diff options
Diffstat (limited to 'beautifulsoup')
| -rw-r--r-- | beautifulsoup/__init__.py | 375 | ||||
| -rw-r--r-- | beautifulsoup/dammit.py | 292 | ||||
| -rw-r--r-- | beautifulsoup/element.py | 851 | ||||
| -rw-r--r-- | beautifulsoup/testing.py | 155 | ||||
| -rw-r--r-- | beautifulsoup/util.py | 29 |
5 files changed, 1702 insertions, 0 deletions
diff --git a/beautifulsoup/__init__.py b/beautifulsoup/__init__.py new file mode 100644 index 0000000..8817164 --- /dev/null +++ b/beautifulsoup/__init__.py @@ -0,0 +1,375 @@ +"""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. + +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) 2004-2009, Leonard Richardson + +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are +met: + + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + + * Redistributions in binary form must reproduce the above + copyright notice, this list of conditions and the following + disclaimer in the documentation and/or other materials provided + with the distribution. + + * Neither the name of the the Beautiful Soup Consortium and All + Night Kosher Bakery nor the names of its contributors may be + used to endorse or promote products derived from this software + without specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR +CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, +EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, +PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR +PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF +LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING +NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS +SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE, DAMMIT. + +""" +from __future__ import generators + +__author__ = "Leonard Richardson (leonardr@segfault.org)" +__version__ = "4.0.0" +__copyright__ = "Copyright (c) 2004-2009 Leonard Richardson" +__license__ = "New-style BSD" + +__all__ = ['BeautifulSoup', + + # Stuff imported from other packages + 'Entities', + + 'BeautifulStoneSoup', + 'ICantBelieveItsBeautifulSoup'] + +import re + +from util import isList, isString, buildSet +from dammit import UnicodeDammit +from element import Entities, NavigableString, Tag + + +class BeautifulStoneSoup(Tag): + """ + This class defines the basic interface called by the tree builders. + + These methods will be called by the parser: + reset() + feed(markup) + + The tree builder may call these methods from its feed() implementation: + handle_starttag(name, attrs) # See note about return value + handle_endtag(name) + handle_data(data) # Appends to the current data node + endData(containerClass=NavigableString) # Ends the current data node + + No matter how complicated the underlying parser is, you should be + able to build a tree using 'start tag' events, 'end tag' events, + 'data' events, and "done with data" events. + + If you encounter a self-closing tag, call handle_starttag and then + handle_endtag, but note that the tag will not be displayed as a + self-closing tag unless you also have your builder's + isSelfClosingTag() implementation return True when passed the tag + name. + """ + ROOT_TAG_NAME = u'[document]' + + # Used to detect the charset in a META tag; see handleSpecialMetaTag + CHARSET_RE = re.compile("((^|;)\s*charset=)([^;]*)", re.M) + + # 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, } + + @classmethod + def default_builder(self): + from lxml import etree + from builder.lxml_builder import LXMLTreeBuilder + return LXMLTreeBuilder(parser_class=etree.XMLParser) + + def __init__(self, markup="", builder=None, parseOnlyThese=None, + fromEncoding=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.""" + + if builder is None: + builder = self.default_builder() + self.builder = builder + self.builder.soup = self + + self.parseOnlyThese = parseOnlyThese + self.fromEncoding = fromEncoding + + self.reset() + + if hasattr(markup, 'read'): # It's a file-type object. + markup = markup.read() + self.markup = markup + try: + self._feed(isHTML=self.builder.assume_html) + except StopParsing: + pass + self.markup = None # The markup can now be GCed. + self.builder.soup = None + self.builder = None # So can the builder. + + def _feed(self, inDocumentEncoding=None, isHTML=False): + # 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.builder.smart_quotes_to, isHTML=isHTML) + markup = dammit.unicode + self.originalEncoding = dammit.originalEncoding + self.declaredHTMLEncoding = dammit.declaredHTMLEncoding + self.builder.reset() + + self.builder.feed(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 reset(self): + Tag.