From 1c4a73ca019df936d06128443cdc67bddefdda21 Mon Sep 17 00:00:00 2001 From: Ethan Dalool Date: Sat, 3 Aug 2019 00:24:52 -0700 Subject: [PATCH] Add cleanerupper. --- cleanerupper/README.md | 8 + cleanerupper/bs4/__init__.py | 529 +++++++ cleanerupper/bs4/builder/__init__.py | 333 +++++ cleanerupper/bs4/builder/_html5lib.py | 426 ++++++ cleanerupper/bs4/builder/_htmlparser.py | 314 ++++ cleanerupper/bs4/builder/_lxml.py | 258 ++++ cleanerupper/bs4/dammit.py | 842 +++++++++++ cleanerupper/bs4/diagnose.py | 219 +++ cleanerupper/bs4/element.py | 1808 +++++++++++++++++++++++ cleanerupper/bs4/testing.py | 770 ++++++++++ cleanerupper/plugin.py | 286 ++++ cleanerupper/plugin.xml | 11 + 12 files changed, 5804 insertions(+) create mode 100644 cleanerupper/README.md create mode 100644 cleanerupper/bs4/__init__.py create mode 100644 cleanerupper/bs4/builder/__init__.py create mode 100644 cleanerupper/bs4/builder/_html5lib.py create mode 100644 cleanerupper/bs4/builder/_htmlparser.py create mode 100644 cleanerupper/bs4/builder/_lxml.py create mode 100644 cleanerupper/bs4/dammit.py create mode 100644 cleanerupper/bs4/diagnose.py create mode 100644 cleanerupper/bs4/element.py create mode 100644 cleanerupper/bs4/testing.py create mode 100644 cleanerupper/plugin.py create mode 100644 cleanerupper/plugin.xml diff --git a/cleanerupper/README.md b/cleanerupper/README.md new file mode 100644 index 0000000..04c45fc --- /dev/null +++ b/cleanerupper/README.md @@ -0,0 +1,8 @@ +The Great Cleaner Upper +======================= + +This is my plugin for aggressively cleaning the absolute heck out of books. You may enjoy this plugin if your books consist almost entirely of plain text with minimal formatting elements and you don't mind removing most/all CSS. + +Because this plugin removes `calibre` CSS classes, some of which may be used for simulating bold, italic, headers, blockquotes, etc., you should use find-and-replace to turn those elements into real ``, ``, `

`, `
`, etc. tags before running the cleaner. + +BeautifulSoup comes from here: https://www.crummy.com/software/BeautifulSoup diff --git a/cleanerupper/bs4/__init__.py b/cleanerupper/bs4/__init__.py new file mode 100644 index 0000000..62e9e5d --- /dev/null +++ b/cleanerupper/bs4/__init__.py @@ -0,0 +1,529 @@ +"""Beautiful Soup +Elixir and Tonic +"The Screen-Scraper's Friend" +http://www.crummy.com/software/BeautifulSoup/ + +Beautiful Soup uses a pluggable XML or HTML parser to parse a +(possibly invalid) document into a tree representation. Beautiful Soup +provides methods and Pythonic idioms that make it easy to navigate, +search, and modify the parse tree. + +Beautiful Soup works with Python 2.7 and up. It works better if lxml +and/or html5lib is installed. + +For more than you ever wanted to know about Beautiful Soup, see the +documentation: +http://www.crummy.com/software/BeautifulSoup/bs4/doc/ + +""" + +# Use of this source code is governed by a BSD-style license that can be +# found in the LICENSE file. + +__author__ = "Leonard Richardson (leonardr@segfault.org)" +__version__ = "4.6.0" +__copyright__ = "Copyright (c) 2004-2017 Leonard Richardson" +__license__ = "MIT" + +__all__ = ['BeautifulSoup'] + +import os +import re +import traceback +import warnings + +from .builder import builder_registry, ParserRejectedMarkup +from .dammit import UnicodeDammit +from .element import ( + CData, + Comment, + DEFAULT_OUTPUT_ENCODING, + Declaration, + Doctype, + NavigableString, + PageElement, + ProcessingInstruction, + ResultSet, + SoupStrainer, + Tag, + ) + +# The very first thing we do is give a useful error if someone is +# running this code under Python 3 without converting it. +'You are trying to run the Python 2 version of Beautiful Soup under Python 3. This will not work.'!='You need to convert the code, either by installing it (`python setup.py install`) or by running 2to3 (`2to3 -w bs4`).' + +class BeautifulSoup(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 an empty-element tag (aka a self-closing tag, + like HTML's
tag), call handle_starttag and then + handle_endtag. + """ + ROOT_TAG_NAME = '[document]' + + # If the end-user gives no indication which tree builder they + # want, look for one with these features. + DEFAULT_BUILDER_FEATURES = ['html', 'fast'] + + ASCII_SPACES = '\x20\x0a\x09\x0c\x0d' + + NO_PARSER_SPECIFIED_WARNING = "No parser was explicitly specified, so I'm using the best available %(markup_type)s parser for this system (\"%(parser)s\"). This usually isn't a problem, but if you run this code on another system, or in a different virtual environment, it may use a different parser and behave differently.\n\nThe code that caused this warning is on line %(line_number)s of the file %(filename)s. To get rid of this warning, change code that looks like this:\n\n BeautifulSoup(YOUR_MARKUP})\n\nto this:\n\n BeautifulSoup(YOUR_MARKUP, \"%(parser)s\")\n" + + def __init__(self, markup="", features=None, builder=None, + parse_only=None, from_encoding=None, exclude_encodings=None, + **kwargs): + """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 'convertEntities' in kwargs: + warnings.warn( + "BS4 does not respect the convertEntities argument to the " + "BeautifulSoup constructor. Entities are always converted " + "to Unicode characters.") + + if 'markupMassage' in kwargs: + del kwargs['markupMassage'] + warnings.warn( + "BS4 does not respect the markupMassage argument to the " + "BeautifulSoup constructor. The tree builder is responsible " + "for any necessary markup massage.") + + if 'smartQuotesTo' in kwargs: + del kwargs['smartQuotesTo'] + warnings.warn( + "BS4 does not respect the smartQuotesTo argument to the " + "BeautifulSoup constructor. Smart quotes are always converted " + "to Unicode characters.") + + if 'selfClosingTags' in kwargs: + del kwargs['selfClosingTags'] + warnings.warn( + "BS4 does not respect the selfClosingTags argument to the " + "BeautifulSoup constructor. The tree builder is responsible " + "for understanding self-closing tags.") + + if 'isHTML' in kwargs: + del kwargs['isHTML'] + warnings.warn( + "BS4 does not respect the isHTML argument to the " + "BeautifulSoup constructor. Suggest you use " + "features='lxml' for HTML and features='lxml-xml' for " + "XML.") + + def deprecated_argument(old_name, new_name): + if old_name in kwargs: + warnings.warn( + 'The "%s" argument to the BeautifulSoup constructor ' + 'has been renamed to "%s."' % (old_name, new_name)) + value = kwargs[old_name] + del kwargs[old_name] + return value + return None + + parse_only = parse_only or deprecated_argument( + "parseOnlyThese", "parse_only") + + from_encoding = from_encoding or deprecated_argument( + "fromEncoding", "from_encoding") + + if from_encoding and isinstance(markup, str): + warnings.warn("You provided Unicode markup but also provided a value for from_encoding. Your from_encoding will be ignored.") + from_encoding = None + + if len(kwargs) > 0: + arg = list(kwargs.keys()).pop() + raise TypeError( + "__init__() got an unexpected keyword argument '%s'" % arg) + + if builder is None: + original_features = features + if isinstance(features, str): + features = [features] + if features is None or len(features) == 0: + features = self.DEFAULT_BUILDER_FEATURES + builder_class = builder_registry.lookup(*features) + if builder_class is None: + raise FeatureNotFound( + "Couldn't find a tree builder with the features you " + "requested: %s. Do you need to install a parser library?" + % ",".join(features)) + builder = builder_class() + if not (original_features == builder.NAME or + original_features in builder.ALTERNATE_NAMES): + if builder.is_xml: + markup_type = "XML" + else: + markup_type = "HTML" + + caller = traceback.extract_stack()[0] + filename = caller[0] + line_number = caller[1] + warnings.warn(self.NO_PARSER_SPECIFIED_WARNING % dict( + filename=filename, + line_number=line_number, + parser=builder.NAME, + markup_type=markup_type)) + + self.builder = builder + self.is_xml = builder.is_xml + self.known_xml = self.is_xml + self.builder.soup = self + + self.parse_only = parse_only + + if hasattr(markup, 'read'): # It's a file-type object. + markup = markup.read() + elif len(markup) <= 256 and ( + (isinstance(markup, bytes) and not b'<' in markup) + or (isinstance(markup, str) and not '<' in markup) + ): + # Print out warnings for a couple beginner problems + # involving passing non-markup to Beautiful Soup. + # Beautiful Soup will still parse the input as markup, + # just in case that's what the user really wants. + if (isinstance(markup, str) + and not os.path.supports_unicode_filenames): + possible_filename = markup.encode("utf8") + else: + possible_filename = markup + is_file = False + try: + is_file = os.path.exists(possible_filename) + except Exception as e: + # This is almost certainly a problem involving + # characters not valid in filenames on this + # system. Just let it go. + pass + if is_file: + if isinstance(markup, str): + markup = markup.encode("utf8") + warnings.warn( + '"%s" looks like a filename, not markup. You should' + ' probably open this file and pass the filehandle into' + ' Beautiful Soup.' % markup) + self._check_markup_is_url(markup) + + for (self.markup, self.original_encoding, self.declared_html_encoding, + self.contains_replacement_characters) in ( + self.builder.prepare_markup( + markup, from_encoding, exclude_encodings=exclude_encodings)): + self.reset() + try: + self._feed() + break + except ParserRejectedMarkup: + pass + + # Clear out the markup and remove the builder's circular + # reference to this object. + self.markup = None + self.builder.soup = None + + def __copy__(self): + copy = type(self)( + self.encode('utf-8'), builder=self.builder, from_encoding='utf-8' + ) + + # Although we encoded the tree to UTF-8, that may not have + # been the encoding of the original markup. Set the copy's + # .original_encoding to reflect the original object's + # .original_encoding. + copy.original_encoding = self.original_encoding + return copy + + def __getstate__(self): + # Frequently a tree builder can't be pickled. + d = dict(self.__dict__) + if 'builder' in d and not self.builder.picklable: + d['builder'] = None + return d + + @staticmethod + def _check_markup_is_url(markup): + """ + Check if markup looks like it's actually a url and raise a warning + if so. Markup can be unicode or str (py2) / bytes (py3). + """ + if isinstance(markup, bytes): + space = b' ' + cant_start_with = (b"http:", b"https:") + elif isinstance(markup, str): + space = ' ' + cant_start_with = ("http:", "https:") + else: + return + + if any(markup.startswith(prefix) for prefix in cant_start_with): + if not space in markup: + if isinstance(markup, bytes): + decoded_markup = markup.decode('utf-8', 'replace') + else: + decoded_markup = markup + warnings.warn( + '"%s" looks like a URL. Beautiful Soup is not an' + ' HTTP client. You should probably use an HTTP client like' + ' requests to get the document behind the URL, and feed' + ' that document to Beautiful Soup.' % decoded_markup + ) + + def _feed(self): + # Convert the document to Unicode. + self.builder.reset() + + self.builder.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 reset(self): + Tag.