[or-cvs] r14583: A small fix concerning the previous commit (in torflow/branches/gsoc2008: . tools tools/BeautifulSoup tools/SocksiPy)
aleksei at seul.org
aleksei at seul.org
Sun May 11 10:27:34 UTC 2008
Author: aleksei
Date: 2008-05-11 06:27:34 -0400 (Sun, 11 May 2008)
New Revision: 14583
Added:
torflow/branches/gsoc2008/getExitsToPort.py
torflow/branches/gsoc2008/soat.py
torflow/branches/gsoc2008/tools/BeautifulSoup/
torflow/branches/gsoc2008/tools/BeautifulSoup/BeautifulSoup.py
torflow/branches/gsoc2008/tools/SocksiPy/
torflow/branches/gsoc2008/tools/SocksiPy/socks.py
Log:
A small fix concerning the previous commit
Added: torflow/branches/gsoc2008/getExitsToPort.py
===================================================================
--- torflow/branches/gsoc2008/getExitsToPort.py (rev 0)
+++ torflow/branches/gsoc2008/getExitsToPort.py 2008-05-11 10:27:34 UTC (rev 14583)
@@ -0,0 +1,47 @@
+#!/usr/bin/python
+
+import sys
+import socket
+from TorCtl import PathSupport, TorCtl, TorUtil
+from TorCtl.TorUtil import *
+from TorCtl.PathSupport import *
+from TorCtl.TorCtl import Connection
+
+control_port = 9051
+
+def main(argv):
+ try:
+ port = int(argv[1])
+ except IndexError:
+ plog('INFO', 'No port number supplied as an argument. Using 80')
+ port = 80
+ except:
+ plog('ERROR', 'Port number must be an integer')
+ exit()
+
+ s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
+ try:
+ s.connect(("127.0.0.1", control_port))
+ except:
+ plog('ERROR', 'Couldn\'t connect to the control port. Port ' + control_port + ' was used')
+ exit()
+
+ c = Connection(s)
+ c.debug(file("exits.log", "w"))
+ c.authenticate()
+
+ plog('INFO', 'Connection established')
+
+ routers = c.read_routers(c.get_network_status())
+ good_exits = []
+
+ exit_restriction = FlagsRestriction(["Exit"],[])
+ port_restriction = OrNodeRestriction(
+ [ExitPolicyRestriction('255.255.255.255', port)])
+ for router in routers:
+ if exit_restriction and port_restriction.r_is_ok(router):
+ good_exits.append(router)
+ plog('INFO', 'Total routers with given exit policy: ' + len(good_exits))
+
+if __name__ == "__main__":
+ main(sys.argv)
Property changes on: torflow/branches/gsoc2008/getExitsToPort.py
___________________________________________________________________
Name: svn:executable
+ *
Added: torflow/branches/gsoc2008/soat.py
===================================================================
--- torflow/branches/gsoc2008/soat.py (rev 0)
+++ torflow/branches/gsoc2008/soat.py 2008-05-11 10:27:34 UTC (rev 14583)
@@ -0,0 +1,243 @@
+import httplib
+import os
+import random
+import re
+from sets import Set
+import socket
+import string
+import sys
+import urllib
+import urllib2
+
+from TorCtl import TorUtil
+from TorCtl.TorUtil import *
+
+sys.path.append("./tools/BeautifulSoup/")
+from BeautifulSoup import BeautifulSoup
+
+sys.path.append("./tools/SocksiPy")
+import socks
+
+# config stuff
+
+meta_port = 9052
+meta_host = '127.0.0.1'
+
+user_agent = 'Mozilla/5.0 (Windows; U; Windows NT 5.1; de; rv:1.8.1) Gecko/20061010 Firefox/2.0'
+
+wordlist_file = './wordlist.txt';
+allowed_filetypes = ['all','pdf']
+result_per_type = 1
+
+same_origin_policy = True
+
+# links of interest
+
+doc_urls = ['http://www.torproject.org']
+doc_https = []
+
+# constants
+
+linebreak = '\r\n'
+
+# a simple interface to handle a socket connection
+# with readline and writeline capability
+class Client:
+
+ def __init__(self, host, port):
+ self.s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
+ self.s.connect((host, port))
+ self.buffer = self.s.makefile('rb')
+
+ def writeline(self, line):
+ self.s.send(line + linebreak)
+
+ def readline(self):
+ s = self.buffer.readline()
+ if not s:
+ raise EOFError
+ if s[-2:] == linebreak:
+ s = s[:-2]
+ elif s[-1:] in linebreak:
+ s = s[:-1]
+ return s
+
+# The scanner class
+class ExitNodeScanner:
+
+ def __init__(self,meta_host,meta_port):
+ # establish a connection
+ plog('INFO', 'ExitNodeScanner starting up...')
+ try:
+ self.__client = Client(meta_host, meta_port)
+ except socket.error:
+ plog('ERROR', 'Couldn\'t connect to metatroller. Is it on?')
