 'display', 'elevation', 'float', 'font', + 'font-family', 'font-size', 'font-style', 'font-variant', 'font-weight', + 'height', 'letter-spacing', 'line-height', 'overflow', 'pause', + 'pause-after', 'pause-before', 'pitch', 'pitch-range', 'richness', + 'speak', 'speak-header', 'speak-numeral', 'speak-punctuation', + 'speech-rate', 'stress', 'text-align', 'text-decoration', 'text-indent', + 'unicode-bidi', 'vertical-align', 'voice-family', 'volume', + 'white-space', 'width'] + + # survey of common keywords found in feeds + acceptable_css_keywords = ['auto', 'aqua', 'black', 'block', 'blue', + 'bold', 'both', 'bottom', 'brown', 'center', 'collapse', 'dashed', + 'dotted', 'fuchsia', 'gray', 'green', '!important', 'italic', 'left', + 'lime', 'maroon', 'medium', 'none', 'navy', 'normal', 'nowrap', 'olive', + 'pointer', 'purple', 'red', 'right', 'solid', 'silver', 'teal', 'top', + 'transparent', 'underline', 'white', 'yellow'] + + valid_css_values = re.compile('^(#[0-9a-f]+|rgb\(\d+%?,\d*%?,?\d*%?\)?|' + + '\d{0,2}\.?\d{0,2}(cm|em|ex|in|mm|pc|pt|px|%|,|\))?)$') + + mathml_elements = ['annotation', 'annotation-xml', 'maction', 'math', + 'merror', 'mfenced', 'mfrac', 'mi', 'mmultiscripts', 'mn', 'mo', 'mover', 'mpadded', + 'mphantom', 'mprescripts', 'mroot', 'mrow', 'mspace', 'msqrt', 'mstyle', + 'msub', 'msubsup', 'msup', 'mtable', 'mtd', 'mtext', 'mtr', 'munder', + 'munderover', 'none', 'semantics'] + + mathml_attributes = ['actiontype', 'align', 'columnalign', 'columnalign', + 'columnalign', 'close', 'columnlines', 'columnspacing', 'columnspan', 'depth', + 'display', 'displaystyle', 'encoding', 'equalcolumns', 'equalrows', + 'fence', 'fontstyle', 'fontweight', 'frame', 'height', 'linethickness', + 'lspace', 'mathbackground', 'mathcolor', 'mathvariant', 'mathvariant', + 'maxsize', 'minsize', 'open', 'other', 'rowalign', 'rowalign', 'rowalign', + 'rowlines', 'rowspacing', 'rowspan', 'rspace', 'scriptlevel', 'selection', + 'separator', 'separators', 'stretchy', 'width', 'width', 'xlink:href', + 'xlink:show', 'xlink:type', 'xmlns', 'xmlns:xlink'] + + # svgtiny - foreignObject + linearGradient + radialGradient + stop + svg_elements = ['a', 'animate', 'animateColor', 'animateMotion', + 'animateTransform', 'circle', 'defs', 'desc', 'ellipse', 'foreignObject', + 'font-face', 'font-face-name', 'font-face-src', 'g', 'glyph', 'hkern', + 'linearGradient', 'line', 'marker', 'metadata', 'missing-glyph', 'mpath', + 'path', 'polygon', 'polyline', 'radialGradient', 'rect', 'set', 'stop', + 'svg', 'switch', 'text', 'title', 'tspan', 'use'] + + # svgtiny + class + opacity + offset + xmlns + xmlns:xlink + svg_attributes = ['accent-height', 'accumulate', 'additive', 'alphabetic', + 'arabic-form', 'ascent', 'attributeName', 'attributeType', + 'baseProfile', 'bbox', 'begin', 'by', 'calcMode', 'cap-height', + 'class', 'color', 'color-rendering', 'content', 'cx', 'cy', 'd', 'dx', + 'dy', 'descent', 'display', 'dur', 'end', 'fill', 'fill-opacity', + 'fill-rule', 'font-family', 'font-size', 'font-stretch', 'font-style', + 'font-variant', 'font-weight', 'from', 'fx', 'fy', 'g1', 'g2', + 'glyph-name', 'gradientUnits', 'hanging', 'height', 'horiz-adv-x', + 'horiz-origin-x', 'id', 'ideographic', 'k', 'keyPoints', 'keySplines', + 'keyTimes', 'lang', 'mathematical', 'marker-end', 'marker-mid', + 'marker-start', 'markerHeight', 'markerUnits', 'markerWidth', 'max', + 'min', 'name', 'offset', 'opacity', 'orient', 'origin', + 'overline-position', 'overline-thickness', 'panose-1', 'path', + 'pathLength', 'points', 'preserveAspectRatio', 'r', 'refX', 'refY', + 'repeatCount', 'repeatDur', 'requiredExtensions', 'requiredFeatures', + 'restart', 'rotate', 'rx', 'ry', 'slope', 'stemh', 'stemv', + 'stop-color', 'stop-opacity', 'strikethrough-position', + 'strikethrough-thickness', 'stroke', 'stroke-dasharray', + 'stroke-dashoffset', 'stroke-linecap', 'stroke-linejoin', + 'stroke-miterlimit', 'stroke-opacity', 'stroke-width', 'systemLanguage', + 'target', 'text-anchor', 'to', 'transform', 'type', 'u1', 'u2', + 'underline-position', 'underline-thickness', 'unicode', 'unicode-range', + 'units-per-em', 'values', 'version', 'viewBox', 'visibility', 'width', + 'widths', 'x', 'x-height', 'x1', 'x2', 'xlink:actuate', 'xlink:arcrole', + 'xlink:href', 'xlink:role', 'xlink:show', 'xlink:title', 'xlink:type', + 'xml:base', 'xml:lang', 'xml:space', 'xmlns', 'xmlns:xlink', 'y', 'y1', + 'y2', 'zoomAndPan'] + + svg_attr_map = None + svg_elem_map = None + + acceptable_svg_properties = [ 'fill', 'fill-opacity', 'fill-rule', + 'stroke', 'stroke-width', 'stroke-linecap', 'stroke-linejoin', + 'stroke-opacity'] + + def reset(self): + _BaseHTMLProcessor.reset(self) + self.unacceptablestack = 0 + self.mathmlOK = 0 + self.svgOK = 0 + + def unknown_starttag(self, tag, attrs): + acceptable_attributes = self.acceptable_attributes + keymap = {} + if not tag in self.acceptable_elements or self.svgOK: + if tag in self.unacceptable_elements_with_end_tag: + self.unacceptablestack += 1 + + # add implicit namespaces to html5 inline svg/mathml + if self.type.endswith('html'): + if not dict(attrs).get('xmlns'): + if tag=='svg': + attrs.append( ('xmlns','http://www.w3.org/2000/svg') ) + if tag=='math': + attrs.append( ('xmlns','http://www.w3.org/1998/Math/MathML') ) + + # not otherwise acceptable, perhaps it is MathML or SVG? + if tag=='math' and ('xmlns','http://www.w3.org/1998/Math/MathML') in attrs: + self.mathmlOK += 1 + if tag=='svg' and ('xmlns','http://www.w3.org/2000/svg') in attrs: + self.svgOK += 1 + + # chose acceptable attributes based on tag class, else bail + if self.mathmlOK and tag in self.mathml_elements: + acceptable_attributes = self.mathml_attributes + elif self.svgOK and tag in self.svg_elements: + # for most vocabularies, lowercasing is a good idea. Many + # svg elements, however, are camel case + if not self.svg_attr_map: + lower=[attr.lower() for attr in self.svg_attributes] + mix=[a for a in self.svg_attributes if a not in lower] + self.svg_attributes = lower + self.svg_attr_map = dict([(a.lower(),a) for a in mix]) + + lower=[attr.lower() for attr in self.svg_elements] + mix=[a for a in self.svg_elements if a not in lower] + self.svg_elements = lower + self.svg_elem_map = dict([(a.lower(),a) for a in mix]) + acceptable_attributes = self.svg_attributes + tag = self.svg_elem_map.get(tag,tag) + keymap = self.svg_attr_map + elif not tag in self.acceptable_elements: + return + + # declare xlink namespace, if needed + if self.mathmlOK or self.svgOK: + if filter(lambda (n,v): n.startswith('xlink:'),attrs): + if not ('xmlns:xlink','http://www.w3.org/1999/xlink') in attrs: + attrs.append(('xmlns:xlink','http://www.w3.org/1999/xlink')) + + clean_attrs = [] + for key, value in self.normalize_attrs(attrs): + if key in acceptable_attributes: + key=keymap.get(key,key) + clean_attrs.append((key,value)) + elif key=='style': + clean_value = self.sanitize_style(value) + if clean_value: clean_attrs.append((key,clean_value)) + _BaseHTMLProcessor.unknown_starttag(self, tag, clean_attrs) + + def unknown_endtag(self, tag): + if not tag in self.acceptable_elements: + if tag in self.unacceptable_elements_with_end_tag: + self.unacceptablestack -= 1 + if self.mathmlOK and tag in self.mathml_elements: + if tag == 'math' and self.mathmlOK: self.mathmlOK -= 1 + elif self.svgOK and tag in self.svg_elements: + tag = self.svg_elem_map.get(tag,tag) + if tag == 'svg' and self.svgOK: self.svgOK -= 1 + else: + return + _BaseHTMLProcessor.unknown_endtag(self, tag) + + def handle_pi(self, text): + pass + + def handle_decl(self, text): + pass + + def handle_data(self, text): + if not self.unacceptablestack: + _BaseHTMLProcessor.handle_data(self, text) + + def sanitize_style(self, style): + # disallow urls + style=re.compile('url\s*\(\s*[^\s)]+?\s*\)\s*').sub(' ',style) + + # gauntlet + if not re.match("""^([:,;#%.\sa-zA-Z0-9!]|\w-\w|'[\s\w]+'|"[\s\w]+"|\([\d,\s]+\))*$""", style): return '' + # This replaced a regexp that used re.match and was prone to pathological back-tracking. + if re.sub("\s*[-\w]+\s*:\s*[^:;]*;?", '', style).strip(): return '' + + clean = [] + for prop,value in re.findall("([-\w]+)\s*:\s*([^:;]*)",style): + if not value: continue + if prop.lower() in self.acceptable_css_properties: + clean.append(prop + ': ' + value + ';') + elif prop.split('-')[0].lower() in ['background','border','margin','padding']: + for keyword in value.split(): + if not keyword in self.acceptable_css_keywords and \ + not self.valid_css_values.match(keyword): + break + else: + clean.append(prop + ': ' + value + ';') + elif self.svgOK and prop.lower() in self.acceptable_svg_properties: + clean.append(prop + ': ' + value + ';') + + return ' '.join(clean) + + +def _sanitizeHTML(htmlSource, encoding, type): + p = _HTMLSanitizer(encoding, type) + p.feed(htmlSource) + data = p.output() + if TIDY_MARKUP: + # loop through list of preferred Tidy interfaces looking for one that's installed, + # then set up a common _tidy function to wrap the interface-specific API. + _tidy = None + for tidy_interface in PREFERRED_TIDY_INTERFACES: + try: + if tidy_interface == "uTidy": + from tidy import parseString as _utidy + def _tidy(data, **kwargs): + return str(_utidy(data, **kwargs)) + break + elif tidy_interface == "mxTidy": + from mx.Tidy import Tidy as _mxtidy + def _tidy(data, **kwargs): + nerrors, nwarnings, data, errordata = _mxtidy.tidy(data, **kwargs) + return data + break + except: + pass + if _tidy: + utf8 = type(data) == type(u'') + if utf8: + data = data.encode('utf-8') + data = _tidy(data, output_xhtml=1, numeric_entities=1, wrap=0, char_encoding="utf8") + if utf8: + data = unicode(data, 'utf-8') + if data.count(''): + data = data.split('>', 1)[1] + if data.count('= '2.3.3' + assert base64 != None + user, passw = base64.decodestring(req.headers['Authorization'].split(' ')[1]).split(':') + realm = re.findall('realm="([^"]*)"', headers['WWW-Authenticate'])[0] + self.add_password(realm, host, user, passw) + retry = self.http_error_auth_reqed('www-authenticate', host, req, headers) + self.reset_retry_count() + return retry + except: + return self.http_error_default(req, fp, code, msg, headers) + +def _open_resource(url_file_stream_or_string, etag, modified, agent, referrer, handlers): + """URL, filename, or string --> stream + + This function lets you define parsers that take any input source + (URL, pathname to local or network file, or actual data as a string) + and deal with it in a uniform manner. Returned object is guaranteed + to have all the basic stdio read methods (read, readline, readlines). + Just .close() the object when you're done with it. + + If the etag argument is supplied, it will be used as the value of an + If-None-Match request header. + + If the modified argument is supplied, it can be a tuple of 9 integers + (as returned by gmtime() in the standard Python time module) or a date + string in any format supported by feedparser. Regardless, it MUST + be in GMT (Greenwich Mean Time). It will be reformatted into an + RFC 1123-compliant date and used as the value of an If-Modified-Since + request header. + + If the agent argument is supplied, it will be used as the value of a + User-Agent request header. + + If the referrer argument is supplied, it will be used as the value of a + Referer[sic] request header. + + If handlers is supplied, it is a list of handlers used to build a + urllib2 opener. + """ + + if hasattr(url_file_stream_or_string, 'read'): + return url_file_stream_or_string + + if url_file_stream_or_string == '-': + return sys.stdin + + if urlparse.urlparse(url_file_stream_or_string)[0] in ('http', 'https', 'ftp'): + if not agent: + agent = USER_AGENT + # test for inline user:password for basic auth + auth = None + if base64: + urltype, rest = urllib.splittype(url_file_stream_or_string) + realhost, rest = urllib.splithost(rest) + if realhost: + user_passwd, realhost = urllib.splituser(realhost) + if user_passwd: + url_file_stream_or_string = '%s://%s%s' % (urltype, realhost, rest) + auth = base64.encodestring(user_passwd).strip() + + # iri support + try: + if isinstance(url_file_stream_or_string,unicode): + url_file_stream_or_string = url_file_stream_or_string.encode('idna') + else: + url_file_stream_or_string = url_file_stream_or_string.decode('utf-8').encode('idna') + except: + pass + + # try to open with urllib2 (to use optional headers) + request = urllib2.Request(url_file_stream_or_string) + request.add_header('User-Agent', agent) + if etag: + request.add_header('If-None-Match', etag) + if type(modified) == type(''): + modified = _parse_date(modified) + if modified: + # format into an RFC 1123-compliant timestamp. We can't use + # time.strftime() since the %a and %b directives can be affected + # by the current locale, but RFC 2616 states that dates must be + # in English. + short_weekdays = ['Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat', 'Sun'] + months = ['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'] + request.add_header('If-Modified-Since', '%s, %02d %s %04d %02d:%02d:%02d GMT' % (short_weekdays[modified[6]], modified[2], months[modified[1] - 1], modified[0], modified[3], modified[4], modified[5])) + if referrer: + request.add_header('Referer', referrer) + if gzip and zlib: + request.add_header('Accept-encoding', 'gzip, deflate') + elif gzip: + request.add_header('Accept-encoding', 'gzip') + elif zlib: + request.add_header('Accept-encoding', 'deflate') + else: + request.add_header('Accept-encoding', '') + if auth: + request.add_header('Authorization', 'Basic %s' % auth) + if ACCEPT_HEADER: + request.add_header('Accept', ACCEPT_HEADER) + request.add_header('A-IM', 'feed') # RFC 3229 support + opener = apply(urllib2.build_opener, tuple([_FeedURLHandler()] + handlers)) + opener.addheaders = [] # RMK - must clear so we only send our custom User-Agent + try: + return opener.open(request) + finally: + opener.close() # JohnD + + # try to open with native open function (if url_file_stream_or_string is a filename) + try: + return open(url_file_stream_or_string) + except: + pass + + # treat url_file_stream_or_string as string + return _StringIO(str(url_file_stream_or_string)) + +_date_handlers = [] +def registerDateHandler(func): + '''Register a date handler function (takes string, returns 9-tuple date in GMT)''' + _date_handlers.insert(0, func) + +# ISO-8601 date parsing routines written by Fazal Majid. +# The ISO 8601 standard is very convoluted and irregular - a full ISO 8601 +# parser is beyond the scope of feedparser and would be a worthwhile addition +# to the Python library. +# A single regular expression cannot parse ISO 8601 date formats into groups +# as the standard is highly irregular (for instance is 030104 2003-01-04 or +# 0301-04-01), so we use templates instead. +# Please note the order in templates is significant because we need a +# greedy match. +_iso8601_tmpl = ['YYYY-?MM-?DD', 'YYYY-0MM?-?DD', 'YYYY-MM', 'YYYY-?OOO', + 'YY-?MM-?DD', 'YY-?OOO', 'YYYY', + '-YY-?MM', '-OOO', '-YY', + '--MM-?DD', '--MM', + '---DD', + 'CC', ''] +_iso8601_re = [ + tmpl.replace( + 'YYYY', r'(?P\d{4})').replace( + 'YY', r'(?P\d\d)').replace( + 'MM', r'(?P[01]\d)').replace( + 'DD', r'(?P[0123]\d)').replace( + 'OOO', r'(?P[0123]\d\d)').replace( + 'CC', r'(?P\d\d$)') + + r'(T?(?P\d{2}):(?P\d{2})' + + r'(:(?P\d{2}))?' + + r'(\.(?P\d+))?' + + r'(?P[+-](?P\d{2})(:(?P\d{2}))?|Z)?)?' + for tmpl in _iso8601_tmpl] +del tmpl +_iso8601_matches = [re.compile(regex).match for regex in _iso8601_re] +del regex +def _parse_date_iso8601(dateString): + '''Parse a variety of ISO-8601-compatible formats like 20040105''' + m = None + for _iso8601_match in _iso8601_matches: + m = _iso8601_match(dateString) + if m: break + if not m: return + if m.span() == (0, 0): return + params = m.groupdict() + ordinal = params.get('ordinal', 0) + if ordinal: + ordinal = int(ordinal) + else: + ordinal = 0 + year = params.get('year', '--') + if not year or year == '--': + year = time.gmtime()[0] + elif len(year) == 2: + # ISO 8601 assumes current century, i.e. 93 -> 2093, NOT 1993 + year = 100 * int(time.gmtime()[0] / 100) + int(year) + else: + year = int(year) + month = params.get('month', '-') + if not month or month == '-': + # ordinals are NOT normalized by mktime, we simulate them + # by setting month=1, day=ordinal + if ordinal: + month = 1 + else: + month = time.gmtime()[1] + month = int(month) + day = params.get('day', 0) + if not day: + # see above + if ordinal: + day = ordinal + elif params.get('century', 0) or \ + params.get('year', 0) or params.get('month', 0): + day = 1 + else: + day = time.gmtime()[2] + else: + day = int(day) + # special case of the century - is the first year of the 21st century + # 2000 or 2001 ? The debate goes on... + if 'century' in params.keys(): + year = (int(params['century']) - 1) * 100 + 1 + # in ISO 8601 most fields are optional + for field in ['hour', 'minute', 'second', 'tzhour', 'tzmin']: + if not params.get(field, None): + params[field] = 0 + hour = int(params.get('hour', 0)) + minute = int(params.get('minute', 0)) + second = int(float(params.get('second', 0))) + # weekday is normalized by mktime(), we can ignore it + weekday = 0 + daylight_savings_flag = -1 + tm = [year, month, day, hour, minute, second, weekday, + ordinal, daylight_savings_flag] + # ISO 8601 time zone adjustments + tz = params.get('tz') + if tz and tz != 'Z': + if tz[0] == '-': + tm[3] += int(params.get('tzhour', 0)) + tm[4] += int(params.get('tzmin', 0)) + elif tz[0] == '+': + tm[3] -= int(params.get('tzhour', 0)) + tm[4] -= int(params.get('tzmin', 0)) + else: + return None + # Python's time.mktime() is a wrapper around the ANSI C mktime(3c) + # which is guaranteed to normalize d/m/y/h/m/s. + # Many implementations have bugs, but we'll pretend they don't. + return time.localtime(time.mktime(tm)) +registerDateHandler(_parse_date_iso8601) + +# 8-bit date handling routines written by ytrewq1. +_korean_year = u'\ub144' # b3e2 in euc-kr +_korean_month = u'\uc6d4' # bff9 in euc-kr +_korean_day = u'\uc77c' # c0cf in euc-kr +_korean_am = u'\uc624\uc804' # bfc0 c0fc in euc-kr +_korean_pm = u'\uc624\ud6c4' # bfc0 c8c4 in euc-kr + +_korean_onblog_date_re = \ + re.compile('(\d{4})%s\s+(\d{2})%s\s+(\d{2})%s\s+(\d{2}):(\d{2}):(\d{2})' % \ + (_korean_year, _korean_month, _korean_day)) +_korean_nate_date_re = \ + re.compile(u'(\d{4})-(\d{2})-(\d{2})\s+(%s|%s)\s+(\d{,2}):(\d{,2}):(\d{,2})' % \ + (_korean_am, _korean_pm)) +def _parse_date_onblog(dateString): + '''Parse a string according to the OnBlog 8-bit date format''' + m = _korean_onblog_date_re.match(dateString) + if not m: return + w3dtfdate = '%(year)s-%(month)s-%(day)sT%(hour)s:%(minute)s:%(second)s%(zonediff)s' % \ + {'year': m.group(1), 'month': m.group(2), 'day': m.group(3),\ + 'hour': m.group(4), 'minute': m.group(5), 'second': m.group(6),\ + 'zonediff': '+09:00'} + if _debug: sys.stderr.write('OnBlog date parsed as: %s\n' % w3dtfdate) + return _parse_date_w3dtf(w3dtfdate) +registerDateHandler(_parse_date_onblog) + +def _parse_date_nate(dateString): + '''Parse a string according to the Nate 8-bit date format''' + m = _korean_nate_date_re.match(dateString) + if not m: return + hour = int(m.group(5)) + ampm = m.group(4) + if (ampm == _korean_pm): + hour += 12 + hour = str(hour) + if len(hour) == 1: + hour = '0' + hour + w3dtfdate = '%(year)s-%(month)s-%(day)sT%(hour)s:%(minute)s:%(second)s%(zonediff)s' % \ + {'year': m.group(1), 'month': m.group(2), 'day': m.group(3),\ + 'hour': hour, 'minute': m.group(6), 'second': m.group(7),\ + 'zonediff': '+09:00'} + if _debug: sys.stderr.write('Nate date parsed as: %s\n' % w3dtfdate) + return _parse_date_w3dtf(w3dtfdate) +registerDateHandler(_parse_date_nate) + +_mssql_date_re = \ + re.compile('(\d{4})-(\d{2})-(\d{2})\s+(\d{2}):(\d{2}):(\d{2})(\.\d+)?') +def _parse_date_mssql(dateString): + '''Parse a string according to the MS SQL date format''' + m = _mssql_date_re.match(dateString) + if not m: return + w3dtfdate = '%(year)s-%(month)s-%(day)sT%(hour)s:%(minute)s:%(second)s%(zonediff)s' % \ + {'year': m.group(1), 'month': m.group(2), 'day': m.group(3),\ + 'hour': m.group(4), 'minute': m.group(5), 'second': m.group(6),\ + 'zonediff': '+09:00'} + if _debug: sys.stderr.write('MS SQL date parsed as: %s\n' % w3dtfdate) + return _parse_date_w3dtf(w3dtfdate) +registerDateHandler(_parse_date_mssql) + +# Unicode strings for Greek date strings +_greek_months = \ + { \ + u'\u0399\u03b1\u03bd': u'Jan', # c9e1ed in iso-8859-7 + u'\u03a6\u03b5\u03b2': u'Feb', # d6e5e2 in iso-8859-7 + u'\u039c\u03ac\u03ce': u'Mar', # ccdcfe in iso-8859-7 + u'\u039c\u03b1\u03ce': u'Mar', # cce1fe in iso-8859-7 + u'\u0391\u03c0\u03c1': u'Apr', # c1f0f1 in iso-8859-7 + u'\u039c\u03ac\u03b9': u'May', # ccdce9 in iso-8859-7 + u'\u039c\u03b1\u03ca': u'May', # cce1fa in iso-8859-7 + u'\u039c\u03b1\u03b9': u'May', # cce1e9 in iso-8859-7 + u'\u0399\u03bf\u03cd\u03bd': u'Jun', # c9effded in iso-8859-7 + u'\u0399\u03bf\u03bd': u'Jun', # c9efed in iso-8859-7 + u'\u0399\u03bf\u03cd\u03bb': u'Jul', # c9effdeb in iso-8859-7 + u'\u0399\u03bf\u03bb': u'Jul', # c9f9eb in iso-8859-7 + u'\u0391\u03cd\u03b3': u'Aug', # c1fde3 in iso-8859-7 + u'\u0391\u03c5\u03b3': u'Aug', # c1f5e3 in iso-8859-7 + u'\u03a3\u03b5\u03c0': u'Sep', # d3e5f0 in iso-8859-7 + u'\u039f\u03ba\u03c4': u'Oct', # cfeaf4 in iso-8859-7 + u'\u039d\u03bf\u03ad': u'Nov', # cdefdd in iso-8859-7 + u'\u039d\u03bf\u03b5': u'Nov', # cdefe5 in iso-8859-7 + u'\u0394\u03b5\u03ba': u'Dec', # c4e5ea in iso-8859-7 + } + +_greek_wdays = \ + { \ + u'\u039a\u03c5\u03c1': u'Sun', # caf5f1 in iso-8859-7 + u'\u0394\u03b5\u03c5': u'Mon', # c4e5f5 in iso-8859-7 + u'\u03a4\u03c1\u03b9': u'Tue', # d4f1e9 in iso-8859-7 + u'\u03a4\u03b5\u03c4': u'Wed', # d4e5f4 in iso-8859-7 + u'\u03a0\u03b5\u03bc': u'Thu', # d0e5ec in iso-8859-7 + u'\u03a0\u03b1\u03c1': u'Fri', # d0e1f1 in iso-8859-7 + u'\u03a3\u03b1\u03b2': u'Sat', # d3e1e2 in iso-8859-7 + } + +_greek_date_format_re = \ + re.compile(u'([^,]+),\s+(\d{2})\s+([^\s]+)\s+(\d{4})\s+(\d{2}):(\d{2}):(\d{2})\s+([^\s]+)') + +def _parse_date_greek(dateString): + '''Parse a string according to a Greek 8-bit date format.''' + m = _greek_date_format_re.match(dateString) + if not m: return + try: + wday = _greek_wdays[m.group(1)] + month = _greek_months[m.group(3)] + except: + return + rfc822date = '%(wday)s, %(day)s %(month)s %(year)s %(hour)s:%(minute)s:%(second)s %(zonediff)s' % \ + {'wday': wday, 'day': m.group(2), 'month': month, 'year': m.group(4),\ + 'hour': m.group(5), 'minute': m.group(6), 'second': m.group(7),\ + 'zonediff': m.group(8)} + if _debug: sys.stderr.write('Greek date parsed as: %s\n' % rfc822date) + return _parse_date_rfc822(rfc822date) +registerDateHandler(_parse_date_greek) + +# Unicode strings for Hungarian date strings +_hungarian_months = \ + { \ + u'janu\u00e1r': u'01', # e1 in iso-8859-2 + u'febru\u00e1ri': u'02', # e1 in iso-8859-2 + u'm\u00e1rcius': u'03', # e1 in iso-8859-2 + u'\u00e1prilis': u'04', # e1 in iso-8859-2 + u'm\u00e1ujus': u'05', # e1 in iso-8859-2 + u'j\u00fanius': u'06', # fa in iso-8859-2 + u'j\u00falius': u'07', # fa in iso-8859-2 + u'augusztus': u'08', + u'szeptember': u'09', + u'okt\u00f3ber': u'10', # f3 in iso-8859-2 + u'november': u'11', + u'december': u'12', + } + +_hungarian_date_format_re = \ + re.compile(u'(\d{4})-([^-]+)-(\d{,2})T(\d{,2}):(\d{2})((\+|-)(\d{,2}:\d{2}))') + +def _parse_date_hungarian(dateString): + '''Parse a string according to a Hungarian 8-bit date format.''' + m = _hungarian_date_format_re.match(dateString) + if not m: return + try: + month = _hungarian_months[m.group(2)] + day = m.group(3) + if len(day) == 1: + day = '0' + day + hour = m.group(4) + if len(hour) == 1: + hour = '0' + hour + except: + return + w3dtfdate = '%(year)s-%(month)s-%(day)sT%(hour)s:%(minute)s%(zonediff)s' % \ + {'year': m.group(1), 'month': month, 'day': day,\ + 'hour': hour, 'minute': m.group(5),\ + 'zonediff': m.group(6)} + if _debug: sys.stderr.write('Hungarian date parsed as: %s\n' % w3dtfdate) + return _parse_date_w3dtf(w3dtfdate) +registerDateHandler(_parse_date_hungarian) + +# W3DTF-style date parsing adapted from PyXML xml.utils.iso8601, written by +# Drake and licensed under the Python license. Removed all range checking +# for month, day, hour, minute, and second, since mktime will normalize +# these later +def _parse_date_w3dtf(dateString): + def __extract_date(m): + year = int(m.group('year')) + if year < 100: + year = 100 * int(time.gmtime()[0] / 100) + int(year) + if year < 1000: + return 0, 0, 0 + julian = m.group('julian') + if julian: + julian = int(julian) + month = julian / 30 + 1 + day = julian % 30 + 1 + jday = None + while jday != julian: + t = time.mktime((year, month, day, 0, 0, 0, 0, 0, 0)) + jday = time.gmtime(t)[-2] + diff = abs(jday - julian) + if jday > julian: + if diff < day: + day = day - diff + else: + month = month - 1 + day = 31 + elif jday < julian: + if day + diff < 28: + day = day + diff + else: + month = month + 1 + return year, month, day + month = m.group('month') + day = 1 + if month is None: + month = 1 + else: + month = int(month) + day = m.group('day') + if day: + day = int(day) + else: + day = 1 + return year, month, day + + def __extract_time(m): + if not m: + return 0, 0, 0 + hours = m.group('hours') + if not hours: + return 0, 0, 0 + hours = int(hours) + minutes = int(m.group('minutes')) + seconds = m.group('seconds') + if seconds: + seconds = int(seconds) + else: + seconds = 0 + return hours, minutes, seconds + + def __extract_tzd(m): + '''Return the Time Zone Designator as an offset in seconds from UTC.''' + if not m: + return 0 + tzd = m.group('tzd') + if not tzd: + return 0 + if tzd == 'Z': + return 0 + hours = int(m.group('tzdhours')) + minutes = m.group('tzdminutes') + if minutes: + minutes = int(minutes) + else: + minutes = 0 + offset = (hours*60 + minutes) * 60 + if tzd[0] == '+': + return -offset + return offset + + __date_re = ('(?P\d\d\d\d)' + '(?:(?P-|)' + '(?:(?P\d\d\d)' + '|(?P\d\d)(?:(?P=dsep)(?P\d\d))?))?') + __tzd_re = '(?P[-+](?P\d\d)(?::?(?P\d\d))|Z)' + __tzd_rx = re.compile(__tzd_re) + __time_re = ('(?P\d\d)(?P:|)(?P\d\d)' + '(?:(?P=tsep)(?P\d\d(?:[.,]\d+)?))?' + + __tzd_re) + __datetime_re = '%s(?:T%s)?' % (__date_re, __time_re) + __datetime_rx = re.compile(__datetime_re) + m = __datetime_rx.match(dateString) + if (m is None) or (m.group() != dateString): return + gmt = __extract_date(m) + __extract_time(m) + (0, 0, 0) + if gmt[0] == 0: return + return time.gmtime(time.mktime(gmt) + __extract_tzd(m) - time.timezone) +registerDateHandler(_parse_date_w3dtf) + +def _parse_date_rfc822(dateString): + '''Parse an RFC822, RFC1123, RFC2822, or asctime-style date''' + data = dateString.split() + if data[0][-1] in (',', '.') or data[0].lower() in rfc822._daynames: + del data[0] + if len(data) == 4: + s = data[3] + i = s.find('+') + if i > 0: + data[3:] = [s[:i], s[i+1:]] + else: + data.append('') + dateString = " ".join(data) + if len(data) < 5: + dateString += ' 00:00:00 GMT' + tm = rfc822.parsedate_tz(dateString) + if tm: + return time.gmtime(rfc822.mktime_tz(tm)) +# rfc822.py defines several time zones, but we define some extra ones. +# 'ET' is equivalent to 'EST', etc. +_additional_timezones = {'AT': -400, 'ET': -500, 'CT': -600, 'MT': -700, 'PT': -800} +rfc822._timezones.update(_additional_timezones) +registerDateHandler(_parse_date_rfc822) + +def _parse_date_perforce(aDateString): + """parse a date in yyyy/mm/dd hh:mm:ss TTT format""" + # Fri, 2006/09/15 08:19:53 EDT + _my_date_pattern = re.compile( \ + r'(\w{,3}), (\d{,4})/(\d{,2})/(\d{2}) (\d{,2}):(\d{2}):(\d{2}) (\w{,3})') + + dow, year, month, day, hour, minute, second, tz = \ + _my_date_pattern.search(aDateString).groups() + months = ['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec'] + dateString = "%s, %s %s %s %s:%s:%s %s" % (dow, day, months[int(month) - 1], year, hour, minute, second, tz) + tm = rfc822.parsedate_tz(dateString) + if tm: + return time.gmtime(rfc822.mktime_tz(tm)) +registerDateHandler(_parse_date_perforce) + +def _parse_date(dateString): + '''Parses a variety of date formats into a 9-tuple in GMT''' + for handler in _date_handlers: + try: + date9tuple = handler(dateString) + if not date9tuple: continue + if len(date9tuple) != 9: + if _debug: sys.stderr.write('date handler function must return 9-tuple\n') + raise ValueError + map(int, date9tuple) + return date9tuple + except Exception, e: + if _debug: sys.stderr.write('%s raised %s\n' % (handler.__name__, repr(e))) + pass + return None + +def _getCharacterEncoding(http_headers, xml_data): + '''Get the character encoding of the XML document + + http_headers is a dictionary + xml_data is a raw string (not Unicode) + + This is so much trickier than it sounds, it's not even funny. + According to RFC 3023 ('XML Media Types'), if the HTTP Content-Type + is application/xml, application/*+xml, + application/xml-external-parsed-entity, or application/xml-dtd, + the encoding given in the charset parameter of the HTTP Content-Type + takes precedence over the encoding given in the XML prefix within the + document, and defaults to 'utf-8' if neither are specified. But, if + the HTTP Content-Type is text/xml, text/*+xml, or + text/xml-external-parsed-entity, the encoding given in the XML prefix + within the document is ALWAYS IGNORED and only the encoding given in + the charset parameter of the HTTP Content-Type header should be + respected, and it defaults to 'us-ascii' if not specified. + + Furthermore, discussion on the atom-syntax mailing list with the + author of RFC 3023 leads me to the conclusion that any document + served with a Content-Type of text/* and no charset parameter + must be treated as us-ascii. (We now do this.) And also that it + must always be flagged as non-well-formed. (We now do this too.) + + If Content-Type is unspecified (input was local file or non-HTTP source) + or unrecognized (server just got it totally wrong), then go by the + encoding given in the XML prefix of the document and default to + 'iso-8859-1' as per the HTTP specification (RFC 2616). + + Then, assuming we didn't find a character encoding in the HTTP headers + (and the HTTP Content-type allowed us to look in the body), we need + to sniff the first few bytes of the XML data and try to determine + whether the encoding is ASCII-compatible. Section F of the XML + specification shows the way here: + http://www.w3.org/TR/REC-xml/#sec-guessing-no-ext-info + + If the sniffed encoding is not ASCII-compatible, we need to make it + ASCII compatible so that we can sniff further into the XML declaration + to find the encoding attribute, which will tell us the true encoding. + + Of course, none of this guarantees that we will be able to parse the + feed in the declared character encoding (assuming it was declared + correctly, which many are not). CJKCodecs and iconv_codec help a lot; + you should definitely install them if you can. + http://cjkpython.i18n.org/ + ''' + + def _parseHTTPContentType(content_type): + '''takes HTTP Content-Type header and returns (content type, charset) + + If no charset is specified, returns (content type, '') + If no content type is specified, returns ('', '') + Both return parameters are guaranteed to be lowercase strings + ''' + content_type = content_type or '' + content_type, params = cgi.parse_header(content_type) + return content_type, params.get('charset', '').replace("'", '') + + sniffed_xml_encoding = '' + xml_encoding = '' + true_encoding = '' + http_content_type, http_encoding = _parseHTTPContentType(http_headers.get('content-type')) + # Must sniff for non-ASCII-compatible character encodings before + # searching for XML declaration. This heuristic is defined in + # section F of the XML specification: + # http://www.w3.org/TR/REC-xml/#sec-guessing-no-ext-info + try: + if xml_data[:4] == '\x4c\x6f\xa7\x94': + # EBCDIC + xml_data = _ebcdic_to_ascii(xml_data) + elif xml_data[:4] == '\x00\x3c\x00\x3f': + # UTF-16BE + sniffed_xml_encoding = 'utf-16be' + xml_data = unicode(xml_data, 'utf-16be').encode('utf-8') + elif (len(xml_data) >= 4) and (xml_data[:2] == '\xfe\xff') and (xml_data[2:4] != '\x00\x00'): + # UTF-16BE with BOM + sniffed_xml_encoding = 'utf-16be' + xml_data = unicode(xml_data[2:], 'utf-16be').encode('utf-8') + elif xml_data[:4] == '\x3c\x00\x3f\x00': + # UTF-16LE + sniffed_xml_encoding = 'utf-16le' + xml_data = unicode(xml_data, 'utf-16le').encode('utf-8') + elif (len(xml_data) >= 4) and (xml_data[:2] == '\xff\xfe') and (xml_data[2:4] != '\x00\x00'): + # UTF-16LE with BOM + sniffed_xml_encoding = 'utf-16le' + xml_data = unicode(xml_data[2:], 'utf-16le').encode('utf-8') + elif xml_data[:4] == '\x00\x00\x00\x3c': + # UTF-32BE + sniffed_xml_encoding = 'utf-32be' + xml_data = unicode(xml_data, 'utf-32be').encode('utf-8') + elif xml_data[:4] == '\x3c\x00\x00\x00': + # UTF-32LE + sniffed_xml_encoding = 'utf-32le' + xml_data = unicode(xml_data, 'utf-32le').encode('utf-8') + elif xml_data[:4] == '\x00\x00\xfe\xff': + # UTF-32BE with BOM + sniffed_xml_encoding = 'utf-32be' + xml_data = unicode(xml_data[4:], 'utf-32be').encode('utf-8') + elif xml_data[:4] == '\xff\xfe\x00\x00': + # UTF-32LE with BOM + sniffed_xml_encoding = 'utf-32le' + xml_data = unicode(xml_data[4:], 'utf-32le').encode('utf-8') + elif xml_data[:3] == '\xef\xbb\xbf': + # UTF-8 with BOM + sniffed_xml_encoding = 'utf-8' + xml_data = unicode(xml_data[3:], 'utf-8').encode('utf-8') + else: + # ASCII-compatible + pass + xml_encoding_match = re.compile('^<\?.*encoding=[\'"](.*?)[\'"].*\?>').match(xml_data) + except: + xml_encoding_match = None + if xml_encoding_match: + xml_encoding = xml_encoding_match.groups()[0].lower() + if sniffed_xml_encoding and (xml_encoding in ('iso-10646-ucs-2', 'ucs-2', 'csunicode', 'iso-10646-ucs-4', 'ucs-4', 'csucs4', 'utf-16', 'utf-32', 'utf_16', 'utf_32', 'utf16', 'u16')): + xml_encoding = sniffed_xml_encoding + acceptable_content_type = 0 + application_content_types = ('application/xml', 'application/xml-dtd', 'application/xml-external-parsed-entity') + text_content_types = ('text/xml', 'text/xml-external-parsed-entity') + if (http_content_type in application_content_types) or \ + (http_content_type.startswith('application/') and http_content_type.endswith('+xml')): + acceptable_content_type = 1 + true_encoding = http_encoding or xml_encoding or 'utf-8' + elif (http_content_type in text_content_types) or \ + (http_content_type.startswith('text/')) and http_content_type.endswith('+xml'): + acceptable_content_type = 1 + true_encoding = http_encoding or 'us-ascii' + elif http_content_type.startswith('text/'): + true_encoding = http_encoding or 'us-ascii' + elif http_headers and (not http_headers.has_key('content-type')): + true_encoding = xml_encoding or 'iso-8859-1' + else: + true_encoding = xml_encoding or 'utf-8' + # some feeds claim to be gb2312 but are actually gb18030. + # apparently MSIE and Firefox both do the following switch: + if true_encoding.lower() == 'gb2312': + true_encoding = 'gb18030' + return true_encoding, http_encoding, xml_encoding, sniffed_xml_encoding, acceptable_content_type + +def _toUTF8(data, encoding): + '''Changes an XML data stream on the fly to specify a new encoding + + data is a raw sequence of bytes (not Unicode) that is presumed to be in %encoding already + encoding is a string recognized by encodings.aliases + ''' + if _debug: sys.stderr.write('entering _toUTF8, trying encoding %s\n' % encoding) + # strip Byte Order Mark (if present) + if (len(data) >= 4) and (data[:2] == '\xfe\xff') and (data[2:4] != '\x00\x00'): + if _debug: + sys.stderr.write('stripping BOM\n') + if encoding != 'utf-16be': + sys.stderr.write('trying utf-16be instead\n') + encoding = 'utf-16be' + data = data[2:] + elif (len(data) >= 4) and (data[:2] == '\xff\xfe') and (data[2:4] != '\x00\x00'): + if _debug: + sys.stderr.write('stripping BOM\n') + if encoding != 'utf-16le': + sys.stderr.write('trying utf-16le instead\n') + encoding = 'utf-16le' + data = data[2:] + elif data[:3] == '\xef\xbb\xbf': + if _debug: + sys.stderr.write('stripping BOM\n') + if encoding != 'utf-8': + sys.stderr.write('trying utf-8 instead\n') + encoding = 'utf-8' + data = data[3:] + elif data[:4] == '\x00\x00\xfe\xff': + if _debug: + sys.stderr.write('stripping BOM\n') + if encoding != 'utf-32be': + sys.stderr.write('trying utf-32be instead\n') + encoding = 'utf-32be' + data = data[4:] + elif data[:4] == '\xff\xfe\x00\x00': + if _debug: + sys.stderr.write('stripping BOM\n') + if encoding != 'utf-32le': + sys.stderr.write('trying utf-32le instead\n') + encoding = 'utf-32le' + data = data[4:] + newdata = unicode(data, encoding) + if _debug: sys.stderr.write('successfully converted %s data to unicode\n' % encoding) + declmatch = re.compile('^<\?xml[^>]*?