__init__(self, self, self.builder, self.ROOT_TAG_NAME) + self.hidden = 1 + self.builder.reset() + self.currentData = [] + self.currentTag = None + self.tagStack = [] + 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 = u''.join(self.currentData) + if (currentData.translate(self.STRIP_ASCII_SPACES) == '' and + not buildSet([tag.name for tag in self.tagStack]).intersection( + self.builder.preserve_whitespace_tags)): + 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 handle_starttag(self, name, attrs): + """Push a start tag on to the stack. + + If this method returns None, the tag was rejected by the + SoupStrainer. You should proceed as if the tag had not occured + in the document. For instance, if this was a self-closing tag, + don't call handle_endtag. + """ + + #print "Start tag %s: %s" % (name, attrs) + self.endData() + + if (self.parseOnlyThese and len(self.tagStack) <= 1 + and (self.parseOnlyThese.text + or not self.parseOnlyThese.searchTag(name, attrs))): + return None + + containsSubstitutions = False + if name == 'meta' and self.builder.assume_html: + containsSubstitutions = self.handleSpecialMetaTag(attrs) + + tag = Tag(self, self.builder, name, attrs, self.currentTag, + self.previous) + tag.containsSubstitutions = containsSubstitutions + if self.previous: + self.previous.next = tag + self.previous = tag + self.pushTag(tag) + return tag + + def handle_endtag(self, name): + #print "End tag: " + name + self.endData() + self._popToTag(name) + + def handle_data(self, data): + self.currentData.append(data) + + def handleSpecialMetaTag(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. Neither lxml nor html5lib does + this, so the feature is still here.""" + httpEquiv = None + contentType = None + contentTypeIndex = None + tagNeedsEncodingSubstitution = False + + if isinstance(attrs, dict): + httpEquiv = attrs.get('http-equiv') + contentType = attrs.get('content') + else: + # XXX do we need this? + 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 (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) + if isinstance(attrs, dict): + attrs['content'] = newAttr + else: + 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 + return tagNeedsEncodingSubstitution + + +class BeautifulSoup(BeautifulStoneSoup): + """A convenience class for parsing HTML without creating a builder.""" + + @classmethod + def default_builder(self): + try: + from builder.html5_builder import HTML5TreeBuilder + return HTML5TreeBuilder() + except ImportError: + from builder.lxml_builder import LXMLTreeBuilder + return LXMLTreeBuilder() + + +class StopParsing(Exception): + pass + + +#By default, act as an HTML pretty-printer. +if __name__ == '__main__': + import sys + soup = BeautifulSoup(sys.stdin) + print soup.prettify() diff --git a/beautifulsoup/dammit.py b/beautifulsoup/dammit.py new file mode 100644 index 0000000..78bd4b2 --- /dev/null +++ b/beautifulsoup/dammit.py @@ -0,0 +1,292 @@ +"""Beautiful Soup 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's Beautiful Soup's job. +""" + +import codecs +import re +import types + +# Autodetects character encodings. +# Download from http://chardet.feedparser.org/ +try: + import chardet +# import chardet.constants +# chardet.constants._debug = 1 +except ImportError: + 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 ImportError: + pass +try: + import iconv_codec +except ImportError: + 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', isHTML=False): + self.declaredHTMLEncoding = None + self.markup, documentEncoding, sniffedEncoding = \ + self._detectEncoding(markup, isHTML) + 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, match): + """Changes a MS smart quote character to an XML or HTML + entity.""" + orig = match.group(1) + sub = self.MS_CHARS.get(orig) + if type(sub) == types.TupleType: + if self.smartQuotesTo == 'xml': + sub = '&#x'.encode() + sub[1].encode() + ';'.encode() + else: + sub = '&'.encode() + sub[0].encode() + ';'.encode() + else: + sub = sub.encode() + 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"): + smart_quotes_re = "([\x80-\x9f])" + smart_quotes_compiled = re.compile(smart_quotes_re) + markup = smart_quotes_compiled.sub(self._subMSChar, 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, isHTML=False): + """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 + except: + xml_encoding_match = None + xml_encoding_re = '^<\?.