__init__(self, self, self.builder, self.ROOT_TAG_NAME) + self.hidden = 1 + self.builder.reset() + self.current_data = [] + self.currentTag = None + self.tagStack = [] + self.preserve_whitespace_tag_stack = [] + self.pushTag(self) + + def new_tag(self, name, namespace=None, nsprefix=None, **attrs): + """Create a new tag associated with this soup.""" + return Tag(None, self.builder, name, namespace, nsprefix, attrs) + + def new_string(self, s, subclass=NavigableString): + """Create a new NavigableString associated with this soup.""" + return subclass(s) + + def insert_before(self, successor): + raise NotImplementedError("BeautifulSoup objects don't support insert_before().") + + def insert_after(self, successor): + raise NotImplementedError("BeautifulSoup objects don't support insert_after().") + + def popTag(self): + tag = self.tagStack.pop() + if self.preserve_whitespace_tag_stack and tag == self.preserve_whitespace_tag_stack[-1]: + self.preserve_whitespace_tag_stack.pop() + #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] + if tag.name in self.builder.preserve_whitespace_tags: + self.preserve_whitespace_tag_stack.append(tag) + + def endData(self, containerClass=NavigableString): + if self.current_data: + current_data = ''.join(self.current_data) + # If whitespace is not preserved, and this string contains + # nothing but ASCII spaces, replace it with a single space + # or newline. + if not self.preserve_whitespace_tag_stack: + strippable = True + for i in current_data: + if i not in self.ASCII_SPACES: + strippable = False + break + if strippable: + if '\n' in current_data: + current_data = '\n' + else: + current_data = ' ' + + # Reset the data collector. + self.current_data = [] + + # Should we add this string to the tree at all? + if self.parse_only and len(self.tagStack) <= 1 and \ + (not self.parse_only.text or \ + not self.parse_only.search(current_data)): + return + + o = containerClass(current_data) + self.object_was_parsed(o) + + def object_was_parsed(self, o, parent=None, most_recent_element=None): + """Add an object to the parse tree.""" + parent = parent or self.currentTag + previous_element = most_recent_element or self._most_recent_element + + next_element = previous_sibling = next_sibling = None + if isinstance(o, Tag): + next_element = o.next_element + next_sibling = o.next_sibling + previous_sibling = o.previous_sibling + if not previous_element: + previous_element = o.previous_element + + o.setup(parent, previous_element, next_element, previous_sibling, next_sibling) + + self._most_recent_element = o + parent.contents.append(o) + + if parent.next_sibling: + # This node is being inserted into an element that has + # already been parsed. Deal with any dangling references. + index = len(parent.contents)-1 + while index >= 0: + if parent.contents[index] is o: + break + index -= 1 + else: + raise ValueError( + "Error building tree: supposedly %r was inserted " + "into %r after the fact, but I don't see it!" % ( + o, parent + ) + ) + if index == 0: + previous_element = parent + previous_sibling = None + else: + previous_element = previous_sibling = parent.contents[index-1] + if index == len(parent.contents)-1: + next_element = parent.next_sibling + next_sibling = None + else: + next_element = next_sibling = parent.contents[index+1] + + o.previous_element = previous_element + if previous_element: + previous_element.next_element = o + o.next_element = next_element + if next_element: + next_element.previous_element = o + o.next_sibling = next_sibling + if next_sibling: + next_sibling.previous_sibling = o + o.previous_sibling = previous_sibling + if previous_sibling: + previous_sibling.next_sibling = o + + def _popToTag(self, name, nsprefix=None, 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: + # The BeautifulSoup object itself can never be popped. + return + + most_recently_popped = None + + stack_size = len(self.tagStack) + for i in range(stack_size - 1, 0, -1): + t = self.tagStack[i] + if (name == t.name and nsprefix == t.prefix): + if inclusivePop: + most_recently_popped = self.popTag() + break + most_recently_popped = self.popTag() + + return most_recently_popped + + def handle_starttag(self, name, namespace, nsprefix, 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 occurred + 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.parse_only and len(self.tagStack) <= 1 + and (self.parse_only.text + or not self.parse_only.search_tag(name, attrs))): + return None + + tag = Tag(self, self.builder, name, namespace, nsprefix, attrs, + self.currentTag, self._most_recent_element) + if tag is None: + return tag + if self._most_recent_element: + self._most_recent_element.next_element = tag + self._most_recent_element = tag + self.pushTag(tag) + return tag + + def handle_endtag(self, name, nsprefix=None): + #print "End tag: " + name + self.endData() + self._popToTag(name, nsprefix) + + def handle_data(self, data): + self.current_data.append(data) + + def decode(self, pretty_print=False, + eventual_encoding=DEFAULT_OUTPUT_ENCODING, + formatter="minimal"): + """Returns a string or Unicode representation of this document. + To get Unicode, pass None for encoding.""" + + if self.is_xml: + # Print the XML declaration + encoding_part = '' + if eventual_encoding != None: + encoding_part = ' encoding="%s"' % eventual_encoding + prefix = '\n' % encoding_part + else: + prefix = '' + if not pretty_print: + indent_level = None + else: + indent_level = 0 + return prefix + super(BeautifulSoup, self).decode( + indent_level, eventual_encoding, formatter) + +# Alias to make it easier to type import: 'from bs4 import _soup' +_s = BeautifulSoup +_soup = BeautifulSoup + +class BeautifulStoneSoup(BeautifulSoup): + """Deprecated interface to an XML parser.""" + + def __init__(self, *args, **kwargs): + kwargs['features'] = 'xml' + warnings.warn( + 'The BeautifulStoneSoup class is deprecated. Instead of using ' + 'it, pass features="xml" into the BeautifulSoup constructor.') + super(BeautifulStoneSoup, self).__init__(*args, **kwargs) + + +class StopParsing(Exception): + pass + +class FeatureNotFound(ValueError): + pass + + +#By default, act as an HTML pretty-printer. +if __name__ == '__main__': + import sys + soup = BeautifulSoup(sys.stdin) + print(soup.prettify()) diff --git a/cleanerupper/bs4/builder/__init__.py b/cleanerupper/bs4/builder/__init__.py new file mode 100644 index 0000000..4d6f96d --- /dev/null +++ b/cleanerupper/bs4/builder/__init__.py @@ -0,0 +1,333 @@ +# Use of this source code is governed by a BSD-style license that can be +# found in the LICENSE file. + +from collections import defaultdict +import itertools +import sys +from bs4.element import ( + CharsetMetaAttributeValue, + ContentMetaAttributeValue, + HTMLAwareEntitySubstitution, + whitespace_re + ) + +__all__ = [ + 'HTMLTreeBuilder', + 'SAXTreeBuilder', + 'TreeBuilder', + 'TreeBuilderRegistry', + ] + +# Some useful features for a TreeBuilder to have. +FAST = 'fast' +PERMISSIVE = 'permissive' +STRICT = 'strict' +XML = 'xml' +HTML = 'html' +HTML_5 = 'html5' + + +class TreeBuilderRegistry(object): + + def __init__(self): + self.builders_for_feature = defaultdict(list) + self.builders = [] + + def register(self, treebuilder_class): + """Register a treebuilder based on its advertised features.""" + for feature in treebuilder_class.features: + self.builders_for_feature[feature].insert(0, treebuilder_class) + self.builders.insert(0, treebuilder_class) + + def lookup(self, *features): + if len(self.builders) == 0: + # There are no builders at all. + return None + + if len(features) == 0: + # They didn't ask for any features. Give them the most + # recently registered builder. + return self.builders[0] + + # Go down the list of features in order, and eliminate any builders + # that don't match every feature. + features = list(features) + features.reverse() + candidates = None + candidate_set = None + while len(features) > 0: + feature = features.pop() + we_have_the_feature = self.builders_for_feature.get(feature, []) + if len(we_have_the_feature) > 0: + if candidates is None: + candidates = we_have_the_feature + candidate_set = set(candidates) + else: + # Eliminate any candidates that don't have this feature. + candidate_set = candidate_set.intersection( + set(we_have_the_feature)) + + # The only valid candidates are the ones in candidate_set. + # Go through the original list of candidates and pick the first one + # that's in candidate_set. + if candidate_set is None: + return None + for candidate in candidates: + if candidate in candidate_set: + return candidate + return None + +# The BeautifulSoup class will take feature lists from developers and use them +# to look up builders in this registry. +builder_registry = TreeBuilderRegistry() + +class TreeBuilder(object): + """Turn a document into a Beautiful Soup object tree.""" + + NAME = "[Unknown tree builder]" + ALTERNATE_NAMES = [] + features = [] + + is_xml = False + picklable = False + preserve_whitespace_tags = set() + empty_element_tags = None # A tag will be considered an empty-element + # tag when and only when it has no contents. + + # A value for these tag/attribute combinations is a space- or + # comma-separated list of CDATA, rather than a single CDATA. + cdata_list_attributes = {} + + + def __init__(self): + self.soup = None + + def reset(self): + pass + + def can_be_empty_element(self, tag_name): + """Might a tag with this name be an empty-element tag? + + The final markup may or may not actually present this tag as + self-closing. + + For instance: an HTMLBuilder does not consider a

tag to be + an empty-element tag (it's not in + HTMLBuilder.empty_element_tags). This means an empty

tag + will be presented as "

", not "

". + + The default implementation has no opinion about which tags are + empty-element tags, so a tag will be presented as an + empty-element tag if and only if it has no contents. + "" will become "", and "bar" will + be left alone. + """ + if self.empty_element_tags is None: + return True + return tag_name in self.empty_element_tags + + def feed(self, markup): + raise NotImplementedError() + + def prepare_markup(self, markup, user_specified_encoding=None, + document_declared_encoding=None): + return markup, None, None, False + + def test_fragment_to_document(self, fragment): + """Wrap an HTML fragment to make it look like a document. + + Different parsers do this differently. For instance, lxml + introduces an empty tag, and html5lib + doesn't. Abstracting this away lets us write simple tests + which run HTML fragments through the parser and compare the + results against other HTML fragments. + + This method should not be used outside of tests. + """ + return fragment + + def set_up_substitutions(self, tag): + return False + + def _replace_cdata_list_attribute_values(self, tag_name, attrs): + """Replaces class="foo bar" with class=["foo", "bar"] + + Modifies its input in place. + """ + if not attrs: + return attrs + if self.cdata_list_attributes: + universal = self.cdata_list_attributes.get('*', []) + tag_specific = self.cdata_list_attributes.get( + tag_name.lower(), None) + for attr in list(attrs.keys()): + if attr in universal or (tag_specific and attr in tag_specific): + # We have a "class"-type attribute whose string + # value is a whitespace-separated list of + # values. Split it into a list. + value = attrs[attr] + if isinstance(value, str): + values = whitespace_re.split(value) + else: + # html5lib sometimes calls setAttributes twice + # for the same tag when rearranging the parse + # tree. On the second call