+ exit()
+
+ # skip two lines of metatroller introduction
+ data = self.__client.readline()
+ data = self.__client.readline()
+
+ # configure metatroller
+ commands = [
+ 'PATHLEN ',
+ 'PERCENTFAST 88',
+ 'USEALLEXITS 1',
+ 'UNIFORM 0',
+ 'BWCUTOFF 1',
+ 'ORDEREXITS',
+ 'GUARDNODES 0',
+ 'RESETSTATS']
+ plog('INFO', 'Executing preliminary configuration commands')
+ for c in commands:
+ self.__client.writeline(c)
+ reply = self.__client.readline()
+ if reply[:3] != '250': # first three chars indicate the reply code
+ reply += self.__client.readline()
+ plog('ERROR', reply)
+ plog('INFO', 'ExitNodeScanner up and ready')
+
+ def get_exit_node(self):
+ self.__client.writeline("GETLASTEXIT" + linebreak)
+ reply = self.__client.readline()
+
+ if reply[:3] != '250':
+ reply += self.__client.readline()
+ plog('ERROR', reply)
+ return 0
+
+ p = re.compile('250 LASTEXIT=[\S]+')
+ m = p.match(reply)
+ self.__exit = m.group()[13:] # drop the irrelevant characters
+ plog('NOTICE','Current node: ' + self.__exit)
+ return self.__exit
+
+ def check_http(self, address):
+ request = urllib2.Request(address)
+ request.add_header('User-Agent', user_agent)
+
+ try:
+ f = urllib2.urlopen(request)
+ except urllib2.URLError:
+ plog('ERROR', 'The requested page ' + address + ' doesn\'t exist')
+ return 0
+ except:
+ plog('ERROR', 'Opening ' + address + ' failed')
+ return 0
+
+ content = f.read()
+ content = content.decode('ascii', 'ignore')
+
+ socks.setdefaultproxy(socks.PROXY_TYPE_SOCKS5, "127.0.0.1:9060")
+ socks.socket = socks.socksocket
+
+ try:
+ g = urllib2.urlopen(request)
+ except:
+ plog('ERROR', 'Opening ' + address + ' via tor failed')
+ return 0
+
+ pcontent = g.read()
+
+ return 0
+
+ def check_openssh(self, address):
+ return 0
+
+ def check_openssl(self, address):
+ return 0
+
+
+# some helpful methods
+
+'''
+construct a list of urls based on the wordlist and filetypes of interest
+'''
+def load_urls():
+ plog('INFO', 'Loading url list')
+
+ wordlist = []
+ fh = open(wordlist_file, 'r')
+ try:
+ for line in fh:
+ wordlist.append(line[:-1]) # get rid of the linebreaks
+ finally:
+ fh.close()
+
+ urllist = []
+ for ft in allowed_filetypes:
+ type_urls = []
+
+ while len(type_urls) < result_per_type:
+ # probably the discover_urls method should consider moving along the search result pages
+ type_urls.extend(discover_urls(ft,
+ wordlist[int(random.random() * len(wordlist))]))
+ type_urls = list(Set(type_urls))
+
+ plog('INFO', 'URL list for ' + ft + ': ' + '\n'.join(type_urls) + '\n')
+ urllist.extend(type_urls)
+
+ return urllist
+
+'''
+Find links to files related to a query
+'''
+def discover_urls(self, filetype, query):
+ # search google for relevant pages
+ # note: google only accepts requests from idenitified browsers
+ if filetype != 'all':
+ query += ':' + filetype
+
+ host = 'www.google.com'
+ params = urllib.urlencode({'q' : query})
+ headers = {'User-Agent' : user_agent}
+ search_url = '/search' + '?' + params
+
+ connection = httplib.HTTPConnection(host)
+ connection.request("GET", search_url, {}, headers) # can't add params here for some reason
+
+ response = connection.getresponse()
+ if response.status == 200:
+ # if everything went well, start parsing
+ urls = []
+
+ content = response.read()
+ content = content.decode('ascii', 'ignore')
+
+ soup = BeautifulSoup(content)
+
+ # l is the class for relevant reply links
+ # probably not the best criterion to rely on, so maybe some other solution needed
+ for link in soup.findAll('a', {'class' : 'l'}):
+ urls.append(link['href'])
+
+ # filter for filetypes if needed
+ if filetype != 'all':
+ urls = [u for u in urls if u[-len(filetype):] == filetype]
+
+ return urls
+ else:
+ plog('ERROR', 'Google search failed: ' +
+ response.status + ' ' + response.reason)
+ return []
+
+# main logic
+
+def main(argv):
+ scanner = ExitNodeScanner(meta_host, meta_port)
+ scanner.get_exit_node()
+ scanner.check_http("http://www.ee.ee")
+
+ '''
+ global doc_urls
+ doc_urls.extend(load_url_list())
+ doc_urls = list(Set(doc_urls))
+
+ plog('NOTICE', 'Final URL list: ' + '\n'.join(doc_urls) + '\n')
+ plog('INFO', 'Beginning scan loop... some day?')