*encoding=[\'"](.*?)[\'"].*\?>'.encode() + xml_encoding_match = re.compile(xml_encoding_re).match(xml_data) + if not xml_encoding_match and isHTML: + meta_re = '<\s*meta[^>]+charset=([^>]*?)[;\'">]'.encode() + regexp = re.compile(meta_re, re.I) + xml_encoding_match = regexp.search(xml_data) + if xml_encoding_match is not None: + xml_encoding = xml_encoding_match.groups()[0].decode( + 'ascii').lower() + if isHTML: + self.declaredHTMLEncoding = xml_encoding + 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', ''),} diff --git a/beautifulsoup/element.py b/beautifulsoup/element.py new file mode 100644 index 0000000..7649b4c --- /dev/null +++ b/beautifulsoup/element.py @@ -0,0 +1,851 @@ +import re +import types +try: + from htmlentitydefs import name2codepoint +except ImportError: + name2codepoint = {} + +from util import isString, isList + +DEFAULT_OUTPUT_ENCODING = "utf-8" + +class Entities: + """A mixin class that knows about XML entities.""" + + HTML_ENTITIES = "html" + XML_ENTITIES = "xml" + XHTML_ENTITIES = "xhtml" + + 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) + +class PageElement: + """Contains the navigational information for some part of the page + (either a tag or a piece of text)""" + + def setup(self, parent=None, previous=None): + """Sets up the initial relations between this element and + other elements.""" + self.parent = parent + self.previous = previous + self.next = None + self.previousSibling = None + self.nextSibling = None + if self.parent and self.parent.contents: + self.previousSibling = self.parent.contents[-1] + self.previousSibling.nextSibling = self + + def replaceWith(self, replaceWith): + oldParent = self.parent + myIndex = self.parent.contents.index(self) + if hasattr(replaceWith, 'parent') and replaceWith.parent == self.parent: + # We're replacing this element with one of its siblings. + index = self.parent.contents.index(replaceWith) + if index and index < myIndex: + # Furthermore, it comes before this element. That + # means that when we extract it, the index of this + # element will change. + myIndex = myIndex - 1 + self.extract() + oldParent.insert(myIndex, replaceWith) + + def extract(self): + """Destructively rips this element out of the tree.""" + if self.parent: + try: + self.parent.contents.remove(self) + except ValueError: + pass + + #Find the two elements that would be next to each other if + #this element (and any children) hadn't been parsed. Connect + #the two. + lastChild = self._lastRecursiveChild() + nextElement = lastChild.next + + if self.previous: + self.previous.next = nextElement + if nextElement: + nextElement.previous = self.previous + self.previous = None + lastChild.next = None + + self.parent = None + if self.previousSibling: + self.previousSibling.nextSibling = self.nextSibling + if self.nextSibling: + self.nextSibling.previousSibling = self.previousSibling + self.previousSibling = self.nextSibling = None + return self + + def _lastRecursiveChild(self): + "Finds the last element beneath this object to be parsed." + lastChild = self + while hasattr(lastChild, 'contents') and lastChild.contents: + lastChild = lastChild.contents[-1] + return lastChild + + def insert(self, position, newChild): + if (isinstance(newChild, basestring) + or isinstance(newChild, unicode)) \ + and not isinstance(newChild, NavigableString): + newChild = NavigableString(newChild) + + position = min(position, len(self.contents)) + if hasattr(newChild, 'parent') and newChild.parent != None: + # We're 'inserting' an element that's already one + # of this object's children. + if newChild.parent == self: + index = self.find(newChild) + if index and index < position: + # Furthermore we're moving it further down the + # list of this object's children. That means that + # when we extract this element, our target index + # will jump down one. + position = position - 1 + newChild.extract() + + newChild.parent = self + previousChild = None + if position == 0: + newChild.previousSibling = None + newChild.previous = self + else: + previousChild = self.contents[position-1] + newChild.previousSibling = previousChild + newChild.previousSibling.nextSibling = newChild + newChild.previous = previousChild._lastRecursiveChild() + if