From fc6dd61021384ef0593f3dc51816b496417dd0d7 Mon Sep 17 00:00:00 2001 From: Deepak Sarda Date: Sat, 18 Jul 2009 11:32:16 +0800 Subject: [PATCH 01/39] Initial repo conversion to mercurial --- LICENSE | 51 + README | 55 + app/BeautifulSoup.py | 1931 +++++++++++++++++++ app/app.yaml | 26 + app/dumble/dumble.js | 376 ++++ app/dumble/images/1_stars.png | Bin 0 -> 1303 bytes app/dumble/images/2_stars.png | Bin 0 -> 1371 bytes app/dumble/images/3_stars.png | Bin 0 -> 1431 bytes app/dumble/images/4_stars.png | Bin 0 -> 1440 bytes app/dumble/images/5_stars.png | Bin 0 -> 1370 bytes app/dumble/images/feed-icon.png | Bin 0 -> 689 bytes app/dumble/index.html | 445 +++++ app/dumble/jquery.cookie.js | 51 + app/feedparser.py | 2858 +++++++++++++++++++++++++++++ app/index.jinja | 257 +++ app/jinja/__init__.py | 72 + app/jinja/_native.py | 85 + app/jinja/constants.py | 32 + app/jinja/datastructure.py | 708 +++++++ app/jinja/debugger.py | 213 +++ app/jinja/defaults.py | 16 + app/jinja/environment.py | 395 ++++ app/jinja/exceptions.py | 91 + app/jinja/filters.py | 986 ++++++++++ app/jinja/lexer.py | 513 ++++++ app/jinja/loaders.py | 822 +++++++++ app/jinja/nodes.py | 792 ++++++++ app/jinja/parser.py | 1187 ++++++++++++ app/jinja/plugin.py | 166 ++ app/jinja/tests.py | 143 ++ app/jinja/translators/__init__.py | 29 + app/jinja/translators/python.py | 1177 ++++++++++++ app/jinja/utils.py | 644 +++++++ app/main.py | 134 ++ app/markdown.py | 1929 +++++++++++++++++++ app/provider/__init__.py | 7 + app/provider/base.py | 34 + app/provider/linkprovider.py | 163 ++ app/provider/oembedprovider.py | 97 + app/provider/photoprovider.py | 203 ++ app/provider/utils.py | 72 + app/provider/videoprovider.py | 301 +++ app/static/amazon_json.xsl | 116 ++ app/static/favicon.ico | Bin 0 -> 1150 bytes app/static/robots.txt | 2 + 45 files changed, 17179 insertions(+) create mode 100644 LICENSE create mode 100644 README create mode 100644 app/BeautifulSoup.py create mode 100644 app/app.yaml create mode 100644 app/dumble/dumble.js create mode 100644 app/dumble/images/1_stars.png create mode 100644 app/dumble/images/2_stars.png create mode 100644 app/dumble/images/3_stars.png create mode 100644 app/dumble/images/4_stars.png create mode 100644 app/dumble/images/5_stars.png create mode 100644 app/dumble/images/feed-icon.png create mode 100644 app/dumble/index.html create mode 100644 app/dumble/jquery.cookie.js create mode 100755 app/feedparser.py create mode 100644 app/index.jinja create mode 100644 app/jinja/__init__.py create mode 100644 app/jinja/_native.py create mode 100644 app/jinja/constants.py create mode 100644 app/jinja/datastructure.py create mode 100644 app/jinja/debugger.py create mode 100644 app/jinja/defaults.py create mode 100644 app/jinja/environment.py create mode 100644 app/jinja/exceptions.py create mode 100644 app/jinja/filters.py create mode 100644 app/jinja/lexer.py create mode 100644 app/jinja/loaders.py create mode 100644 app/jinja/nodes.py create mode 100644 app/jinja/parser.py create mode 100644 app/jinja/plugin.py create mode 100644 app/jinja/tests.py create mode 100644 app/jinja/translators/__init__.py create mode 100644 app/jinja/translators/python.py create mode 100644 app/jinja/utils.py create mode 100644 app/main.py create mode 100644 app/markdown.py create mode 100644 app/provider/__init__.py create mode 100644 app/provider/base.py create mode 100644 app/provider/linkprovider.py create mode 100644 app/provider/oembedprovider.py create mode 100644 app/provider/photoprovider.py create mode 100644 app/provider/utils.py create mode 100644 app/provider/videoprovider.py create mode 100644 app/static/amazon_json.xsl create mode 100644 app/static/favicon.ico create mode 100644 app/static/robots.txt diff --git a/LICENSE b/LICENSE new file mode 100644 index 0000000..b3b1d22 --- /dev/null +++ b/LICENSE @@ -0,0 +1,51 @@ +oohEmbed +======== +Copyright (c) 2008, Deepak Sarda + +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are +met: + + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above + copyright notice, this list of conditions and the following + disclaimer in the documentation and/or other materials provided + with the distribution. + + * Neither the name of the oohEmbed project nor the names of its + contributors may be used to endorse or promote products derived + from this software without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR +CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, +EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, +PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR +PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF +LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING +NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS +SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +Jinja +===== +Jinja is a template language that oohEmbed uses for generating the +homepage at oohembed.com. Jinja is licensed under a BSD License. +See http://jinja.pocoo.org/ for more details + +Python-Markdown +=============== +Python-Markdown is an implementation of Markdown markup syntax +in Python. It is under a GPL2 license or BSD license. +See http://www.freewisdom.org/projects/python-markdown/ + +Beautiful Soup +============== +Beautiful Soup is a HTML/XML parser designed for quick turnaround +projects like screen-scraping. It is licensed under a BSD style license. +See http://www.crummy.com/software/BeautifulSoup/ for more. diff --git a/README b/README new file mode 100644 index 0000000..3182895 --- /dev/null +++ b/README @@ -0,0 +1,55 @@ +# Welcome to oohEmbed! + +## oohEmbed What? + +In a nutshell: __oohEmbed is an oEmbed compatible provider of HTML embed codes for various web sites.__ + +What is this oEmbed? From [oembed.com](http://www.oembed.com/): + +> oEmbed is a format for allowing an embedded representation of a URL on third party sites. +> The simple API allows a website to display embedded content (such as photos or videos) when +> a user posts a link to that resource, without having to parse the resource directly. + +Still don't get it? Perhaps an example will make things clear. If you make a URL request like this: + + http://oohembed.com/oohembed/?url=http%3A//www.amazon.com/Myths-Innovation-Scott-Berkun/dp/0596527055/ + +You will get this as the response: + +{ + "asin": "0596527055", + "title": "The Myths of Innovation", + "url": "http://ecx.images-amazon.com/images/I/31%2BfVjL2nqL.jpg", + "thumbnail_width": "48", + "height": "500", + "width": "317", + "version": "1.0", + "author_name": "Scott Berkun", + "provider_name": "Amazon Product Image", + "thumbnail_url": "http://ecx.images-amazon.com/images/I/31%2BfVjL2nqL._SL75_.jpg", + "type": "photo", + "thumbnail_height": "75", + "author_url": "http://www.amazon.com/gp/redirect.html%3FASIN=0596527055%26location=/Myths-Innovation-Scott-Berkun/dp/0596527055" +} + +That should make everything clear. No? Then head over to http://oohEmbed.com/ for more details! + +## What's in the source? + +If you are reading this file, you've chanced upon the source code that powers oohEmbed.com + +Yes, the code running behind oohEmbed.com is open source (BSD License) and is available at +http://code.google.com/p/oohembed/ + +## Brief Overview + +oohEmbed is running on Google App Engine and so what you see in the source files is a Python +app meant to run on App Engine. Thus, some familiarity with App Engine would be a good thing +to have before jumping into the code. + +Although oohEmbed is a Google App Engine app, there is very little App Engine specific in the +code itself. Apart from the use of `urlfetch`, everything else is standard Python and should +run anywhere that Python runs. + +The codebase is so small that hopefully, it is self-explanatory! If you have any questions, +please feel free to ask! diff --git a/app/BeautifulSoup.py b/app/BeautifulSoup.py new file mode 100644 index 0000000..0e55aba --- /dev/null +++ b/app/BeautifulSoup.py @@ -0,0 +1,1931 @@ +"""Beautiful Soup +Elixir and Tonic +"The Screen-Scraper's Friend" +http://www.crummy.com/software/BeautifulSoup/ + +Beautiful Soup parses a (possibly invalid) XML or HTML document into a +tree representation. It provides methods and Pythonic idioms that make +it easy to navigate, search, and modify the tree. + +A well-formed XML/HTML document yields a well-formed data +structure. An ill-formed XML/HTML document yields a correspondingly +ill-formed data structure. If your document is only locally +well-formed, you can use this library to find and process the +well-formed part of it. + +Beautiful Soup works with Python 2.2 and up. It has no external +dependencies, but you'll have more success at converting data to UTF-8 +if you also install these three packages: + +* chardet, for auto-detecting character encodings + http://chardet.feedparser.org/ +* cjkcodecs and iconv_codec, which add more encodings to the ones supported + by stock Python. + http://cjkpython.i18n.org/ + +Beautiful Soup defines classes for two main parsing strategies: + + * BeautifulStoneSoup, for parsing XML, SGML, or your domain-specific + language that kind of looks like XML. + + * BeautifulSoup, for parsing run-of-the-mill HTML code, be it valid + or invalid. This class has web browser-like heuristics for + obtaining a sensible parse tree in the face of common HTML errors. + +Beautiful Soup also defines a class (UnicodeDammit) for autodetecting +the encoding of an HTML or XML document, and converting it to +Unicode. Much of this code is taken from Mark Pilgrim's Universal Feed Parser. + +For more than you ever wanted to know about Beautiful Soup, see the +documentation: +http://www.crummy.com/software/BeautifulSoup/documentation.html + +Here, have some legalese: + +Copyright (c) 2004-2007, Leonard Richardson + +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are +met: + + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above + copyright notice, this list of conditions and the following + disclaimer in the documentation and/or other materials provided + with the distribution. + + * Neither the name of the the Beautiful Soup Consortium and All + Night Kosher Bakery nor the names of its contributors may be + used to endorse or promote products derived from this software + without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR +CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, +EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, +PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR +PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF +LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING +NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS +SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE, DAMMIT. + +""" +from __future__ import generators + +__author__ = "Leonard Richardson (leonardr@segfault.org)" +__version__ = "3.0.6" +__copyright__ = "Copyright (c) 2004-2008 Leonard Richardson" +__license__ = "New-style BSD" + +from sgmllib import SGMLParser, SGMLParseError +import codecs +import types +import re +import sgmllib +try: + from htmlentitydefs import name2codepoint +except ImportError: + name2codepoint = {} + +#This hack makes Beautiful Soup able to parse XML with namespaces +sgmllib.tagfind = re.compile('[a-zA-Z][-_.:a-zA-Z0-9]*') + +DEFAULT_OUTPUT_ENCODING = "utf-8" + +# First, the classes that represent markup elements. + +class PageElement: + """Contains the navigational information for some part of the page + (either a tag or a piece of text)""" + + def setup(self, parent=None, previous=None): + """Sets up the initial relations between this element and + other elements.""" + self.parent = parent + self.previous = previous + self.next = None + self.previousSibling = None + self.nextSibling = None + if self.parent and self.parent.contents: + self.previousSibling = self.parent.contents[-1] + self.previousSibling.nextSibling = self + + def replaceWith(self, replaceWith): + oldParent = self.parent + myIndex = self.parent.contents.index(self) + if hasattr(replaceWith, 'parent') and replaceWith.parent == self.parent: + # We're replacing this element with one of its siblings. + index = self.parent.contents.index(replaceWith) + if index and index < myIndex: + # Furthermore, it comes before this element. That + # means that when we extract it, the index of this + # element will change. + myIndex = myIndex - 1 + self.extract() + oldParent.insert(myIndex, replaceWith) + + def extract(self): + """Destructively rips this element out of the tree.""" + if self.parent: + try: + self.parent.contents.remove(self) + except ValueError: + pass + + #Find the two elements that would be next to each other if + #this element (and any children) hadn't been parsed. Connect + #the two. + lastChild = self._lastRecursiveChild() + nextElement = lastChild.next + + if self.previous: + self.previous.next = nextElement + if nextElement: + nextElement.previous = self.previous + self.previous = None + lastChild.next = None + + self.parent = None + if self.previousSibling: + self.previousSibling.nextSibling = self.nextSibling + if self.nextSibling: + self.nextSibling.previousSibling = self.previousSibling + self.previousSibling = self.nextSibling = None + return self + + def _lastRecursiveChild(self): + "Finds the last element beneath this object to be parsed." + lastChild = self + while hasattr(lastChild, 'contents') and lastChild.contents: + lastChild = lastChild.contents[-1] + return lastChild + + def insert(self, position, newChild): + if (isinstance(newChild, basestring) + or isinstance(newChild, unicode)) \ + and not isinstance(newChild, NavigableString): + newChild = NavigableString(newChild) + + position = min(position, len(self.contents)) + if hasattr(newChild, 'parent') and newChild.parent != None: + # We're 'inserting' an element that's already one + # of this object's children. + if newChild.parent == self: + index = self.find(newChild) + if index and index < position: + # Furthermore we're moving it further down the + # list of this object's children. That means that + # when we extract this element, our target index + # will jump down one. + position = position - 1 + newChild.extract() + + newChild.parent = self + previousChild = None + if position == 0: + newChild.previousSibling = None + newChild.previous = self + else: + previousChild = self.contents[position-1] + newChild.previousSibling = previousChild + newChild.previousSibling.nextSibling = newChild + newChild.previous = previousChild._lastRecursiveChild() + if newChild.previous: + newChild.previous.next = newChild + + newChildsLastElement = newChild._lastRecursiveChild() + + if position >= len(self.contents): + newChild.nextSibling = None + + parent = self + parentsNextSibling = None + while not parentsNextSibling: + parentsNextSibling = parent.nextSibling + parent = parent.parent + if not parent: # This is the last element in the document. + break + if parentsNextSibling: + newChildsLastElement.next = parentsNextSibling + else: + newChildsLastElement.next = None + else: + nextChild = self.contents[position] + newChild.nextSibling = nextChild + if newChild.nextSibling: + newChild.nextSibling.previousSibling = newChild + newChildsLastElement.next = nextChild + + if