+ '''
+
+# initiate the program
+
+if __name__ == '__main__':
+ main(sys.argv)
Property changes on: torflow/branches/gsoc2008/soat.py
___________________________________________________________________
Name: svn:executable
+ *
Added: torflow/branches/gsoc2008/tools/BeautifulSoup/BeautifulSoup.py
===================================================================
--- torflow/branches/gsoc2008/tools/BeautifulSoup/BeautifulSoup.py (rev 0)
+++ torflow/branches/gsoc2008/tools/BeautifulSoup/BeautifulSoup.py 2008-05-11 10:27:34 UTC (rev 14583)
@@ -0,0 +1,1920 @@
+"""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 at segfault.org)"
+__version__ = "3.0.5"
+__copyright__ = "Copyright (c) 2004-2007 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
+
+ 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
+ before after Tag in the document."""
+ return self._findAll(name, attrs, text, limit, self.nextGenerator)
+
+ 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 unicode(str(self))
+
+ 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 "<![CDATA[%s]]>" % 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 "<?%s?>" % self.toEncoding(output, encoding)
+
+class Comment(NavigableString):
+ def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING):
+ return "<!--%s-->" % NavigableString.__str__(self, encoding)
+
+class Declaration(NavigableString):
+ def __str__(self, encoding=DEFAULT_OUTPUT_ENCODING):
+ return "<!%s>" % 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 = '</%s>' % 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 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, "<foo><bar></foo>" actually means
+ "<foo><bar></bar></foo>".
+
+ [Another possible explanation is "<foo><bar /></foo>", 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('<!\s+([^<>]*)>'),
+ lambda x: '<!' + x.group(1) + '>')
+ ]
+
+ 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:
+
+ <br/> (No space between name of closing tag and tag close)
+ <! --Comment--> (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:
+ <p>Foo<b>Bar *<p>* should pop to 'p', not 'b'.
+ <p>Foo<table>Bar *<p>* should pop to 'table', not 'p'.
+ <p>Foo<table><tr>Bar *<p>* should pop to 'tr', not 'p'.
+
+ <li><ul><li> *<li>* should pop to 'ul', not the first 'li'.
+ <tr><table><tr> *<tr>* should pop to 'table', not the first 'tr'
+ <td><tr><td> *<td>* 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 "</%s> is not real!" % name
+ self.handle_data('</%s>' % 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] == '<![CDATA[':
+ k = self.rawdata.find(']]>', 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 <p> tag should implicitly close the previous <p> tag.
+
+ <p>Para1<p>Para2
+ should be transformed into:
+ <p>Para1</p><p>Para2
+
+ Some tags can be nested arbitrarily. For instance, the occurance
+ of a <blockquote> tag should _not_ implicitly close the previous
+ <blockquote> tag.
+
+ Alice said: <blockquote>Bob said: <blockquote>Blah
+ should NOT be transformed into:
+ Alice said: <blockquote>Bob said: </blockquote><blockquote>Blah
+
+ Some tags can be nested, but the nesting is reset by the
+ interposition of other tags. For instance, a <tr> tag should
+ implicitly close the previous <tr> tag within the same <table>,
+ but not close a <tr> tag in another table.
+
+ <table><tr>Blah<tr>Blah
+ should be transformed into:
+ <table><tr>Blah</tr><tr>Blah
+ but,
+ <tr>Blah<table><tr>Blah
+ should NOT be transformed into
+ <tr>Blah<table></tr><tr>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%", value)
+ 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:
+
+ <b>Foo<b>Bar</b></b>
+
+ 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
+ "<b>Foo<b>Bar", 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 '</b></b>' 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
+ <script> tags contain Javascript and should not be parsed, that
+ META tags may contain encoding information, and so on.
+
+ This also makes it better for subclassing than BeautifulStoneSoup
+ or BeautifulSoup."""
+
+ RESET_NESTING_TAGS = buildTagMap('noscript')
+ NESTABLE_TAGS = {}
+
+class BeautifulSOAP(BeautifulStoneSoup):
+ """This class will push a tag with only a single string child into
+ the tag's parent as an attribute. The attribute's name is the tag
+ name, and the value is the string child. An example should give
+ the flavor of the change:
+
+ <foo><bar>baz</bar></foo>
+ =>
+ <foo bar="baz"><bar>baz</bar></foo>
+
+ You can then access fooTag['bar'] instead of fooTag.barTag.string.
+
+ This is, of course, useful for scraping structures that tend to
+ use subelements instead of attributes, such as SOAP messages. Note
+ that it modifies its input, so don't print the modified version
+ out.
+
+ I'm not sure how many people really want to use this class; let me
+ know if you do. Mainly I like the name."""