newChild.previous: + newChild.previous.next = newChild + + newChildsLastElement = newChild._lastRecursiveChild() + + if position >= len(self.contents): + newChild.nextSibling = None + + parent = self + parentsNextSibling = None + while not parentsNextSibling: + parentsNextSibling = parent.nextSibling + parent = parent.parent + if not parent: # This is the last element in the document. + break + if parentsNextSibling: + newChildsLastElement.next = parentsNextSibling + else: + newChildsLastElement.next = None + else: + nextChild = self.contents[position] + newChild.nextSibling = nextChild + if newChild.nextSibling: + newChild.nextSibling.previousSibling = newChild + newChildsLastElement.next = nextChild + + if newChildsLastElement.next: + newChildsLastElement.next.previous = newChildsLastElement + self.contents.insert(position, newChild) + + def append(self, tag): + """Appends the given tag to the contents of this tag.""" + self.insert(len(self.contents), tag) + + def findNext(self, name=None, attrs={}, text=None, **kwargs): + """Returns the first item that matches the given criteria and + appears after this Tag in the document.""" + return self._findOne(self.findAllNext, name, attrs, text, **kwargs) + + def findAllNext(self, name=None, attrs={}, text=None, limit=None, + **kwargs): + """Returns all items that match the given criteria and appear + after this Tag in the document.""" + return self._findAll(name, attrs, text, limit, self.nextGenerator, + **kwargs) + + def findNextSibling(self, name=None, attrs={}, text=None, **kwargs): + """Returns the closest sibling to this Tag that matches the + given criteria and appears after this Tag in the document.""" + return self._findOne(self.findNextSiblings, name, attrs, text, + **kwargs) + + def findNextSiblings(self, name=None, attrs={}, text=None, limit=None, + **kwargs): + """Returns the siblings of this Tag that match the given + criteria and appear after this Tag in the document.""" + return self._findAll(name, attrs, text, limit, + self.nextSiblingGenerator, **kwargs) + fetchNextSiblings = findNextSiblings # Compatibility with pre-3.x + + def findPrevious(self, name=None, attrs={}, text=None, **kwargs): + """Returns the first item that matches the given criteria and + appears before this Tag in the document.""" + return self._findOne(self.findAllPrevious, name, attrs, text, **kwargs) + + def findAllPrevious(self, name=None, attrs={}, text=None, limit=None, + **kwargs): + """Returns all items that match the given criteria and appear + before this Tag in the document.""" + return self._findAll(name, attrs, text, limit, self.previousGenerator, + **kwargs) + fetchPrevious = findAllPrevious # Compatibility with pre-3.x + + def findPreviousSibling(self, name=None, attrs={}, text=None, **kwargs): + """Returns the closest sibling to this Tag that matches the + given criteria and appears before this Tag in the document.""" + return self._findOne(self.findPreviousSiblings, name, attrs, text, + **kwargs) + + def findPreviousSiblings(self, name=None, attrs={}, text=None, + limit=None, **kwargs): + """Returns the siblings of this Tag that match the given + criteria and appear before this Tag in the document.""" + return self._findAll(name, attrs, text, limit, + self.previousSiblingGenerator, **kwargs) + fetchPreviousSiblings = findPreviousSiblings # Compatibility with pre-3.x + + def findParent(self, name=None, attrs={}, **kwargs): + """Returns the closest parent of this Tag that matches the given + criteria.""" + # NOTE: We can't use _findOne because findParents takes a different + # set of arguments. + r = None + l = self.findParents(name, attrs, 1) + if l: + r = l[0] + return r + + def findParents(self, name=None, attrs={}, limit=None, **kwargs): + """Returns the parents of this Tag that match the given + criteria.""" + + return self._findAll(name, attrs, None, limit, self.parentGenerator, + **kwargs) + fetchParents = findParents # Compatibility with pre-3.x + + #These methods do the real heavy lifting. + + def _findOne(self, method, name, attrs, text, **kwargs): + r = None + l = method(name, attrs, text, 1, **kwargs) + if l: + r = l[0] + return r + + def _findAll(self, name, attrs, text, limit, generator, **kwargs): + "Iterates over a generator looking for things that match." + + if isinstance(name, SoupStrainer): + strainer = name + else: + # Build a SoupStrainer + strainer = SoupStrainer(name, attrs, text, **kwargs) + results = ResultSet(strainer) + g = generator() + while True: + try: + i = g.next() + except StopIteration: + break + if i: + found = strainer.search(i) + if found: + results.append(found) + if limit and len(results) >= limit: + break + return results + + #These Generators can be used to navigate starting from both + #NavigableStrings and Tags. + def nextGenerator(self): + i = self + while i: + i = i.next + yield i + + def nextSiblingGenerator(self): + i = self + while i: + i = i.nextSibling + yield i + + def previousGenerator(self): + i = self + while i: + i = i.previous + yield i + + def previousSiblingGenerator(self): + i = self + while i: + i = i.previousSibling + yield i + + def parentGenerator(self): + i = self + while i: + i = i.parent + yield i + + # Utility methods + def substituteEncoding(self, str, encoding=None): + encoding = encoding or "utf-8" + return str.replace("%SOUP-ENCODING%", encoding) + + def toEncoding(self, s, encoding=None): + """Encodes an object to a string in some encoding, or to Unicode. + .""" + if isinstance(s, unicode): + if encoding: + s = s.encode(encoding) + elif isinstance(s, str): + if encoding: + s = s.encode(encoding) + else: + s = unicode(s) + else: + if encoding: + s = self.toEncoding(str(s), encoding) + else: + s = unicode(s) + return s + +class NavigableString(unicode, PageElement): + + def __new__(cls, value): + """Create a new NavigableString. + + When unpickling a NavigableString, this method is called with + the string in DEFAULT_OUTPUT_ENCODING. That encoding needs to be + passed in to the superclass's __new__ or the superclass won't know + how to handle non-ASCII characters. + """ + if isinstance(value, unicode): + return unicode.__new__(cls, value) + return unicode.__new__(cls, value, DEFAULT_OUTPUT_ENCODING) + + def __getnewargs__(self): + return (unicode(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 encode(self, encoding=DEFAULT_OUTPUT_ENCODING): + return self.decode().encode(encoding) + + def decodeGivenEventualEncoding(self, eventualEncoding): + return self + +class CData(NavigableString): + + def decodeGivenEventualEncoding(self, eventualEncoding): + return u'<![CDATA[' + self + u']]>' + +class ProcessingInstruction(NavigableString): + + def decodeGivenEventualEncoding(self, eventualEncoding): + output = self + if u'%SOUP-ENCODING%' in output: + output = self.substituteEncoding(output, eventualEncoding) + return u'<?' + output + u'?>' + +class Comment(NavigableString): + def decodeGivenEventualEncoding(self, eventualEncoding): + return u'<!--' + self + u'-->' + +class Declaration(NavigableString): + def decodeGivenEventualEncoding(self, eventualEncoding): + return u'<!' + self + u'>' + +class Tag(PageElement, Entities): + + """Represents a found HTML tag with its attributes and contents.""" + + def _convertEntities(self, builder, 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 builder.convert_html_entities and x in name2codepoint: + return unichr(name2codepoint[x]) + elif x in self.XML_ENTITIES_TO_SPECIAL_CHARS: + if builder.convert_xml_entities: + 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, builder, 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.name = name + self.isSelfClosing = builder.isSelfClosingTag(name) + if attrs == None: + attrs = [] + if isinstance(attrs, types.DictType): + self.attrMap = attrs + self.attrs = attrs + self.contents = [] + self.setup(parent, previous) + self.hidden = False + self.containsSubstitutions = False + self.escapeUnrecognizedEntities = parser.escapeUnrecognizedEntities + + # Convert any HTML, XML, or numeric entities in the attribute values. + convert_one = lambda x: self._convertEntities(parser.builder, x) + def convert(kval): + k, val = kval + if val is None: + return kval + return (k, re.sub("&(#\d+|#x[0-9a-fA-F]+|\w+);", convert_one, val)) + if isinstance(attrs, types.DictType): + self.attrs = [convert(kv) for kv in attrs.items()] + else: + self.attrs = map(convert, 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.decode(eventualEncoding=encoding) + + 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 __unicode__(self): + return self.decode() + + def __str__(self): + return self.encode() + + def encode(self, encoding=DEFAULT_OUTPUT_ENCODING, + prettyPrint=False, indentLevel=0): + return self.decode(prettyPrint, indentLevel, encoding).encode(encoding) + + def