newChildsLastElement.next: + newChildsLastElement.next.previous = newChildsLastElement + self.contents.insert(position, newChild) + + def append(self, tag): + """Appends the given tag to the contents of this tag.""" + self.insert(len(self.contents), tag) + + def findNext(self, name=None, attrs={}, text=None, **kwargs): + """Returns the first item that matches the given criteria and + appears after this Tag in the document.""" + return self._findOne(self.findAllNext, name, attrs, text, **kwargs) + + def findAllNext(self, name=None, attrs={}, text=None, limit=None, + **kwargs): + """Returns all items that match the given criteria and appear + after this Tag in the document.""" + return self._findAll(name, attrs, text, limit, self.nextGenerator, + **kwargs) + + def findNextSibling(self, name=None, attrs={}, text=None, **kwargs): + """Returns the closest sibling to this Tag that matches the + given criteria and appears after this Tag in the document.""" + return self._findOne(self.findNextSiblings, name, attrs, text, + **kwargs) + + def findNextSiblings(self, name=None, attrs={}, text=None, limit=None, + **kwargs): + """Returns the siblings of this Tag that match the given + criteria and appear after this Tag in the document.""" + return self._findAll(name, attrs, text, limit, + self.nextSiblingGenerator, **kwargs) + fetchNextSiblings = findNextSiblings # Compatibility with pre-3.x + + def findPrevious(self, name=None, attrs={}, text=None, **kwargs): + """Returns the first item that matches the given criteria and + appears before this Tag in the document.""" + return self._findOne(self.findAllPrevious, name, attrs, text, **kwargs) + + def findAllPrevious(self, name=None, attrs={}, text=None, limit=None, + **kwargs): + """Returns all items that match the given criteria and appear + before this Tag in the document.""" + return self._findAll(name, attrs, text, limit, self.previousGenerator, + **kwargs) + fetchPrevious = findAllPrevious # Compatibility with pre-3.x + + def findPreviousSibling(self, name=None, attrs={}, text=None, **kwargs): + """Returns the closest sibling to this Tag that matches the + given criteria and appears before this Tag in the document.""" + return self._findOne(self.findPreviousSiblings, name, attrs, text, + **kwargs) + + def findPreviousSiblings(self, name=None, attrs={}, text=None, + limit=None, **kwargs): + """Returns the siblings of this Tag that match the given + criteria and appear before this Tag in the document.""" + return self._findAll(name, attrs, text, limit, + self.previousSiblingGenerator, **kwargs) + fetchPreviousSiblings = findPreviousSiblings # Compatibility with pre-3.x + + def findParent(self, name=None, attrs={}, **kwargs): + """Returns the closest parent of this Tag that matches the given + criteria.""" + # NOTE: We can't use _findOne because findParents takes a different + # set of arguments. + r = None + l = self.findParents(name, attrs, 1) + if l: + r = l[0] + return r + + def findParents(self, name=None, attrs={}, limit=None, **kwargs): + """Returns the parents of this Tag that match the given + criteria.""" + + return self._findAll(name, attrs, None, limit, self.parentGenerator, + **kwargs) + fetchParents = findParents # Compatibility with pre-3.x + + #These methods do the real heavy lifting. + + def _findOne(self, method, name, attrs, text, **kwargs): + r = None + l = method(name, attrs, text, 1, **kwargs) + if l: + r = l[0] + return r + + def _findAll(self, name, attrs, text, limit, generator, **kwargs): + "Iterates over a generator looking for things that match." + + if isinstance(name, SoupStrainer): + strainer = name + else: + # Build a SoupStrainer + strainer = SoupStrainer(name, attrs, text, **kwargs) + results = ResultSet(strainer) + g = generator() + while True: + try: + i = g.next() + except StopIteration: + break + if i: + found = strainer.search(i) + if found: + results.append(found) + if limit and len(results) >= limit: + break + return results + + #These Generators can be used to navigate starting from both + #NavigableStrings and Tags. + def nextGenerator(self): + i = self + while i: + i = i.next + yield i + + def nextSiblingGenerator(self): + i = self + while i: + i = i.nextSibling + yield i + + def previousGenerator(self): + i = self + while i: + i = i.previous + yield i + + def previousSiblingGenerator(self): + i = self + while i: + i = i.previousSibling + yield i + + def parentGenerator(self): + i = self + while i: + i = i.parent + yield i + + # Utility methods + def substituteEncoding(self, str, encoding=None): + encoding = encoding or "utf-8" + return str.replace("%SOUP-ENCODING%", encoding) + + def toEncoding(self, s, encoding=None): + """Encodes an object to a string in some encoding, or to Unicode. + .""" + if isinstance(s, unicode): + if encoding: + s = s.encode(encoding) + elif isinstance(s, str): + if encoding: + s = s.encode(encoding) + else: + s = unicode(s) + else: + if encoding: + s = self.toEncoding(str(s), encoding) + else: + s = unicode(s) + return s + +class NavigableString(unicode, PageElement): + + def __getnewargs__(self): + return (NavigableString.__str__(self),) + + def __getattr__(self, attr): + """text.string gives you text. This is for backwards + compatibility for Navigable*String, but for CData* it lets you + get the string without the CData wrapper.""" + if attr == 'string': + return self + else: + raise AttributeError, "'%s' object has no attribute '%s'" % (self.__class__.__name__, attr) + + def __unicode__(self): + return str(self).decode(DEFAULT_OUTPUT_ENCODING) + + def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING): + if encoding: + return self.encode(encoding) + else: + return self + +class CData(NavigableString): + + def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING): + return "" % NavigableString.__str__(self, encoding) + +class ProcessingInstruction(NavigableString): + def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING): + output = self + if "%SOUP-ENCODING%" in output: + output = self.substituteEncoding(output, encoding) + return "" % self.toEncoding(output, encoding) + +class Comment(NavigableString): + def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING): + return "" % NavigableString.__str__(self, encoding) + +class Declaration(NavigableString): + def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING): + return "" % NavigableString.__str__(self, encoding) + +class Tag(PageElement): + + """Represents a found HTML tag with its attributes and contents.""" + + def _invert(h): + "Cheap function to invert a hash." + i = {} + for k,v in h.items(): + i[v] = k + return i + + XML_ENTITIES_TO_SPECIAL_CHARS = { "apos" : "'", + "quot" : '"', + "amp" : "&", + "lt" : "<", + "gt" : ">" } + + XML_SPECIAL_CHARS_TO_ENTITIES = _invert(XML_ENTITIES_TO_SPECIAL_CHARS) + + def _convertEntities(self, match): + """Used in a call to re.sub to replace HTML, XML, and numeric + entities with the appropriate Unicode characters. If HTML + entities are being converted, any unrecognized entities are + escaped.""" + x = match.group(1) + if self.convertHTMLEntities and x in name2codepoint: + return unichr(name2codepoint[x]) + elif x in self.XML_ENTITIES_TO_SPECIAL_CHARS: + if self.convertXMLEntities: + return self.XML_ENTITIES_TO_SPECIAL_CHARS[x] + else: + return u'&%s;' % x + elif len(x) > 0 and x[0] == '#': + # Handle numeric entities + if len(x) > 1 and x[1] == 'x': + return unichr(int(x[2:], 16)) + else: + return unichr(int(x[1:])) + + elif self.escapeUnrecognizedEntities: + return u'&%s;' % x + else: + return u'&%s;' % x + + def __init__(self, parser, name, attrs=None, parent=None, + previous=None): + "Basic constructor." + + # We don't actually store the parser object: that lets extracted + # chunks be garbage-collected + self.parserClass = parser.__class__ + self.isSelfClosing = parser.isSelfClosingTag(name) + self.name = name + if attrs == None: + attrs = [] + self.attrs = attrs + self.contents = [] + self.setup(parent, previous) + self.hidden = False + self.containsSubstitutions = False + self.convertHTMLEntities = parser.convertHTMLEntities + self.convertXMLEntities = parser.convertXMLEntities + self.escapeUnrecognizedEntities = parser.escapeUnrecognizedEntities + + # Convert any HTML, XML, or numeric entities in the attribute values. + convert = lambda(k, val): (k, + re.sub("&(#\d+|#x[0-9a-fA-F]+|\w+);", + self._convertEntities, + val)) + self.attrs = map(convert, self.attrs) + + def get(self, key, default=None): + """Returns the value of the 'key' attribute for the tag, or + the value given for 'default' if it doesn't have that + attribute.""" + return self._getAttrMap().get(key, default) + + def has_key(self, key): + return self._getAttrMap().has_key(key) + + def __getitem__(self, key): + """tag[key] returns the value of the 'key' attribute for the tag, + and throws an exception if it's not there.""" + return self._getAttrMap()[key] + + def __iter__(self): + "Iterating over a tag iterates over its contents." + return iter(self.contents) + + def __len__(self): + "The length of a tag is the length of its list of contents." + return len(self.contents) + + def __contains__(self, x): + return x in self.contents + + def __nonzero__(self): + "A tag is non-None even if it has no contents." + return True + + def __setitem__(self, key, value): + """Setting tag[key] sets the value of the 'key' attribute for the + tag.""" + self._getAttrMap() + self.attrMap[key] = value + found = False + for i in range(0, len(self.attrs)): + if self.attrs[i][0] == key: + self.attrs[i] = (key, value) + found = True + if not found: + self.attrs.append((key, value)) + self._getAttrMap()[key] = value + + def __delitem__(self, key): + "Deleting tag[key] deletes all 'key' attributes for the tag." + for item in self.attrs: + if item[0] == key: + self.attrs.remove(item) + #We don't break because bad HTML can define the same + #attribute multiple times. + self._getAttrMap() + if self.attrMap.has_key(key): + del self.attrMap[key] + + def __call__(self, *args, **kwargs): + """Calling a tag like a function is the same as calling its + findAll() method. Eg. tag('a') returns a list of all the A tags + found within this tag.""" + return apply(self.findAll, args, kwargs) + + def __getattr__(self, tag): + #print "Getattr %s.%s" % (self.__class__, tag) + if len(tag) > 3 and tag.rfind('Tag') == len(tag)-3: + return self.find(tag[:-3]) + elif tag.find('__') != 0: + return self.find(tag) + raise AttributeError, "'%s' object has no attribute '%s'" % (self.__class__, tag) + + def __eq__(self, other): + """Returns true iff this tag has the same name, the same attributes, + and the same contents (recursively) as the given tag. + + NOTE: right now this will return false if two tags have the + same attributes in a different order. Should this be fixed?""" + if not hasattr(other, 'name') or not hasattr(other, 'attrs') or not hasattr(other, 'contents') or self.name != other.name or self.attrs != other.attrs or len(self) != len(other): + return False + for i in range(0, len(self.contents)): + if self.contents[i] != other.contents[i]: + return False + return True + + def __ne__(self, other): + """Returns true iff this tag is not identical to the other tag, + as defined in __eq__.""" + return not self == other + + def __repr__(self, encoding=DEFAULT_OUTPUT_ENCODING): + """Renders this tag as a string.""" + return self.__str__(encoding) + + def __unicode__(self): + return self.__str__(None) + + BARE_AMPERSAND_OR_BRACKET = re.compile("([<>]|" + + "&(?!#\d+;|#x[0-9a-fA-F]+;|\w+;)" + + ")") + + def _sub_entity(self, x): + """Used with a regular expression to substitute the + appropriate XML entity for an XML special character.""" + return "&" + self.XML_SPECIAL_CHARS_TO_ENTITIES[x.group(0)[0]] + ";" + + def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING, + prettyPrint=False, indentLevel=0): + """Returns a string or Unicode representation of this tag and + its contents. To get Unicode, pass None for encoding. + + NOTE: since Python's HTML parser consumes whitespace, this + method is not certain to reproduce the whitespace present in + the original string.""" + + encodedName = self.toEncoding(self.name, encoding) + + attrs = [] + if self.attrs: + for key, val in self.attrs: + fmt = '%s="%s"' + if isString(val): + if self.containsSubstitutions and '%SOUP-ENCODING%' in val: + val = self.substituteEncoding(val, encoding) + + # The attribute value either: + # + # * Contains no embedded double quotes or single quotes. + # No problem: we enclose it in double quotes. + # * Contains embedded single quotes. No problem: + # double quotes work here too. + # * Contains embedded double quotes. No problem: + # we enclose it in single quotes. + # * Embeds both single _and_ double quotes. This + # can't happen naturally, but it can happen if + # you modify an attribute value after parsing + # the document. Now we have a bit of a + # problem. We solve it by enclosing the + # attribute in single quotes, and escaping any + # embedded single quotes to XML entities. + if '"' in val: + fmt = "%s='%s'" + if "'" in val: + # TODO: replace with apos when + # appropriate. + val = val.replace("'", "&squot;") + + # Now we're okay w/r/t quotes. But the attribute + # value might also contain angle brackets, or + # ampersands that aren't part of entities. We need + # to escape those to XML entities too. + val = self.BARE_AMPERSAND_OR_BRACKET.sub(self._sub_entity, val) + + attrs.append(fmt % (self.toEncoding(key, encoding), + self.toEncoding(val, encoding))) + close = '' + closeTag = '' + if self.isSelfClosing: + close = ' /' + else: + closeTag = '' % encodedName + + indentTag, indentContents = 0, 0 + if prettyPrint: + indentTag = indentLevel + space = (' ' * (indentTag-1)) + indentContents = indentTag + 1 + contents = self.renderContents(encoding, prettyPrint, indentContents) + if self.hidden: + s = contents + else: + s = [] + attributeString = '' + if attrs: + attributeString = ' ' + ' '.join(attrs) + if prettyPrint: + s.append(space) + s.append('<%s%s%s>' % (encodedName, attributeString, close)) + if prettyPrint: + s.append("\n") + s.append(contents) + if prettyPrint and contents and contents[-1] != "\n": + s.append("\n") + if prettyPrint and closeTag: + s.append(space) + s.append(closeTag) + if prettyPrint and closeTag and self.nextSibling: + s.append("\n") + s = ''.join(s) + return s + + def decompose(self): + """Recursively destroys the contents of this tree.""" + contents = [i for i in self.contents] + for i in contents: + if isinstance(i, Tag): + i.decompose() + else: + i.extract() + self.extract() + + def prettify(self, encoding=DEFAULT_OUTPUT_ENCODING): + return self.__str__(encoding, True) + + def renderContents(self, encoding=DEFAULT_OUTPUT_ENCODING, + prettyPrint=False, indentLevel=0): + """Renders the contents of this tag as a string in the given + encoding. If encoding is None, returns a Unicode string..""" + s=[] + for c in self: + text = None + if isinstance(c, NavigableString): + text = c.__str__(encoding) + elif isinstance(c, Tag): + s.append(c.