+
+ def popTag(self):
+ if len(self.tagStack) > 1:
+ tag = self.tagStack[-1]
+ parent = self.tagStack[-2]
+ parent._getAttrMap()
+ if (isinstance(tag, Tag) and len(tag.contents) == 1 and
+ isinstance(tag.contents[0], NavigableString) and
+ not parent.attrMap.has_key(tag.name)):
+ parent[tag.name] = tag.contents[0]
+ BeautifulStoneSoup.popTag(self)
+
+#Enterprise class names! It has come to our attention that some people
+#think the names of the Beautiful Soup parser classes are too silly
+#and "unprofessional" for use in enterprise screen-scraping. We feel
+#your pain! For such-minded folk, the Beautiful Soup Consortium And
+#All-Night Kosher Bakery recommends renaming this file to
+#"RobustParser.py" (or, in cases of extreme enterprisiness,
+#"RobustParserBeanInterface.class") and using the following
+#enterprise-friendly class aliases:
+class RobustXMLParser(BeautifulStoneSoup):
+ pass
+class RobustHTMLParser(BeautifulSoup):
+ pass
+class RobustWackAssHTMLParser(ICantBelieveItsBeautifulSoup):
+ pass
+class RobustInsanelyWackAssHTMLParser(MinimalSoup):
+ pass
+class SimplifyingSOAPParser(BeautifulSOAP):
+ pass
+
+######################################################
+#
+# Bonus library: Unicode, Dammit
+#
+# This class forces XML data into a standard format (usually to UTF-8
+# or Unicode). It is heavily based on code from Mark Pilgrim's
+# Universal Feed Parser. It does not rewrite the XML or HTML to
+# reflect a new encoding: that happens in BeautifulStoneSoup.handle_pi
+# (XML) and BeautifulSoup.start_meta (HTML).
+
+# Autodetects character encodings.
+# Download from http://chardet.feedparser.org/
+try:
+ import chardet
+# import chardet.constants
+# chardet.constants._debug = 1
+except:
+ chardet = None
+chardet = None
+
+# cjkcodecs and iconv_codec make Python know about more character encodings.
+# Both are available from http://cjkpython.i18n.org/
+# They're built in if you use Python 2.4.
+try:
+ import cjkcodecs.aliases
+except:
+ pass
+try:
+ import iconv_codec
+except:
+ pass
+
+class UnicodeDammit:
+ """A class for detecting the encoding of a *ML document and
+ converting it to a Unicode string. If the source encoding is
+ windows-1252, can replace MS smart quotes with their HTML or XML
+ equivalents."""
+
+ # This dictionary maps commonly seen values for "charset" in HTML
+ # meta tags to the corresponding Python codec names. It only covers
+ # values that aren't in Python's aliases and can't be determined
+ # by the heuristics in find_codec.
+ CHARSET_ALIASES = { "macintosh" : "mac-roman",
+ "x-sjis" : "shift-jis" }
+
+ def __init__(self, markup, overrideEncodings=[],
+ smartQuotesTo='xml'):
+ self.markup, documentEncoding, sniffedEncoding = \
+ self._detectEncoding(markup)
+ self.smartQuotesTo = smartQuotesTo
+ self.triedEncodings = []
+ if markup == '' or isinstance(markup, unicode):
+ self.originalEncoding = None
+ self.unicode = unicode(markup)
+ return
+
+ u = None
+ for proposedEncoding in overrideEncodings:
+ u = self._convertFrom(proposedEncoding)
+ if u: break
+ if not u:
+ for proposedEncoding in (documentEncoding, sniffedEncoding):
+ u = self._convertFrom(proposedEncoding)
+ if u: break
+
+ # If no luck and we have auto-detection library, try that:
+ if not u and chardet and not isinstance(self.markup, unicode):
+ u = self._convertFrom(chardet.detect(self.markup)['encoding'])
+
+ # As a last resort, try utf-8 and windows-1252:
+ if not u:
+ for proposed_encoding in ("utf-8", "windows-1252"):
+ u = self._convertFrom(proposed_encoding)
+ if u: break
+ self.unicode = u
+ if not u: self.originalEncoding = None
+
+ def _subMSChar(self, orig):
+ """Changes a MS smart quote character to an XML or HTML
+ entity."""
+ sub = self.MS_CHARS.get(orig)
+ if type(sub) == types.TupleType:
+ if self.smartQuotesTo == 'xml':
+ sub = '&#x%s;' % sub[1]
+ else:
+ sub = '&%s;' % sub[0]
+ return sub
+
+ def _convertFrom(self, proposed):
+ proposed = self.find_codec(proposed)
+ if not proposed or proposed in self.triedEncodings:
+ return None
+ self.triedEncodings.append(proposed)
+ markup = self.markup
+
+ # Convert smart quotes to HTML if coming from an encoding
+ # that might have them.
+ if self.smartQuotesTo and proposed.lower() in("windows-1252",
+ "iso-8859-1",
+ "iso-8859-2"):
+ markup = re.compile("([\x80-\x9f])").sub \
+ (lambda(x): self._subMSChar(x.group(1)),
+ markup)
+
+ try:
+ # print "Trying to convert document to %s" % proposed
+ u = self._toUnicode(markup, proposed)
+ self.markup = u
+ self.originalEncoding = proposed
+ except Exception, e:
+ # print "That didn't work!"
+ # print e
+ return None
+ #print "Correct encoding: %s" % proposed
+ return self.markup
+
+ def _toUnicode(self, data, encoding):
+ '''Given a string and its encoding, decodes the string into Unicode.