decode(self, prettyPrint=False, indentLevel=0, + eventualEncoding=DEFAULT_OUTPUT_ENCODING): + """Returns a string or Unicode representation of this tag and + its contents. To get Unicode, pass None for encoding.""" + + attrs = [] + if self.attrs: + for key, val in self.attrs: + fmt = '%s="%s"' + if isString(val): + if (self.containsSubstitutions + and eventualEncoding is not None + and '%SOUP-ENCODING%' in val): + val = self.substituteEncoding(val, eventualEncoding) + + # 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) + if val is None: + # Handle boolean attributes. + decoded = key + else: + decoded = fmt % (key, val) + attrs.append(decoded) + close = '' + closeTag = '' + if self.isSelfClosing: + close = ' /' + else: + closeTag = '</%s>' % self.name + + indentTag, indentContents = 0, 0 + if prettyPrint: + indentTag = indentLevel + space = (' ' * (indentTag-1)) + indentContents = indentTag + 1 + contents = self.decodeContents(prettyPrint, indentContents, + eventualEncoding) + if self.hidden: + s = contents + else: + s = [] + attributeString = '' + if attrs: + attributeString = ' ' + ' '.join(attrs) + if prettyPrint: + s.append(space) + s.append('<%s%s%s>' % (self.name, attributeString, close)) + if prettyPrint: + s.append("\n") + s.append(contents) + if prettyPrint and contents and contents[-1] != "\n": + s.append("\n") + if prettyPrint and closeTag: + s.append(space) + s.append(closeTag) + if prettyPrint and closeTag and self.nextSibling: + s.append("\n") + s = ''.join(s) + return s + + def decompose(self): + """Recursively destroys the contents of this tree.""" + contents = [i for i in self.contents] + for i in contents: + if isinstance(i, Tag): + i.decompose() + else: + i.extract() + self.extract() + + def prettify(self, encoding=DEFAULT_OUTPUT_ENCODING): + return self.encode(encoding, True) + + def encodeContents(self, encoding=DEFAULT_OUTPUT_ENCODING, + prettyPrint=False, indentLevel=0): + return self.decodeContents(prettyPrint, indentLevel).encode(encoding) + + def decodeContents(self, prettyPrint=False, indentLevel=0, + eventualEncoding=DEFAULT_OUTPUT_ENCODING): + """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.decodeGivenEventualEncoding(eventualEncoding) + elif isinstance(c, Tag): + s.append(c.decode(prettyPrint, indentLevel, eventualEncoding)) + 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 + + #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): + if not len(self.contents): + raise StopIteration + stopNode = self._lastRecursiveChild().next + current = self.contents[0] + while current is not stopNode: + yield current + current = current.next + +# 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 is not None 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) + and (markup is not None or not isString(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 diff --git a/beautifulsoup/testing.py b/beautifulsoup/testing.py new file mode 100644 index 0000000..eea14f0 --- /dev/null +++ b/beautifulsoup/testing.py @@ -0,0 +1,155 @@ +"""Helper classes for tests.""" + +import unittest +from beautifulsoup import BeautifulSoup +from beautifulsoup.element import Comment, SoupStrainer +from beautifulsoup.builder.lxml_builder import LXMLTreeBuilder + +class SoupTest(unittest.TestCase): + + @property + def default_builder(self): + return LXMLTreeBuilder() + + def soup(self, markup, **kwargs): + """Build a Beautiful Soup object from markup.""" + return BeautifulSoup(markup, builder=self.default_builder, **kwargs) + + def document_for(self, markup): + """Turn an HTML fragment into a document. + + The details depend on the builder. + """ + return self.default_builder.test_fragment_to_document(markup) + + def assertSoupEquals(self, to_parse, compare_parsed_to=None): + builder = self.default_builder + obj = BeautifulSoup(to_parse, builder=builder) + if compare_parsed_to is None: + compare_parsed_to = to_parse + + self.assertEquals(obj.decode(), self.document_for(compare_parsed_to)) + + + +class BuilderSmokeTest(SoupTest): + """A generic smoke test for tree builders. + + Subclasses of this test ensure that all of Beautiful Soup's tree + builders generate more or less the same trees. It's okay for trees + to differ, especially when given invalid markup--just override the + appropriate test method to demonstrate how one tree builder + differs from others. + """ + + def test_bare_string(self): + # A bare string is