__str__(encoding, prettyPrint, indentLevel)) + if text and prettyPrint: + text = text.strip() + if text: + if prettyPrint: + s.append(" " * (indentLevel-1)) + s.append(text) + if prettyPrint: + s.append("\n") + return ''.join(s) + + #Soup methods + + def find(self, name=None, attrs={}, recursive=True, text=None, + **kwargs): + """Return only the first child of this Tag matching the given + criteria.""" + r = None + l = self.findAll(name, attrs, recursive, text, 1, **kwargs) + if l: + r = l[0] + return r + findChild = find + + def findAll(self, name=None, attrs={}, recursive=True, text=None, + limit=None, **kwargs): + """Extracts a list of Tag objects that match the given + criteria. You can specify the name of the Tag and any + attributes you want the Tag to have. + + The value of a key-value pair in the 'attrs' map can be a + string, a list of strings, a regular expression object, or a + callable that takes a string and returns whether or not the + string matches for some custom definition of 'matches'. The + same is true of the tag name.""" + generator = self.recursiveChildGenerator + if not recursive: + generator = self.childGenerator + return self._findAll(name, attrs, text, limit, generator, **kwargs) + findChildren = findAll + + # Pre-3.x compatibility methods + first = find + fetch = findAll + + def fetchText(self, text=None, recursive=True, limit=None): + return self.findAll(text=text, recursive=recursive, limit=limit) + + def firstText(self, text=None, recursive=True): + return self.find(text=text, recursive=recursive) + + #Private methods + + def _getAttrMap(self): + """Initializes a map representation of this tag's attributes, + if not already initialized.""" + if not getattr(self, 'attrMap'): + self.attrMap = {} + for (key, value) in self.attrs: + self.attrMap[key] = value + return self.attrMap + + #Generator methods + def childGenerator(self): + for i in range(0, len(self.contents)): + yield self.contents[i] + raise StopIteration + + def recursiveChildGenerator(self): + stack = [(self, 0)] + while stack: + tag, start = stack.pop() + if isinstance(tag, Tag): + for i in range(start, len(tag.contents)): + a = tag.contents[i] + yield a + if isinstance(a, Tag) and tag.contents: + if i < len(tag.contents) - 1: + stack.append((tag, i+1)) + stack.append((a, 0)) + break + raise StopIteration + +# Next, a couple classes to represent queries and their results. +class SoupStrainer: + """Encapsulates a number of ways of matching a markup element (tag or + text).""" + + def __init__(self, name=None, attrs={}, text=None, **kwargs): + self.name = name + if isString(attrs): + kwargs['class'] = attrs + attrs = None + if kwargs: + if attrs: + attrs = attrs.copy() + attrs.update(kwargs) + else: + attrs = kwargs + self.attrs = attrs + self.text = text + + def __str__(self): + if self.text: + return self.text + else: + return "%s|%s" % (self.name, self.attrs) + + def searchTag(self, markupName=None, markupAttrs={}): + found = None + markup = None + if isinstance(markupName, Tag): + markup = markupName + markupAttrs = markup + callFunctionWithTagData = callable(self.name) \ + and not isinstance(markupName, Tag) + + if (not self.name) \ + or callFunctionWithTagData \ + or (markup and self._matches(markup, self.name)) \ + or (not markup and self._matches(markupName, self.name)): + if callFunctionWithTagData: + match = self.name(markupName, markupAttrs) + else: + match = True + markupAttrMap = None + for attr, matchAgainst in self.attrs.items(): + if not markupAttrMap: + if hasattr(markupAttrs, 'get'): + markupAttrMap = markupAttrs + else: + markupAttrMap = {} + for k,v in markupAttrs: + markupAttrMap[k] = v + attrValue = markupAttrMap.get(attr) + if not self._matches(attrValue, matchAgainst): + match = False + break + if match: + if markup: + found = markup + else: + found = markupName + return found + + def search(self, markup): + #print 'looking for %s in %s' % (self, markup) + found = None + # If given a list of items, scan it for a text element that + # matches. + if isList(markup) and not isinstance(markup, Tag): + for element in markup: + if isinstance(element, NavigableString) \ + and self.search(element): + found = element + break + # If it's a Tag, make sure its name or attributes match. + # Don't bother with Tags if we're searching for text. + elif isinstance(markup, Tag): + if not self.text: + found = self.searchTag(markup) + # If it's text, make sure the text matches. + elif isinstance(markup, NavigableString) or \ + isString(markup): + if self._matches(markup, self.text): + found = markup + else: + raise Exception, "I don't know how to match against a %s" \ + % markup.__class__ + return found + + def _matches(self, markup, matchAgainst): + #print "Matching %s against %s" % (markup, matchAgainst) + result = False + if matchAgainst == True and type(matchAgainst) == types.BooleanType: + result = markup != None + elif callable(matchAgainst): + result = matchAgainst(markup) + else: + #Custom match methods take the tag as an argument, but all + #other ways of matching match the tag name as a string. + if isinstance(markup, Tag): + markup = markup.name + if markup and not isString(markup): + markup = unicode(markup) + #Now we know that chunk is either a string, or None. + if hasattr(matchAgainst, 'match'): + # It's a regexp object. + result = markup and matchAgainst.search(markup) + elif isList(matchAgainst): + result = markup in matchAgainst + elif hasattr(matchAgainst, 'items'): + result = markup.has_key(matchAgainst) + elif matchAgainst and isString(markup): + if isinstance(markup, unicode): + matchAgainst = unicode(matchAgainst) + else: + matchAgainst = str(matchAgainst) + + if not result: + result = matchAgainst == markup + return result + +class ResultSet(list): + """A ResultSet is just a list that keeps track of the SoupStrainer + that created it.""" + def __init__(self, source): + list.__init__([]) + self.source = source + +# Now, some helper functions. + +def isList(l): + """Convenience method that works with all 2.x versions of Python + to determine whether or not something is listlike.""" + return hasattr(l, '__iter__') \ + or (type(l) in (types.ListType, types.TupleType)) + +def isString(s): + """Convenience method that works with all 2.x versions of Python + to determine whether or not something is stringlike.""" + try: + return isinstance(s, unicode) or isinstance(s, basestring) + except NameError: + return isinstance(s, str) + +def buildTagMap(default, *args): + """Turns a list of maps, lists, or scalars into a single map. + Used to build the SELF_CLOSING_TAGS, NESTABLE_TAGS, and + NESTING_RESET_TAGS maps out of lists and partial maps.""" + built = {} + for portion in args: + if hasattr(portion, 'items'): + #It's a map. Merge it. + for k,v in portion.items(): + built[k] = v + elif isList(portion): + #It's a list. Map each item to the default. + for k in portion: + built[k] = default + else: + #It's a scalar. Map it to the default. + built[portion] = default + return built + +# Now, the parser classes. + +class BeautifulStoneSoup(Tag, SGMLParser): + + """This class contains the basic parser and search code. It defines + a parser that knows nothing about tag behavior except for the + following: + + You can't close a tag without closing all the tags it encloses. + That is, "" actually means + "". + + [Another possible explanation is "", but since + this class defines no SELF_CLOSING_TAGS, it will never use that + explanation.] + + This class is useful for parsing XML or made-up markup languages, + or when BeautifulSoup makes an assumption counter to what you were + expecting.""" + + SELF_CLOSING_TAGS = {} + NESTABLE_TAGS = {} + RESET_NESTING_TAGS = {} + QUOTE_TAGS = {} + + MARKUP_MASSAGE = [(re.compile('(<[^<>]*)/>'), + lambda x: x.group(1) + ' />'), + (re.compile(']*)>'), + lambda x: '') + ] + + ROOT_TAG_NAME = u'[document]' + + HTML_ENTITIES = "html" + XML_ENTITIES = "xml" + XHTML_ENTITIES = "xhtml" + # TODO: This only exists for backwards-compatibility + ALL_ENTITIES = XHTML_ENTITIES + + # Used when determining whether a text node is all whitespace and + # can be replaced with a single space. A text node that contains + # fancy Unicode spaces (usually non-breaking) should be left + # alone. + STRIP_ASCII_SPACES = { 9: None, 10: None, 12: None, 13: None, 32: None, } + + def __init__(self, markup="", parseOnlyThese=None, fromEncoding=None, + markupMassage=True, smartQuotesTo=XML_ENTITIES, + convertEntities=None, selfClosingTags=None): + """The Soup object is initialized as the 'root tag', and the + provided markup (which can be a string or a file-like object) + is fed into the underlying parser. + + sgmllib will process most bad HTML, and the BeautifulSoup + class has some tricks for dealing with some HTML that kills + sgmllib, but Beautiful Soup can nonetheless choke or lose data + if your data uses self-closing tags or declarations + incorrectly. + + By default, Beautiful Soup uses regexes to sanitize input, + avoiding the vast majority of these problems. If the problems + don't apply to you, pass in False for markupMassage, and + you'll get better performance. + + The default parser massage techniques fix the two most common + instances of invalid HTML that choke sgmllib: + +
(No space between name of closing tag and tag close) + (Extraneous whitespace in declaration) + + You can pass in a custom list of (RE object, replace method) + tuples to get Beautiful Soup to scrub your input the way you + want.""" + + self.parseOnlyThese = parseOnlyThese + self.fromEncoding = fromEncoding + self.smartQuotesTo = smartQuotesTo + self.convertEntities = convertEntities + # Set the rules for how we'll deal with the entities we + # encounter + if self.convertEntities: + # It doesn't make sense to convert encoded characters to + # entities even while you're converting entities to Unicode. + # Just convert it all to Unicode. + self.smartQuotesTo = None + if convertEntities == self.HTML_ENTITIES: + self.convertXMLEntities = False + self.convertHTMLEntities = True + self.escapeUnrecognizedEntities = True + elif convertEntities == self.XHTML_ENTITIES: + self.convertXMLEntities = True + self.convertHTMLEntities = True + self.escapeUnrecognizedEntities = False + elif convertEntities == self.XML_ENTITIES: + self.convertXMLEntities = True + self.convertHTMLEntities = False + self.escapeUnrecognizedEntities = False + else: + self.convertXMLEntities = False + self.convertHTMLEntities = False + self.escapeUnrecognizedEntities = False + + self.instanceSelfClosingTags = buildTagMap(None, selfClosingTags) + SGMLParser.__init__(self) + + if hasattr(markup, 'read'): # It's a file-type object. + markup = markup.read() + self.markup = markup + self.markupMassage = markupMassage + try: + self._feed() + except StopParsing: + pass + self.markup = None # The markup can now be GCed + + def convert_charref(self, name): + """This method fixes a bug in Python's SGMLParser.""" + try: + n = int(name) + except ValueError: + return + if not 0 <= n <= 127 : # ASCII ends at 127, not 255 + return + return self.convert_codepoint(n) + + def _feed(self, inDocumentEncoding=None): + # Convert the document to Unicode. + markup = self.markup + if isinstance(markup, unicode): + if not hasattr(self, 'originalEncoding'): + self.originalEncoding = None + else: + dammit = UnicodeDammit\ + (markup, [self.fromEncoding, inDocumentEncoding], + smartQuotesTo=self.smartQuotesTo) + markup = dammit.unicode + self.originalEncoding = dammit.originalEncoding + if markup: + if self.markupMassage: + if not isList(self.markupMassage): + self.markupMassage = self.MARKUP_MASSAGE + for fix, m in self.markupMassage: + markup = fix.sub(m, markup) + # TODO: We get rid of markupMassage so that the + # soup object can be deepcopied later on. Some + # Python installations can't copy regexes. If anyone + # was relying on the existence of markupMassage, this + # might cause problems. + del(self.markupMassage) + self.reset() + + SGMLParser.feed(self, markup) + # Close out any unfinished strings and close all the open tags. + self.endData() + while self.currentTag.name != self.ROOT_TAG_NAME: + self.popTag() + + def __getattr__(self, methodName): + """This method routes method call requests to either the SGMLParser + superclass or the Tag superclass, depending on the method name.""" + #print "__getattr__ called on %s.%s" % (self.__class__, methodName) + + if methodName.find('start_') == 0 or methodName.find('end_') == 0 \ + or methodName.find('do_') == 0: + return SGMLParser.__getattr__(self, methodName) + elif methodName.find('__') != 0: + return Tag.__getattr__(self, methodName) + else: + raise AttributeError + + def isSelfClosingTag(self, name): + """Returns true iff the given string is the name of a + self-closing tag according to this parser.""" + return self.SELF_CLOSING_TAGS.has_key(name) \ + or self.instanceSelfClosingTags.has_key(name) + + def reset(self): + Tag.__init__(self, self, self.ROOT_TAG_NAME) + self.hidden = 1 + SGMLParser.reset(self) + self.currentData = [] + self.currentTag = None + self.tagStack = [] + self.quoteStack = [] + self.pushTag(self) + + def popTag(self): + tag = self.tagStack.pop() + # Tags with just one string-owning child get the child as a + # 'string' property, so that soup.tag.string is shorthand for + # soup.tag.contents[0] + if len(self.currentTag.contents) == 1 and \ + isinstance(self.currentTag.contents[0], NavigableString): + self.currentTag.string = self.currentTag.contents[0] + + #print "Pop", tag.name + if self.tagStack: + self.currentTag = self.tagStack[-1] + return self.currentTag + + def pushTag(self, tag): + #print "Push", tag.name + if self.currentTag: + self.currentTag.contents.append(tag) + self.tagStack.append(tag) + self.currentTag = self.tagStack[-1] + + def endData(self, containerClass=NavigableString): + if self.currentData: + currentData = ''.join(self.currentData) + if not currentData.translate(self.STRIP_ASCII_SPACES): + if '\n' in currentData: + currentData = '\n' + else: + currentData = ' ' + self.currentData = [] + if self.parseOnlyThese and len(self.tagStack) <= 1 and \ + (not self.parseOnlyThese.text or \ + not self.parseOnlyThese.search(currentData)): + return + o = containerClass(currentData) + o.setup(self.currentTag, self.previous) + if self.previous: + self.previous.next = o + self.previous = o + self.currentTag.contents.append(o) + + + def _popToTag(self, name, inclusivePop=True): + """Pops the tag stack up to and including the most recent + instance of the given tag. If inclusivePop is false, pops the tag + stack up to but *not* including the most recent instqance of + the given tag.""" + #print "Popping to %s" % name + if name == self.ROOT_TAG_NAME: + return + + numPops = 0 + mostRecentTag = None + for i in range(len(self.tagStack)-1, 0, -1): + if name == self.tagStack[i].name: + numPops = len(self.tagStack)-i + break + if not inclusivePop: + numPops = numPops - 1 + + for i in range(0, numPops): + mostRecentTag = self.popTag() + return mostRecentTag + + def _smartPop(self, name): + + """We need to pop up to the previous tag of this type, unless + one of this tag's nesting reset triggers comes between this + tag and the previous tag of this type, OR unless this tag is a + generic nesting trigger and another generic nesting trigger + comes between this tag and the previous tag of this type. + + Examples: +