the attribute value + # here is already a list. If this happens, + # leave the value alone rather than trying to + # split it again. + values = value + attrs[attr] = values + return attrs + +class SAXTreeBuilder(TreeBuilder): + """A Beautiful Soup treebuilder that listens for SAX events.""" + + def feed(self, markup): + raise NotImplementedError() + + def close(self): + pass + + def startElement(self, name, attrs): + attrs = dict((key[1], value) for key, value in list(attrs.items())) + #print "Start %s, %r" % (name, attrs) + self.soup.handle_starttag(name, attrs) + + def endElement(self, name): + #print "End %s" % name + self.soup.handle_endtag(name) + + def startElementNS(self, nsTuple, nodeName, attrs): + # Throw away (ns, nodeName) for now. + self.startElement(nodeName, attrs) + + def endElementNS(self, nsTuple, nodeName): + # Throw away (ns, nodeName) for now. + self.endElement(nodeName) + #handler.endElementNS((ns, node.nodeName), node.nodeName) + + def startPrefixMapping(self, prefix, nodeValue): + # Ignore the prefix for now. + pass + + def endPrefixMapping(self, prefix): + # Ignore the prefix for now. + # handler.endPrefixMapping(prefix) + pass + + def characters(self, content): + self.soup.handle_data(content) + + def startDocument(self): + pass + + def endDocument(self): + pass + + +class HTMLTreeBuilder(TreeBuilder): + """This TreeBuilder knows facts about HTML. + + Such as which tags are empty-element tags. + """ + + preserve_whitespace_tags = HTMLAwareEntitySubstitution.preserve_whitespace_tags + empty_element_tags = set([ + # These are from HTML5. + 'area', 'base', 'br', 'col', 'embed', 'hr', 'img', 'input', 'keygen', 'link', 'menuitem', 'meta', 'param', 'source', 'track', 'wbr', + + # These are from HTML4, removed in HTML5. + 'spacer', 'frame' + ]) + + # The HTML standard defines these attributes as containing a + # space-separated list of values, not a single value. That is, + # class="foo bar" means that the 'class' attribute has two values, + # 'foo' and 'bar', not the single value 'foo bar'. When we + # encounter one of these attributes, we will parse its value into + # a list of values if possible. Upon output, the list will be + # converted back into a string. + cdata_list_attributes = { + "*" : ['class', 'accesskey', 'dropzone'], + "a" : ['rel', 'rev'], + "link" : ['rel', 'rev'], + "td" : ["headers"], + "th" : ["headers"], + "td" : ["headers"], + "form" : ["accept-charset"], + "object" : ["archive"], + + # These are HTML5 specific, as are *.accesskey and *.dropzone above. + "area" : ["rel"], + "icon" : ["sizes"], + "iframe" : ["sandbox"], + "output" : ["for"], + } + + def set_up_substitutions(self, tag): + # We are only interested in tags + if tag.name != 'meta': + return False + + http_equiv = tag.get('http-equiv') + content = tag.get('content') + charset = tag.get('charset') + + # We are interested in tags that say what encoding the + # document was originally in. This means HTML 5-style + # tags that provide the "charset" attribute. It also means + # HTML 4-style tags that provide the "content" + # attribute and have "http-equiv" set to "content-type". + # + # In both cases we will replace the value of the appropriate + # attribute with a standin object that can take on any + # encoding. + meta_encoding = None + if charset is not None: + # HTML 5 style: + # + meta_encoding = charset + tag['charset'] = CharsetMetaAttributeValue(charset) + + elif (content is not None and http_equiv is not None + and http_equiv.lower() == 'content-type'): + # HTML 4 style: + # + tag['content'] = ContentMetaAttributeValue(content) + + return (meta_encoding is not None) + +def register_treebuilders_from(module): + """Copy TreeBuilders from the given module into this module.""" + # I'm fairly sure this is not the best way to do this. + this_module = sys.modules['bs4.builder'] + for name in module.__all__: + obj = getattr(module, name) + + if issubclass(obj, TreeBuilder): + setattr(this_module, name, obj) + this_module.__all__.append(name) + # Register the builder while we're at it. + this_module.builder_registry.register(obj) + +class ParserRejectedMarkup(Exception): + pass + +# Builders are registered in reverse order of priority, so that custom +# builder registrations will take precedence. In general, we want lxml +# to take precedence over html5lib, because it's faster. And we only +# want to use HTMLParser as a last result. +from . import _htmlparser +register_treebuilders_from(_htmlparser) +try: + from . import _html5lib + register_treebuilders_from(_html5lib) +except ImportError: + # They don't have html5lib installed. + pass +try: + from . import _lxml + register_treebuilders_from(_lxml) +except ImportError: + # They don't have lxml installed. + pass diff --git a/cleanerupper/bs4/builder/_html5lib.py b/cleanerupper/bs4/builder/_html5lib.py new file mode 100644 index 0000000..d9d468f --- /dev/null +++ b/cleanerupper/bs4/builder/_html5lib.py @@ -0,0 +1,426 @@ +# Use of this source code is governed by a BSD-style license that can be +# found in the LICENSE file. + +__all__ = [ + 'HTML5TreeBuilder', + ] + +import warnings +import re +from bs4.builder import ( + PERMISSIVE, + HTML, + HTML_5, + HTMLTreeBuilder, + ) +from bs4.element import ( + NamespacedAttribute, + whitespace_re, +) +import html5lib +from html5lib.constants import ( + namespaces, + prefixes, + ) +from bs4.element import ( + Comment, + Doctype, + NavigableString, + Tag, + ) + +try: + # Pre-0.99999999 + from html5lib.treebuilders import _base as treebuilder_base + new_html5lib = False +except ImportError as e: + # 0.99999999 and up + from html5lib.treebuilders import base as treebuilder_base + new_html5lib = True + +class HTML5TreeBuilder(HTMLTreeBuilder): + """Use html5lib to build a tree.""" + + NAME = "html5lib" + + features = [NAME, PERMISSIVE, HTML_5, HTML] + + def prepare_markup(self, markup, user_specified_encoding, + document_declared_encoding=None, exclude_encodings=None): + # Store the user-specified encoding for use later on. + self.user_specified_encoding = user_specified_encoding + + # document_declared_encoding and exclude_encodings aren't used + # ATM because the html5lib TreeBuilder doesn't use + # UnicodeDammit. + if exclude_encodings: + warnings.warn("You provided a value for exclude_encoding, but the html5lib tree builder doesn't support exclude_encoding.") + yield (markup, None, None, False) + + # These methods are defined by Beautiful Soup. + def feed(self, markup): + if self.soup.parse_only is not None: + warnings.warn("You provided a value for parse_only, but the html5lib tree builder doesn't support parse_only. The entire document will be parsed.") + parser = html5lib.HTMLParser(tree=self.create_treebuilder) + + extra_kwargs = dict() + if not isinstance(markup, str): + if new_html5lib: + extra_kwargs['override_encoding'] = self.user_specified_encoding + else: + extra_kwargs['encoding'] = self.user_specified_encoding + doc = parser.parse(markup, **extra_kwargs) + + # Set the character encoding detected by the tokenizer. + if isinstance(markup, str): + # We need to special-case this because html5lib sets + # charEncoding to UTF-8 if it gets Unicode input. + doc.original_encoding = None + else: + original_encoding = parser.tokenizer.stream.charEncoding[0] + if not isinstance(original_encoding, str): + # In 0.99999999 and up, the encoding is an html5lib + # Encoding object. We want to use a string for compatibility + # with other tree builders. + original_encoding = original_encoding.name + doc.original_encoding = original_encoding + + def create_treebuilder(self, namespaceHTMLElements): + self.underlying_builder = TreeBuilderForHtml5lib( + namespaceHTMLElements, self.soup) + return self.underlying_builder + + def test_fragment_to_document(self, fragment): + """See `TreeBuilder`.""" + return '%s' % fragment + + +class TreeBuilderForHtml5lib(treebuilder_base.TreeBuilder): + + def __init__(self, namespaceHTMLElements, soup=None): + if soup: + self.soup = soup + else: + from bs4 import BeautifulSoup + self.soup = BeautifulSoup("", "html.parser") + super(TreeBuilderForHtml5lib, self).__init__(namespaceHTMLElements) + + def documentClass(self): + self.soup.reset() + return Element(self.soup, self.soup, None) + + def insertDoctype(self, token): + name = token["name"] + publicId = token["publicId"] + systemId = token["systemId"] + + doctype = Doctype.for_name_and_ids(name, publicId, systemId) + self.soup.object_was_parsed(doctype) + + def elementClass(self, name, namespace): + tag = self.soup.new_tag(name, namespace) + return Element(tag, self.soup, namespace) + + def commentClass(self, data): + return TextNode(Comment(data), self.soup) + + def fragmentClass(self): + from bs4 import BeautifulSoup + self.soup = BeautifulSoup("", "html.parser") + self.soup.name = "[document_fragment]" + return Element(self.soup, self.soup, None) + + def appendChild(self, node): + # XXX This code is not covered by the BS4 tests. + self.soup.append(node.element) + + def getDocument(self): + return self.soup + + def getFragment(self): + return treebuilder_base.TreeBuilder.getFragment(self).element + + def testSerializer(self, element): + from bs4 import BeautifulSoup + rv = [] + doctype_re = re.compile(r'^(.*?)(?: PUBLIC "(.*?)"(?: "(.*?)")?| SYSTEM "(.*?)")?$') + + def serializeElement(element, indent=0): + if isinstance(element, BeautifulSoup): + pass + if isinstance(element, Doctype): + m = doctype_re.match(element) + if m: + name = m.group(1) + if m.lastindex > 1: + publicId = m.group(2) or "" + systemId = m.group(3) or m.group(4) or "" + rv.append("""|%s""" % + (' ' * indent, name, publicId, systemId)) + else: + rv.append("|%s" % (' ' * indent, name)) + else: + rv.append("|%s" % (' ' * indent,)) + elif isinstance(element, Comment): + rv.append("|%s" % (' ' * indent, element)) + elif isinstance(element, NavigableString): + rv.append("|%s\"%s\"" % (' ' * indent, element)) + else: + if element.namespace: + name = "%s %s" % (prefixes[element.namespace], + element.name) + else: + name = element.name + rv.append("|%s<%s>" % (' ' * indent, name)) + if element.attrs: + attributes = [] + for name, value in list(element.attrs.items()): + if isinstance(name, NamespacedAttribute): + name = "%s %s" % (prefixes[name.namespace], name.name) + if isinstance(value, list): + value = " ".join(value) + attributes.append((name, value)) + + for name, value in sorted(attributes): + rv.append('|%s%s="%s"' % (' ' * (indent + 2), name, value)) + indent += 2 + for child in element.children: + serializeElement(child, indent) + serializeElement(element, 0) + + return "\n".join(rv) + +class AttrList(object): + def __init__(self, element): + self.element = element + self.attrs = dict(self.element.attrs) + def __iter__(self): + return list(self.attrs.items()).__iter__() + def __setitem__(self, name, value): + # If this attribute is a multi-valued attribute for this element, + # turn its value into a list. + list_attr = HTML5TreeBuilder.cdata_list_attributes + if (name in list_attr['*'] + or (self.element.name in list_attr + and name in list_attr[self.element.name])): + # A node that is being cloned may have already undergone + # this procedure. + if not isinstance(value, list): + value = whitespace_re.split(value) + self.element[name] = value + def items(self): + return list(self.attrs.items()) + def keys(self): + return list(self.attrs.keys()) + def __len__(self): + return len(self.attrs) + def __getitem__(self, name): + return self.attrs[name] + def __contains__(self, name): + return name in list(self.attrs.keys()) + + +class Element(treebuilder_base.Node): + def __init__(self, element, soup, namespace): + treebuilder_base.Node.