>') + newdecl = '''''' + if declmatch.search(newdata): + newdata = declmatch.sub(newdecl, newdata) + else: + newdata = newdecl + u'\n' + newdata + return newdata.encode('utf-8') + +def _stripDoctype(data): + '''Strips DOCTYPE from XML document, returns (rss_version, stripped_data) + + rss_version may be 'rss091n' or None + stripped_data is the same XML document, minus the DOCTYPE + ''' + start = re.search('<\w',data) + start = start and start.start() or -1 + head,data = data[:start+1], data[start+1:] + + entity_pattern = re.compile(r'^\s*]*?)>', re.MULTILINE) + entity_results=entity_pattern.findall(head) + head = entity_pattern.sub('', head) + doctype_pattern = re.compile(r'^\s*]*?)>', re.MULTILINE) + doctype_results = doctype_pattern.findall(head) + doctype = doctype_results and doctype_results[0] or '' + if doctype.lower().count('netscape'): + version = 'rss091n' + else: + version = None + + # only allow in 'safe' inline entity definitions + replacement='' + if len(doctype_results)==1 and entity_results: + safe_pattern=re.compile('\s+(\w+)\s+"(&#\w+;|[^&"]*)"') + safe_entities=filter(lambda e: safe_pattern.match(e),entity_results) + if safe_entities: + replacement='\n]>' % '>\n ', 'larr':'<-', 'middot':'*', +'ndash':'-', 'oelig':'oe', 'aelig':'ae', +'agrave':'a', 'aacute':'a', 'acirc':'a', 'atilde':'a', 'auml':'a', 'aring':'a', +'egrave':'e', 'eacute':'e', 'ecirc':'e', 'euml':'e', +'igrave':'i', 'iacute':'i', 'icirc':'i', 'iuml':'i', +'ograve':'o', 'oacute':'o', 'ocirc':'o', 'otilde':'o', 'ouml':'o', +'ugrave':'u', 'uacute':'u', 'ucirc':'u', 'uuml':'u'} + +unifiable_n = {} + +for k in unifiable.keys(): + unifiable_n[name2cp(k)] = unifiable[k] + +def charref(name): + if name[0] in ['x','X']: + c = int(name[1:], 16) + else: + c = int(name) + + if not UNICODE_SNOB and c in unifiable_n.keys(): + return unifiable_n[c] + else: + return unichr(c) + +def entityref(c): + if not UNICODE_SNOB and c in unifiable.keys(): + return unifiable[c] + else: + try: name2cp(c) + except KeyError: return "&" + c + else: return unichr(name2cp(c)) + +def replaceEntities(s): + s = s.group(1) + if s[0] == "#": + return charref(s[1:]) + else: return entityref(s) + +r_unescape = re.compile(r"&(#?[xX]?(?:[0-9a-fA-F]+|\w{1,8}));") +def unescape(s): + return r_unescape.sub(replaceEntities, s) + +def fixattrs(attrs): + # Fix bug in sgmllib.py + if not attrs: return attrs + newattrs = [] + for attr in attrs: + newattrs.append((attr[0], unescape(attr[1]))) + return newattrs + +### End Entity Nonsense ### + +def onlywhite(line): + """Return true if the line does only consist of whitespace characters.""" + for c in line: + if c is not ' ' and c is not ' ': + return c is ' ' + return line + +def optwrap(text): + """Wrap all paragraphs in the provided text.""" + if not BODY_WIDTH: + return text + + assert wrap, "Requires Python 2.3." + result = '' + newlines = 0 + for para in text.split("\n"): + if len(para) > 0: + if para[0] is not ' ' and para[0] is not '-' and para[0] is not '*': + for line in wrap(para, BODY_WIDTH): + result += line + "\n" + result += "\n" + newlines = 2 + else: + if not onlywhite(para): + result += para + "\n" + newlines = 1 + else: + if newlines < 2: + result += "\n" + newlines += 1 + return result + +def hn(tag): + if tag[0] == 'h' and len(tag) == 2: + try: + n = int(tag[1]) + if n in range(1, 10): return n + except ValueError: return 0 + +class _html2text(sgmllib.SGMLParser): + def __init__(self, out=None, baseurl=''): + sgmllib.SGMLParser.__init__(self) + + if out is None: self.out = self.outtextf + else: self.out = out + self.outtext = u'' + self.quiet = 0 + self.p_p = 0 + self.outcount = 0 + self.start = 1 + self.space = 0 + self.a = [] + self.astack = [] + self.acount = 0 + self.list = [] + self.blockquote = 0 + self.pre = 0 + self.startpre = 0 + self.lastWasNL = 0 + self.abbr_title = None # current abbreviation definition + self.abbr_data = None # last inner HTML (for abbr being defined) + self.abbr_list = {} # stack of abbreviations to write later + self.baseurl = baseurl + + def outtextf(self, s): + self.outtext += s + + def close(self): + sgmllib.SGMLParser.close(self) + + self.pbr() + self.o('', 0, 'end') + + return self.outtext + + def handle_charref(self, c): + self.o(charref(c)) + + def handle_entityref(self, c): + self.o(entityref(c)) + + def unknown_starttag(self, tag, attrs): + self.handle_tag(tag, attrs, 1) + + def unknown_endtag(self, tag): + self.handle_tag(tag, None, 0) + + def previousIndex(self, attrs): + """ returns the index of certain set of attributes (of a link) in the + self.a list + + If the set of attributes is not found, returns None + """ + if not attrs.has_key('href'): return None + + i = -1 + for a in self.a: + i += 1 + match = 0 + + if a.has_key('href') and a['href'] == attrs['href']: + if a.has_key('title') or attrs.has_key('title'): + if (a.has_key('title') and attrs.has_key('title') and + a['title'] == attrs['title']): + match = True + else: + match = True + + if match: return i + + def handle_tag(self, tag, attrs, start): + attrs = fixattrs(attrs) + + if hn(tag): + self.p() + if start: self.o(hn(tag)*"#" + ' ') + + if tag in ['p', 'div']: self.p() + + if tag == "br" and start: self.o(" \n") + + if tag == "hr" and start: + self.p() + self.o("* * *") + self.p() + + if tag in ["head", "style", 'script']: + if start: self.quiet += 1 + else: self.quiet -= 1 + + if tag in ["body"]: + self.quiet = 0 # sites like 9rules.com never close + + if tag == "blockquote": + if start: + self.p(); self.o('> ', 0, 1); self.start = 1 + self.blockquote += 1 + else: + self.blockquote -= 1 + self.p() + + if tag in ['em', 'i', 'u']: self.o("_") + if tag in ['strong', 'b']: self.o("**") + if tag == "code" and not self.pre: self.o('`') #TODO: `` `this` `` + if tag == "abbr": + if start: + attrsD = {} + for (x, y) in attrs: attrsD[x] = y + attrs = attrsD + + self.abbr_title = None + self.abbr_data = '' + if attrs.has_key('title'): + self.abbr_title = attrs['title'] + else: + if self.abbr_title != None: + self.abbr_list[self.abbr_data] = self.abbr_title + self.abbr_title = None + self.abbr_data = '' + + if tag == "a": + if start: + attrsD = {} + for (x, y) in attrs: attrsD[x] = y + attrs = attrsD + if attrs.has_key('href') and not (SKIP_INTERNAL_LINKS and attrs['href'].startswith('#')): + self.astack.append(attrs) + self.o("[") + else: + self.astack.append(None) + else: + if self.astack: + a = self.astack.pop() + if a: + i = self.previousIndex(a) + if i is not None: + a = self.a[i] + else: + self.acount += 1 + a['count'] = self.acount + a['outcount'] = self.outcount + self.a.append(a) + self.o("][" + `a['count']` + "]") + + if tag == "img" and start: + attrsD = {} + for (x, y) in attrs: attrsD[x] = y + attrs = attrsD + if attrs.has_key('src'): + attrs['href'] = attrs['src'] + alt = attrs.get('alt', '') + i = self.previousIndex(attrs) + if i is not None: + attrs = self.a[i] + else: + self.acount += 1 + attrs['count'] = self.acount + attrs['outcount'] = self.outcount + self.a.append(attrs) + self.o("![") + self.o(alt) + self.o("]["+`attrs['count']`+"]") + + if tag == 'dl' and start: self.p() + if tag == 'dt' and not start: self.pbr() + if tag == 'dd' and start: self.o(' ') + if tag == 'dd' and not start: self.pbr() + + if tag in ["ol", "ul"]: + if start: + self.list.append({'name':tag, 'num':0}) + else: + if self.list: self.list.pop() + + self.p() + + if tag == 'li': + if start: + self.pbr() + if self.list: li = self.list[-1] + else: li = {'name':'ul', 'num':0} + self.o(" "*len(self.list)) #TODO: line up
  1. s > 9 correctly. + if li['name'] == "ul": self.o("* ") + elif li['name'] == "ol": + li['num'] += 1 + self.o(`li['num']`+". ") + self.start = 1 + else: + self.pbr() + + if tag in ["table", "tr"] and start: self.p() + if tag == 'td': self.pbr() + + if tag == "pre": + if start: + self.startpre = 1 + self.pre = 1 + else: + self.pre = 0 + self.p() + + def pbr(self): + if self.p_p == 0: self.p_p = 1 + + def p(self): self.p_p = 2 + + def o(self, data, puredata=0, force=0): + if self.abbr_data is not None: self.abbr_data += data + + if not self.quiet: + if puredata and not self.pre: + data = re.sub('\s+', ' ', data) + if data and data[0] == ' ': + self.space = 1 + data = data[1:] + if not data and not force: return + + if self.startpre: + #self.out(" :") #TODO: not output when already one there + self.startpre = 0 + + bq = (">" * self.blockquote) + if not (force and data and data[0] == ">") and self.blockquote: bq += " " + + if self.pre: + bq += " " + data = data.replace("\n", "\n"+bq) + + if self.start: + self.space = 0 + self.p_p = 0 + self.start = 0 + + if force == 'end': + # It's the end. + self.p_p = 0 + self.out("\n") + self.space = 0 + + + if self.p_p: + self.out(('\n'+bq)*self.p_p) + self.space = 0 + + if self.space: + if not self.lastWasNL: self.out(' ') + self.space = 0 + + if self.a and ((self.p_p == 2 and LINKS_EACH_PARAGRAPH) or force == "end"): + if force == "end": self.out("\n") + + newa = [] + for link in self.a: + if self.outcount > link['outcount']: + self.out(" ["+`link['count']`+"]: " + urlparse.urljoin(self.baseurl, link['href'])) + if link.has_key('title'): self.out(" ("+link['title']+")") + self.out("\n") + else: + newa.append(link) + + if self.a != newa: self.out("\n") # Don't need an extra line when nothing was done. + + self.a = newa + + if self.abbr_list and force == "end": + for abbr, definition in self.abbr_list.items(): + self.out(" *[" + abbr + "]: " + definition + "\n") + + self.p_p = 0 + self.out(data) + self.lastWasNL = data and data[-1] == '\n' + self.outcount += 1 + + def handle_data(self, data): + if r'\/script>' in data: self.quiet -= 1 + self.o(data, 1) + + def unknown_decl(self, data): pass + +def wrapwrite(text): sys.stdout.write(text.encode('utf8')) + +def html2text_file(html, out=wrapwrite, baseurl=''): + h = _html2text(out, baseurl) + h.feed(html) + h.feed("") + return h.close() + +def html2text(html, baseurl=''): + return optwrap(html2text_file(html, None, baseurl)) + +if __name__ == "__main__": + baseurl = '' + if sys.argv[1:]: + arg = sys.argv[1] + if arg.startswith('http://'): + baseurl = arg + j = urllib.urlopen(baseurl) + try: + from feedparser import _getCharacterEncoding as enc + except ImportError: + enc = lambda x, y: ('utf-8', 1) + text = j.read() + encoding = enc(j.headers, text)[0] + if encoding == 'us-ascii': encoding = 'utf-8' + data = text.decode(encoding) + + else: + encoding = 'utf8' + if len(sys.argv) > 2: + encoding = sys.argv[2] + data = open(arg, 'r').read().decode(encoding) + else: + data = sys.stdin.read().decode('utf8') + wrapwrite(html2text(data, baseurl)) + diff --git a/r2e b/r2e new file mode 100644 index 0000000..e582fae --- /dev/null +++ b/r2e @@ -0,0 +1,2 @@ +#!/bin/sh +python rss2email.py feeds.dat $* diff --git a/r2e.bat b/r2e.bat new file mode 100644 index 0000000..d80fcdf --- /dev/null +++ b/r2e.bat @@ -0,0 +1 @@ +@python rss2email.py feeds.dat %1 %2 %3 %4 %5 %6 %7 %8 %9 diff --git a/readme.html b/readme.html new file mode 100644 index 0000000..224fbbe --- /dev/null +++ b/readme.html @@ -0,0 +1,163 @@ + + +Getting Started With rss2email + + +