FooBar *

* should pop to 'p', not 'b'. +

FooBar *

* should pop to 'table', not 'p'. +

Foo

Bar *

* should pop to 'tr', not 'p'. + +

    • *
    • * should pop to 'ul', not the first 'li'. +
    • ** should pop to 'table', not the first 'tr' + tag should + implicitly close the previous tag within the same
    ** should pop to 'tr', not the first 'td' + """ + + nestingResetTriggers = self.NESTABLE_TAGS.get(name) + isNestable = nestingResetTriggers != None + isResetNesting = self.RESET_NESTING_TAGS.has_key(name) + popTo = None + inclusive = True + for i in range(len(self.tagStack)-1, 0, -1): + p = self.tagStack[i] + if (not p or p.name == name) and not isNestable: + #Non-nestable tags get popped to the top or to their + #last occurance. + popTo = name + break + if (nestingResetTriggers != None + and p.name in nestingResetTriggers) \ + or (nestingResetTriggers == None and isResetNesting + and self.RESET_NESTING_TAGS.has_key(p.name)): + + #If we encounter one of the nesting reset triggers + #peculiar to this tag, or we encounter another tag + #that causes nesting to reset, pop up to but not + #including that tag. + popTo = p.name + inclusive = False + break + p = p.parent + if popTo: + self._popToTag(popTo, inclusive) + + def unknown_starttag(self, name, attrs, selfClosing=0): + #print "Start tag %s: %s" % (name, attrs) + if self.quoteStack: + #This is not a real tag. + #print "<%s> is not real!" % name + attrs = ''.join(map(lambda(x, y): ' %s="%s"' % (x, y), attrs)) + self.handle_data('<%s%s>' % (name, attrs)) + return + self.endData() + + if not self.isSelfClosingTag(name) and not selfClosing: + self._smartPop(name) + + if self.parseOnlyThese and len(self.tagStack) <= 1 \ + and (self.parseOnlyThese.text or not self.parseOnlyThese.searchTag(name, attrs)): + return + + tag = Tag(self, name, attrs, self.currentTag, self.previous) + if self.previous: + self.previous.next = tag + self.previous = tag + self.pushTag(tag) + if selfClosing or self.isSelfClosingTag(name): + self.popTag() + if name in self.QUOTE_TAGS: + #print "Beginning quote (%s)" % name + self.quoteStack.append(name) + self.literal = 1 + return tag + + def unknown_endtag(self, name): + #print "End tag %s" % name + if self.quoteStack and self.quoteStack[-1] != name: + #This is not a real end tag. + #print " is not real!" % name + self.handle_data('' % name) + return + self.endData() + self._popToTag(name) + if self.quoteStack and self.quoteStack[-1] == name: + self.quoteStack.pop() + self.literal = (len(self.quoteStack) > 0) + + def handle_data(self, data): + self.currentData.append(data) + + def _toStringSubclass(self, text, subclass): + """Adds a certain piece of text to the tree as a NavigableString + subclass.""" + self.endData() + self.handle_data(text) + self.endData(subclass) + + def handle_pi(self, text): + """Handle a processing instruction as a ProcessingInstruction + object, possibly one with a %SOUP-ENCODING% slot into which an + encoding will be plugged later.""" + if text[:3] == "xml": + text = u"xml version='1.0' encoding='%SOUP-ENCODING%'" + self._toStringSubclass(text, ProcessingInstruction) + + def handle_comment(self, text): + "Handle comments as Comment objects." + self._toStringSubclass(text, Comment) + + def handle_charref(self, ref): + "Handle character references as data." + if self.convertEntities: + data = unichr(int(ref)) + else: + data = '&#%s;' % ref + self.handle_data(data) + + def handle_entityref(self, ref): + """Handle entity references as data, possibly converting known + HTML and/or XML entity references to the corresponding Unicode + characters.""" + data = None + if self.convertHTMLEntities: + try: + data = unichr(name2codepoint[ref]) + except KeyError: + pass + + if not data and self.convertXMLEntities: + data = self.XML_ENTITIES_TO_SPECIAL_CHARS.get(ref) + + if not data and self.convertHTMLEntities and \ + not self.XML_ENTITIES_TO_SPECIAL_CHARS.get(ref): + # TODO: We've got a problem here. We're told this is + # an entity reference, but it's not an XML entity + # reference or an HTML entity reference. Nonetheless, + # the logical thing to do is to pass it through as an + # unrecognized entity reference. + # + # Except: when the input is "&carol;" this function + # will be called with input "carol". When the input is + # "AT&T", this function will be called with input + # "T". We have no way of knowing whether a semicolon + # was present originally, so we don't know whether + # this is an unknown entity or just a misplaced + # ampersand. + # + # The more common case is a misplaced ampersand, so I + # escape the ampersand and omit the trailing semicolon. + data = "&%s" % ref + if not data: + # This case is different from the one above, because we + # haven't already gone through a supposedly comprehensive + # mapping of entities to Unicode characters. We might not + # have gone through any mapping at all. So the chances are + # very high that this is a real entity, and not a + # misplaced ampersand. + data = "&%s;" % ref + self.handle_data(data) + + def handle_decl(self, data): + "Handle DOCTYPEs and the like as Declaration objects." + self._toStringSubclass(data, Declaration) + + def parse_declaration(self, i): + """Treat a bogus SGML declaration as raw data. Treat a CDATA + declaration as a CData object.""" + j = None + if self.rawdata[i:i+9] == '', i) + if k == -1: + k = len(self.rawdata) + data = self.rawdata[i+9:k] + j = k+3 + self._toStringSubclass(data, CData) + else: + try: + j = SGMLParser.parse_declaration(self, i) + except SGMLParseError: + toHandle = self.rawdata[i:] + self.handle_data(toHandle) + j = i + len(toHandle) + return j + +class BeautifulSoup(BeautifulStoneSoup): + + """This parser knows the following facts about HTML: + + * Some tags have no closing tag and should be interpreted as being + closed as soon as they are encountered. + + * The text inside some tags (ie. 'script') may contain tags which + are not really part of the document and which should be parsed + as text, not tags. If you want to parse the text as tags, you can + always fetch it and parse it explicitly. + + * Tag nesting rules: + + Most tags can't be nested at all. For instance, the occurance of + a

    tag should implicitly close the previous

    tag. + +

    Para1

    Para2 + should be transformed into: +

    Para1

    Para2 + + Some tags can be nested arbitrarily. For instance, the occurance + of a

    tag should _not_ implicitly close the previous +
    tag. + + Alice said:
    Bob said:
    Blah + should NOT be transformed into: + Alice said:
    Bob said:
    Blah + + Some tags can be nested, but the nesting is reset by the + interposition of other tags. For instance, a
    , + but not close a tag in another table. + +
    BlahBlah + should be transformed into: +
    BlahBlah + but, + Blah
    Blah + should NOT be transformed into + Blah
    Blah + + Differing assumptions about tag nesting rules are a major source + of problems with the BeautifulSoup class. If BeautifulSoup is not + treating as nestable a tag your page author treats as nestable, + try ICantBelieveItsBeautifulSoup, MinimalSoup, or + BeautifulStoneSoup before writing your own subclass.""" + + def __init__(self, *args, **kwargs): + if not kwargs.has_key('smartQuotesTo'): + kwargs['smartQuotesTo'] = self.HTML_ENTITIES + BeautifulStoneSoup.__init__(self, *args, **kwargs) + + SELF_CLOSING_TAGS = buildTagMap(None, + ['br' , 'hr', 'input', 'img', 'meta', + 'spacer', 'link', 'frame', 'base']) + + QUOTE_TAGS = {'script' : None, 'textarea' : None} + + #According to the HTML standard, each of these inline tags can + #contain another tag of the same type. Furthermore, it's common + #to actually use these tags this way. + NESTABLE_INLINE_TAGS = ['span', 'font', 'q', 'object', 'bdo', 'sub', 'sup', + 'center'] + + #According to the HTML standard, these block tags can contain + #another tag of the same type. Furthermore, it's common + #to actually use these tags this way. + NESTABLE_BLOCK_TAGS = ['blockquote', 'div', 'fieldset', 'ins', 'del'] + + #Lists can contain other lists, but there are restrictions. + NESTABLE_LIST_TAGS = { 'ol' : [], + 'ul' : [], + 'li' : ['ul', 'ol'], + 'dl' : [], + 'dd' : ['dl'], + 'dt' : ['dl'] } + + #Tables can contain other tables, but there are restrictions. + NESTABLE_TABLE_TAGS = {'table' : [], + 'tr' : ['table', 'tbody', 'tfoot', 'thead'], + 'td' : ['tr'], + 'th' : ['tr'], + 'thead' : ['table'], + 'tbody' : ['table'], + 'tfoot' : ['table'], + } + + NON_NESTABLE_BLOCK_TAGS = ['address', 'form', 'p', 'pre'] + + #If one of these tags is encountered, all tags up to the next tag of + #this type are popped. + RESET_NESTING_TAGS = buildTagMap(None, NESTABLE_BLOCK_TAGS, 'noscript', + NON_NESTABLE_BLOCK_TAGS, + NESTABLE_LIST_TAGS, + NESTABLE_TABLE_TAGS) + + NESTABLE_TAGS = buildTagMap([], NESTABLE_INLINE_TAGS, NESTABLE_BLOCK_TAGS, + NESTABLE_LIST_TAGS, NESTABLE_TABLE_TAGS) + + # Used to detect the charset in a META tag; see start_meta + CHARSET_RE = re.compile("((^|;)\s*charset=)([^;]*)") + + def start_meta(self, attrs): + """Beautiful Soup can detect a charset included in a META tag, + try to convert the document to that charset, and re-parse the + document from the beginning.""" + httpEquiv = None + contentType = None + contentTypeIndex = None + tagNeedsEncodingSubstitution = False + + for i in range(0, len(attrs)): + key, value = attrs[i] + key = key.lower() + if key == 'http-equiv': + httpEquiv = value + elif key == 'content': + contentType = value + contentTypeIndex = i + + if httpEquiv and contentType: # It's an interesting meta tag. + match = self.CHARSET_RE.search(contentType) + if match: + if getattr(self, 'declaredHTMLEncoding') or \ + (self.originalEncoding == self.fromEncoding): + # This is our second pass through the document, or + # else an encoding was specified explicitly and it + # worked. Rewrite the meta tag. + newAttr = self.CHARSET_RE.sub\ + (lambda(match):match.group(1) + + "%SOUP-ENCODING%", contentType) + attrs[contentTypeIndex] = (attrs[contentTypeIndex][0], + newAttr) + tagNeedsEncodingSubstitution = True + else: + # This is our first pass through the document. + # Go through it again with the new information. + newCharset = match.group(3) + if newCharset and newCharset != self.originalEncoding: + self.declaredHTMLEncoding = newCharset + self._feed(self.declaredHTMLEncoding) + raise StopParsing + tag = self.unknown_starttag("meta", attrs) + if tag and tagNeedsEncodingSubstitution: + tag.containsSubstitutions = True + +class StopParsing(Exception): + pass + +class ICantBelieveItsBeautifulSoup(BeautifulSoup): + + """The BeautifulSoup class is oriented towards skipping over + common HTML errors like unclosed tags. However, sometimes it makes + errors of its own. For instance, consider this fragment: + + FooBar + + This is perfectly valid (if bizarre) HTML. However, the + BeautifulSoup class will implicitly close the first b tag when it + encounters the second 'b'. It will think the author wrote + "FooBar", and didn't close the first 'b' tag, because + there's no real-world reason to bold something that's already + bold. When it encounters '' it will close two more 'b' + tags, for a grand total of three tags closed instead of two. This + can throw off the rest of your document structure. The same is + true of a number of other tags, listed below. + + It's much more common for someone to forget to close a 'b' tag + than to actually use nested 'b' tags, and the BeautifulSoup class + handles the common case. This class handles the not-co-common + case: where you can't believe someone wrote what they did, but + it's valid HTML and BeautifulSoup screwed up by assuming it + wouldn't be.""" + + I_CANT_BELIEVE_THEYRE_NESTABLE_INLINE_TAGS = \ + ['em', 'big', 'i', 'small', 'tt', 'abbr', 'acronym', 'strong', + 'cite', 'code', 'dfn', 'kbd', 'samp', 'strong', 'var', 'b', + 'big'] + + I_CANT_BELIEVE_THEYRE_NESTABLE_BLOCK_TAGS = ['noscript'] + + NESTABLE_TAGS = buildTagMap([], BeautifulSoup.NESTABLE_TAGS, + I_CANT_BELIEVE_THEYRE_NESTABLE_BLOCK_TAGS, + I_CANT_BELIEVE_THEYRE_NESTABLE_INLINE_TAGS) + +class MinimalSoup(BeautifulSoup): + """The MinimalSoup class is for parsing HTML that contains + pathologically bad markup. It makes no assumptions about tag + nesting, but it does know which tags are self-closing, that + + + + + + + + + + + +
    + +

    Session History

      +

      + Clear? +

      +
      + +
      +
      +

      What is Dumble?

      +
      Dumble is an oohEmbed powered experiment in creating a tumblelog from a user's delicious.com links.

      To create a tumblelog for a different delicious user, just modify the details below. +
      +
      + Username: + A tag (optional): + + +
      +
      +

      +

        +
        + +
        + Need a link to this page?   +  Or get the feed!  RSS Feed icon

        + There's also a View in Dumble bookmarklet and a similar greasemonkey script to integrate Dumble with delicious. +
        +
        +