+ %encoding is a string recognized by encodings.aliases'''
+
+ # strip Byte Order Mark (if present)
+ if (len(data) >= 4) and (data[:2] == '\xfe\xff') \
+ and (data[2:4] != '\x00\x00'):
+ encoding = 'utf-16be'
+ data = data[2:]
+ elif (len(data) >= 4) and (data[:2] == '\xff\xfe') \
+ and (data[2:4] != '\x00\x00'):
+ encoding = 'utf-16le'
+ data = data[2:]
+ elif data[:3] == '\xef\xbb\xbf':
+ encoding = 'utf-8'
+ data = data[3:]
+ elif data[:4] == '\x00\x00\xfe\xff':
+ encoding = 'utf-32be'
+ data = data[4:]
+ elif data[:4] == '\xff\xfe\x00\x00':
+ encoding = 'utf-32le'
+ data = data[4:]
+ newdata = unicode(data, encoding)
+ return newdata
+
+ def _detectEncoding(self, xml_data):
+ """Given a document, tries to detect its XML encoding."""
+ xml_encoding = sniffed_xml_encoding = None
+ try:
+ if xml_data[:4] == '\x4c\x6f\xa7\x94':
+ # EBCDIC
+ xml_data = self._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:
+ sniffed_xml_encoding = 'ascii'
+ 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
+ return xml_data, xml_encoding, sniffed_xml_encoding
+
+
+ def find_codec(self, charset):
+ return self._codec(self.CHARSET_ALIASES.get(charset, charset)) \
+ or (charset and self._codec(charset.replace("-", ""))) \
+ or (charset and self._codec(charset.replace("-", "_"))) \
+ or charset
+
+ def _codec(self, charset):
+ if not charset: return charset
+ codec = None
+ try:
+ codecs.lookup(charset)
+ codec = charset
+ except (LookupError, ValueError):
+ pass
+ return codec
+
+ EBCDIC_TO_ASCII_MAP = None
+ def _ebcdic_to_ascii(self, s):
+ c = self.__class__
+ if not c.EBCDIC_TO_ASCII_MAP:
+ emap = (0,1,2,3,156,9,134,127,151,141,142,11,12,13,14,15,
+ 16,17,18,19,157,133,8,135,24,25,146,143,28,29,30,31,
+ 128,129,130,131,132,10,23,27,136,137,138,139,140,5,6,7,
+ 144,145,22,147,148,149,150,4,152,153,154,155,20,21,158,26,
+ 32,160,161,162,163,164,165,166,167,168,91,46,60,40,43,33,
+ 38,169,170,171,172,173,174,175,176,177,93,36,42,41,59,94,
+ 45,47,178,179,180,181,182,183,184,185,124,44,37,95,62,63,
+ 186,187,188,189,190,191,192,193,194,96,58,35,64,39,61,34,
+ 195,97,98,99,100,101,102,103,104,105,196,197,198,199,200,
+ 201,202,106,107,108,109,110,111,112,113,114,203,204,205,
+ 206,207,208,209,126,115,116,117,118,119,120,121,122,210,
+ 211,212,213,214,215,216,217,218,219,220,221,222,223,224,
+ 225,226,227,228,229,230,231,123,65,66,67,68,69,70,71,72,
+ 73,232,233,234,235,236,237,125,74,75,76,77,78,79,80,81,
+ 82,238,239,240,241,242,243,92,159,83,84,85,86,87,88,89,
+ 90,244,245,246,247,248,249,48,49,50,51,52,53,54,55,56,57,
+ 250,251,252,253,254,255)
+ import string
+ c.EBCDIC_TO_ASCII_MAP = string.maketrans( \
+ ''.join(map(chr, range(256))), ''.join(map(chr, emap)))
+ return s.translate(c.EBCDIC_TO_ASCII_MAP)
+
+ MS_CHARS = { '\x80' : ('euro', '20AC'),
+ '\x81' : ' ',
+ '\x82' : ('sbquo', '201A'),
+ '\x83' : ('fnof', '192'),
+ '\x84' : ('bdquo', '201E'),
+ '\x85' : ('hellip', '2026'),
+ '\x86' : ('dagger', '2020'),
+ '\x87' : ('Dagger', '2021'),
+ '\x88' : ('circ', '2C6'),
+ '\x89' : ('permil', '2030'),
+ '\x8A' : ('Scaron', '160'),
+ '\x8B' : ('lsaquo', '2039'),
+ '\x8C' : ('OElig', '152'),
+ '\x8D' : '?',
+ '\x8E' : ('#x17D', '17D'),
+ '\x8F' : '?',
+ '\x90' : '?',
+ '\x91' : ('lsquo', '2018'),
+ '\x92' : ('rsquo', '2019'),
+ '\x93' : ('ldquo', '201C'),
+ '\x94' : ('rdquo', '201D'),
+ '\x95' : ('bull', '2022'),
+ '\x96' : ('ndash', '2013'),
+ '\x97' : ('mdash', '2014'),
+ '\x98' : ('tilde', '2DC'),
+ '\x99' : ('trade', '2122'),
+ '\x9a' : ('scaron', '161'),
+ '\x9b' : ('rsaquo', '203A'),
+ '\x9c' : ('oelig', '153'),
+ '\x9d' : '?',
+ '\x9e' : ('#x17E', '17E'),
+ '\x9f' : ('Yuml', ''),}
+
+#######################################################################
+
+
+#By default, act as an HTML pretty-printer.