__init__(self, element.name) + self.element = element + self.soup = soup + self.namespace = namespace + + def appendChild(self, node): + string_child = child = None + if isinstance(node, str): + # Some other piece of code decided to pass in a string + # instead of creating a TextElement object to contain the + # string. + string_child = child = node + elif isinstance(node, Tag): + # Some other piece of code decided to pass in a Tag + # instead of creating an Element object to contain the + # Tag. + child = node + elif node.element.__class__ == NavigableString: + string_child = child = node.element + node.parent = self + else: + child = node.element + node.parent = self + + if not isinstance(child, str) and child.parent is not None: + node.element.extract() + + if (string_child and self.element.contents + and self.element.contents[-1].__class__ == NavigableString): + # We are appending a string onto another string. + # TODO This has O(n^2) performance, for input like + # "aaa..." + old_element = self.element.contents[-1] + new_element = self.soup.new_string(old_element + string_child) + old_element.replace_with(new_element) + self.soup._most_recent_element = new_element + else: + if isinstance(node, str): + # Create a brand new NavigableString from this string. + child = self.soup.new_string(node) + + # Tell Beautiful Soup to act as if it parsed this element + # immediately after the parent's last descendant. (Or + # immediately after the parent, if it has no children.) + if self.element.contents: + most_recent_element = self.element._last_descendant(False) + elif self.element.next_element is not None: + # Something from further ahead in the parse tree is + # being inserted into this earlier element. This is + # very annoying because it means an expensive search + # for the last element in the tree. + most_recent_element = self.soup._last_descendant() + else: + most_recent_element = self.element + + self.soup.object_was_parsed( + child, parent=self.element, + most_recent_element=most_recent_element) + + def getAttributes(self): + if isinstance(self.element, Comment): + return {} + return AttrList(self.element) + + def setAttributes(self, attributes): + + if attributes is not None and len(attributes) > 0: + + converted_attributes = [] + for name, value in list(attributes.items()): + if isinstance(name, tuple): + new_name = NamespacedAttribute(*name) + del attributes[name] + attributes[new_name] = value + + self.soup.builder._replace_cdata_list_attribute_values( + self.name, attributes) + for name, value in list(attributes.items()): + self.element[name] = value + + # The attributes may contain variables that need substitution. + # Call set_up_substitutions manually. + # + # The Tag constructor called this method when the Tag was created, + # but we just set/changed the attributes, so call it again. + self.soup.builder.set_up_substitutions(self.element) + attributes = property(getAttributes, setAttributes) + + def insertText(self, data, insertBefore=None): + text = TextNode(self.soup.new_string(data), self.soup) + if insertBefore: + self.insertBefore(text, insertBefore) + else: + self.appendChild(text) + + def insertBefore(self, node, refNode): + index = self.element.index(refNode.element) + if (node.element.__class__ == NavigableString and self.element.contents + and self.element.contents[index-1].__class__ == NavigableString): + # (See comments in appendChild) + old_node = self.element.contents[index-1] + new_str = self.soup.new_string(old_node + node.element) + old_node.replace_with(new_str) + else: + self.element.insert(index, node.element) + node.parent = self + + def removeChild(self, node): + node.element.extract() + + def reparentChildren(self, new_parent): + """Move all of this tag's children into another tag.""" + # print "MOVE", self.element.contents + # print "FROM", self.element + # print "TO", new_parent.element + + element = self.element + new_parent_element = new_parent.element + # Determine what this tag's next_element will be once all the children + # are removed. + final_next_element = element.next_sibling + + new_parents_last_descendant = new_parent_element._last_descendant(False, False) + if len(new_parent_element.contents) > 0: + # The new parent already contains children. We will be + # appending this tag's children to the end. + new_parents_last_child = new_parent_element.contents[-1] + new_parents_last_descendant_next_element = new_parents_last_descendant.next_element + else: + # The new parent contains no children. + new_parents_last_child = None + new_parents_last_descendant_next_element = new_parent_element.next_element + + to_append = element.contents + if len(to_append) > 0: + # Set the first child's previous_element and previous_sibling + # to elements within the new parent + first_child = to_append[0] + if new_parents_last_descendant: + first_child.previous_element = new_parents_last_descendant + else: + first_child.previous_element = new_parent_element + first_child.previous_sibling = new_parents_last_child + if new_parents_last_descendant: + new_parents_last_descendant.next_element = first_child + else: + new_parent_element.next_element = first_child + if new_parents_last_child: + new_parents_last_child.next_sibling = first_child + + # Find the very last element being moved. It is now the + # parent's last descendant. It has no .next_sibling and + # its .next_element is whatever the previous last + # descendant had. + last_childs_last_descendant = to_append[-1]._last_descendant(False, True) + + last_childs_last_descendant.next_element = new_parents_last_descendant_next_element + if new_parents_last_descendant_next_element: + # TODO: This code has no test coverage and I'm not sure + # how to get html5lib to go through this path, but it's + # just the other side of the previous line. + new_parents_last_descendant_next_element.previous_element = last_childs_last_descendant + last_childs_last_descendant.next_sibling = None + + for child in to_append: + child.parent = new_parent_element + new_parent_element.contents.append(child) + + # Now that this element has no children, change its .next_element. + element.contents = [] + element.next_element = final_next_element + + # print "DONE WITH MOVE" + # print "FROM", self.element + # print "TO", new_parent_element + + def cloneNode(self): + tag = self.soup.new_tag(self.element.name, self.namespace) + node = Element(tag, self.soup, self.namespace) + for key,value in self.attributes: + node.attributes[key] = value + return node + + def hasContent(self): + return self.element.contents + + def getNameTuple(self): + if self.namespace == None: + return namespaces["html"], self.name + else: + return self.namespace, self.name + + nameTuple = property(getNameTuple) + +class TextNode(Element): + def __init__(self, element, soup): + treebuilder_base.Node.__init__(self, None) + self.element = element + self.soup = soup + + def cloneNode(self): + raise NotImplementedError diff --git a/cleanerupper/bs4/builder/_htmlparser.py b/cleanerupper/bs4/builder/_htmlparser.py new file mode 100644 index 0000000..907d355 --- /dev/null +++ b/cleanerupper/bs4/builder/_htmlparser.py @@ -0,0 +1,314 @@ +"""Use the HTMLParser library to parse HTML files that aren't too bad.""" + +# Use of this source code is governed by a BSD-style license that can be +# found in the LICENSE file. + +__all__ = [ + 'HTMLParserTreeBuilder', + ] + +from html.parser import HTMLParser + +try: + from html.parser import HTMLParseError +except ImportError as e: + # HTMLParseError is removed in Python 3.5. Since it can never be + # thrown in 3.5, we can just define our own class as a placeholder. + class HTMLParseError(Exception): + pass + +import sys +import warnings + +# Starting in Python 3.2, the HTMLParser constructor takes a 'strict' +# argument, which we'd like to set to False. Unfortunately, +# http://bugs.python.org/issue13273 makes strict=True a better bet +# before Python 3.2.3. +# +# At the end of this file, we monkeypatch HTMLParser so that +# strict=True works well on Python 3.2.2. +major, minor, release = sys.version_info[:3] +CONSTRUCTOR_TAKES_STRICT = major == 3 and minor == 2 and release >= 3 +CONSTRUCTOR_STRICT_IS_DEPRECATED = major == 3 and minor == 3 +CONSTRUCTOR_TAKES_CONVERT_CHARREFS = major == 3 and minor >= 4 + + +from bs4.element import ( + CData, + Comment, + Declaration, + Doctype, + ProcessingInstruction, + ) +from bs4.dammit import EntitySubstitution, UnicodeDammit + +from bs4.builder import ( + HTML, + HTMLTreeBuilder, + STRICT, + ) + + +HTMLPARSER = 'html.parser' + +class BeautifulSoupHTMLParser(HTMLParser): + + def __init__(self, *args, **kwargs): + HTMLParser.__init__(self, *args, **kwargs) + + # Keep a list of empty-element tags that were encountered + # without an explicit closing tag. If we encounter a closing tag + # of this type, we'll associate it with one of those entries. + # + # This isn't a stack because we don't care about the + # order. It's a list of closing tags we've already handled and + # will ignore, assuming they ever show up. + self.already_closed_empty_element = [] + + def handle_startendtag(self, name, attrs): + # This is only called when the markup looks like + # . + + # is_startend() tells handle_starttag not to close the tag + # just because its name matches a known empty-element tag. We + # know that this is an empty-element tag and we want to call + # handle_endtag ourselves. + tag = self.handle_starttag(name, attrs, handle_empty_element=False) + self.handle_endtag(name) + + def handle_starttag(self, name, attrs, handle_empty_element=True): + # XXX namespace + attr_dict = {} + for key, value in attrs: + # Change None attribute values to the empty string + # for consistency with the other tree builders. + if value is None: + value = '' + attr_dict[key] = value + attrvalue = '""' + #print "START", name + tag = self.soup.handle_starttag(name, None, None, attr_dict) + if tag and tag.is_empty_element and handle_empty_element: + # Unlike other parsers, html.parser doesn't send separate end tag + # events for empty-element tags. (It's handled in + # handle_startendtag, but only if the original markup looked like + # .) + # + # So we need to call handle_endtag() ourselves. Since we + # know the start event is identical to the end event, we + # don't want handle_endtag() to cross off any previous end + # events for tags of this name. + self.handle_endtag(name, check_already_closed=False) + + # But we might encounter an explicit closing tag for this tag + # later on. If so, we want to ignore it. + self.already_closed_empty_element.append(name) + + def handle_endtag(self, name, check_already_closed=True): + #print "END", name + if check_already_closed and name in self.already_closed_empty_element: + # This is a redundant end tag for an empty-element tag. + # We've already called handle_endtag() for it, so just + # check it off the list. + # print "ALREADY CLOSED", name + self.already_closed_empty_element.remove(name) + else: + self.soup.handle_endtag(name) + + def handle_data(self, data): + self.soup.handle_data(data) + + def handle_charref(self, name): + # XXX