        +

          + +
          +
          + +
          +
          + +
          + +
          + +
          + More? +
          + +
          + + + + diff --git a/app/dumble/jquery.cookie.js b/app/dumble/jquery.cookie.js new file mode 100644 index 0000000..fe16025 --- /dev/null +++ b/app/dumble/jquery.cookie.js @@ -0,0 +1,51 @@ +/** + * Cookie plugin + * + * Copyright (c) 2006 Klaus Hartl (stilbuero.de) + * Dual licensed under the MIT and GPL licenses: + * http://www.opensource.org/licenses/mit-license.php + * http://www.gnu.org/licenses/gpl.html + * + */ + +jQuery.cookie = function(name, value, options) { + if (typeof value != 'undefined') { // name and value given, set cookie + options = options || {}; + if (value === null) { + value = ''; + options.expires = -1; + } + var expires = ''; + if (options.expires && (typeof options.expires == 'number' || options.expires.toUTCString)) { + var date; + if (typeof options.expires == 'number') { + date = new Date(); + date.setTime(date.getTime() + (options.expires * 24 * 60 * 60 * 1000)); + } else { + date = options.expires; + } + expires = '; expires=' + date.toUTCString(); // use expires attribute, max-age is not supported by IE + } + // CAUTION: Needed to parenthesize options.path and options.domain + // in the following expressions, otherwise they evaluate to undefined + // in the packed version for some reason... + var path = options.path ? '; path=' + (options.path) : ''; + var domain = options.domain ? '; domain=' + (options.domain) : ''; + var secure = options.secure ? '; secure' : ''; + document.cookie = [name, '=', encodeURIComponent(value), expires, path, domain, secure].join(''); + } else { // only name given, get cookie + var cookieValue = null; + if (document.cookie && document.cookie != '') { + var cookies = document.cookie.split(';'); + for (var i = 0; i < cookies.length; i++) { + var cookie = jQuery.trim(cookies[i]); + // Does this cookie string begin with the name we want? + if (cookie.substring(0, name.length + 1) == (name + '=')) { + cookieValue = decodeURIComponent(cookie.substring(name.length + 1)); + break; + } + } + } + return cookieValue; + } +}; diff --git a/app/feedparser.py b/app/feedparser.py new file mode 100755 index 0000000..bb802df --- /dev/null +++ b/app/feedparser.py @@ -0,0 +1,2858 @@ +#!/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__ = "4.1"# + "$Revision: 1.92 $"[11:15] + "-cvs" +__license__ = """Copyright (c) 2002-2006, 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 "] +_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"] + +# ---------- 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): + data = data.replace('&', '&') + data = data.replace('>', '>') + data = data.replace('<', '<') + 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 + +# ---------- 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('' % (tag, ''.join([' %s="%s"' % t for t in 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: + return self.push(prefix + suffix, 1) + + 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 + '_' + + # call special handler (if defined) or default handler + methodname = '_end_' + prefix + suffix + try: + 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 + 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 + else: + # entity resolution graciously donated by Aaron Swartz + def name2cp(k): + import htmlentitydefs + if hasattr(htmlentitydefs, 'name2codepoint'): # requires Python 2.3 + return htmlentitydefs.name2codepoint[k] + k = htmlentitydefs.entitydefs[k] + if k.startswith('&#') and k.endswith(';'): + return int(k[2:-1]) # not in latin-1 + return ord(k) + try: name2cp(ref) + except KeyError: text = '&%s;' % ref + else: text = unichr(name2cp(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: k = len(self.rawdata) + self.handle_data(_xmlescape(self.rawdata[i+9:k]), 0) + return k+3 + else: + k = self.rawdata.find('>', i) + return k+1 + + 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 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() + 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) + + # remove temporary cruft from contentparams + try: + del self.contentparams['mode'] + except KeyError: + pass + try: + del self.contentparams['base64'] + except KeyError: + pass + + # resolve relative URIs within embedded markup + if self.mapContentType(self.contentparams.get('type', 'text/html')) in self.html_types: + if element in self.can_contain_relative_uris: + output = _resolveRelativeURIs(output, self.baseuri, self.encoding) + + # sanitize embedded markup + if self.mapContentType(self.contentparams.get('type', 'text/html')) in self.html_types: + if element in self.can_contain_dangerous_markup: + output = _sanitizeHTML(output, self.encoding) + + if self.encoding and type(output) != type(u''): + try: + output = unicode(output, self.encoding) + except: + pass + + # categories/tags/keywords/whatever are handled in _end_category + if element == 'category': + 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 + 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 + + def _mapToStandardPrefix(self, name): + colonpos = name.find(':') + if colonpos <> -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 not self.version: + 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): + self.inimage = 1 + self.push('image', 0) + context = self._getContext() + context.setdefault('image', FeedParserDict()) + + def _end_image(self): + self.pop('image') + self.inimage = 0 + + def _start_textinput(self, attrsD): + self.intextinput = 1 + self.push('textinput', 0) + context = self._getContext() + context.setdefault('textinput', FeedParserDict()) + _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['textinput']['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['image']['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['image']['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) + elif self.inimage: + context = self._getContext() + context['image']['href'] = value + elif self.intextinput: + context = self._getContext() + context['textinput']['link'] = 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.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 = context.get(key) + if not author: return + emailmatch = re.search(r'''(([a-zA-Z0-9\_\-\.\+]+)@((\[[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.)|(([a-zA-Z0-9\-]+\.)+))([a-zA-Z]{2,4}|[0-9]{1,3})(\]?))''', author) + if not emailmatch: return + email = emailmatch.group(0) + # probably a better way to do the following, but it passes all the tests + author = author.replace(email, '') + 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() + context.setdefault('%s_detail' % key, FeedParserDict()) + context['%s_detail' % key]['name'] = author + 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 + 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): + self.push('license', 1) + value = self._getAttribute(attrsD, 'rdf:resource') + if value: + self.elementstack[-1][2].append(value) + self.pop('license') + + def _start_creativecommons_license(self, attrsD): + self.push('license', 1) + + def _end_creativecommons_license(self): + self.pop('license') + + 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(FeedParserDict({'term': term, 'scheme': scheme, 'label': label})) + + 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') + attrsD.setdefault('type', 'text/html') + attrsD = self._itsAnHrefDamnIt(attrsD) + if attrsD.has_key('href'): + attrsD['href'] = self.resolveURI(attrsD['href']) + expectingText = self.infeed or self.inentry or self.insource + context = self._getContext() + context.setdefault('links', []) + context['links'].append(FeedParserDict(attrsD)) + if attrsD['rel'] == 'enclosure': + self._start_enclosure(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() + if self.intextinput: + context['textinput']['link'] = value + if self.inimage: + context['image']['link'] = value + _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): + 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): + value = self.popContent('title') + context = self._getContext() + if self.intextinput: + context['textinput']['title'] = value + elif self.inimage: + context['image']['title'] = value + _end_dc_title = _end_title + _end_media_title = _end_title + + 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) + + 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') + context = self._getContext() + if self.intextinput: + context['textinput']['description'] = value + elif self.inimage: + context['image']['description'] = value + self._summaryKey = None + _end_abstract = _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) + self._getContext().setdefault('enclosures', []).append(FeedParserDict(attrsD)) + href = attrsD.get('href') + if href: + context = self._getContext() + if not context.get('id'): + context['id'] = href + + def _start_source(self, attrsD): + self.insource = 1 + + def _end_source(self): + self.insource = 0 + 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 + +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 + + def startPrefixMapping(self, prefix, uri): + self.trackNamespace(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 + if prefix: + localname = prefix + ':' + localname + localname = str(localname).lower() + 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)) + + # 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 = {} + 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 + 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): + elements_no_end_tag = ['area', 'base', 'basefont', 'br', 'col', 'frame', 'hr', + 'img', 'input', 'isindex', 'link', 'meta', 'param'] + + def __init__(self, encoding): + self.encoding = encoding + 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 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) + + def normalize_attrs(self, attrs): + # utility method to be called by descendants + 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] + 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 = []
+        # thanks to Kevin Marks for this breathtaking hack to deal with (valid) high-bit attribute values in UTF-8 feeds
+        for key, value in attrs:
+            if type(value) != type(u''):
+                value = unicode(value, self.encoding)
+            uattrs.append((unicode(key, self.encoding), value))
+        strattrs = u''.join([u' %s="%s"' % (key, value) for key, value in uattrs]).encode(self.encoding)
+        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. + 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. + 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_text, 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 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): + sgmllib.SGMLParser.__init__(self) + _FeedParserMixin.__init__(self, baseuri, baselang, encoding) + + 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(''', ''') + 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 + +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): + _BaseHTMLProcessor.__init__(self, encoding) + self.baseuri = baseuri + + def resolveURI(self, uri): + return _urljoin(self.baseuri, uri) + + def unknown_starttag(self, tag, 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): + if _debug: sys.stderr.write('entering _resolveRelativeURIs\n') + p = _RelativeURIResolver(baseURI, encoding) + p.feed(htmlSource) + return p.output() + +class _HTMLSanitizer(_BaseHTMLProcessor): + acceptable_elements = ['a', 'abbr', 'acronym', 'address', 'area', 'b', 'big', + 'blockquote', 'br', 'button', 'caption', 'center', 'cite', 'code', 'col', + 'colgroup', 'dd', 'del', 'dfn', 'dir', 'div', 'dl', 'dt', 'em', 'fieldset', + 'font', 'form', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'hr', 'i', 'img', 'input', + 'ins', 'kbd', 'label', 'legend', 'li', 'map', 'menu', 'ol', 'optgroup', + 'option', 'p', 'pre', 'q', 's', 'samp', 'select', 'small', 'span', 'strike', + 'strong', 'sub', 'sup', 'table', 'tbody', 'td', 'textarea', 'tfoot', 'th', + 'thead', 'tr', 'tt', 'u', 'ul', 'var'] + + acceptable_attributes = ['abbr', 'accept', 'accept-charset', 'accesskey', + 'action', 'align', 'alt', 'axis', 'border', 'cellpadding', 'cellspacing', + 'char', 'charoff', 'charset', 'checked', 'cite', 'class', 'clear', 'cols', + 'colspan', 'color', 'compact', 'coords', 'datetime', 'dir', 'disabled', + 'enctype', 'for', 'frame', 'headers', 'height', 'href', 'hreflang', 'hspace', + 'id', 'ismap', 'label', 'lang', 'longdesc', 'maxlength', 'media', 'method', + 'multiple', 'name', 'nohref', 'noshade', 'nowrap', 'prompt', 'readonly', + 'rel', 'rev', 'rows', 'rowspan', 'rules', 'scope', 'selected', 'shape', 'size', + 'span', 'src', 'start', 'summary', 'tabindex', 'target', 'title', 'type', + 'usemap', 'valign', 'value', 'vspace', 'width'] + + unacceptable_elements_with_end_tag = ['script', 'applet'] + + def reset(self): + _BaseHTMLProcessor.reset(self) + self.unacceptablestack = 0 + + def unknown_starttag(self, tag, attrs): + if not tag in self.acceptable_elements: + if tag in self.unacceptable_elements_with_end_tag: + self.unacceptablestack += 1 + return + attrs = self.normalize_attrs(attrs) + attrs = [(key, value) for key, value in attrs if key in self.acceptable_attributes] + _BaseHTMLProcessor.unknown_starttag(self, tag, attrs) + + def unknown_endtag(self, tag): + if not tag in self.acceptable_elements: + if tag in self.unacceptable_elements_with_end_tag: + self.unacceptablestack -= 1 + return + _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 _sanitizeHTML(htmlSource, encoding): + p = _HTMLSanitizer(encoding) + 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 must be a tuple of 9 integers + as returned by gmtime() in the standard Python time module. This MUST + be in GMT (Greenwich Mean Time). The formatted date/time will be used + as the value of an If-Modified-Since request header. + + If the 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() + # 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 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-MM', 'YYYY-?OOO', + 'YY-?MM-?DD', 'YY-?OOO', 'YYYY', + '-YY-?MM', '-OOO', '-YY', + '--MM-?DD', '--MM', + '---DD', + 'CC', ''] +_iso8601_re = [ + tmpl.replace( + 'YYYY', r'(?P\d{4})').replace( + 'YY', r'(?P\d\d)').replace( + 'MM', r'(?P[01]\d)').replace( + 'DD', r'(?P[0123]\d)').replace( + 'OOO', r'(?P[0123]\d\d)').replace( + 'CC', r'(?P\d\d$)') + + r'(T?(?P\d{2}):(?P\d{2})' + + r'(:(?P\d{2}))?' + + r'(?P[+-](?P\d{2})(:(?P\d{2}))?|Z)?)?' + for tmpl in _iso8601_tmpl] +del tmpl +_iso8601_matches = [re.compile(regex).match for regex in _iso8601_re] +del regex +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(params.get('second', 0)) + # weekday is normalized by mktime(), we can ignore it + weekday = 0 + # daylight savings is complex, but not needed for feedparser's purposes + # as time zones, if specified, include mention of whether it is active + # (e.g. PST vs. PDT, CET). Using -1 is implementation-dependent and + # and most implementations have DST bugs + daylight_savings_flag = 0 + tm = [year, month, day, hour, minute, second, weekday, + ordinal, daylight_savings_flag] + # ISO 8601 time zone adjustments + tz = params.get('tz') + if tz and tz != 'Z': + if tz[0] == '-': + tm[3] += int(params.get('tzhour', 0)) + tm[4] += int(params.get('tzmin', 0)) + elif tz[0] == '+': + tm[3] -= int(params.get('tzhour', 0)) + tm[4] -= int(params.get('tzmin', 0)) + else: + return None + # Python's time.mktime() is a wrapper around the ANSI C mktime(3c) + # which is guaranteed to normalize d/m/y/h/m/s. + # Many implementations have bugs, but we'll pretend they don't. + return time.localtime(time.mktime(tm)) +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(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' + 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 + ''' + entity_pattern = re.compile(r']*?)>', re.MULTILINE) + data = entity_pattern.sub('', data) + doctype_pattern = re.compile(r']*?)