+if __name__ == '__main__':
+ import sys
+ soup = BeautifulSoup(sys.stdin.read())
+ print soup.prettify()
Added: torflow/branches/gsoc2008/tools/SocksiPy/socks.py
===================================================================
--- torflow/branches/gsoc2008/tools/SocksiPy/socks.py (rev 0)
+++ torflow/branches/gsoc2008/tools/SocksiPy/socks.py 2008-05-11 10:27:34 UTC (rev 14583)
@@ -0,0 +1,387 @@
+"""SocksiPy - Python SOCKS module.
+Version 1.00
+
+Copyright 2006 Dan-Haim. All rights reserved.
+
+Redistribution and use in source and binary forms, with or without modification,
+are permitted provided that the following conditions are met:
+1. Redistributions of source code must retain the above copyright notice, this
+ list of conditions and the following disclaimer.
+2. 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.
+3. Neither the name of Dan Haim nor the names of his contributors may be used
+ to endorse or promote products derived from this software without specific
+ prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY DAN HAIM "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 DAN HAIM OR HIS 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, 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 DAMANGE.
+
+
+This module provides a standard socket-like interface for Python
+for tunneling connections through SOCKS proxies.
+
+"""
+
+import socket
+import struct
+
+PROXY_TYPE_SOCKS4 = 1
+PROXY_TYPE_SOCKS5 = 2
+PROXY_TYPE_HTTP = 3
+
+_defaultproxy = None
+_orgsocket = socket.socket
+
+class ProxyError(Exception):
+ def __init__(self, value):
+ self.value = value
+ def __str__(self):
+ return repr(self.value)
+
+class GeneralProxyError(ProxyError):
+ def __init__(self, value):
+ self.value = value
+ def __str__(self):
+ return repr(self.value)
+
+class Socks5AuthError(ProxyError):
+ def __init__(self, value):
+ self.value = value
+ def __str__(self):
+ return repr(self.value)
+
+class Socks5Error(ProxyError):
+ def __init__(self, value):
+ self.value = value
+ def __str__(self):
+ return repr(self.value)
+
+class Socks4Error(ProxyError):
+ def __init__(self, value):
+ self.value = value
+ def __str__(self):
+ return repr(self.value)
+
+class HTTPError(ProxyError):
+ def __init__(self, value):
+ self.value = value
+ def __str__(self):
+ return repr(self.value)
+
+_generalerrors = ("success",
+ "invalid data",
+ "not connected",
+ "not available",
+ "bad proxy type",
+ "bad input")
+
+_socks5errors = ("succeeded",
+ "general SOCKS server failure",
+ "connection not allowed by ruleset",
+ "Network unreachable",
+ "Host unreachable",
+ "Connection refused",
+ "TTL expired",
+ "Command not supported",
+ "Address type not supported",
+ "Unknown error")
+
+_socks5autherrors = ("succeeded",
+ "authentication is required",
+ "all offered authentication methods were rejected",
+ "unknown username or invalid password",
+ "unknown error")
+
+_socks4errors = ("request granted",
+ "request rejected or failed",
+ "request rejected because SOCKS server cannot connect to identd on the client",
+ "request rejected because the client program and identd report different user-ids",
+ "unknown error")
+
+def setdefaultproxy(proxytype=None,addr=None,port=None,rdns=True,username=None,password=None):
+ """setdefaultproxy(proxytype, addr[, port[, rdns[, username[, password]]]])
+ Sets a default proxy which all further socksocket objects will use,
+ unless explicitly changed.
+ """
+ global _defaultproxy
+ _defaultproxy = (proxytype,addr,port,rdns,username,password)
+
+class socksocket(socket.socket):
+ """socksocket([family[, type[, proto]]]) -> socket object
+
+ Open a SOCKS enabled socket. The parameters are the same as
+ those of the standard socket init. In order for SOCKS to work,
+ you must specify family=AF_INET, type=SOCK_STREAM and proto=0.
+ """
+
+ def __init__(self, family=socket.AF_INET, type=socket.SOCK_STREAM, proto=0, _sock=None):
+ _orgsocket.__init__(self,family,type,proto,_sock)
+ if _defaultproxy != None:
+ self.__proxy = _defaultproxy
+ else:
+ self.__proxy = (None, None, None, None, None, None)
+ self.__proxysockname = None
+ self.__proxypeername = None
+
+ def __recvall(self, bytes):
+ """__recvall(bytes) -> data
+ Receive EXACTLY the number of bytes requested from the socket.
+ Blocks until the required number of bytes have been received.
+ """
+ data = ""
+ while len(data) < bytes:
+ data = data + self.recv(bytes-len(data))
+ return data
+
+ def setproxy(self,proxytype=None,addr=None,port=None,rdns=True,username=None,password=None):
+ """setproxy(proxytype, addr[, port[, rdns[, username[, password]]]])
+ Sets the proxy to be used.