workaround for a bug in HTMLParser. Remove this once + # it's fixed in all supported versions. + # http://bugs.python.org/issue13633 + if name.startswith('x'): + real_name = int(name.lstrip('x'), 16) + elif name.startswith('X'): + real_name = int(name.lstrip('X'), 16) + else: + real_name = int(name) + + try: + data = chr(real_name) + except (ValueError, OverflowError) as e: + data = "\N{REPLACEMENT CHARACTER}" + + self.handle_data(data) + + def handle_entityref(self, name): + character = EntitySubstitution.HTML_ENTITY_TO_CHARACTER.get(name) + if character is not None: + data = character + else: + data = "&%s;" % name + self.handle_data(data) + + def handle_comment(self, data): + self.soup.endData() + self.soup.handle_data(data) + self.soup.endData(Comment) + + def handle_decl(self, data): + self.soup.endData() + if data.startswith("DOCTYPE "): + data = data[len("DOCTYPE "):] + elif data == 'DOCTYPE': + # i.e. "" + data = '' + self.soup.handle_data(data) + self.soup.endData(Doctype) + + def unknown_decl(self, data): + if data.upper().startswith('CDATA['): + cls = CData + data = data[len('CDATA['):] + else: + cls = Declaration + self.soup.endData() + self.soup.handle_data(data) + self.soup.endData(cls) + + def handle_pi(self, data): + self.soup.endData() + self.soup.handle_data(data) + self.soup.endData(ProcessingInstruction) + + +class HTMLParserTreeBuilder(HTMLTreeBuilder): + + is_xml = False + picklable = True + NAME = HTMLPARSER + features = [NAME, HTML, STRICT] + + def __init__(self, *args, **kwargs): + if CONSTRUCTOR_TAKES_STRICT and not CONSTRUCTOR_STRICT_IS_DEPRECATED: + kwargs['strict'] = False + if CONSTRUCTOR_TAKES_CONVERT_CHARREFS: + kwargs['convert_charrefs'] = False + self.parser_args = (args, kwargs) + + def prepare_markup(self, markup, user_specified_encoding=None, + document_declared_encoding=None, exclude_encodings=None): + """ + :return: A 4-tuple (markup, original encoding, encoding + declared within markup, whether any characters had to be + replaced with REPLACEMENT CHARACTER). + """ + if isinstance(markup, str): + yield (markup, None, None, False) + return + + try_encodings = [user_specified_encoding, document_declared_encoding] + dammit = UnicodeDammit(markup, try_encodings, is_html=True, + exclude_encodings=exclude_encodings) + yield (dammit.markup, dammit.original_encoding, + dammit.declared_html_encoding, + dammit.contains_replacement_characters) + + def feed(self, markup): + args, kwargs = self.parser_args + parser = BeautifulSoupHTMLParser(*args, **kwargs) + parser.soup = self.soup + try: + parser.feed(markup) + except HTMLParseError as e: + warnings.warn(RuntimeWarning( + "Python's built-in HTMLParser cannot parse the given document. This is not a bug in Beautiful Soup. The best solution is to install an external parser (lxml or html5lib), and use Beautiful Soup with that parser. See http://www.crummy.com/software/BeautifulSoup/bs4/doc/#installing-a-parser for help.")) + raise e + parser.already_closed_empty_element = [] + +# Patch 3.2 versions of HTMLParser earlier than 3.2.3 to use some +# 3.2.3 code. This ensures they don't treat markup like

as a +# string. +# +# XXX This code can be removed once most Python 3 users are on 3.2.3. +if major == 3 and minor == 2 and not CONSTRUCTOR_TAKES_STRICT: + import re + attrfind_tolerant = re.compile( + r'\s*((?<=[\'"\s])[^\s/>][^\s/=>]*)(\s*=+\s*' + r'(\'[^\']*\'|"[^"]*"|(?![\'"])[^>\s]*))?') + HTMLParserTreeBuilder.attrfind_tolerant = attrfind_tolerant + + locatestarttagend = re.compile(r""" + <[a-zA-Z][-.a-zA-Z0-9:_]* # tag name + (?:\s+ # whitespace before attribute name + (?:[a-zA-Z_][-.:a-zA-Z0-9_]* # attribute name + (?:\s*=\s* # value indicator + (?:'[^']*' # LITA-enclosed value + |\"[^\"]*\" # LIT-enclosed value + |[^'\">\s]+ # bare value + ) + )? + ) + )* + \s* # trailing whitespace +""", re.VERBOSE) + BeautifulSoupHTMLParser.locatestarttagend = locatestarttagend + + from html.parser import tagfind, attrfind + + def parse_starttag(self, i): + self.__starttag_text = None + endpos = self.check_for_whole_start_tag(i) + if endpos < 0: + return endpos + rawdata = self.rawdata + self.__starttag_text = rawdata[i:endpos] + + # Now parse the data between i+1 and j into a tag and attrs + attrs = [] + match = tagfind.match(rawdata, i+1) + assert match, 'unexpected call to parse_starttag()' + k = match.end() + self.lasttag = tag = rawdata[i+1:k].lower() + while k < endpos: + if self.strict: + m = attrfind.match(rawdata, k) + else: + m = attrfind_tolerant.match(rawdata, k) + if not m: + break + attrname, rest, attrvalue = m.group(1, 2, 3) + if not rest: + attrvalue = None + elif attrvalue[:1] == '\'' == attrvalue[-1:] or \ + attrvalue[:1] == '"' == attrvalue[-1:]: + attrvalue = attrvalue[1:-1] + if attrvalue: + attrvalue = self.unescape(attrvalue) + attrs.append((attrname.lower(), attrvalue)) + k = m.end() + + end = rawdata[k:endpos].strip() + if end not in (">", "/>"): + lineno, offset = self.getpos() + if "\n" in self.__starttag_text: + lineno = lineno + self.__starttag_text.count("\n") + offset = len(self.__starttag_text) \ + - self.__starttag_text.rfind("\n") + else: + offset = offset + len(self.__starttag_text) + if self.strict: + self.error("junk characters in start tag: %r" + % (rawdata[k:endpos][:20],)) + self.handle_data(rawdata[i:endpos]) + return endpos + if end.endswith('/>'): + # XHTML-style empty tag: + self.handle_startendtag(tag, attrs) + else: + self.handle_starttag(tag, attrs) + if tag in self.CDATA_CONTENT_ELEMENTS: + self.set_cdata_mode(tag) + return endpos + + def set_cdata_mode(self, elem): + self.cdata_elem = elem.lower() + self.interesting = re.compile(r'' % self.cdata_elem, re.I) + + BeautifulSoupHTMLParser.parse_starttag = parse_starttag + BeautifulSoupHTMLParser.set_cdata_mode = set_cdata_mode + + CONSTRUCTOR_TAKES_STRICT = True diff --git a/cleanerupper/bs4/builder/_lxml.py b/cleanerupper/bs4/builder/_lxml.py new file mode 100644 index 0000000..244d457 --- /dev/null +++ b/cleanerupper/bs4/builder/_lxml.py @@ -0,0 +1,258 @@ +# Use of this source code is governed by a BSD-style license that can be +# found in the LICENSE file. +__all__ = [ + 'LXMLTreeBuilderForXML', + 'LXMLTreeBuilder', + ] + +from io import BytesIO +from io import StringIO +import collections +from lxml import etree +from bs4.element import ( + Comment, + Doctype, + NamespacedAttribute, + ProcessingInstruction, + XMLProcessingInstruction, +) +from bs4.builder import ( + FAST, + HTML, + HTMLTreeBuilder, + PERMISSIVE, + ParserRejectedMarkup, + TreeBuilder, + XML) +from bs4.dammit import EncodingDetector + +LXML = 'lxml' + +class LXMLTreeBuilderForXML(TreeBuilder): + DEFAULT_PARSER_CLASS = etree.XMLParser + + is_xml = True + processing_instruction_class = XMLProcessingInstruction + + NAME = "lxml-xml" + ALTERNATE_NAMES = ["xml"] + + # Well, it's permissive by XML parser standards. + features = [NAME, LXML, XML, FAST, PERMISSIVE] + + CHUNK_SIZE = 512 + + # This namespace mapping is specified in the XML Namespace + # standard. + DEFAULT_NSMAPS = {'http://www.w3.org/XML/1998/namespace' : "xml"} + + def default_parser(self, encoding): + # This can either return a parser object or a class, which + # will be instantiated with default arguments. + if self._default_parser is not None: + return self._default_parser + return etree.XMLParser( + target=self, strip_cdata=False, recover=True, encoding=encoding) + + def parser_for(self, encoding): + # Use the default parser. + parser = self.default_parser(encoding) + + if isinstance(parser, collections.Callable): + # Instantiate the parser with default arguments + parser = parser(target=self, strip_cdata=False, encoding=encoding) + return parser + + def __init__(self, parser=None, empty_element_tags=None): + # TODO: Issue a warning if parser is present but not a + # callable, since that means there's no way to create new + # parsers for different encodings. + self._default_parser = parser + if empty_element_tags is not None: + self.empty_element_tags = set(empty_element_tags) + self.soup = None + self.nsmaps = [self.DEFAULT_NSMAPS] + + def _getNsTag(self, tag): + # Split the namespace URL out of a fully-qualified lxml tag + # name. Copied from lxml's src/lxml/sax.py. + if tag[0] == '{': + return tuple(tag[1:].split('}', 1)) + else: + return (None, tag) + + def prepare_markup(self, markup, user_specified_encoding=None, + exclude_encodings=None, + document_declared_encoding=None): + """ + :yield: A series of 4-tuples. + (markup, encoding, declared encoding, + has undergone character replacement) + + Each 4-tuple represents a strategy for parsing the document. + """ + # Instead of using UnicodeDammit to convert the bytestring to + # Unicode using different encodings, use EncodingDetector to + # iterate over the encodings, and tell lxml to try to parse + # the document as each one in turn. + is_html = not self.is_xml + if is_html: + self.processing_instruction_class = ProcessingInstruction + else: + self.processing_instruction_class = XMLProcessingInstruction + + if isinstance(markup, str): + # We were given Unicode. Maybe lxml can parse Unicode on + # this system? + yield markup, None, document_declared_encoding, False + + if isinstance(markup, str): + # No, apparently not. Convert the Unicode to UTF-8 and + # tell lxml to parse it as UTF-8. + yield (markup.encode("utf8"), "utf8", + document_declared_encoding, False) + + try_encodings = [user_specified_encoding, document_declared_encoding] + detector = EncodingDetector( + markup, try_encodings, is_html, exclude_encodings) + for encoding in detector.encodings: + yield (detector.markup, encoding, document_declared_encoding, False) + + def feed(self, markup): + if isinstance(markup, bytes): + markup = BytesIO(markup) + elif isinstance(markup, str): + markup = StringIO(markup) + + # Call feed() at least once, even if the markup is empty, + # or the parser won't be initialized. + data = markup.read(self.CHUNK_SIZE) + try: + self.parser = self.parser_for(self.soup.original_encoding) + self.parser.feed(data) + while len(data) != 0: + # Now call feed() on the rest of the data, chunk by chunk. + data = markup.read(self.CHUNK_SIZE) + if len(data) != 0: + self.parser.feed(data) + self.parser.close() + except (UnicodeDecodeError, LookupError, etree.ParserError) as e: + raise ParserRejectedMarkup(str(e)) + + def close(self): + self.nsmaps = [self.DEFAULT_NSMAPS] + + def start(self, name, attrs, nsmap={}): + # Make sure attrs is a mutable dict--lxml may send an immutable dictproxy. + attrs = dict(attrs) + nsprefix = None + # Invert each namespace map as it comes in. + if len(self.nsmaps) > 1: + # There are no new namespaces for this tag, but + # non-default namespaces are in play, so we need a + # separate tag stack to know when they end. + self.nsmaps.append(None) + elif len(nsmap) > 0: + # A new namespace mapping has come into play. + inverted_nsmap = dict((value, key) for key, value in list(nsmap.items())) + self.nsmaps.append(inverted_nsmap) + # Also treat the namespace mapping as a set of attributes on the + # tag, so we can recreate it later. + attrs = attrs.copy() + for prefix, namespace in list(nsmap.items()): + attribute = NamespacedAttribute( + "xmlns", prefix, "http://www.w3.org/2000/xmlns/") + attrs[attribute] = namespace + + # Namespaces are in play. Find any attributes that came in + # from lxml with namespaces attached to their