>', re.MULTILINE) + doctype_results = doctype_pattern.findall(data) + doctype = doctype_results and doctype_results[0] or '' + if doctype.lower().count('netscape'): + version = 'rss091n' + else: + version = None + data = doctype_pattern.sub('', data) + return version, data + +def parse(url_file_stream_or_string, etag=None, modified=None, agent=None, referrer=None, handlers=[]): + '''Parse a feed from a URL, file, stream, or string''' + result = FeedParserDict() + result['feed'] = FeedParserDict() + result['entries'] = [] + if _XML_AVAILABLE: + result['bozo'] = 0 + if type(handlers) == types.InstanceType: + handlers = [handlers] + try: + f = _open_resource(url_file_stream_or_string, etag, modified, agent, referrer, handlers) + data = f.read() + except Exception, e: + result['bozo'] = 1 + result['bozo_exception'] = e + data = '' + f = None + + # if feed is gzip-compressed, decompress it + if f and data and hasattr(f, 'headers'): + if gzip and f.headers.get('content-encoding', '') == 'gzip': + try: + data = gzip.GzipFile(fileobj=_StringIO(data)).read() + except Exception, e: + # Some feeds claim to be gzipped but they're not, so + # we get garbage. Ideally, we should re-request the + # feed without the 'Accept-encoding: gzip' header, + # but we don't. + result['bozo'] = 1 + result['bozo_exception'] = e + data = '' + elif zlib and f.headers.get('content-encoding', '') == 'deflate': + try: + data = zlib.decompress(data, -zlib.MAX_WBITS) + except Exception, e: + result['bozo'] = 1 + result['bozo_exception'] = e + data = '' + + # save HTTP headers + if hasattr(f, 'info'): + info = f.info() + result['etag'] = info.getheader('ETag') + last_modified = info.getheader('Last-Modified') + if last_modified: + result['modified'] = _parse_date(last_modified) + if hasattr(f, 'url'): + result['href'] = f.url + result['status'] = 200 + if hasattr(f, 'status'): + result['status'] = f.status + if hasattr(f, 'headers'): + result['headers'] = f.headers.dict + if hasattr(f, 'close'): + f.close() + + # there are four encodings to keep track of: + # - http_encoding is the encoding declared in the Content-Type HTTP header + # - xml_encoding is the encoding declared in the ; changed +# project name +#2.5 - 7/25/2003 - MAP - changed to Python license (all contributors agree); +# removed unnecessary urllib code -- urllib2 should always be available anyway; +# return actual url, status, and full HTTP headers (as result['url'], +# result['status'], and result['headers']) if parsing a remote feed over HTTP -- +# this should pass all the HTTP tests at ; +# added the latest namespace-of-the-week for RSS 2.0 +#2.5.1 - 7/26/2003 - RMK - clear opener.addheaders so we only send our custom +# User-Agent (otherwise urllib2 sends two, which confuses some servers) +#2.5.2 - 7/28/2003 - MAP - entity-decode inline xml properly; added support for +# inline and as used in some RSS 2.0 feeds +#2.5.3 - 8/6/2003 - TvdV - patch to track whether we're inside an image or +# textInput, and also to return the character encoding (if specified) +#2.6 - 1/1/2004 - MAP - dc:author support (MarekK); fixed bug tracking +# nested divs within content (JohnD); fixed missing sys import (JohanS); +# fixed regular expression to capture XML character encoding (Andrei); +# added support for Atom 0.3-style links; fixed bug with textInput tracking; +# added support for cloud (MartijnP); added support for multiple +# category/dc:subject (MartijnP); normalize content model: 'description' gets +# description (which can come from description, summary, or full content if no +# description), 'content' gets dict of base/language/type/value (which can come +# from content:encoded, xhtml:body, content, or fullitem); +# fixed bug matching arbitrary Userland namespaces; added xml:base and xml:lang +# tracking; fixed bug tracking unknown tags; fixed bug tracking content when +# element is not in default namespace (like Pocketsoap feed); +# resolve relative URLs in link, guid, docs, url, comments, wfw:comment, +# wfw:commentRSS; resolve relative URLs within embedded HTML markup in +# description, xhtml:body, content, content:encoded, title, subtitle, +# summary, info, tagline, and copyright; added support for pingback and +# trackback namespaces +#2.7 - 1/5/2004 - MAP - really added support for trackback and pingback +# namespaces, as opposed to 2.6 when I said I did but didn't really; +# sanitize HTML markup within some elements; added mxTidy support (if +# installed) to tidy HTML markup within some elements; fixed indentation +# bug in _parse_date (FazalM); use socket.setdefaulttimeout if available +# (FazalM); universal date parsing and normalization (FazalM): 'created', modified', +# 'issued' are parsed into 9-tuple date format and stored in 'created_parsed', +# 'modified_parsed', and 'issued_parsed'; 'date' is duplicated in 'modified' +# and vice-versa; 'date_parsed' is duplicated in 'modified_parsed' and vice-versa +#2.7.1 - 1/9/2004 - MAP - fixed bug handling " and '. fixed memory +# leak not closing url opener (JohnD); added dc:publisher support (MarekK); +# added admin:errorReportsTo support (MarekK); Python 2.1 dict support (MarekK) +#2.7.4 - 1/14/2004 - MAP - added workaround for improperly formed
          tags in +# encoded HTML (skadz); fixed unicode handling in normalize_attrs (ChrisL); +# fixed relative URI processing for guid (skadz); added ICBM support; added +# base64 support +#2.7.5 - 1/15/2004 - MAP - added workaround for malformed DOCTYPE (seen on many +# blogspot.com sites); added _debug variable +#2.7.6 - 1/16/2004 - MAP - fixed bug with StringIO importing +#3.0b3 - 1/23/2004 - MAP - parse entire feed with real XML parser (if available); +# added several new supported namespaces; fixed bug tracking naked markup in +# description; added support for enclosure; added support for source; re-added +# support for cloud which got dropped somehow; added support for expirationDate +#3.0b4 - 1/26/2004 - MAP - fixed xml:lang inheritance; fixed multiple bugs tracking +# xml:base URI, one for documents that don't define one explicitly and one for +# documents that define an outer and an inner xml:base that goes out of scope +# before the end of the document +#3.0b5 - 1/26/2004 - MAP - fixed bug parsing multiple links at feed level +#3.0b6 - 1/27/2004 - MAP - added feed type and version detection, result['version'] +# will be one of SUPPORTED_VERSIONS.keys() or empty string if unrecognized; +# added support for creativeCommons:license and cc:license; added support for +# full Atom content model in title, tagline, info, copyright, summary; fixed bug +# with gzip encoding (not always telling server we support it when we do) +#3.0b7 - 1/28/2004 - MAP - support Atom-style author element in author_detail +# (dictionary of 'name', 'url', 'email'); map author to author_detail if author +# contains name + email address +#3.0b8 - 1/28/2004 - MAP - added support for contributor +#3.0b9 - 1/29/2004 - MAP - fixed check for presence of dict function; added +# support for summary +#3.0b10 - 1/31/2004 - MAP - incorporated ISO-8601 date parsing routines from +# xml.util.iso8601 +#3.0b11 - 2/2/2004 - MAP - added 'rights' to list of elements that can contain +# dangerous markup; fiddled with decodeEntities (not right); liberalized +# date parsing even further +#3.0b12 - 2/6/2004 - MAP - fiddled with decodeEntities (still not right); +# added support to Atom 0.2 subtitle; added support for Atom content model +# in copyright; better sanitizing of dangerous HTML elements with end tags +# (script, frameset) +#3.0b13 - 2/8/2004 - MAP - better handling of empty HTML tags (br, hr, img, +# etc.) in embedded markup, in either HTML or XHTML form (
          ,
          ,
          ) +#3.0b14 - 2/8/2004 - MAP - fixed CDATA handling in non-wellformed feeds under +# Python 2.1 +#3.0b15 - 2/11/2004 - MAP - fixed bug resolving relative links in wfw:commentRSS; +# fixed bug capturing author and contributor URL; fixed bug resolving relative +# links in author and contributor URL; fixed bug resolvin relative links in +# generator URL; added support for recognizing RSS 1.0; passed Simon Fell's +# namespace tests, and included them permanently in the test suite with his +# permission; fixed namespace handling under Python 2.1 +#3.0b16 - 2/12/2004 - MAP - fixed support for RSS 0.90 (broken in b15) +#3.0b17 - 2/13/2004 - MAP - determine character encoding as per RFC 3023 +#3.0b18 - 2/17/2004 - MAP - always map description to summary_detail (Andrei); +# use libxml2 (if available) +#3.0b19 - 3/15/2004 - MAP - fixed bug exploding author information when author +# name was in parentheses; removed ultra-problematic mxTidy support; patch to +# workaround crash in PyXML/expat when encountering invalid entities +# (MarkMoraes); support for textinput/textInput +#3.0b20 - 4/7/2004 - MAP - added CDF support +#3.0b21 - 4/14/2004 - MAP - added Hot RSS support +#3.0b22 - 4/19/2004 - MAP - changed 'channel' to 'feed', 'item' to 'entries' in +# results dict; changed results dict to allow getting values with results.key +# as well as results[key]; work around embedded illformed HTML with half +# a DOCTYPE; work around malformed Content-Type header; if character encoding +# is wrong, try several common ones before falling back to regexes (if this +# works, bozo_exception is set to CharacterEncodingOverride); fixed character +# encoding issues in BaseHTMLProcessor by tracking encoding and converting +# from Unicode to raw strings before feeding data to sgmllib.SGMLParser; +# convert each value in results to Unicode (if possible), even if using +# regex-based parsing +#3.0b23 - 4/21/2004 - MAP - fixed UnicodeDecodeError for feeds that contain +# high-bit characters in attributes in embedded HTML in description (thanks +# Thijs van de Vossen); moved guid, date, and date_parsed to mapped keys in +# FeedParserDict; tweaked FeedParserDict.has_key to return True if asking +# about a mapped key +#3.0fc1 - 4/23/2004 - MAP - made results.entries[0].links[0] and +# results.entries[0].enclosures[0] into FeedParserDict; fixed typo that could +# cause the same encoding to be tried twice (even if it failed the first time); +# fixed DOCTYPE stripping when DOCTYPE contained entity declarations; +# better textinput and image tracking in illformed RSS 1.0 feeds +#3.0fc2 - 5/10/2004 - MAP - added and passed Sam's amp tests; added and passed +# my blink tag tests +#3.0fc3 - 6/18/2004 - MAP - fixed bug in _changeEncodingDeclaration that +# failed to parse utf-16 encoded feeds; made source into a FeedParserDict; +# duplicate admin:generatorAgent/@rdf:resource in generator_detail.url; +# added support for image; refactored parse() fallback logic to try other +# encodings if SAX parsing fails (previously it would only try other encodings +# if re-encoding failed); remove unichr madness in normalize_attrs now that +# we're properly tracking encoding in and out of BaseHTMLProcessor; set +# feed.language from root-level xml:lang; set entry.id from rdf:about; +# send Accept header +#3.0 - 6/21/2004 - MAP - don't try iso-8859-1 (can't distinguish between +# iso-8859-1 and windows-1252 anyway, and most incorrectly marked feeds are +# windows-1252); fixed regression that could cause the same encoding to be +# tried twice (even if it failed the first time) +#3.0.1 - 6/22/2004 - MAP - default to us-ascii for all text/* content types; +# recover from malformed content-type header parameter with no equals sign +# ('text/xml; charset:iso-8859-1') +#3.1 - 6/28/2004 - MAP - added and passed tests for converting HTML entities +# to Unicode equivalents in illformed feeds (aaronsw); added and +# passed tests for converting character entities to Unicode equivalents +# in illformed feeds (aaronsw); test for valid parsers when setting +# XML_AVAILABLE; make version and encoding available when server returns +# a 304; add handlers parameter to pass arbitrary urllib2 handlers (like +# digest auth or proxy support); add code to parse username/password +# out of url and send as basic authentication; expose downloading-related +# exceptions in bozo_exception (aaronsw); added __contains__ method to +# FeedParserDict (aaronsw); added publisher_detail (aaronsw) +#3.2 - 7/3/2004 - MAP - use cjkcodecs and iconv_codec if available; always +# convert feed to UTF-8 before passing to XML parser; completely revamped +# logic for determining character encoding and attempting XML parsing +# (much faster); increased default timeout to 20 seconds; test for presence +# of Location header on redirects; added tests for many alternate character +# encodings; support various EBCDIC encodings; support UTF-16BE and +# UTF16-LE with or without a BOM; support UTF-8 with a BOM; support +# UTF-32BE and UTF-32LE with or without a BOM; fixed crashing bug if no +# XML parsers are available; added support for 'Content-encoding: deflate'; +# send blank 'Accept-encoding: ' header if neither gzip nor zlib modules +# are available +#3.3 - 7/15/2004 - MAP - optimize EBCDIC to ASCII conversion; fix obscure +# problem tracking xml:base and xml:lang if element declares it, child +# doesn't, first grandchild redeclares it, and second grandchild doesn't; +# refactored date parsing; defined public registerDateHandler so callers +# can add support for additional date formats at runtime; added support +# for OnBlog, Nate, MSSQL, Greek, and Hungarian dates (ytrewq1); added +# zopeCompatibilityHack() which turns FeedParserDict into a regular +# dictionary, required for Zope compatibility, and also makes command- +# line debugging easier because pprint module formats real dictionaries +# better than dictionary-like objects; added NonXMLContentType exception, +# which is stored in bozo_exception when a feed is served with a non-XML +# media type such as 'text/plain'; respect Content-Language as default +# language if not xml:lang is present; cloud dict is now FeedParserDict; +# generator dict is now FeedParserDict; better tracking of xml:lang, +# including support for xml:lang='' to unset the current language; +# recognize RSS 1.0 feeds even when RSS 1.0 namespace is not the default +# namespace; don't overwrite final status on redirects (scenarios: +# redirecting to a URL that returns 304, redirecting to a URL that +# redirects to another URL with a different type of redirect); add +# support for HTTP 303 redirects +#4.0 - MAP - support for relative URIs in xml:base attribute; fixed +# encoding issue with mxTidy (phopkins); preliminary support for RFC 3229; +# support for Atom 1.0; support for iTunes extensions; new 'tags' for +# categories/keywords/etc. as array of dict +# {'term': term, 'scheme': scheme, 'label': label} to match Atom 1.0 +# terminology; parse RFC 822-style dates with no time; lots of other +# bug fixes +#4.1 - MAP - removed socket timeout; added support for chardet library diff --git a/app/index.jinja b/app/index.jinja new file mode 100644 index 0000000..30cf9dd --- /dev/null +++ b/app/index.jinja @@ -0,0 +1,257 @@ + + + + + + + oohEmbed.com - your one-stop oEmbed provider + + + + + + + + + + + +{% filter markdown %} +# Welcome to oohEmbed! +## oohEmbed What? + +In a nutshell: __oohEmbed is an oEmbed compatible provider of HTML embed codes for various web sites.__ + +What is this oEmbed? From [oembed.com](http://www.oembed.com/): + +> oEmbed is a format for allowing an embedded representation of a URL on third party sites. +> The simple API allows a website to display embedded content (such as photos or videos) when +> a user posts a link to that resource, without having to parse the resource directly. + +Don't get it? Perhaps an example will make things clear. If you make a URL request like this: +{% endfilter %} + +
          http://oohembed.com/oohembed/?url=http%3A//www.amazon.com/Myths-Innovation-Scott-Berkun/dp/0596527055/
          + +{% filter markdown %} +You will get this as the response: +{% endfilter %} + +
          {
+     "type": "photo", 
+     "version": "1.0", 
+     "provider_name": "Amazon Product Image", 
+     "url": "http://ecx.images-amazon.com/images/I/31%2BfVjL2nqL.jpg", 
+     "height": "500", 
+     "width": "317", 
+     "thumbnail_url": "http://ecx.images-amazon.com/images/I/31%2BfVjL2nqL._SL75_.jpg", 
+     "thumbnail_height": "75", 
+     "thumbnail_width": "48", 
+     "asin": "0596527055", 
+     "title": "The Myths of Innovation", 
+     "author_name": "Scott Berkun", 
+     "author_url": "http://www.amazon.com/gp/redirect.html%3FASIN=0596527055%26location=/Myths-Innovation-Scott-Berkun/dp/0596527055"
+}