+ proxytype - The type of the proxy to be used. Three types
+ are supported: PROXY_TYPE_SOCKS4 (including socks4a),
+ PROXY_TYPE_SOCKS5 and PROXY_TYPE_HTTP
+ addr - The address of the server (IP or DNS).
+ port - The port of the server. Defaults to 1080 for SOCKS
+ servers and 8080 for HTTP proxy servers.
+ rdns - Should DNS queries be preformed on the remote side
+ (rather than the local side). The default is True.
+ Note: This has no effect with SOCKS4 servers.
+ username - Username to authenticate with to the server.
+ The default is no authentication.
+ password - Password to authenticate with to the server.
+ Only relevant when username is also provided.
+ """
+ self.__proxy = (proxytype,addr,port,rdns,username,password)
+
+ def __negotiatesocks5(self,destaddr,destport):
+ """__negotiatesocks5(self,destaddr,destport)
+ Negotiates a connection through a SOCKS5 server.
+ """
+ # First we'll send the authentication packages we support.
+ if (self.__proxy[4]!=None) and (self.__proxy[5]!=None):
+ # The username/password details were supplied to the
+ # setproxy method so we support the USERNAME/PASSWORD
+ # authentication (in addition to the standard none).
+ self.sendall("\x05\x02\x00\x02")
+ else:
+ # No username/password were entered, therefore we
+ # only support connections with no authentication.
+ self.sendall("\x05\x01\x00")
+ # We'll receive the server's response to determine which
+ # method was selected
+ chosenauth = self.__recvall(2)
+ if chosenauth[0] != "\x05":
+ self.close()
+ raise GeneralProxyError((1,_generalerrors[1]))
+ # Check the chosen authentication method
+ if chosenauth[1] == "\x00":
+ # No authentication is required
+ pass
+ elif chosenauth[1] == "\x02":
+ # Okay, we need to perform a basic username/password
+ # authentication.
+ self.sendall("\x01" + chr(len(self.__proxy[4])) + self.__proxy[4] + chr(len(self.proxy[5])) + self.__proxy[5])
+ authstat = self.__recvall(2)
+ if authstat[0] != "\x01":
+ # Bad response
+ self.close()
+ raise GeneralProxyError((1,_generalerrors[1]))
+ if authstat[1] != "\x00":
+ # Authentication failed
+ self.close()
+ raise Socks5AuthError,((3,_socks5autherrors[3]))
+ # Authentication succeeded
+ else:
+ # Reaching here is always bad
+ self.close()
+ if chosenauth[1] == "\xFF":
+ raise Socks5AuthError((2,_socks5autherrors[2]))
+ else:
+ raise GeneralProxyError((1,_generalerrors[1]))
+ # Now we can request the actual connection
+ req = "\x05\x01\x00"
+ # If the given destination address is an IP address, we'll
+ # use the IPv4 address request even if remote resolving was specified.
+ try:
+ ipaddr = socket.inet_aton(destaddr)
+ req = req + "\x01" + ipaddr
+ except socket.error:
+ # Well it's not an IP number, so it's probably a DNS name.
+ if self.__proxy[3]==True:
+ # Resolve remotely
+ ipaddr = None
+ req = req + "\x03" + chr(len(destaddr)) + destaddr
+ else:
+ # Resolve locally
+ ipaddr = socket.inet_aton(socket.gethostbyname(destaddr))
+ req = req + "\x01" + ipaddr
+ req = req + struct.pack(">H",destport)
+ self.sendall(req)
+ # Get the response
+ resp = self.__recvall(4)
+ if resp[0] != "\x05":
+ self.close()
+ raise GeneralProxyError((1,_generalerrors[1]))
+ elif resp[1] != "\x00":
+ # Connection failed
+ self.close()
+ if ord(resp[1])<=8:
+ raise Socks5Error(ord(resp[1]),_generalerrors[ord(resp[1])])
+ else:
+ raise Socks5Error(9,_generalerrors[9])
+ # Get the bound address/port
+ elif resp[3] == "\x01":
+ boundaddr = self.__recvall(4)
+ elif resp[3] == "\x03":
+ resp = resp + self.recv(1)
+ boundaddr = self.__recvall(resp[4])
+ else:
+ self.close()
+ raise GeneralProxyError((1,_generalerrors[1]))
+ boundport = struct.unpack(">H",self.__recvall(2))[0]
+ self.__proxysockname = (boundaddr,boundport)
+ if ipaddr != None:
+ self.__proxypeername = (socket.inet_ntoa(ipaddr),destport)
+ else:
+ self.__proxypeername = (destaddr,destport)
+
+ def getproxysockname(self):
+ """getsockname() -> address info
+ Returns the bound IP address and port number at the proxy.
+ """
+ return self.__proxysockname
+
+ def getproxypeername(self):
+ """getproxypeername() -> address info
+ Returns the IP and port number of the proxy.
+ """
+ return _orgsocket.getpeername(self)
+
+ def getpeername(self):
+ """getpeername() -> address info
+ Returns the IP address and port number of the destination
+ machine (note: getproxypeername returns the proxy)
+ """
+ return self.__proxypeername
+
+ def __negotiatesocks4(self,destaddr,destport):
+ """__negotiatesocks4(self,destaddr,destport)
+ Negotiates a connection through a SOCKS4 server.
