* | * should pop to 'tr', not the first 'td'
- """
-
- nestingResetTriggers = self.NESTABLE_TAGS.get(name)
- isNestable = nestingResetTriggers != None
- isResetNesting = name in self.RESET_NESTING_TAGS
- popTo = None
- inclusive = True
- for i in xrange(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 is not None
- and p.name in nestingResetTriggers) \
- or (nestingResetTriggers is None and isResetNesting
- and p.name in self.RESET_NESTING_TAGS):
-
- #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([' %s="%s"' % (x, y) for x, y in 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] == '', i)
- if k == -1:
- k = len(self.rawdata)
- data = self.rawdata[i+9:k]
- j = k+3
- self._toStringSubclass(data, CData)
- else:
- try:
- j = sgmllib.SGMLParser.parse_declaration(self, i)
- except sgmllib.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 tag should
- implicitly close the previous tag within the same ,
- but not close a tag in another table.
-
- Blah Blah
- should be transformed into:
- Blah Blah
- but,
- BlahBlah
- should NOT be transformed into
- BlahBlah
-
- 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 'smartQuotesTo' not in kwargs:
- kwargs['smartQuotesTo'] = self.HTML_ENTITIES
- kwargs['isHTML'] = True
- BeautifulStoneSoup.__init__(self, *args, **kwargs)
-
- SELF_CLOSING_TAGS = buildTagMap(None,
- ('br' , 'hr', 'input', 'img', 'meta',
- 'spacer', 'link', 'frame', 'base', 'col'))
-
- PRESERVE_WHITESPACE_TAGS = set(['pre', 'textarea'])
-
- 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(r"((^|;)\s*charset=)([^;]*)", re.M)
-
- 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 xrange(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 (self.declaredHTMLEncoding is not None or
- self.originalEncoding == self.fromEncoding):
- # An HTML encoding was sniffed while converting
- # the document to Unicode, or an HTML encoding was
- # sniffed during a previous pass through the
- # document, or an encoding was specified
- # explicitly and it worked. Rewrite the meta tag.
- def rewrite(match):
- return match.group(1) + "%SOUP-ENCODING%"
- newAttr = self.CHARSET_RE.sub(rewrite, contentType)
- attrs[contentTypeIndex] = (attrs[contentTypeIndex][0],
- newAttr)
- tagNeedsEncodingSubstitution = True
- else:
- # This is our first pass through the document.
- # Go through it again with the encoding information.
- newCharset = match.group(3)
- if newCharset and newCharset != self.originalEncoding:
- self.declaredHTMLEncoding = newCharset
- self._feed(self.declaredHTMLEncoding)
- raise StopParsing
- pass
- 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
- ",
- "XSS::\\\";alert('XSS');//",
- "XSS::1' onerror=alert(String.fromCharCode(88,83,83))>",
- "XSS::var n=0;while(true){n++;}]]>",
- "XSS::",
- "XSS::javascript:alert(/XSS/)",
- "XSS::
|