          + +{% filter markdown %} +There, that should make everything clear. Still no? Then perhaps you should go and read this excellent article introducing [oEmbed & oohEmbed](http://ciaranmcnulty.com/blog/2009/01/embedding-third-party-content-in-your-site-using-oembed) by Ciaran McNulty. Once you are done reading that, come back here for our Q&A! + +## Q & A + +1. How is this useful? + I built oohEmbed to further development of [Dumble](/dumble/?u=antrix&t=dumbletest), my del.icio.us tumblelog. + But I imagine oohEmbed could be useful in several other places. For example, someone could build a + wordpress plugin backed by the oohEmbed web service that makes embedding content into blog posts + easier. (If you do build such a plugin, let me know!) + + _New:_ Scroll down to see a list of [consumer sites and apps](#consumers) that use oohEmbed. + +2. So what sites can oohEmbed provide information for? + Quite a few to start with - scroll down to see [the list](#configuration). I'll be adding more sites as I find time! + +3. Is oohEmbed really oembed.com spec compliant? + Almost. oohEmbed differs in some ways from the spec. + * It usually ignores the maxwidth/maxheight request parameters. + * Only JSON responses are provided. I don't have any plans of supporting XML. + * You can specify a JSON callback. + +4. JSON Callback? + To overcome the [same origin policy](http://en.wikipedia.org/wiki/Same_origin_policy) of web browsers, you can specify a 'callback' parameter as part of your request. + e.g., if you specify 'callback=myCallback' as part of the request, then the JSON response object will be wrapped as a parameter to the myCallback function. + +5. Will this callback expose my site to cross-site attacks? + Possibly. Although you have to trust that I won't be doing anything funny or malicious intentionally, oohEmbed itself is just a proxy to the target website. + So if the target website sends back something bad, all bets are off! + +6. So I can't sue you if anything goes wrong? + No! By using this service, you agree that I will not be held liable for anything that goes wrong. I do appreciate bug reports though. + +7. Is it free to use? + Yes. At the moment, all this service costs me is $10/year for the domain name and several hours of my free time. + Google App Engine kindly hosts the app for free (thanks!) so that's taken care of. + +8. But is it _really free_? + Oh yes, it is _really free_ too! You can get access to the code that powers oohEmbed.com from the [oohEmbed project on Google Code](http://oohembed.googlecode.com). + The code is available under a liberal BSD license so feel free to use it as you wish. Of course, I'd appreciate any code contributions too! + +8. So I can use the site as much as I want? + Yes, but to a limit. If I detect abuse, I reserve the right to throttle your use or worse, ban you from accessing the service. + +9. Great, how do I sign-up?! + Umm.. there is no sign-up. Here's your service, run with it: http://oohembed.com/oohembed/?url= + +10. I'm all fired up. Where's the oEmbed provider configuration for the oohEmbed service to get me going? + Just scroll down this page for the [API endpoint and supported URL schemes](#configuration). + +11. I notice that you also support websites which already have an oEmbed compliant provider. Why? + Two reasons: + * They don't support the 'callback' parameter for JSON requests + * As a developer, wouldn't it be nice if there was just one API endpoint for all sites + instead of having to support one endpoint for each? + +12. I represent one of the sites for which oohEmbed provides responses and I don't like it. + Fair enough. Drop me an email and I'll stop supporting your site. + +13. How do I report bugs? Do you take feature requests? + Sure. For now, just use the [comment form](#comments) at the end of this page. If this thing gets popular, I'll switch to more traditional + channels of bug tracking. + _Update:_ Now that the oohEmbed open source project is live, you can use the [issue tracker](http://code.google.com/p/oohembed/issues/) on Google Code! + +14. Anything else? + Nothing, except that oohEmbed wouldn't be possible without the folks behind the oEmbed.com spec so kudos to them. + And thanks again to Google for creating App Engine! + +15. Who are you? + I am Deepak Sarda and you can find me here. + +## Configuration + +There's just one API endpoint: http://oohembed.com/oohembed/. The supported URL schemes are listed below. If you call the API endpoint with a URL that doesn't confirm to one of the schemes listed below, you will get a 404 response. As mentioned earlier, you can use a `callback` parameter when calling the endpoint. + +Click on any item in the list below to see configuration details for that item. +{% endfilter %} + +
            + {% for p in providers %} +
          1. {{ p.title }} + +
            2. + {% endfor %} +
          + +{% filter markdown %} +## Consumers + +These consumers currently use the oohEmbed service. + +* [Dumble](/dumble/?u=antrix&t=linker) of course, being the first consumer and my reason for creating oohEmbed. +* [280 Slides](http://blog.280north.com/2008/06/25/280-slides-gets-smarter-about-links/), a snazzy web-app to create presentations. +* [Buckybase](http://buckybase.blogspot.com/2008/06/oembed-support.html), a social web database with bidirectional hyperlinks. +* [Indy.com](http://www.indy.com/), a local entertainment site by The Indianapolis Star. +* [Ars Technica](http://arstechnica.com/), a technology news and analysis site. +* [DropAVideo](http://dropavideo.com/), a Brazilian video micro-blogging service. + +Moreover, oohEmbed support is already baked into these libraries: + +* [django-oembed](http://code.google.com/p/django-oembed/) +* [ruby-oembed](http://github.com/judofyr/ruby-oembed/tree/master) +* [oembed_links gem](http://github.com/netshade/oembed_links/tree/master) +* [jquery-oembed](http://code.google.com/p/jquery-oembed/) +* [drupal-oembed](http://github.com/voxpelli/drupal-oembed/tree/master) + +If you are using oohEmbed, please leave a comment below and I'll add you to this list! + +## Changes + +* _6th June, 2009:_ Added missing width, height elements to Youtube response. Added title, author and thumbnail elements to Metacafe response. +* _10th May, 2009:_ New LiveJournal UserPic provider. More data returned in Youtube provider. Removed Pownce provider. Updated Consumers section. +* _31st March, 2009:_ Updates to Consumers section. +* _2nd March, 2009:_ New twitpic.com support. +* _3rd February, 2009:_ New 5min.com, nfb.ca & thedailyshow.com support. +* _14th November, 2008:_ Released oohEmbed source. Fixed mime-type of responses. Moved Dumble to new delicious.com API. +* _23rd July, 2008:_ New Vimeo support. Some text encoding related fixes. +* _30th June, 2008:_ New SlideShare provider. Cleaned up homepage (added consumers). Several minor fixes. +* _12th June, 2008:_ New IMDb.com movie photo/link provider. +* _11th June, 2008:_ New Wordpress.com blogs support. Several fixes in [Dumble](/dumble/). +* _3rd June, 2008:_ New hulu.com & revision3.com support +* _1st June, 2008:_ Launch +{% endfilter %} + +
          + + + +{% if production %} + +
          comments powered by Disqus + + + + +{% endif %} + + + + diff --git a/app/jinja/__init__.py b/app/jinja/__init__.py new file mode 100644 index 0000000..5393c75 --- /dev/null +++ b/app/jinja/__init__.py @@ -0,0 +1,72 @@ +# -*- coding: utf-8 -*- +""" + jinja + ~~~~~ + + Jinja is a `sandboxed`_ template engine written in pure Python. It + provides a `Django`_ like non-XML syntax and compiles templates into + executable python code. It's basically a combination of Django templates + and python code. + + Nutshell + -------- + + Here a small example of a Jinja template:: + + {% extends 'base.html' %} + {% block title %}Memberlist{% endblock %} + {% block content %} + + {% endblock %} + + Philosophy + ---------- + + Application logic is for the controller but don't try to make the life + for the template designer too hard by giving him too few functionality. + + For more informations visit the new `jinja webpage`_ and `documentation`_. + + Note + ---- + + This is the Jinja 1.0 release which is completely incompatible with the + old "pre 1.0" branch. The old branch will still receive security updates + and bugfixes but the 1.0 branch will be the only version that receives + support. + + If you have an application that uses Jinja 0.9 and won't be updated in + the near future the best idea is to ship a Jinja 0.9 checkout together + with the application. + + The `Jinja tip`_ is installable via `easy_install` with ``easy_install + Jinja==dev``. + + .. _sandboxed: http://en.wikipedia.org/wiki/Sandbox_(computer_security) + .. _Django: http://www.djangoproject.com/ + .. _jinja webpage: http://jinja.pocoo.org/ + .. _documentation: http://jinja.pocoo.org/documentation/index.html + .. _Jinja tip: http://dev.pocoo.org/hg/jinja-main/archive/tip.tar.gz#egg=Jinja-dev + + + :copyright: 2007 by Armin Ronacher. + :license: BSD, see LICENSE for more details. +""" + +from jinja.environment import Environment +from jinja.datastructure import Markup +from jinja.plugin import jinja_plugin_factory as template_plugin_factory +from jinja.loaders import FileSystemLoader, PackageLoader, DictLoader, \ + ChoiceLoader, FunctionLoader, MemcachedFileSystemLoader +from jinja.utils import from_string + + +__all__ = ['Environment', 'Markup', 'FileSystemLoader', 'PackageLoader', + 'DictLoader', 'ChoiceLoader', 'FunctionLoader', + 'MemcachedFileSystemLoader', 'from_string'] + +__version__ = '1.2' diff --git a/app/jinja/_native.py b/app/jinja/_native.py new file mode 100644 index 0000000..1d9747a --- /dev/null +++ b/app/jinja/_native.py @@ -0,0 +1,85 @@ +# -*- coding: utf-8 -*- +""" + jinja._native + ~~~~~~~~~~~~~ + + This module implements the native base classes in case of not + having a jinja with the _speedups module compiled. + + Note that if you change semantics here you have to edit the + _speedups.c file to in order to support those changes for jinja + setups with enabled speedup module. + + :copyright: 2007 by Armin Ronacher. + :license: BSD, see LICENSE for more details. +""" +from jinja.datastructure import Deferred +from jinja.utils import deque + + +class BaseContext(object): + + def __init__(self, undefined_singleton, globals, initial): + self._undefined_singleton = undefined_singleton + self.current = current = {} + self._stack = deque([current, initial, globals]) + self.globals = globals + self.initial = initial + + self._push = self._stack.appendleft + self._pop = self._stack.popleft + + def stack(self): + return list(self._stack)[::-1] + stack = property(stack) + + def pop(self): + """Pop the last layer from the stack and return it.""" + rv = self._pop() + self.current = self._stack[0] + return rv + + def push(self, data=None): + """ + Push one layer to the stack and return it. Layer must be + a dict or omitted. + """ + data = data or {} + self._push(data) + self.current = self._stack[0] + return data + + def __getitem__(self, name): + """ + Resolve one item. Restrict the access to internal variables + such as ``'::cycle1'``. Resolve deferreds. + """ + if not name.startswith('::'): + for d in self._stack: + if name in d: + rv = d[name] + if rv.__class__ is Deferred: + rv = rv(self, name) + # never touch the globals! + if d is self.globals: + self.initial[name] = rv + else: + d[name] = rv + return rv + return self._undefined_singleton + + def __setitem__(self, name, value): + """Set a variable in the outermost layer.""" + self.current[name] = value + + def __delitem__(self, name): + """Delete a variable in the outermost layer.""" + if name in self.current: + del self.current[name] + + def __contains__(self, name): + """ Check if the context contains a given variable.""" + for layer in self._stack: + if name in layer: + return True + return False diff --git a/app/jinja/constants.py b/app/jinja/constants.py new file mode 100644 index 0000000..a0b4a63 --- /dev/null +++ b/app/jinja/constants.py @@ -0,0 +1,32 @@ +# -*- coding: utf-8 -*- +""" + jinja.constants + ~~~~~~~~~~~~~~~ + + Various constants. + + :copyright: 2007 by Armin Ronacher. + :license: BSD, see LICENSE for more details. +""" + + +#: list of lorem ipsum words used by the lipsum() helper function +LOREM_IPSUM_WORDS = u'''\ +a ac accumsan ad adipiscing aenean aliquam aliquet amet ante aptent arcu at +auctor augue bibendum blandit class commodo condimentum congue consectetuer +consequat conubia convallis cras cubilia cum curabitur curae cursus dapibus +diam dictum dictumst dignissim dis dolor donec dui duis egestas eget eleifend +elementum elit enim erat eros est et etiam eu euismod facilisi facilisis fames +faucibus felis fermentum feugiat fringilla fusce gravida habitant habitasse hac +hendrerit hymenaeos iaculis id imperdiet in inceptos integer interdum ipsum +justo lacinia lacus laoreet lectus leo libero ligula litora lobortis lorem +luctus maecenas magna magnis malesuada massa mattis mauris metus mi molestie +mollis montes morbi mus nam nascetur natoque nec neque netus nibh nisi nisl non +nonummy nostra nulla nullam nunc odio orci ornare parturient pede pellentesque +penatibus per pharetra phasellus placerat platea porta porttitor posuere +potenti praesent pretium primis proin pulvinar purus quam quis quisque rhoncus +ridiculus risus rutrum sagittis sapien scelerisque sed sem semper senectus sit +sociis sociosqu sodales sollicitudin suscipit suspendisse taciti tellus tempor +tempus tincidunt torquent tortor tristique turpis ullamcorper ultrices +ultricies urna ut varius vehicula vel velit venenatis vestibulum vitae vivamus +viverra volutpat vulputate''' diff --git a/app/jinja/datastructure.py b/app/jinja/datastructure.py new file mode 100644 index 0000000..eb213d4 --- /dev/null +++ b/app/jinja/datastructure.py @@ -0,0 +1,708 @@ +# -*- coding: utf-8 -*- +""" + jinja.datastructure + ~~~~~~~~~~~~~~~~~~~ + + Module that helds several data types used in the template engine. + + :copyright: 2007 by Armin Ronacher. + :license: BSD, see LICENSE for more details. +""" +from jinja.exceptions import TemplateSyntaxError, TemplateRuntimeError + +_missing = object() + + +def contextcallable(f): + """ + Mark a function context callable. + """ + f.jinja_context_callable = True + return f + + +def unsafe(f): + """ + Mark function as unsafe. + """ + f.jinja_unsafe_call = True + return f + + +def make_undefined(implementation): + """ + Creates an undefined singleton based on a given implementation. + It performs some tests that make sure the undefined type implements + everything it should. + """ + self = object.__new__(implementation) + self.__reduce__() + return self + + +class AbstractUndefinedType(object): + """ + Base class for any undefined type. + """ + __slots__ = () + + def __init__(self): + raise TypeError('cannot create %r instances' % + self.__class__.__name__) + + def __setattr__(self, name, value): + raise AttributeError('%r object has no attribute %r' % ( + self.__class__.__name__, + name + )) + + def __eq__(self, other): + return self is other + + def __ne__(self, other): + return self is not other + + def __copy__(self): + return self + __deepcopy__ = __copy__ + + def __repr__(self): + return 'Undefined' + + def __reduce__(self): + raise TypeError('undefined objects have to provide a __reduce__') + + +class SilentUndefinedType(AbstractUndefinedType): + """ + An object that does not exist. + """ + __slots__ = () + + def __add__(self, other): + """Any operator returns the operand.""" + return other + __sub__ = __mul__ = __div__ = __rsub__ = __rmul__ = __div__ = __mod__ =\ + __radd__ = __rmod__ = __add__ + + def __getitem__(self, arg): + """Getting any item returns `Undefined`""" + return self + + def __iter__(self): + """Iterating over `Undefined` returns an empty iterator.""" + if False: + yield None + + def __getattr__(self, arg): + """Getting any attribute returns `Undefined`""" + return self + + def __nonzero__(self): + """`Undefined` is considered boolean `False`""" + return False + + def __len__(self): + """`Undefined` is an empty sequence""" + return 0 + + def __str__(self): + """The string representation is empty.""" + return '' + + def __unicode__(self): + """The unicode representation is empty.""" + return u'' + + def __int__(self): + """Converting `Undefined` to an integer ends up in ``0``""" + return 0 + + def __float__(self): + """Converting `Undefined` to an float ends up in ``0.0``""" + return 0.0 + + def __call__(self, *args, **kwargs): + """Calling `Undefined` returns `Undefined`""" + return self + + def __reduce__(self): + """Helper for pickle.""" + return 'SilentUndefined' + + +class ComplainingUndefinedType(AbstractUndefinedType): + """ + An object that does not exist. + """ + __slots__ = () + + def __len__(self): + """Getting the length raises error.""" + raise TemplateRuntimeError('Operated on undefined object') + + def __iter__(self): + """Iterating over `Undefined` raises an error.""" + raise TemplateRuntimeError('Iterated over undefined object') + + def __nonzero__(self): + """`Undefined` is considered boolean `False`""" + return False + + def __str__(self): + """The string representation raises an error.""" + raise TemplateRuntimeError('Undefined object rendered') + + def __unicode__(self): + """The unicode representation raises an error.""" + self.