turned into some kind of HTML document or + # fragment recognizable as the original string. + # + # In this case, lxml puts a <p> tag around the bare string. + self.assertSoupEquals( + "A bare string", "<p>A bare string</p>") + + def test_mixed_case_tags(self): + # Mixed-case tags are folded to lowercase. + self.assertSoupEquals( + "<a><B><Cd><EFG></efg></CD></b></A>", + "<a><b><cd><efg></efg></cd></b></a>") + + def test_self_closing(self): + # HTML's self-closing tags are recognized as such. + self.assertSoupEquals( + "<p>A <meta> tag</p>", "<p>A <meta /> tag</p>") + + self.assertSoupEquals( + "<p>Foo<br/>bar</p>", "<p>Foo<br />bar</p>") + + def test_comment(self): + # Comments are represented as Comment objects. + markup = "<p>foo<!--foobar-->baz</p>" + self.assertSoupEquals(markup) + + soup = self.soup(markup) + comment = soup.find(text="foobar") + self.assertEquals(comment.__class__, Comment) + + def test_nested_inline_elements(self): + # Inline tags can be nested indefinitely. + b_tag = "<b>Inside a B tag</b>" + self.assertSoupEquals(b_tag) + + nested_b_tag = "<p>A <i>nested <b>tag</b></i></p>" + self.assertSoupEquals(nested_b_tag) + + double_nested_b_tag = "<p>A <a>doubly <i>nested <b>tag</b></i></a></p>" + self.assertSoupEquals(nested_b_tag) + + def test_nested_block_level_elements(self): + soup = self.soup('<blockquote><p><b>Foo</b></p></blockquote>') + blockquote = soup.blockquote + self.assertEqual(blockquote.p.b.string, 'Foo') + self.assertEqual(blockquote.b.string, 'Foo') + + def test_collapsed_whitespace(self): + """In most tags, whitespace is collapsed.""" + self.assertSoupEquals("<p> </p>", "<p> </p>") + + def test_preserved_whitespace_in_pre_and_textarea(self): + """In <pre> and <textarea> tags, whitespace is preserved.""" + self.assertSoupEquals("<pre> </pre>") + self.assertSoupEquals("<textarea> woo </textarea>") + + def test_single_quote_attribute_values_become_double_quotes(self): + self.assertSoupEquals("<foo attr='bar'></foo>", + '<foo attr="bar"></foo>') + + def test_attribute_values_with_nested_quotes_are_left_alone(self): + text = """<foo attr='bar "brawls" happen'>a</foo>""" + self.assertSoupEquals(text) + + def test_attribute_values_with_double_nested_quotes_get_quoted(self): + text = """<foo attr='bar "brawls" happen'>a</foo>""" + soup = self.soup(text) + soup.foo['attr'] = 'Brawls happen at "Bob\'s Bar"' + self.assertSoupEquals( + soup.foo.decode(), + """<foo attr='Brawls happen at "Bob&squot;s Bar"'>a</foo>""") + + def test_ampersand_in_attribute_value_gets_quoted(self): + self.assertSoupEquals('<this is="really messed up & stuff"></this>', + '<this is="really messed up & stuff"></this>') + + +class BuilderInvalidMarkupSmokeTest(SoupTest): + """Tests of invalid markup. + + These are very likely to give different results for different tree + builders. It's not required that a tree builder handle invalid + markup at all. + """ + + def test_unclosed_block_level_elements(self): + # Unclosed block-level elements should be closed. + self.assertSoupEquals( + '<blockquote><p><b>Foo</blockquote><p>Bar', + '<blockquote><p><b>Foo</b></p></blockquote><p>Bar</p>') + + def test_fake_self_closing_tag(self): + # If a self-closing tag presents as a normal tag, the 'open' + # tag is treated as an instance of the self-closing tag and + # the 'close' tag is ignored. + self.assertSoupEquals( + "<item><link>http://foo.com/</link></item>", + "<item><link />http://foo.com/</item>") + + def test_boolean_attribute_with_no_value_gets_empty_value(self): + soup = self.soup("<table><td nowrap>foo</td></table>") + self.assertEquals(soup.table.td['nowrap'], '') + + def test_incorrectly_nested_tables(self): + self.assertSoupEquals( + '<table><tr><table><tr id="nested">', + '<table><tr><table><tr id="nested"></tr></table></tr></table>') + + + diff --git a/beautifulsoup/util.py b/beautifulsoup/util.py new file mode 100644 index 0000000..693a7e2 --- /dev/null +++ b/beautifulsoup/util.py @@ -0,0 +1,29 @@ +# Helper functions and mixin classes for Beautiful Soup + +import types +try: + set +except NameError: + from sets import Set as set + +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__') and not isString(l)) + 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 buildSet(args=None): + """Turns a list or a string into a set.""" + if isinstance(args, str): + return set([args]) + if args is None: + return set() + return set(args) |