+ """
+ # Check if the destination address provided is an IP address
+ rmtrslv = False
+ try:
+ ipaddr = socket.inet_aton(destaddr)
+ except socket.error:
+ # It's a DNS name. Check where it should be resolved.
+ if self.__proxy[3]==True:
+ ipaddr = "\x00\x00\x00\x01"
+ rmtrslv = True
+ else:
+ ipaddr = socket.inet_aton(socket.gethostbyname(destaddr))
+ # Construct the request packet
+ req = "\x04\x01" + struct.pack(">H",destport) + ipaddr
+ # The username parameter is considered userid for SOCKS4
+ if self.__proxy[4] != None:
+ req = req + self.__proxy[4]
+ req = req + "\x00"
+ # DNS name if remote resolving is required
+ # NOTE: This is actually an extension to the SOCKS4 protocol
+ # called SOCKS4A and may not be supported in all cases.
+ if rmtrslv==True:
+ req = req + destaddr + "\x00"
+ self.sendall(req)
+ # Get the response from the server
+ resp = self.__recvall(8)
+ if resp[0] != "\x00":
+ # Bad data
+ self.close()
+ raise GeneralProxyError((1,_generalerrors[1]))
+ if resp[1] != "\x5A":
+ # Server returned an error
+ self.close()
+ if ord(resp[1]) in (91,92,93):
+ self.close()
+ raise Socks4Error((ord(resp[1]),_socks4errors[ord(resp[1])-90]))
+ else:
+ raise Socks4Error((94,_socks4errors[4]))
+ # Get the bound address/port
+ self.__proxysockname = (socket.inet_ntoa(resp[4:]),struct.unpack(">H",resp[2:4])[0])
+ if rmtrslv != None:
+ self.__proxypeername = (socket.inet_ntoa(ipaddr),destport)
+ else:
+ self.__proxypeername = (destaddr,destport)
+
+ def __negotiatehttp(self,destaddr,destport):
+ """__negotiatehttp(self,destaddr,destport)
+ Negotiates a connection through an HTTP server.
+ """
+ # If we need to resolve locally, we do this now
+ if self.__proxy[3] == False:
+ addr = socket.gethostbyname(destaddr)
+ else:
+ addr = destaddr
+ self.sendall("CONNECT " + addr + ":" + str(destport) + " HTTP/1.1\r\n" + "Host: " + destaddr + "\r\n\r\n")
+ # We read the response until we get the string "\r\n\r\n"
+ resp = self.recv(1)
+ while resp.find("\r\n\r\n")==-1:
+ resp = resp + self.recv(1)
+ # We just need the first line to check if the connection
+ # was successful
+ statusline = resp.splitlines()[0].split(" ",2)
+ if statusline[0] not in ("HTTP/1.0","HTTP/1.1"):
+ self.close()
+ raise GeneralProxyError((1,_generalerrors[1]))
+ try:
+ statuscode = int(statusline[1])
+ except ValueError:
+ self.close()
+ raise GeneralProxyError((1,_generalerrors[1]))
+ if statuscode != 200:
+ self.close()
+ raise HTTPError((statuscode,statusline[2]))
+ self.__proxysockname = ("0.0.0.0",0)
+ self.__proxypeername = (addr,destport)
+
+ def connect(self,destpair):
+ """connect(self,despair)
+ Connects to the specified destination through a proxy.
+ destpar - A tuple of the IP/DNS address and the port number.
+ (identical to socket's connect).
+ To select the proxy server use setproxy().
+ """
+ # Do a minimal input check first
+ if (type(destpair) in (list,tuple)==False) or (len(destpair)<2) or (type(destpair[0])!=str) or (type(destpair[1])!=int):
+ raise GeneralProxyError((5,_generalerrors[5]))
+ if self.__proxy[0] == PROXY_TYPE_SOCKS5:
+ if self.__proxy[2] != None:
+ portnum = self.__proxy[2]
+ else:
+ portnum = 1080
+ _orgsocket.connect(self,(self.__proxy[1],portnum))
+ self.__negotiatesocks5(destpair[0],destpair[1])
+ elif self.__proxy[0] == PROXY_TYPE_SOCKS4:
+ if self.__proxy[2] != None:
+ portnum = self.__proxy[2]
+ else:
+ portnum = 1080
+ _orgsocket.connect(self,(self.__proxy[1],portnum))
+ self.__negotiatesocks4(destpair[0],destpair[1])
+ elif self.__proxy[0] == PROXY_TYPE_HTTP:
+ if self.__proxy[2] != None:
+ portnum = self.__proxy[2]
+ else:
+ portnum = 8080
+ _orgsocket.connect(self,(self.__proxy[1],portnum))
+ self.__negotiatehttp(destpair[0],destpair[1])
+ elif self.__proxy[0] == None:
+ _orgsocket.connect(self,(destpair[0],destpair[1]))
+ else:
+ raise GeneralProxyError((4,_generalerrors[4]))
More information about the tor-commits
mailing list