__str__() + + def __call__(self, *args, **kwargs): + """Calling `Undefined` returns `Undefined`""" + raise TemplateRuntimeError('Undefined object called') + + def __reduce__(self): + """Helper for pickle.""" + return 'ComplainingUndefined' + + +#: the singleton instances for the undefined objects +SilentUndefined = make_undefined(SilentUndefinedType) +ComplainingUndefined = make_undefined(ComplainingUndefinedType) + +#: jinja 1.0 compatibility +Undefined = SilentUndefined +UndefinedType = SilentUndefinedType + + +class FakeTranslator(object): + """ + Default null translator. + """ + + def gettext(self, s): + """ + Translate a singular string. + """ + return s + + def ngettext(self, s, p, n): + """ + Translate a plural string. + """ + if n == 1: + return s + return p + + +class Deferred(object): + """ + Object marking an deferred value. Deferred objects are + objects that are called first access in the context. + """ + + def __init__(self, factory): + self.factory = factory + + def __call__(self, context, name): + return self.factory(context.environment, context, name) + + +class Markup(unicode): + """ + Compatibility for Pylons and probably some other frameworks. + + It's only used in Jinja environments with `auto_escape` set + to true. + """ + + def __html__(self): + return unicode(self) + + +class TemplateData(Markup): + """ + Subclass of unicode to mark objects that are coming from the + template. The autoescape filter can use that. + """ + + +# import these here because those modules import Deferred and Undefined +# from this module. +try: + # try to use the c implementation of the base context if available + from jinja._speedups import BaseContext +except ImportError: + # if there is no c implementation we go with a native python one + from jinja._native import BaseContext + + +class Context(BaseContext): + """ + Dict like object containing the variables for the template. + """ + + def __init__(self, *args, **kwargs): + environment = args[0] + if not kwargs and len(args) == 2 and isinstance(args[1], dict): + base = args[1] + else: + base = dict(*args[1:], **kwargs) + super(Context, self).__init__(environment.undefined_singleton, + environment.globals, base) + self._translate_func = None + self.cache = {} + self.environment = environment + + def to_dict(self): + """ + Convert the context into a dict. This skips the globals. + """ + result = {} + for layer in self.stack[1:]: + for key, value in layer.iteritems(): + if key.startswith('::'): + continue + result[key] = value + return result + + def set_nonlocal(self, name, value): + """ + Set a value in an outer scope. + """ + for layer in self.stack[:0:-1]: + if name in layer: + layer[name] = value + return + self.initial[name] = value + + def translate_func(self): + """ + The translation function for this context. It takes + 4 parameters. The singular string, the optional plural one, + The name of the variable in the replacements dict and the + replacements dict. This is only used by the i18n system + internally the simplified version (just one argument) is + available in the template for the user too. + """ + if self._translate_func is not None: + return self._translate_func + translator = self.environment.get_translator(self) + gettext = translator.gettext + ngettext = translator.ngettext + def translate(s, p=None, n=None, r=None): + if p is None: + s = gettext(s) + else: + s = ngettext(s, p, r[n]) + # apply replacement substitution only if replacements + # are given. This is the case for {% trans %}...{% endtrans %} + # but for the "_()" syntax and a trans tag without a body. + if r is not None: + return s % r + return s + translate.__doc__ = Context.translate_func.__doc__ + self._translate_func = translate + return translate + translate_func = property(translate_func, doc=translate_func.__doc__) + + def __repr__(self): + """ + String representation of the context. + """ + return 'Context(%r)' % self.to_dict() + + def __pretty__(self, p, cycle): + if cycle: + return p.text('Context({...})') + p.begin_group(9, 'Context({') + for idx, (key, value) in enumerate(self.to_dict().iteritems()): + if idx: + p.text(',') + p.breakable() + p.pretty(key) + p.text(': ') + p.pretty(value) + p.end_group(9, '})') + + +class LoopContext(object): + """ + Simple class that provides special loop variables. + Used by `Environment.iterate`. + """ + + jinja_allowed_attributes = ['index', 'index0', 'length', 'parent', + 'even', 'odd', 'revindex0', 'revindex', + 'first', 'last'] + + def __init__(self, seq, parent, loop_function): + self.loop_function = loop_function + self.parent = parent + self._stack = [] + if loop_function is None: + self.push(seq) + + def push(self, seq): + """ + Push a sequence to the loop stack. This is used by the + recursive for loop. + """ + # iteration over None is catched, but we don't catch iteration + # over undefined because that behavior is handled in the + # undefined singleton + if seq is None: + seq = () + length = 0 + else: + try: + length = len(seq) + except (AttributeError, TypeError): + seq = list(seq) + length = len(seq) + self._stack.append({ + 'index': -1, + 'seq': seq, + 'length': length + }) + return self + + def pop(self): + """Remove the last layer from the loop stack.""" + return self._stack.pop() + + iterated = property(lambda s: s._stack[-1]['index'] > -1) + index0 = property(lambda s: s._stack[-1]['index']) + index = property(lambda s: s._stack[-1]['index'] + 1) + revindex0 = property(lambda s: s._stack[-1]['length'] - + s._stack[-1]['index'] - 1) + revindex = property(lambda s: s._stack[-1]['length'] - + s._stack[-1]['index']) + length = property(lambda s: s._stack[-1]['length']) + even = property(lambda s: s._stack[-1]['index'] % 2 == 1) + odd = property(lambda s: s._stack[-1]['index'] % 2 == 0) + first = property(lambda s: s._stack[-1]['index'] == 0) + last = property(lambda s: s._stack[-1]['index'] == + s._stack[-1]['length'] - 1) + + def __iter__(self): + s = self._stack[-1] + for idx, item in enumerate(s['seq']): + s['index'] = idx + yield item + + def __len__(self): + return self._stack[-1]['length'] + + def __call__(self, seq): + if self.loop_function is not None: + return self.loop_function(seq) + raise TemplateRuntimeError('In order to make loops callable you have ' + 'to define them with the "recursive" ' + 'modifier.') + + def __repr__(self): + if self._stack: + return '' % ( + self.index, + self.length, + self.loop_function is not None and ' recursive' or '' + ) + return '' + + +class CycleContext(object): + """ + Helper class used for cycling. + """ + + def __init__(self, seq=None): + self.pos = -1 + # bind the correct helper function based on the constructor signature + if seq is not None: + self.seq = seq + self.length = len(seq) + self.cycle = self.cycle_static + else: + self.cycle = self.cycle_dynamic + + def cycle_static(self): + """Helper function for static cycling.""" + self.pos = (self.pos + 1) % self.length + return self.seq[self.pos] + + def cycle_dynamic(self, seq): + """Helper function for dynamic cycling.""" + self.pos = pos = (self.pos + 1) % len(seq) + return seq[pos] + + +class SuperBlock(object): + """ + Helper class for ``{{ super() }}``. + """ + jinja_allowed_attributes = ['name'] + + def __init__(self, name, blocks, level, context): + self.name = name + self.context = context + if name in blocks: + self.stack = blocks[name] + self.level = level + else: + self.stack = None + + def __call__(self, offset=1): + if self.stack is not None: + level = self.level + (offset - 1) + if level < len(self.stack): + return self.stack[level](self.context) + raise TemplateRuntimeError('no super block for %r' % self.name) + + def __repr__(self): + return '' % self.name + + +class StateTest(object): + """ + Wrapper class for basic lambdas in order to simplify + debugging in the parser. It also provides static helper + functions that replace some lambda expressions + """ + + def __init__(self, func, msg): + self.func = func + self.msg = msg + + def __call__(self, token): + return self.func(token) + + def expect_token(*types, **kw): + """Scans until one of the given tokens is found.""" + msg = kw.pop('msg', None) + if kw: + raise TypeError('unexpected keyword argument %r' % iter(kw).next()) + if len(types) == 1: + if msg is None: + msg = "expected '%s'" % types[0] + return StateTest(lambda t: t.type == types[0], msg) + if msg is None: + msg = 'expected one of %s' % ', '.join(["'%s'" % type + for type in types]) + return StateTest(lambda t: t.type in types, msg) + expect_token = staticmethod(expect_token) + + +class Token(object): + """ + Token class. + """ + __slots__ = ('lineno', 'type', 'value') + + def __init__(self, lineno, type, value): + self.lineno = lineno + self.type = intern(str(type)) + self.value = value + + def __str__(self): + from jinja.lexer import keywords, reverse_operators + if self.type in keywords: + return self.type + elif self.type in reverse_operators: + return reverse_operators[self.type] + return self.value + + def __repr__(self): + return 'Token(%r, %r, %r)' % ( + self.lineno, + self.type, + self.value + ) + + +class TokenStreamIterator(object): + """ + The iterator for tokenstreams. Iterate over the stream + until the eof token is reached. + """ + + def __init__(self, stream): + self._stream = stream + + def __iter__(self): + return self + + def next(self): + token = self._stream.current + if token.type == 'eof': + self._stream.close() + raise StopIteration() + self._stream.next() + return token + + +class TokenStream(object): + """ + A token stream wraps a generator and supports pushing tokens back. + It also provides some functions to expect tokens and similar stuff. + + Important note: Do never push more than one token back to the + stream. Although the stream object won't stop you + from doing so, the behavior is undefined. Multiple + pushed tokens are only used internally! + """ + + def __init__(self, generator, filename): + self._next = generator.next + self._pushed = [] + self.current = Token(1, 'initial', '') + self.filename = filename + self.next() + + def __iter__(self): + return TokenStreamIterator(self) + + def bound(self): + """Return True if the token stream is bound to a parser.""" + return self.parser is not None + bound = property(bound, doc=bound.__doc__) + + def lineno(self): + """The current line number.""" + return self.current.lineno + lineno = property(lineno, doc=lineno.__doc__) + + def __nonzero__(self): + """Are we at the end of the tokenstream?""" + return bool(self._pushed) or self.current.type != 'eof' + + eos = property(lambda x: not x.__nonzero__(), doc=__nonzero__.__doc__) + + def look(self): + """See what's the next token.""" + if self._pushed: + return self._pushed[-1] + old_token = self.current + self.next() + new_token = self.current + self.current = old_token + self.push(new_token) + return new_token + + def push(self, token): + """Push a token back to the stream.""" + self._pushed.append(token) + + def skip(self, n): + """Got n tokens ahead.""" + for x in xrange(n): + self.next() + + def shift(self, token): + """ + Push one token into the stream. + """ + old_current = self.current + self.next() + self.push(self.current) + self.push(old_current) + self.push(token) + self.next() + + def next(self): + """Go one token ahead.""" + if self._pushed: + self.current = self._pushed.pop() + elif self.current.type != 'eof': + try: + self.current = self._next() + except StopIteration: + self.close() + + def read_whitespace(self): + """Read all the whitespace, up to the next tag.""" + lineno = self.current.lineno + buf = [] + while self.current.type == 'data' and not \ + self.current.value.strip(): + buf.append(self.current.value) + self.next() + if buf: + return Token(lineno, 'data', u''.join(buf)) + + def close(self): + """Close the stream.""" + self.current = Token(self.current.lineno, 'eof', '') + self._next = None + + def expect(self, token_type, token_value=_missing): + """Expect a given token type and return it""" + if self.current.type != token_type: + raise TemplateSyntaxError("expected token %r, got %r" % + (token_type, self.current.type), + self.current.lineno, + self.filename) + elif token_value is not _missing and \ + self.current.value != token_value: + raise TemplateSyntaxError("expected %r, got %r" % + (token_value, self.current.value), + self.current.lineno, + self.filename) + try: + return self.current + finally: + self.next() + + +class TemplateStream(object): + """ + Wraps a genererator for outputing template streams. + """ + + def __init__(self, gen): + self._gen = gen + self._next = gen.next + self.buffered = False + + def disable_buffering(self): + """ + Disable the output buffering. + """ + self._next = self._gen.next + self.buffered = False + + def enable_buffering(self, size=5): + """ + Enable buffering. Buffer `size` items before + yielding them. + """ + if size <= 1: + raise ValueError('buffer size too small') + self.buffered = True + + def buffering_next(): + buf = [] + c_size = 0 + push = buf.append + next = self._gen.next + + try: + while True: + item = next() + if item: + push(item) + c_size += 1 + if c_size >= size: + raise StopIteration() + except StopIteration: + if not c_size: + raise + return u''.join(buf) + + self._next = buffering_next + + def __iter__(self): + return self + + def next(self): + return self._next() diff --git a/app/jinja/debugger.py b/app/jinja/debugger.py new file mode 100644 index 0000000..932fb1c --- /dev/null +++ b/app/jinja/debugger.py @@ -0,0 +1,213 @@ +# -*- coding: utf-8 -*- +""" + jinja.debugger + ~~~~~~~~~~~~~~ + + This module implements helper function Jinja uses to give the users a + possibility to develop Jinja templates like they would debug python code. + It seamlessly integreates into the python traceback system, in fact it + just modifies the trackback stack so that the line numbers are correct + and the frame information are bound to the context and not the frame of + the template evaluation loop. + + To achive this it raises the exception it cought before in an isolated + namespace at a given line. The locals namespace is set to the current + template context. + + The traceback generated by raising that exception is then either returned + or linked with the former traceback if the `jinja._debugger` module is + available. Because it's not possible to modify traceback objects from the + python space this module is needed for this process. + + If it's not available it just ignores the other frames. Because this can + lead to actually harder to debug code there is a setting on the jinja + environment to disable the debugging system. + + The isolated namespace which is used to raise the exception also contains + a `__loader__` name that helds a reference to a PEP 302 compatible loader. + Because there are currently some traceback systems (such as the paste + evalexception debugger) that do not provide the frame globals when + retrieving the source from the linecache module, Jinja injects the source + to the linecache module itself and changes the filename to a URL style + "virtual filename" so that Jinja doesn't acidentally override other files + in the linecache. + + :copyright: 2007 by Armin Ronacher. + :license: BSD, see LICENSE for more details. +""" + +import sys +from random import randrange + +# if we have extended debugger support we should really use it +try: + from jinja._debugger import * + has_extended_debugger = True +except ImportError: + has_extended_debugger = False + +# we need the RUNTIME_EXCEPTION_OFFSET to skip the not used frames +from jinja.utils import reversed, RUNTIME_EXCEPTION_OFFSET + + +def fake_template_exception(exc_type, exc_value, tb, filename, lineno, + source, context_or_env, tb_back=None): + """ + Raise an exception "in a template". Return a traceback + object. This is used for runtime debugging, not compile time. + """ + # some traceback systems allow to skip frames + __traceback_hide__ = True + + # create the namespace which will be the local namespace + # of the new frame then. Some debuggers show local variables + # so we better inject the context and not the evaluation loop context. + from jinja.datastructure import Context + if isinstance(context_or_env, Context): + env = context_or_env.environment + namespace = context_or_env.to_dict() + else: + env = context_or_env + namespace = {} + + # no unicode for filenames + if isinstance(filename, unicode): + filename = filename.encode('utf-8') + + # generate an jinja unique filename used so that linecache + # gets data that doesn't interfere with other modules + if filename is None: + vfilename = 'jinja://~%d' % randrange(0, 10000) + filename = '' + else: + vfilename = 'jinja://%s' % filename + + # now create the used loaded and update the linecache + loader = TracebackLoader(env, source, filename) + loader.update_linecache(vfilename) + globals = { + '__name__': vfilename, + '__file__': vfilename, + '__loader__': loader + } + + # use the simple debugger to reraise the exception in the + # line where the error originally occoured + globals['__exception_to_raise__'] = (exc_type, exc_value) + offset = '\n' * (lineno - 1) + code = compile(offset + 'raise __exception_to_raise__[0], ' + '__exception_to_raise__[1]', + vfilename or '