    Getting Started With rss2email

    + +

    We highly recommend that you subscribe to the rss2email project feed so you can keep up to date with the latest version, bugfixes and features: http://feeds.feedburner.com/allthingsrss/hJBr

    +

    Instructions for Windows Users
    +Instructions for UNIX Users
    +Customizing rss2email

    + + +

    Instructions for Windows Users

    + +

    Requirements

    + +

    Before you install rss2email, you'll need to make sure that a few things are in place. First, is that a version of Python 2.x installed. Second, determine your outgoing email server's address. That should be all you need.

    + +

    Download

    + +
      +
    1. Create a new folder
      2. +
    2. Download the latest rss2email .ZIP file and unzip to the new folder +
    + +

    Configure

    + +

    Edit the config.py file and fill in your outoing email server's details. If your server requires you to login, change "AUTHREQUIRED = 0" to "AUTHREQUIRED = 1" and enter your email username and password.

    + +

    Install

    + +

    From the command line, change to the folder you created. Now create a new feed database to send updates to your email address:

    + +
    +

    r2e new you@yourdomain.com

    +
    + +

    Subscribe to some feeds:

    + +
    +

    r2e add http://feeds.feedburner.com/allthingsrss/hJBr

    +
    + +

    That's the feed to be notified when there's a new version of rss2email. Repeat this for each feed you want to subscribe to.

    + +

    When you run rss2email, it emails you about every story it hasn't seen before. But the first time you run it, that will be every story. To avoid this, you can ask rss2email not to send you any stories the first time you run it:

    + +
    +

    r2e run --no-send

    +
    + +

    Then later, you can ask it to email you new stories:

    + +
    +

    r2e run

    + +
    + +

    If you get an error message "Sender domain must exist", add a line to config.py like this:

    + +
    +

    DEFAULT_FROM = rss2email@yoursite.com

    +
    + +

    You can make the email address whatever you want, but your mail server requires that the yoursite.com part actually exists.