names, and + # turn then into NamespacedAttribute objects. + new_attrs = {} + for attr, value in list(attrs.items()): + namespace, attr = self._getNsTag(attr) + if namespace is None: + new_attrs[attr] = value + else: + nsprefix = self._prefix_for_namespace(namespace) + attr = NamespacedAttribute(nsprefix, attr, namespace) + new_attrs[attr] = value + attrs = new_attrs + + namespace, name = self._getNsTag(name) + nsprefix = self._prefix_for_namespace(namespace) + self.soup.handle_starttag(name, namespace, nsprefix, attrs) + + def _prefix_for_namespace(self, namespace): + """Find the currently active prefix for the given namespace.""" + if namespace is None: + return None + for inverted_nsmap in reversed(self.nsmaps): + if inverted_nsmap is not None and namespace in inverted_nsmap: + return inverted_nsmap[namespace] + return None + + def end(self, name): + self.soup.endData() + completed_tag = self.soup.tagStack[-1] + namespace, name = self._getNsTag(name) + nsprefix = None + if namespace is not None: + for inverted_nsmap in reversed(self.nsmaps): + if inverted_nsmap is not None and namespace in inverted_nsmap: + nsprefix = inverted_nsmap[namespace] + break + self.soup.handle_endtag(name, nsprefix) + if len(self.nsmaps) > 1: + # This tag, or one of its parents, introduced a namespace + # mapping, so pop it off the stack. + self.nsmaps.pop() + + def pi(self, target, data): + self.soup.endData() + self.soup.handle_data(target + ' ' + data) + self.soup.endData(self.processing_instruction_class) + + def data(self, content): + self.soup.handle_data(content) + + def doctype(self, name, pubid, system): + self.soup.endData() + doctype = Doctype.for_name_and_ids(name, pubid, system) + self.soup.object_was_parsed(doctype) + + def comment(self, content): + "Handle comments as Comment objects." + self.soup.endData() + self.soup.handle_data(content) + self.soup.endData(Comment) + + def test_fragment_to_document(self, fragment): + """See `TreeBuilder`.""" + return '\n%s' % fragment + + +class LXMLTreeBuilder(HTMLTreeBuilder, LXMLTreeBuilderForXML): + + NAME = LXML + ALTERNATE_NAMES = ["lxml-html"] + + features = ALTERNATE_NAMES + [NAME, HTML, FAST, PERMISSIVE] + is_xml = False + processing_instruction_class = ProcessingInstruction + + def default_parser(self, encoding): + return etree.HTMLParser + + def feed(self, markup): + encoding = self.soup.original_encoding + try: + self.parser = self.parser_for(encoding) + self.parser.feed(markup) + self.parser.close() + except (UnicodeDecodeError, LookupError, etree.ParserError) as e: + raise ParserRejectedMarkup(str(e)) + + + def test_fragment_to_document(self, fragment): + """See `TreeBuilder`.""" + return '%s' % fragment diff --git a/cleanerupper/bs4/dammit.py b/cleanerupper/bs4/dammit.py new file mode 100644 index 0000000..8e399e0 --- /dev/null +++ b/cleanerupper/bs4/dammit.py @@ -0,0 +1,842 @@ +# -*- coding: utf-8 -*- +"""Beautiful Soup bonus library: Unicode, Dammit + +This library converts a bytestream to Unicode through any means +necessary. It is heavily based on code from Mark Pilgrim's Universal +Feed Parser. It works best on XML and HTML, but it does not rewrite the +XML or HTML to reflect a new encoding; that's the tree builder's job. +""" +# Use of this source code is governed by a BSD-style license that can be +# found in the LICENSE file. +__license__ = "MIT" + +import codecs +from html.entities import codepoint2name +import re +import logging +import string + +# Import a library to autodetect character encodings. +chardet_type = None +try: + # First try the fast C implementation. + # PyPI package: cchardet + import cchardet + def chardet_dammit(s): + return cchardet.detect(s)['encoding'] +except ImportError: + try: + # Fall back to the pure Python implementation + # Debian package: python-chardet + # PyPI package: chardet + import chardet + def chardet_dammit(s): + return chardet.detect(s)['encoding'] + #import chardet.constants + #chardet.constants._debug = 1 + except ImportError: + # No chardet available. + def chardet_dammit(s): + return None + +# Available from http://cjkpython.i18n.org/. +try: + import iconv_codec +except ImportError: + pass + +xml_encoding_re = re.compile( + '^<\?.*encoding=[\'"](.*?)[\'"].*\?>'.encode(), re.I) +html_meta_re = re.compile( + '<\s*meta[^>]+charset\s*=\s*["\']?([^>]*?)[ /;\'">]'.encode(), re.I) + +class EntitySubstitution(object): + + """Substitute XML or HTML entities for the corresponding characters.""" + + def _populate_class_variables(): + lookup = {} + reverse_lookup = {} + characters_for_re = [] + for codepoint, name in list(codepoint2name.items()): + character = chr(codepoint) + if codepoint != 34: + # There's no point in turning the quotation mark into + # ", unless it happens within an attribute value, which + # is handled elsewhere. + characters_for_re.append(character) + lookup[character] = name + # But we do want to turn " into the quotation mark. + reverse_lookup[name] = character + re_definition = "[%s]" % "".join(characters_for_re) + return lookup, reverse_lookup, re.compile(re_definition) + (CHARACTER_TO_HTML_ENTITY, HTML_ENTITY_TO_CHARACTER, + CHARACTER_TO_HTML_ENTITY_RE) = _populate_class_variables() + + CHARACTER_TO_XML_ENTITY = { + "'": "apos", + '"': "quot", + "&": "amp", + "<": "lt", + ">": "gt", + } + + BARE_AMPERSAND_OR_BRACKET = re.compile("([<>]|" + "&(?!#\d+;|#x[0-9a-fA-F]+;|\w+;)" + ")") + + AMPERSAND_OR_BRACKET = re.compile("([<>&])") + + @classmethod + def _substitute_html_entity(cls, matchobj): + entity = cls.CHARACTER_TO_HTML_ENTITY.get(matchobj.group(0)) + return "&%s;" % entity + + @classmethod + def _substitute_xml_entity(cls, matchobj): + """Used with a regular expression to substitute the + appropriate XML entity for an XML special character.""" + entity = cls.CHARACTER_TO_XML_ENTITY[matchobj.group(0)] + return "&%s;" % entity + + @classmethod + def quoted_attribute_value(self, value): + """Make a value into a quoted XML attribute, possibly escaping it. + + Most strings will be quoted using double quotes. + + Bob's Bar -> "Bob's Bar" + + If a string contains double quotes, it will be quoted using + single quotes. + + Welcome to "my bar" -> 'Welcome to "my bar"' + + If a string contains both single and double quotes, the + double quotes will be escaped, and the string will be quoted + using double quotes. + + Welcome to "Bob's Bar" -> "Welcome to "Bob's bar" + """ + quote_with = '"' + if '"' in value: + if "'" in value: + # The string contains both single and double + # quotes. Turn the double quotes into + # entities. We quote the double quotes rather than + # the single quotes because the entity name is + # """ whether this is HTML or XML. If we + # quoted the single quotes, we'd have to decide + # between ' and &squot;. + replace_with = """ + value = value.replace('"', replace_with) + else: + # There are double quotes but no single quotes. + # We can use single quotes to quote the attribute. + quote_with = "'" + return quote_with + value + quote_with + + @classmethod + def substitute_xml(cls, value, make_quoted_attribute=False): + """Substitute XML entities for special XML characters. + + :param value: A string to be substituted. The less-than sign + will become <, the greater-than sign will become >, + and any ampersands will become &. If you want ampersands + that appear to be part of an entity definition to be left + alone, use substitute_xml_containing_entities() instead. + + :param make_quoted_attribute: If True, then the string will be + quoted, as befits an attribute value. + """ + # Escape angle brackets and ampersands. + value = cls.AMPERSAND_OR_BRACKET.sub( + cls._substitute_xml_entity, value) + + if make_quoted_attribute: + value = cls.quoted_attribute_value(value) + return value + + @classmethod + def substitute_xml_containing_entities( + cls, value, make_quoted_attribute=False): + """Substitute XML entities for special XML characters. + + :param value: A string to be substituted. The less-than sign will + become <, the greater-than sign will become >, and any + ampersands that are not part of an entity defition will + become &. + + :param make_quoted_attribute: If True, then the string will be + quoted, as befits an attribute value. + """ + # Escape angle brackets, and ampersands that aren't part of + # entities. + value = cls.BARE_AMPERSAND_OR_BRACKET.sub( + cls._substitute_xml_entity, value) + + if make_quoted_attribute: + value = cls.quoted_attribute_value(value) + return value + + @classmethod + def substitute_html(cls, s): + """Replace certain Unicode characters with named HTML entities. + + This differs from data.encode(encoding, 'xmlcharrefreplace') + in that the goal is to make the result more readable (to those + with ASCII displays) rather than to recover from + errors. There's absolutely nothing wrong with a UTF-8 string + containg a LATIN SMALL LETTER E WITH ACUTE, but replacing that + character with "é" will make it more readable to some + people. + """ + return cls.CHARACTER_TO_HTML_ENTITY_RE.sub( + cls._substitute_html_entity, s) + + +class EncodingDetector: + """Suggests a number of possible encodings for a bytestring. + + Order of precedence: + + 1. Encodings you specifically tell EncodingDetector to try first + (the override_encodings argument to the constructor). + + 2. An encoding declared within the bytestring itself, either in an + XML declaration (if the bytestring is to be interpreted as an XML + document), or in a tag (if the bytestring is to be + interpreted as an HTML document.) + + 3. An encoding detected through textual analysis by chardet, + cchardet, or a similar external library. + + 4. UTF-8. + + 5. Windows-1252. + """ + def __init__(self, markup, override_encodings=None, is_html=False, + exclude_encodings=None): + self.override_encodings = override_encodings or [] + exclude_encodings = exclude_encodings or [] + self.exclude_encodings = set([x.lower() for x in exclude_encodings]) + self.chardet_encoding = None + self.is_html = is_html + self.declared_encoding = None + + # First order of business: strip a byte-order mark. + self.markup, self.sniffed_encoding = self.strip_byte_order_mark(markup) + + def _usable(self, encoding, tried): + if encoding is not None: + encoding = encoding.lower() + if encoding in self.exclude_encodings: + return False + if encoding not in tried: + tried.add(encoding) + return True + return False + + @property + def encodings(self): + """Yield a number of encodings that might work for this markup.""" + tried = set() + for e in self.override_encodings: + if self._usable(e, tried): + yield e + + # Did the document originally start with a byte-order mark + # that indicated its encoding? + if self._usable(self.sniffed_encoding, tried): + yield self.sniffed_encoding + + # Look within the document for an XML or HTML encoding + # declaration. + if self.declared_encoding is None: + self.declared_encoding = self.find_declared_encoding( + self.markup, self.is_html) + if self._usable(self.declared_encoding, tried): + yield self.declared_encoding + + # Use third-party character set detection to guess at the + # encoding. + if self.chardet_encoding is None: + self.chardet_encoding = chardet_dammit(self.markup) + if self._usable(self.chardet_encoding, tried): + yield self.chardet_encoding + + # As a last-ditch effort, try utf-8 and windows-1252. + for e in ('utf-8', 'windows-1252'): + if self._usable(e, tried): + yield