    + +

    Automating rss2email

    + +

    More than likely you will want rss2email to run automatically at a regular interval. Under Windows this is can be easily accomplished using the Windows Task Scheduler. This site has a nice tutorial on it. Just select r2e.bat as the program to run. Once you've created the task, double click on it in the task list and change the Run entry so that "run" comes after r2e.bat. For example, if you installed rss2email in the C:\rss2email folder, then you would change the Run entry from "C:\rss2email\r2e.bat" to "C:\rss2email\r2e.bat run".

    + +

    Now jump down to the section on customizing rss2email to your needs.

    + +

    Upgrading to a new version

    +

    Simply replace all of the files from the .ZIP package to your install directory EXCEPT config.py

    + +

    Instructions for UNIX/Linux Users

    + +

    Requirements

    + +

    Before you install rss2email, you'll need to make sure that a few things are in place. First, is a version of Python 2.x installed. Second, is whether you have sendmail (or a compatible replacement like postfix) installed. If sendmail isn't installed, determine your outgoing email server's address. That should be all you need.

    + +

    Download

    + +

    A quick way to get rss2email going is using pre-made packages. Here are releases for Debian Linux, Ubuntu Linux and NetBSD.

    + +

    If you are unable to use these packages or you want the latest and greatest version, here's what you do:

    + +
    +Unarchive (probably 'tar -xzf') the rss2email .tar.gz package to [folder where you want rss2email files to live]
    +cd [yourfolder]
    +chmod +x r2e +
    + +

    Install

    + +

    Create a new feed database with your target email address:

    + +
    +

    ./r2e new you@yourdomain.com

    +
    + +

    Subscribe to some feeds:

    + +
    +

    ./r2e add http://feeds.feedburner.com/allthingsrss/hJBr

    + +
    + +

    That's the feed to be notified when there's a new version of rss2email. Repeat this for each feed you want to subscribe to.

    + +

    When you run rss2email, it emails you about every story it hasn't seen before. But the first time you run it, that will be every story. To avoid this, you can ask rss2email not to send you any stories the first time you run it:

    + +
    +

    ./r2e run --no-send

    +
    + +

    Then later, you can ask it to email you new stories:

    + +
    +

    ./r2e run

    +
    + +

    You probably want to set things up so that this command is run repeatedly. (One good way is via a cron job.)

    + +

    If you get an error message "Sender domain must exist", add a line to config.py like this:

    + +
    +

    DEFAULT_FROM = rss2email@yoursite.com

    +
    + +

    You can make the email address whatever you want, but your mail server requires that the yoursite.com part actually exists.

    + +

    Upgrading to a new version

    +

    Simply replace all of the files from the .tar.gz package to your install directory EXCEPT config.py

    + + + +

    Customize rss2email

    + +

    There are a number of options, described in full at the top of rss2email.py file, to customize the way rss2email behaves. If you want to change something, edit the config.py file. If you're not using rss2email under Windows, you'll have to create this file if it doesn't already exist.

    + +

    For example, if you want to receive HTML mail, instead of having entries converted to plain text:

    + +
    +

    HTML_MAIL = 1

    + +
    + +

    To be notified every time a post changes, instead of just when it's first posted:

    + +
    +

    TRUST_GUID = 0

    +
    + +

    And to make the emails look as if they were sent when the item was posted:

    + +
    +

    DATE_HEADER = 1

    + + + + \ No newline at end of file diff --git a/rss2email.py b/rss2email.py new file mode 100644 index 0000000..40b4b4d --- /dev/null +++ b/rss2email.py @@ -0,0 +1,956 @@ +#!/usr/bin/python +"""rss2email: get RSS feeds emailed to you +http://rss2email.infogami.com + +Usage: + new [emailaddress] (create new feedfile) + email newemailaddress (update default email) + run [--no-send] [num] + add feedurl [emailaddress] + list + reset + delete n + pause n + unpause n + opmlexport + opmlimport filename +""" +__version__ = "2.70" +__author__ = "Lindsey Smith (lindsey@allthingsrss.com)" +__copyright__ = "(C) 2004 Aaron Swartz. GNU GPL 2 or 3." +___contributors__ = ["Dean Jackson", "Brian Lalor", "Joey Hess", + "Matej Cepl", "Martin 'Joey' Schulze", + "Marcel Ackermann (http://www.DreamFlasher.de)", + "Lindsey Smith (maintainer)", "Erik Hetzner", "Aaron Swartz (original author)" ] + +import urllib2 +urllib2.install_opener(urllib2.build_opener()) + +### Vaguely Customizable Options ### + +# The email address messages are from by default: +DEFAULT_FROM = "bozo@dev.null.invalid" + +# 1: Send text/html messages when possible. +# 0: Convert HTML to plain text. +HTML_MAIL = 0 + +# 1: Only use the DEFAULT_FROM address. +# 0: Use the email address specified by the feed, when possible. +FORCE_FROM = 0 + +# 1: Receive one email per post. +# 0: Receive an email every time a post changes. +TRUST_GUID = 1 + +# 1: Generate Date header based on item's date, when possible. +# 0: Generate Date header based on time sent. +DATE_HEADER = 0 + +# A tuple consisting of some combination of +# ('issued', 'created', 'modified', 'expired') +# expressing ordered list of preference in dates +# to use for the Date header of the email. +DATE_HEADER_ORDER = ('modified', 'issued', 'created') + +# 1: Apply Q-P conversion (required for some MUAs). +# 0: Send message in 8-bits. +# http://cr.yp.to/smtp/8bitmime.html +#DEPRECATED +QP_REQUIRED = 0 +#DEPRECATED + +# 1: Name feeds as they're being processed. +# 0: Keep quiet. +VERBOSE = 0 + +# 1: Use the publisher's email if you can't find the author's. +# 0: Just use the DEFAULT_FROM email instead. +USE_PUBLISHER_EMAIL = 0 + +# 1: Use SMTP_SERVER to send mail. +# 0: Call /usr/sbin/sendmail to send mail. +SMTP_SEND = 0 + +SMTP_SERVER = "smtp.yourisp.net:25" +AUTHREQUIRED = 0 # if you need to use SMTP AUTH set to 1 +SMTP_USER = 'username' # for SMTP AUTH, set SMTP username here +SMTP_PASS = 'password' # for SMTP AUTH, set SMTP password here + +# Connect to the SMTP server using SSL +SMTP_SSL = 0 + +# Set this to add a bonus header to all emails (start with '\n'). +BONUS_HEADER = '' +# Example: BONUS_HEADER = '\nApproved: joe@bob.org' + +# Set this to override From addresses. Keys are feed URLs, values are new titles. +OVERRIDE_FROM = {} + +# Set this to override From email addresses. Keys are feed URLs, values are new emails. +OVERRIDE_EMAIL = {} + +# Set this to default From email addresses. Keys are feed URLs, values are new email addresses. +DEFAULT_EMAIL = {} + +# Only use the email from address rather than friendly name plus email address +NO_FRIENDLY_NAME = 0 + +# Set this to override the timeout (in seconds) for feed server response +FEED_TIMEOUT = 60 + +# Optional CSS styling +USE_CSS_STYLING = 0 +STYLE_SHEET='h1 {font: 18pt Georgia, "Times New Roman";} body {font: 12pt Arial;} a:link {font: 12pt Arial; font-weight: bold; color: #0000cc} blockquote {font-family: monospace; } .header { background: #e0ecff; border-bottom: solid 4px #c3d9ff; padding: 5px; margin-top: 0px; color: red;} .header a { font-size: 20px; text-decoration: none; } .footer { background: #c3d9ff; border-top: solid 4px #c3d9ff; padding: 5px; margin-bottom: 0px; } #entry {border: solid 4px #c3d9ff; } #body { margin-left: 5px; margin-right: 5px; }' + +# If you have an HTTP Proxy set this in the format 'http://your.proxy.here:8080/' +PROXY="" + +# To most correctly encode emails with international characters, we iterate through the list below and use the first character set that works +# Eventually (and theoretically) ISO-8859-1 and UTF-8 are our catch-all failsafes +CHARSET_LIST='US-ASCII', 'BIG5', 'ISO-2022-JP', 'ISO-8859-1', 'UTF-8' + +from email.MIMEText import MIMEText +from email.Header import Header +from email.Utils import parseaddr, formataddr + +# Note: You can also override the send function. + +def send(sender, recipient, subject, body, contenttype, extraheaders=None, smtpserver=None): + """Send an email. + + All arguments should be Unicode strings (plain ASCII works as well). + + Only the real name part of sender and recipient addresses may contain + non-ASCII characters. + + The email will be properly MIME encoded and delivered though SMTP to + localhost port 25. This is easy to change if you want something different. + + The charset of the email will be the first one out of the list + that can represent all the characters occurring in the email. + """ + + # Header class is smart enough to try US-ASCII, then the charset we + # provide, then fall back to UTF-8. + header_charset = 'ISO-8859-1' + + # We must choose the body charset manually + for body_charset in CHARSET_LIST: + try: + body.encode(body_charset) + except (UnicodeError, LookupError): + pass + else: + break + + # Split real name (which is optional) and email address parts + sender_name, sender_addr = parseaddr(sender) + recipient_name, recipient_addr = parseaddr(recipient) + + # We must always pass Unicode strings to Header, otherwise it will + # use RFC 2047 encoding even on plain ASCII strings. + sender_name = str(Header(unicode(sender_name), header_charset)) + recipient_name = str(Header(unicode(recipient_name), header_charset)) + + # Make sure email addresses do not contain non-ASCII characters + sender_addr = sender_addr.encode('ascii') + recipient_addr = recipient_addr.encode('ascii') + + # Create the message ('plain' stands for Content-Type: text/plain) + msg = MIMEText(body.encode(body_charset), contenttype, body_charset) + msg['To'] = formataddr((recipient_name, recipient_addr)) + msg['Subject'] = Header(unicode(subject), header_charset) + for hdr in extraheaders.keys(): + try: + msg[hdr] = Header(unicode(extraheaders[hdr], header_charset)) + except: + msg[hdr] = Header(extraheaders[hdr]) + + fromhdr = formataddr((sender_name, sender_addr)) + msg['From'] = fromhdr + + msg_as_string = msg.as_string() +#DEPRECATED if QP_REQUIRED: +#DEPRECATED ins, outs = SIO(msg_as_string), SIO() +#DEPRECATED mimify.mimify(ins, outs) +#DEPRECATED msg_as_string = outs.getvalue() + + if SMTP_SEND: + if not smtpserver: + import smtplib + + try: + if SMTP_SSL: + smtpserver = smtplib.SMTP_SSL() + else: + smtpserver = smtplib.SMTP() + smtpserver.connect(SMTP_SERVER) + except KeyboardInterrupt: + raise + except Exception, e: + print >>warn, "" + print >>warn, ('Fatal error: could not connect to mail server "%s"' % SMTP_SERVER) + print >>warn, ('Check your config.py file to confirm that SMTP_SERVER and other mail server settings are configured properly') + if hasattr(e, 'reason'): + print >>warn, "Reason:", e.reason + sys.exit(1) + + if AUTHREQUIRED: + try: + smtpserver.ehlo() + if not SMTP_SSL: smtpserver.starttls() + smtpserver.ehlo() + smtpserver.login(SMTP_USER, SMTP_PASS) + except KeyboardInterrupt: + raise + except Exception, e: + print >>warn, "" + print >>warn, ('Fatal error: could not authenticate with mail server "%s" as user "%s"' % (SMTP_SERVER, SMTP_USER)) + print >>warn, ('Check your config.py file to confirm that SMTP_SERVER and other mail server settings are configured properly') + if hasattr(e, 'reason'): + print >>warn, "Reason:", e.reason + sys.exit(1) + + smtpserver.sendmail(sender, recipient, msg_as_string) + return smtpserver + + else: + try: + p = subprocess.Popen(["/usr/sbin/sendmail", recipient], stdin=subprocess.PIPE, stdout=subprocess.PIPE) + p.communicate(msg_as_string) + status = p.returncode + assert status != None, "just a sanity check" + if status != 0: + print >>warn, "" + print >>warn, ('Fatal error: sendmail exited with code %s' % status) + sys.exit(1) + except: + print '''Error attempting to send email via sendmail. Possibly you need to configure your config.py to use a SMTP server? Please refer to the rss2email documentation or website (http://rss2email.infogami.com) for complete documentation of config.py. The options below may suffice for configuring email: +# 1: Use SMTP_SERVER to send mail. +# 0: Call /usr/sbin/sendmail to send mail. +SMTP_SEND = 0 + +SMTP_SERVER = "smtp.yourisp.net:25" +AUTHREQUIRED = 0 # if you need to use SMTP AUTH set to 1 +SMTP_USER = 'username' # for SMTP AUTH, set SMTP username here +SMTP_PASS = 'password' # for SMTP AUTH, set SMTP password here +''' + sys.exit(1) + return None + +## html2text options ## + +# Use Unicode characters instead of their ascii psuedo-replacements +UNICODE_SNOB = 0 + +# Put the links after each paragraph instead of at the end. +LINKS_EACH_PARAGRAPH = 0 + +# Wrap long lines at position. 0 for no wrapping. (Requires Python 2.3.) +BODY_WIDTH = 0 + +### Load the Options ### + +# Read options from config file if present. +import sys +sys.path.insert(0,".") +try: + from config import * +except: + pass + +warn = sys.stderr + +if QP_REQUIRED: + print >>warn, "QP_REQUIRED has been deprecated in rss2email." + +### Import Modules ### + +import cPickle as pickle, time, os, traceback, sys, types, subprocess +hash = () +try: + import hashlib + hash = hashlib.md5 +except ImportError: + import md5 + hash = md5.new + +unix = 0 +try: + import fcntl +# A pox on SunOS file locking methods + if (sys.platform.find('sunos') == -1): + unix = 1 +except: + pass + +import socket; socket_errors = [] +for e in ['error', 'gaierror']: + if hasattr(socket, e): socket_errors.append(getattr(socket, e)) + +#DEPRECATED import mimify +#DEPRECATED from StringIO import StringIO as SIO +#DEPRECATED mimify.CHARSET = 'utf-8' + +import feedparser +feedparser.USER_AGENT = "rss2email/"+__version__+ " +http://www.allthingsrss.com/rss2email/" + +import html2text as h2t + +h2t.UNICODE_SNOB = UNICODE_SNOB +h2t.LINKS_EACH_PARAGRAPH = LINKS_EACH_PARAGRAPH +h2t.BODY_WIDTH = BODY_WIDTH +html2text = h2t.html2text + +from types import * + +### Utility Functions ### + +import threading +class TimeoutError(Exception): pass + +class InputError(Exception): pass + +def timelimit(timeout, function): +# def internal(function): + def internal2(*args, **kw): + """ + from http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/473878 + """ + class Calculator(threading.Thread): + def __init__(self): + threading.Thread.__init__(self) + self.result = None + self.error = None + + def run(self): + try: + self.result = function(*args, **kw) + except: + self.error = sys.exc_info() + + c = Calculator() + c.setDaemon(True) # don't hold up exiting + c.start() + c.join(timeout) + if c.isAlive(): + raise TimeoutError + if c.error: + raise c.error[0], c.error[1] + return c.result + return internal2 +# return internal + + +def isstr(f): return isinstance(f, type('')) or isinstance(f, type(u'')) +def ishtml(t): return type(t) is type(()) +def contains(a,b): return a.find(b) != -1 +def unu(s): # I / freakin' hate / that unicode + if type(s) is types.UnicodeType: return s.encode('utf-8') + else: return s + +### Parsing Utilities ### + +def getContent(entry, HTMLOK=0): + """Select the best content from an entry, deHTMLizing if necessary. + If raw HTML is best, an ('HTML', best) tuple is returned. """ + + # How this works: + # * We have a bunch of potential contents. + # * We go thru looking for our first choice. + # (HTML or text, depending on HTMLOK) + # * If that doesn't work, we go thru looking for our second choice. + # * If that still doesn't work, we just take the first one. + # + # Possible future improvement: + # * Instead of just taking the first one + # pick the one in the "best" language. + # * HACK: hardcoded HTMLOK, should take a tuple of media types + + conts = entry.get('content', []) + + if entry.get('summary_detail', {}): + conts += [entry.summary_detail] + + if conts: + if HTMLOK: + for c in conts: + if contains(c.type, 'html'): return ('HTML', c.value) + + if not HTMLOK: # Only need to convert to text if HTML isn't OK + for c in conts: + if contains(c.type, 'html'): + return html2text(c.value) + + for c in conts: + if c.type == 'text/plain': return c.value + + return conts[0].value + + return "" + +def getID(entry): + """Get best ID from an entry.""" + if TRUST_GUID: + if 'id' in entry and entry.id: + # Newer versions of feedparser could return a dictionary + if type(entry.id) is DictType: + return entry.id.values()[0] + + return entry.id + + content = getContent(entry) + if content and content != "\n": return hash(unu(content)).hexdigest() + if 'link' in entry: return entry.link + if 'title' in entry: return hash(unu(entry.title)).hexdigest() + +def getName(r, entry): + """Get the best name.""" + + if NO_FRIENDLY_NAME: return '' + + feed = r.feed + if hasattr(r, "url") and r.url in OVERRIDE_FROM.keys(): + return OVERRIDE_FROM[r.url] + + name = feed.get('title', '') + + if 'name' in entry.get('author_detail', []): # normally {} but py2.1 + if entry.author_detail.name: + if name: name += ": " + det=entry.author_detail.name + try: + name += entry.author_detail.name + except UnicodeDecodeError: + name += unicode(entry.author_detail.name, 'utf-8') + + elif 'name' in feed.get('author_detail', []): + if feed.author_detail.name: + if name: name += ", " + name += feed.author_detail.name + + return name + +def validateEmail(email, planb): + """Do a basic quality check on email address, but return planb if email doesn't appear to be well-formed""" + email_parts = email.split('@') + if len(email_parts) != 2: + return planb + return email + +def getEmail(r, entry): + """Get the best email_address. If the best guess isn't well-formed (something@somthing.com), use DEFAULT_FROM instead""" + + feed = r.feed + + if FORCE_FROM: return DEFAULT_FROM + + if r.url in OVERRIDE_EMAIL.keys(): + return validateEmail(OVERRIDE_EMAIL[r.url], DEFAULT_FROM) + + if 'email' in entry.get('author_detail', []): + return validateEmail(entry.author_detail.email, DEFAULT_FROM) + + if 'email' in feed.get('author_detail', []): + return validateEmail(feed.author_detail.email, DEFAULT_FROM) + + if USE_PUBLISHER_EMAIL: + if 'email' in feed.get('publisher_detail', []): + return validateEmail(feed.publisher_detail.email, DEFAULT_FROM) + + if feed.get("errorreportsto", ''): + return validateEmail(feed.errorreportsto, DEFAULT_FROM) + + if r.url in DEFAULT_EMAIL.keys(): + return DEFAULT_EMAIL[r.url] + return DEFAULT_FROM + +### Simple Database of Feeds ### + +class Feed: + def __init__(self, url, to): + self.url, self.etag, self.modified, self.seen = url, None, None, {} + self.active = True + self.to = to + +def load(lock=1): + if not os.path.exists(feedfile): + print 'Feedfile "%s" does not exist. If you\'re using r2e for the first time, you' % feedfile + print "have to run 'r2e new' first." + sys.exit(1) + try: + feedfileObject = open(feedfile, 'r') + except IOError, e: + print "Feedfile could not be opened: %s" % e + sys.exit(1) + feeds = pickle.load(feedfileObject) + + if lock: + locktype = 0 + if unix: + locktype = fcntl.LOCK_EX + fcntl.flock(feedfileObject.fileno(), locktype) + #HACK: to deal with lock caching + feedfileObject = open(feedfile, 'r') + feeds = pickle.load(feedfileObject) + if unix: + fcntl.flock(feedfileObject.fileno(), locktype) + if feeds: + for feed in feeds[1:]: + if not hasattr(feed, 'active'): + feed.active = True + + return feeds, feedfileObject + +def unlock(feeds, feedfileObject): + if not unix: + pickle.dump(feeds, open(feedfile, 'w')) + else: + pickle.dump(feeds, open(feedfile+'.tmp', 'w')) + os.rename(feedfile+'.tmp', feedfile) + fcntl.flock(feedfileObject.fileno(), fcntl.LOCK_UN) + +#@timelimit(FEED_TIMEOUT) +def parse(url, etag, modified): + if PROXY == '': + return feedparser.parse(url, etag, modified) + else: + proxy = urllib2.ProxyHandler( {"http":PROXY} ) + return feedparser.parse(url, etag, modified, handlers = [proxy]) + + +### Program Functions ### + +def add(*args): + if len(args) == 2 and contains(args[1], '@') and not contains(args[1], '://'): + urls, to = [args[0]], args[1] + else: + urls, to = args, None + + feeds, feedfileObject = load() + if (feeds and not isstr(feeds[0]) and to is None) or (not len(feeds) and to is None): + print "No email address has been defined. Please run 'r2e email emailaddress' or" + print "'r2e add url emailaddress'." + sys.exit(1) + for url in urls: feeds.append(Feed(url, to)) + unlock(feeds, feedfileObject) + +def run(num=None): + feeds, feedfileObject = load() + smtpserver = None + try: + # We store the default to address as the first item in the feeds list. + # Here we take it out and save it for later. + default_to = "" + if feeds and isstr(feeds[0]): default_to = feeds[0]; ifeeds = feeds[1:] + else: ifeeds = feeds + + if num: ifeeds = [feeds[num]] + feednum = 0 + + for f in ifeeds: + try: + feednum += 1 + if not f.active: continue + + if VERBOSE: print >>warn, 'I: Processing [%d] "%s"' % (feednum, f.url) + r = {} + try: + r = timelimit(FEED_TIMEOUT, parse)(f.url, f.etag, f.modified) + except TimeoutError: + print >>warn, 'W: feed [%d] "%s" timed out' % (feednum, f.url) + continue + + # Handle various status conditions, as required + if 'status' in r: + if r.status == 301: f.url = r['url'] + elif r.status == 410: + print >>warn, "W: feed gone; deleting", f.url + feeds.remove(f) + continue + + http_status = r.get('status', 200) + if VERBOSE > 1: print >>warn, "I: http status", http_status + http_headers = r.get('headers', { + 'content-type': 'application/rss+xml', + 'content-length':'1'}) + exc_type = r.get("bozo_exception", Exception()).__class__ + if http_status != 304 and not r.entries and not r.get('version', ''): + if http_status not in [200, 302]: + print >>warn, "W: error %d [%d] %s" % (http_status, feednum, f.url) + + elif contains(http_headers.get('content-type', 'rss'), 'html'): + print >>warn, "W: looks like HTML [%d] %s" % (feednum, f.url) + + elif http_headers.get('content-length', '1') == '0': + print >>warn, "W: empty page [%d] %s" % (feednum, f.url) + + elif hasattr(socket, 'timeout') and exc_type == socket.timeout: + print >>warn, "W: timed out on [%d] %s" % (feednum, f.url) + + elif exc_type == IOError: + print >>warn, 'W: "%s" [%d] %s' % (r.bozo_exception, feednum, f.url) + + elif hasattr(feedparser, 'zlib') and exc_type == feedparser.zlib.error: + print >>warn, "W: broken compression [%d] %s" % (feednum, f.url) + + elif exc_type in socket_errors: + exc_reason = r.bozo_exception.args[1] + print >>warn, "W: %s [%d] %s" % (exc_reason, feednum, f.url) + + elif exc_type == urllib2.URLError: + if r.bozo_exception.reason.__class__ in socket_errors: + exc_reason = r.bozo_exception.reason.args[1] + else: + exc_reason = r.bozo_exception.reason + print >>warn, "W: %s [%d] %s" % (exc_reason, feednum, f.url) + + elif exc_type == AttributeError: + print >>warn, "W: %s [%d] %s" % (r.bozo_exception, feednum, f.url) + + elif exc_type == KeyboardInterrupt: + raise r.bozo_exception + + elif r.bozo: + print >>warn, 'E: error in [%d] "%s" feed (%s)' % (feednum, f.url, r.get("bozo_exception", "can't process")) + + else: + print >>warn, "=== rss2email encountered a problem with this feed ===" + print >>warn, "=== See the rss2email FAQ at http://www.allthingsrss.com/rss2email/ for assistance ===" + print >>warn, "=== If this occurs repeatedly, send this to lindsey@allthingsrss.com ===" + print >>warn, "E:", r.get("bozo_exception", "can't process"), f.url + print >>warn, r + print >>warn, "rss2email", __version__ + print >>warn, "feedparser", feedparser.__version__ + print >>warn, "html2text", h2t.__version__ + print >>warn, "Python", sys.version + print >>warn, "=== END HERE ===" + continue + + r.entries.reverse() + + for entry in r.entries: + id = getID(entry) + + # If TRUST_GUID isn't set, we get back hashes of the content. + # Instead of letting these run wild, we put them in context + # by associating them with the actual ID (if it exists). + + frameid = entry.get('id') + if not(frameid): frameid = id + if type(frameid) is DictType: + frameid = frameid.values()[0] + + # If this item's ID is in our database + # then it's already been sent + # and we don't need to do anything more. + + if frameid in f.seen: + if f.seen[frameid] == id: continue + + if not (f.to or default_to): + print "No default email address defined. Please run 'r2e email emailaddress'" + print "Ignoring feed %s" % f.url + break + + if 'title_detail' in entry and entry.title_detail: + title = entry.title_detail.value + if contains(entry.title_detail.type, 'html'): + title = html2text(title) + else: + title = getContent(entry)[:70] + + title = title.replace("\n", " ").strip() + + datetime = time.gmtime() + + if DATE_HEADER: + for datetype in DATE_HEADER_ORDER: + kind = datetype+"_parsed" + if kind in entry and entry[kind]: datetime = entry[kind] + + link = entry.get('link', "") + + from_addr = getEmail(r, entry) + + name = h2t.unescape(getName(r, entry)) + fromhdr = formataddr((name, from_addr,)) + tohdr = (f.to or default_to) + subjecthdr = title + datehdr = time.strftime("%a, %d %b %Y %H:%M:%S -0000", datetime) + useragenthdr = "rss2email" + + # Add post tags, if available + tagline = "" + if 'tags' in entry: + tags = entry.get('tags') + taglist = [] + if tags: + for tag in tags: + taglist.append(tag['term']) + if taglist: + tagline = ",".join(taglist) + + extraheaders = {'Date': datehdr, 'User-Agent': useragenthdr, 'X-RSS-Feed': f.url, 'X-RSS-ID': id, 'X-RSS-URL': link, 'X-RSS-TAGS' : tagline} + if BONUS_HEADER != '': + for hdr in BONUS_HEADER.strip().splitlines(): + pos = hdr.strip().find(':') + if pos > 0: + extraheaders[hdr[:pos]] = hdr[pos+1:].strip() + else: + print >>warn, "W: malformed BONUS HEADER", BONUS_HEADER + + entrycontent = getContent(entry, HTMLOK=HTML_MAIL) + contenttype = 'plain' + content = '' + if USE_CSS_STYLING and HTML_MAIL: + contenttype = 'html' + content = "\n" + content += '\n' + content += '\n' + content += '
    \n' + content += ''+subjecthdr+'\n' + if ishtml(entrycontent): + body = entrycontent[1].strip() + else: + body = entrycontent.strip() + if body != '': + content += '
    \n' + body + '
    \n' + content += '\n

    URL: '+link+'' + if hasattr(entry,'enclosures'): + for enclosure in entry.enclosures: + if (hasattr(enclosure, 'url') and enclosure.url != ""): + content += ('
    Enclosure: '+enclosure.url+"\n") + if (hasattr(enclosure, 'src') and enclosure.src != ""): + content += ('
    Enclosure: '+enclosure.src+'
    '+extralink['title']+'\n' + content += '

    \n' + content += "\n\n" + else: + if ishtml(entrycontent): + contenttype = 'html' + content = "\n" + content = ("\n\n" + + '

    '+subjecthdr+'

    \n\n' + + entrycontent[1].strip() + # drop type tag (HACK: bad abstraction) + '

    URL: '+link+'

    ' ) + + if hasattr(entry,'enclosures'): + for enclosure in entry.enclosures: + if enclosure.url != "": + content += ('Enclosure: '+enclosure.url+"
    \n") + if 'links' in entry: + for extralink in entry.links: + if ('rel' in extralink) and extralink['rel'] == u'via': + content += 'Via: '+extralink['title']+'
    \n' + + content += ("\n") + else: + content = entrycontent.strip() + "\n\nURL: "+link + if hasattr(entry,'enclosures'): + for enclosure in entry.enclosures: + if enclosure.url != "": + content += ('\nEnclosure: ' + enclosure.url + "\n") + if 'links' in entry: + for extralink in entry.links: + if ('rel' in extralink) and extralink['rel'] == u'via': + content += 'Via: '+extralink['title']+'\n' + + smtpserver = send(fromhdr, tohdr, subjecthdr, content, contenttype, extraheaders, smtpserver) + + f.seen[frameid] = id + + f.etag, f.modified = r.get('etag', None), r.get('modified', None) + except (KeyboardInterrupt, SystemExit): + raise + except: + print >>warn, "=== rss2email encountered a problem with this feed ===" + print >>warn, "=== See the rss2email FAQ at http://www.allthingsrss.com/rss2email/ for assistance ===" + print >>warn, "=== If this occurs repeatedly, send this to lindsey@allthingsrss.com ===" + print >>warn, "E: could not parse", f.url + traceback.print_exc(file=warn) + print >>warn, "rss2email", __version__ + print >>warn, "feedparser", feedparser.__version__ + print >>warn, "html2text", h2t.__version__ + print >>warn, "Python", sys.version + print >>warn, "=== END HERE ===" + continue + + finally: + unlock(feeds, feedfileObject) + if smtpserver: + smtpserver.quit() + +def list(): + feeds, feedfileObject = load(lock=0) + default_to = "" + + if feeds and isstr(feeds[0]): + default_to = feeds[0]; ifeeds = feeds[1:]; i=1 + print "default email:", default_to + else: ifeeds = feeds; i = 0 + for f in ifeeds: + active = ('[ ]', '[*]')[f.active] + print `i`+':',active, f.url, '('+(f.to or ('default: '+default_to))+')' + if not (f.to or default_to): + print " W: Please define a default address with 'r2e email emailaddress'" + i+= 1 + +def opmlexport(): + import xml.sax.saxutils + feeds, feedfileObject = load(lock=0) + + if feeds: + print '\n\n\nrss2email OPML export\n\n' + for f in feeds[1:]: + url = xml.sax.saxutils.escape(f.url) + print '' % (url, url) + print '\n' + +def opmlimport(importfile): + importfileObject = None + print 'Importing feeds from', importfile + if not os.path.exists(importfile): + print 'OPML import file "%s" does not exist.' % feedfile + try: + importfileObject = open(importfile, 'r') + except IOError, e: + print "OPML import file could not be opened: %s" % e + sys.exit(1) + try: + import xml.dom.minidom + dom = xml.dom.minidom.parse(importfileObject) + newfeeds = dom.getElementsByTagName('outline') + except: + print 'E: Unable to parse OPML file' + sys.exit(1) + + feeds, feedfileObject = load(lock=1) + + import xml.sax.saxutils + + for f in newfeeds: + if f.hasAttribute('xmlUrl'): + feedurl = f.getAttribute('xmlUrl') + print 'Adding %s' % xml.sax.saxutils.unescape(feedurl) + feeds.append(Feed(feedurl, None)) + + unlock(feeds, feedfileObject) + +def delete(n): + feeds, feedfileObject = load() + if (n == 0) and (feeds and isstr(feeds[0])): + print >>warn, "W: ID has to be equal to or higher than 1" + elif n >= len(feeds): + print >>warn, "W: no such feed" + else: + print >>warn, "W: deleting feed %s" % feeds[n].url + feeds = feeds[:n] + feeds[n+1:] + if n != len(feeds): + print >>warn, "W: feed IDs have changed, list before deleting again" + unlock(feeds, feedfileObject) + +def toggleactive(n, active): + feeds, feedfileObject = load() + if (n == 0) and (feeds and isstr(feeds[0])): + print >>warn, "W: ID has to be equal to or higher than 1" + elif n >= len(feeds): + print >>warn, "W: no such feed" + else: + action = ('Pausing', 'Unpausing')[active] + print >>warn, "%s feed %s" % (action, feeds[n].url) + feeds[n].active = active + unlock(feeds, feedfileObject) + +def reset(): + feeds, feedfileObject = load() + if feeds and isstr(feeds[0]): + ifeeds = feeds[1:] + else: ifeeds = feeds + for f in ifeeds: + if VERBOSE: print "Resetting %d already seen items" % len(f.seen) + f.seen = {} + f.etag = None + f.modified = None + + unlock(feeds, feedfileObject) + +def email(addr): + feeds, feedfileObject = load() + if feeds and isstr(feeds[0]): feeds[0] = addr + else: feeds = [addr] + feeds + unlock(feeds, feedfileObject) + +if __name__ == '__main__': + args = sys.argv + try: + if len(args) < 3: raise InputError, "insufficient args" + feedfile, action, args = args[1], args[2], args[3:] + + if action == "run": + if args and args[0] == "--no-send": + def send(sender, recipient, subject, body, contenttype, extraheaders=None, smtpserver=None): + if VERBOSE: print 'Not sending:', unu(subject) + + if args and args[-1].isdigit(): run(int(args[-1])) + else: run() + + elif action == "email": + if not args: + raise InputError, "Action '%s' requires an argument" % action + else: + email(args[0]) + + elif action == "add": add(*args) + + elif action == "new": + if len(args) == 1: d = [args[0]] + else: d = [] + pickle.dump(d, open(feedfile, 'w')) + + elif action == "list": list() + + elif action in ("help", "--help", "-h"): print __doc__ + + elif action == "delete": + if not args: + raise InputError, "Action '%s' requires an argument" % action + elif args[0].isdigit(): + delete(int(args[0])) + else: + raise InputError, "Action '%s' requires a number as its argument" % action + + elif action in ("pause", "unpause"): + if not args: + raise InputError, "Action '%s' requires an argument" % action + elif args[0].isdigit(): + active = (action == "unpause") + toggleactive(int(args[0]), active) + else: + raise InputError, "Action '%s' requires a number as its argument" % action + + elif action == "reset": reset() + + elif action == "opmlexport": opmlexport() + + elif action == "opmlimport": + if not args: + raise InputError, "OPML import '%s' requires a filename argument" % action + opmlimport(args[0]) + + else: + raise InputError, "Invalid action" + + except InputError, e: + print "E:", e + print + print __doc__ +