e + + @classmethod + def strip_byte_order_mark(cls, data): + """If a byte-order mark is present, strip it and return the encoding it implies.""" + encoding = None + if isinstance(data, str): + # Unicode data cannot have a byte-order mark. + return data, encoding + if (len(data) >= 4) and (data[:2] == b'\xfe\xff') \ + and (data[2:4] != '\x00\x00'): + encoding = 'utf-16be' + data = data[2:] + elif (len(data) >= 4) and (data[:2] == b'\xff\xfe') \ + and (data[2:4] != '\x00\x00'): + encoding = 'utf-16le' + data = data[2:] + elif data[:3] == b'\xef\xbb\xbf': + encoding = 'utf-8' + data = data[3:] + elif data[:4] == b'\x00\x00\xfe\xff': + encoding = 'utf-32be' + data = data[4:] + elif data[:4] == b'\xff\xfe\x00\x00': + encoding = 'utf-32le' + data = data[4:] + return data, encoding + + @classmethod + def find_declared_encoding(cls, markup, is_html=False, search_entire_document=False): + """Given a document, tries to find its declared encoding. + + An XML encoding is declared at the beginning of the document. + + An HTML encoding is declared in a tag, hopefully near the + beginning of the document. + """ + if search_entire_document: + xml_endpos = html_endpos = len(markup) + else: + xml_endpos = 1024 + html_endpos = max(2048, int(len(markup) * 0.05)) + + declared_encoding = None + declared_encoding_match = xml_encoding_re.search(markup, endpos=xml_endpos) + if not declared_encoding_match and is_html: + declared_encoding_match = html_meta_re.search(markup, endpos=html_endpos) + if declared_encoding_match is not None: + declared_encoding = declared_encoding_match.groups()[0].decode( + 'ascii', 'replace') + if declared_encoding: + return declared_encoding.lower() + return None + +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"} + + ENCODINGS_WITH_SMART_QUOTES = [ + "windows-1252", + "iso-8859-1", + "iso-8859-2", + ] + + def __init__(self, markup, override_encodings=[], + smart_quotes_to=None, is_html=False, exclude_encodings=[]): + self.smart_quotes_to = smart_quotes_to + self.tried_encodings = [] + self.contains_replacement_characters = False + self.is_html = is_html + self.log = logging.getLogger(__name__) + self.detector = EncodingDetector( + markup, override_encodings, is_html, exclude_encodings) + + # Short-circuit if the data is in Unicode to begin with. + if isinstance(markup, str) or markup == '': + self.markup = markup + self.unicode_markup = str(markup) + self.original_encoding = None + return + + # The encoding detector may have stripped a byte-order mark. + # Use the stripped markup from this point on. + self.markup = self.detector.markup + + u = None + for encoding in self.detector.encodings: + markup = self.detector.markup + u = self._convert_from(encoding) + if u is not None: + break + + if not u: + # None of the encodings worked. As an absolute last resort, + # try them again with character replacement. + + for encoding in self.detector.encodings: + if encoding != "ascii": + u = self._convert_from(encoding, "replace") + if u is not None: + self.log.warning( + "Some characters could not be decoded, and were " + "replaced with REPLACEMENT CHARACTER." + ) + self.contains_replacement_characters = True + break + + # If none of that worked, we could at this point force it to + # ASCII, but that would destroy so much data that I think + # giving up is better. + self.unicode_markup = u + if not u: + self.original_encoding = None + + def _sub_ms_char(self, match): + """Changes a MS smart quote character to an XML or HTML + entity, or an ASCII character.""" + orig = match.group(1) + if self.smart_quotes_to == 'ascii': + sub = self.MS_CHARS_TO_ASCII.get(orig).encode() + else: + sub = self.MS_CHARS.get(orig) + if type(sub) == tuple: + if self.smart_quotes_to == 'xml': + sub = '&#x'.encode() + sub[1].encode() + ';'.encode() + else: + sub = '&'.encode() + sub[0].encode() + ';'.encode() + else: + sub = sub.encode() + return sub + + def _convert_from(self, proposed, errors="strict"): + proposed = self.find_codec(proposed) + if not proposed or (proposed, errors) in self.tried_encodings: + return None + self.tried_encodings.append((proposed, errors)) + markup = self.markup + # Convert smart quotes to HTML if coming from an encoding + # that might have them. + if (self.smart_quotes_to is not None + and proposed in self.ENCODINGS_WITH_SMART_QUOTES): + smart_quotes_re = b"([\x80-\x9f])" + smart_quotes_compiled = re.compile(smart_quotes_re) + markup = smart_quotes_compiled.sub(self._sub_ms_char, markup) + + try: + #print "Trying to convert document to %s (errors=%s)" % ( + # proposed, errors) + u = self._to_unicode(markup, proposed, errors) + self.markup = u + self.original_encoding = proposed + except Exception as e: + #print "That didn't work!" + #print e + return None + #print "Correct encoding: %s" % proposed + return self.markup + + def _to_unicode(self, data, encoding, errors="strict"): + '''Given a string and its encoding, decodes the string into Unicode. + %encoding is a string recognized by encodings.aliases''' + return str(data, encoding, errors) + + @property + def declared_html_encoding(self): + if not self.is_html: + return None + return self.detector.declared_encoding + + def find_codec(self, charset): + value = (self._codec(self.CHARSET_ALIASES.get(charset, charset)) + or (charset and self._codec(charset.replace("-", ""))) + or (charset and self._codec(charset.replace("-", "_"))) + or (charset and charset.lower()) + or charset + ) + if value: + return value.lower() + return None + + def _codec(self, charset): + if not charset: + return charset + codec = None + try: + codecs.lookup(charset) + codec = charset + except (LookupError, ValueError): + pass + return codec + + + # A partial mapping of ISO-Latin-1 to HTML entities/XML numeric entities. + MS_CHARS = {b'\x80': ('euro', '20AC'), + b'\x81': ' ', + b'\x82': ('sbquo', '201A'), + b'\x83': ('fnof', '192'), + b'\x84': ('bdquo', '201E'), + b'\x85': ('hellip', '2026'), + b'\x86': ('dagger', '2020'), + b'\x87': ('Dagger', '2021'), + b'\x88': ('circ', '2C6'), + b'\x89': ('permil', '2030'), + b'\x8A': ('Scaron', '160'), + b'\x8B': ('lsaquo', '2039'), + b'\x8C': ('OElig', '152'), + b'\x8D': '?', + b'\x8E': ('#x17D', '17D'), + b'\x8F': '?', + b'\x90': '?', + b'\x91': ('lsquo', '2018'), + b'\x92': ('rsquo', '2019'), + b'\x93': ('ldquo', '201C'), + b'\x94': ('rdquo', '201D'), + b'\x95': ('bull', '2022'), + b'\x96': ('ndash', '2013'), + b'\x97': ('mdash', '2014'), + b'\x98': ('tilde', '2DC'), + b'\x99': ('trade', '2122'), + b'\x9a': ('scaron', '161'), + b'\x9b': ('rsaquo', '203A'), + b'\x9c': ('oelig', '153'), + b'\x9d': '?', + b'\x9e': ('#x17E', '17E'), + b'\x9f': ('Yuml', ''),} + + # A parochial partial mapping of ISO-Latin-1 to ASCII. Contains + # horrors like stripping diacritical marks to turn á into a, but also + # contains non-horrors like turning “ into ". + MS_CHARS_TO_ASCII = { + b'\x80' : 'EUR', + b'\x81' : ' ', + b'\x82' : ',', + b'\x83' : 'f', + b'\x84' : ',,', + b'\x85' : '...', + b'\x86' : '+', + b'\x87' : '++', + b'\x88' : '^', + b'\x89' : '%', + b'\x8a' : 'S', + b'\x8b' : '<', + b'\x8c' : 'OE', + b'\x8d' : '?', + b'\x8e' : 'Z', + b'\x8f' : '?', + b'\x90' : '?', + b'\x91' : "'", + b'\x92' : "'", + b'\x93' : '"', + b'\x94' : '"', + b'\x95' : '*', + b'\x96' : '-', + b'\x97' : '--', + b'\x98' : '~', + b'\x99' : '(TM)', + b'\x9a' : 's', + b'\x9b' : '>', + b'\x9c' : 'oe', + b'\x9d' : '?', + b'\x9e' : 'z', + b'\x9f' : 'Y', + b'\xa0' : ' ', + b'\xa1' : '!', + b'\xa2' : 'c', + b'\xa3' : 'GBP', + b'\xa4' : '$', #This approximation is especially parochial--this is the + #generic currency symbol. + b'\xa5' : 'YEN', + b'\xa6' : '|', + b'\xa7' : 'S', + b'\xa8' : '..', + b'\xa9' : '', + b'\xaa' : '(th)', + b'\xab' : '<<', + b'\xac' : '!', + b'\xad' : ' ', + b'\xae' : '(R)', + b'\xaf' : '-', + b'\xb0' : 'o', + b'\xb1' : '+-', + b'\xb2' : '2', + b'\xb3' : '3', + b'\xb4' : ("'", 'acute'), + b'\xb5' : 'u', + b'\xb6' : 'P', + b'\xb7' : '*', + b'\xb8' : ',', + b'\xb9' : '1', + b'\xba' : '(th)', + b'\xbb' : '>>', + b'\xbc' : '1/4', + b'\xbd' : '1/2', + b'\xbe' : '3/4', + b'\xbf' : '?', + b'\xc0' : 'A', + b'\xc1' : 'A', + b'\xc2' : 'A', + b'\xc3' : 'A', + b'\xc4' : 'A', + b'\xc5' : 'A', + b'\xc6' : 'AE', + b'\xc7' : 'C', + b'\xc8' : 'E', + b'\xc9' : 'E', + b'\xca' : 'E', + b'\xcb' : 'E', + b'\xcc' : 'I', + b'\xcd' : 'I', + b'\xce' : 'I', + b'\xcf' : 'I', + b'\xd0' : 'D', + b'\xd1' : 'N', + b'\xd2' : 'O', + b'\xd3' : 'O', + b'\xd4' : 'O', + b'\xd5' : 'O', + b'\xd6' : 'O', + b'\xd7' : '*', + b'\xd8' : 'O', + b'\xd9' : 'U', + b'\xda' : 'U', + b'\xdb' : 'U', + b'\xdc' : 'U', + b'\xdd' : 'Y', + b'\xde' : 'b', + b'\xdf' : 'B', + b'\xe0' : 'a', + b'\xe1' : 'a', + b'\xe2' : 'a', + b'\xe3' : 'a', + b'\xe4' : 'a', + b'\xe5' : 'a', + b'\xe6' : 'ae', + b'\xe7' : 'c', + b'\xe8' : 'e', + b'\xe9' : 'e', + b'\xea' : 'e', + b'\xeb' : 'e', + b'\xec' : 'i', + b'\xed' : 'i', + b'\xee' : 'i', + b'\xef' : 'i', + b'\xf0' : 'o', + b'\xf1' : 'n', + b'\xf2' : 'o', + b'\xf3' : 'o', + b'\xf4' : 'o', + b'\xf5' : 'o', + b'\xf6' : 'o', + b'\xf7' : '/', + b'\xf8' : 'o', + b'\xf9' : 'u', + b'\xfa' : 'u', + b'\xfb' : 'u', + b'\xfc' : 'u', + b'\xfd' : 'y', + b'\xfe' : 'b', + b'\xff' : 'y', + } + + # A map used when removing rogue Windows-1252/ISO-8859-1 + # characters in otherwise UTF-8 documents. + # + # Note that \x81, \x8d, \x8f, \x90, and \x9d are undefined in + # Windows-1252. + WINDOWS_1252_TO_UTF8 = { + 0x80 : b'\xe2\x82\xac', # € + 0x82 : b'\xe2\x80\x9a', # ‚ + 0x83 : b'\xc6\x92', # ƒ + 0x84 : b'\xe2\x80\x9e', # „ + 0x85 : b'\xe2\x80\xa6', # … + 0x86 : b'\xe2\x80\xa0', # † + 0x87 : b'\xe2\x80\xa1', # ‡ + 0x88 : b'\xcb\x86', # ˆ + 0x89 : b'\xe2\x80\xb0', # ‰ + 0x8a : b'\xc5\xa0', # Š + 0x8b : b'\xe2\x80\xb9', # ‹ + 0x8c : b'\xc5\x92', # Œ + 0x8e : b'\xc5\xbd', # Ž + 0x91 : b'\xe2\x80\x98', # ‘ + 0x92 : b'\xe2\x80\x99', # ’ + 0x93 : b'\xe2\x80\x9c', # “ + 0x94 : b'\xe2\x80\x9d', # ” + 0x95 : b'\xe2\x80\xa2', # • + 0x96 : b'\xe2\x80\x93', # – + 0x97 : b'\xe2\x80\x94', # — + 0x98 : b'\xcb\x9c', # ˜ + 0x99 : b'\xe2\x84\xa2', # ™ + 0x9a : b'\xc5\xa1', # š + 0x9b : b'\xe2\x80\xba', # › + 0x9c : b'\xc5\x93', # œ + 0x9e : b'\xc5\xbe', # ž + 0x9f : b'\xc5\xb8', # Ÿ + 0xa0 : b'\xc2\xa0', #   + 0xa1 : b'\xc2\xa1', # ¡ + 0xa2 : b'\xc2\xa2', # ¢ + 0xa3 : b'\xc2\xa3', # £ + 0xa4 : b'\xc2\xa4', # ¤ + 0xa5 : b'\xc2\xa5', # ¥ + 0xa6 : b'\xc2\xa6', # ¦ + 0xa7 : b'\xc2\xa7', # § + 0xa8 : b'\xc2\xa8', # ¨ + 0xa9 : b'\xc2\xa9', # © + 0xaa : b'\xc2\xaa', # ª + 0xab : b'\xc2\xab', # « + 0xac : b'\xc2\xac', # ¬ + 0xad : b'\xc2\xad', # ­ + 0xae : b'\xc2\xae', # ® + 0xaf : b'\xc2\xaf', # ¯ + 0xb0 : b'\xc2\xb0', # ° + 0xb1 : b'\xc2\xb1', # ± + 0xb2 : b'\xc2\xb2', # ² + 0xb3 : b'\xc2\xb3', # ³ + 0xb4 : b'\xc2\xb4', # ´ + 0xb5 : b'\xc2\xb5', # µ + 0xb6 : b'\xc2\xb6', # ¶ + 0xb7 : b'\xc2\xb7', # · + 0xb8 : b'\xc2\xb8', # ¸ + 0xb9 : b'\xc2\xb9', # ¹ + 0xba : b'\xc2\xba', # º + 0xbb : b'\xc2\xbb', # » + 0xbc : b'\xc2\xbc', # ¼ + 0xbd : b'\xc2\xbd', # ½ + 0xbe : b'\xc2\xbe', # ¾ + 0xbf : b'\xc2\xbf', # ¿ + 0xc0 : b'\xc3\x80', # À + 0xc1 : b'\xc3\x81', # Á + 0xc2 : b'\xc3\x82', #  + 0xc3 : b'\xc3\x83', # à + 0xc4 : b'\xc3\x84', # Ä + 0xc5 : b'\xc3\x85', # Å + 0xc6 : b'\xc3\x86', # Æ + 0xc7 : b'\xc3\x87', # Ç + 0xc8 : b'\xc3\x88', # È + 0xc9 : b'\xc3\x89', # É + 0xca : b'\xc3\x8a', # Ê + 0xcb : b'\xc3\x8b', # Ë + 0xcc : b'\xc3\x8c', # Ì + 0xcd : b'\xc3\x8d', # Í + 0xce : b'\xc3\x8e', # Î + 0xcf : b'\xc3\x8f', # Ï + 0xd0 : b'\xc3\x90', # Ð + 0xd1 : b'\xc3\x91', # Ñ + 0xd2 : b'\xc3\x92', # Ò + 0xd3 : b'\xc3\x93', # Ó + 0xd4 : b'\xc3\x94', # Ô + 0xd5 : b'\xc3\x95', # Õ + 0xd6 : b'\xc3\x96', # Ö + 0xd7 : b'\xc3\x97', # × + 0xd8 : b'\xc3\x98', # Ø + 0xd9 : b'\xc3\x99', # Ù + 0xda : b'\xc3\x9a', # Ú + 0xdb : b'\xc3\x9b', # Û + 0xdc : b'\xc3\x9c', # Ü + 0xdd : b'\xc3\x9d', # Ý + 0xde : b'\xc3\x9e', # Þ + 0xdf : b'\xc3\x9f', # ß + 0xe0 : b'\xc3\xa0', # à + 0xe1 : b'\xa1', # á + 0xe2 : b'\xc3\xa2', # â + 0xe3 : b'\xc3\xa3', # ã + 0xe4 : b'\xc3\xa4', # ä + 0xe5 : b'\xc3\xa5', # å + 0xe6 : b'\xc3\xa6', # æ + 0xe7 : b'\xc3\xa7', # ç + 0xe8 : b'\xc3\xa8', # è + 0xe9 : b'\xc3\xa9', # é + 0xea : b'\xc3\xaa', # ê + 0xeb : b'\xc3\xab', # ë + 0xec : b'\xc3\xac', # ì + 0xed : b'\xc3\xad', # í + 0xee : b'\xc3\xae', # î + 0xef : b'\xc3\xaf', # ï + 0xf0 : b'\xc3\xb0', # ð + 0xf1 : b'\xc3\xb1', # ñ + 0xf2 : b'\xc3\xb2', # ò + 0xf3 : b'\xc3\xb3', # ó + 0xf4 : b'\xc3\xb4', # ô + 0xf5 : b'\xc3\xb5', # õ + 0xf6 : b'\xc3\xb6', # ö + 0xf7 : b'\xc3\xb7', # ÷ + 0xf8 : b'\xc3\xb8', # ø + 0xf9 : b'\xc3\xb9', # ù + 0xfa : b'\xc3\xba', # ú + 0xfb : b'\xc3\xbb', # û + 0xfc : b'\xc3\xbc', # ü + 0xfd : b'\xc3\xbd', # ý + 0xfe : b'\xc3\xbe', # þ + } + + MULTIBYTE_MARKERS_AND_SIZES = [ + (0xc2, 0xdf, 2), # 2-byte characters start with a byte C2-DF + (0xe0, 0xef, 3), # 3-byte characters start with E0-EF + (0xf0, 0xf4, 4), # 4-byte characters start with F0-F4 + ] + + FIRST_MULTIBYTE_MARKER = MULTIBYTE_MARKERS_AND_SIZES[0][0] + LAST_MULTIBYTE_MARKER = MULTIBYTE_MARKERS_AND_SIZES[-1][1] + + @classmethod + def detwingle(cls, in_bytes, main_encoding="utf8", + embedded_encoding="windows-1252"): + """Fix characters from one encoding embedded in some other encoding. + + Currently the only situation supported is Windows-1252 (or its + subset ISO-8859-1), embedded in UTF-8. + + The input must be a bytestring. If you've already converted + the document to Unicode, you're too late. + + The output is a bytestring in which `embedded_encoding` + characters have been converted to their `main_encoding` + equivalents. + """ + if embedded_encoding.replace('_', '-').lower() not in ( + 'windows-1252', 'windows_1252'): + raise NotImplementedError( + "Windows-1252 and ISO-8859-1 are the only currently supported " + "embedded encodings.") + + if main_encoding.lower() not in ('utf8', 'utf-8'): + raise NotImplementedError( + "UTF-8 is the only currently supported main encoding.") + + byte_chunks = [] + + chunk_start = 0 + pos = 0 + while pos < len(in_bytes): + byte = in_bytes[pos] + if not isinstance(byte, int): + # Python 2.x + byte = ord(byte) + if (byte >= cls.FIRST_MULTIBYTE_MARKER + and byte <= cls.LAST_MULTIBYTE_MARKER): + # This is the start of a UTF-8 multibyte character. Skip + # to the end. + for start, end, size in cls.MULTIBYTE_MARKERS_AND_SIZES: + if byte >= start and byte <= end: + pos += size + break + elif byte >= 0x80 and byte in cls.WINDOWS_1252_TO_UTF8: + # We found a Windows-1252 character! + # Save the string up to this point as a chunk. + byte_chunks.append(in_bytes[chunk_start:pos]) + + # Now translate the Windows-1252 character into UTF-8 + # and add it as another, one-byte chunk. + byte_chunks.append(cls.WINDOWS_1252_TO_UTF8[byte]) + pos += 1 + chunk_start = pos + else: + # Go on to the next character. + pos += 1 + if chunk_start == 0: + # The string is unchanged. + return in_bytes + else: + # Store the final chunk. + byte_chunks.append(in_bytes[chunk_start:]) + return b''.join(byte_chunks) + diff --git a/cleanerupper/bs4/diagnose.py b/cleanerupper/bs4/diagnose.py new file mode 100644 index 0000000..1254861 --- /dev/null +++ b/cleanerupper/bs4/diagnose.py @@ -0,0 +1,219 @@ +"""Diagnostic functions, mainly for use when doing tech support.""" + +# Use of this source code is governed by a BSD-style license that can be +# found in the LICENSE file. +__license__ = "MIT" + +import cProfile +from io import StringIO +from html.parser import HTMLParser +import bs4 +from bs4 import BeautifulSoup, __version__ +from bs4.builder import builder_registry + +import os +import pstats +import random +import tempfile +import time +import traceback +import sys +import cProfile + +def diagnose(data): + """Diagnostic suite for isolating common problems.""" + print("Diagnostic running on Beautiful Soup %s" % __version__) + print("Python version %s" % sys.version) + + basic_parsers = ["html.parser", "html5lib", "lxml"] + for name in basic_parsers: + for builder in builder_registry.builders: + if name in builder.features: + break + else: + basic_parsers.remove(name) + print(( + "I noticed that %s is not installed. Installing it may help." % + name)) + + if 'lxml' in basic_parsers: + basic_parsers.append(["lxml", "xml"]) + try: + from lxml import etree + print("Found lxml version %s" % ".".join(map(str,etree.LXML_VERSION))) + except ImportError as e: + print ( + "lxml is not installed or couldn't be imported.") + + + if 'html5lib' in basic_parsers: + try: + import html5lib + print("Found html5lib version %s" % html5lib.__version__) + except ImportError as e: + print ( + "html5lib is not installed or couldn't be imported.") + + if hasattr(data, 'read'): + data = data.read() + elif os.path.exists(data): + print('"%s" looks like a filename. Reading data from the file.' % data) + with open(data) as fp: + data = fp.read() + elif data.startswith("http:") or data.startswith("https:"): + print('"%s" looks like a URL. Beautiful Soup is not an HTTP client.' % data) + print("You need to use some other library to get the document behind the URL, and feed that document to Beautiful Soup.") + return + print() + + for parser in basic_parsers: + print("Trying to parse your markup with %s" % parser) + success = False + try: + soup = BeautifulSoup(data, parser) + success = True + except Exception as e: + print("%s could not parse the markup." % parser) + traceback.print_exc() + if success: + print("Here's what %s did with the markup:" % parser) + print(soup.prettify()) + + print("-" * 80) + +def lxml_trace(data, html=True, **kwargs): + """Print out the lxml events that occur during parsing. + + This lets you see how lxml parses a document when no Beautiful + Soup code is running. + """ + from lxml import etree + for event, element in etree.iterparse(StringIO(data), html=html, **kwargs): + print(("%s, %4s, %s" % (event, element.tag, element.text))) + +class AnnouncingParser(HTMLParser): + """Announces HTMLParser parse events, without doing anything else.""" + + def _p(self, s): + print(s) + + def handle_starttag(self, name, attrs): + self._p("%s START" % name) + + def handle_endtag(self, name): + self._p("%s END" % name) + + def handle_data(self, data): + self._p("%s DATA" % data) + + def handle_charref(self, name): + self._p("%s CHARREF" % name) + + def handle_entityref(self, name): + self._p("%s ENTITYREF" % name) + + def handle_comment(self, data): + self._p("%s COMMENT" % data) + + def handle_decl(self, data): + self._p("%s DECL" % data) + + def unknown_decl(self, data): + self._p("%s UNKNOWN-DECL" % data) + + def handle_pi(self, data): + self._p("%s PI" % data) + +def htmlparser_trace(data): + """Print out the HTMLParser events that occur during parsing. + + This lets you see how HTMLParser parses a document when no + Beautiful Soup code is running. + """ + parser = AnnouncingParser() + parser.feed(data) + +_vowels = "aeiou" +_consonants = "bcdfghjklmnpqrstvwxyz" + +def rword(length=5): + "Generate a random word-like string." + s = '' + for i in range(length): + if i % 2 == 0: + t = _consonants + else: + t = _vowels + s += random.choice(t) + return s + +def rsentence(length=4): + "Generate a random sentence-like string." + return " ".join(rword(random.randint(4,9)) for i in range(length)) + +def rdoc(num_elements=1000): + """Randomly generate an invalid HTML document.""" + tag_names = ['p', 'div', 'span', 'i', 'b', 'script', 'table'] + elements = [] + for i in range(num_elements): + choice = random.randint(0,3) + if choice == 0: + # New tag. + tag_name = random.choice(tag_names) + elements.append("<%s>" % tag_name) + elif choice == 1: + elements.append(rsentence(random.randint(1,4))) + elif choice == 2: + # Close a tag. + tag_name = random.choice(tag_names) + elements.append("" % tag_name) + return "" + "\n".join(elements) + "" + +def benchmark_parsers(num_elements=100000): + """Very basic head-to-head performance benchmark.""" + print("Comparative parser benchmark on Beautiful Soup %s" % __version__) + data = rdoc(num_elements) + print("Generated a large invalid HTML document (%d bytes)." % len(data)) + + for parser in ["lxml", ["lxml", "html"], "html5lib", "html.parser"]: + success = False + try: + a = time.time() + soup = BeautifulSoup(data, parser) + b = time.time() + success = True + except Exception as e: + print("%s could not parse the markup." % parser) + traceback.print_exc() + if success: + print("BS4+%s parsed the markup in %.2fs." % (parser, b-a)) + + from lxml import etree + a = time.time() + etree.HTML(data) + b = time.time() + print("Raw lxml parsed the markup in %.2fs." % (b-a)) + + import html5lib + parser = html5lib.HTMLParser() + a = time.time() + parser.parse(data) + b = time.time() + print("Raw html5lib parsed the markup in %.2fs." % (b-a)) + +def profile(num_elements=100000, parser="lxml"): + + filehandle = tempfile.NamedTemporaryFile() + filename = filehandle.name + + data = rdoc(num_elements) + vars = dict(bs4=bs4, data=data, parser=parser) + cProfile.runctx('bs4.BeautifulSoup(data, parser)' , vars, vars, filename) + + stats = pstats.Stats(filename) + # stats.strip_dirs() + stats.sort_stats("cumulative") + stats.print_stats('_html5lib|bs4', 50) + +if __name__ == '__main__': + diagnose(sys.stdin.read()) diff --git a/cleanerupper/bs4/element.py b/cleanerupper/bs4/element.py new file mode 100644 index 0000000..a4a750d --- /dev/null +++ b/cleanerupper/bs4/element.py @@ -0,0 +1,1808 @@ +# Use of this source code is governed by a BSD-style license that can be +# found in the LICENSE file. +__license__ = "MIT" + +import collections +import re +import shlex +import sys +import warnings +from bs4.dammit import EntitySubstitution + +DEFAULT_OUTPUT_ENCODING = "utf-8" +PY3K = (sys.version_info[0] > 2) + +whitespace_re = re.compile("\s+") + +def _alias(attr): + """Alias one attribute name to another for backward compatibility""" + @property + def alias(self): + return getattr(self, attr) + + @alias.setter + def alias(self): + return setattr(self, attr) + return alias + + +class NamespacedAttribute(str): + + def __new__(cls, prefix, name, namespace=None): + if name is None: + obj = str.__new__(cls, prefix) + elif prefix is None: + # Not really namespaced. + obj = str.__new__(cls, name) + else: + obj = str.__new__(cls, prefix + ":" + name) + obj.prefix = prefix + obj.name = name + obj.namespace = namespace + return obj + +class AttributeValueWithCharsetSubstitution(str): + """A stand-in object for a character encoding specified in HTML.""" + +class CharsetMetaAttributeValue(AttributeValueWithCharsetSubstitution): + """A generic stand-in for the value of a meta tag's 'charset' attribute. + + When Beautiful Soup parses the markup '', the + value of the 'charset' attribute will be one of these objects. + """ + + def __new__(cls, original_value): + obj = str.__new__(cls, original_value) + obj.original_value = original_value + return obj + + def encode(self, encoding): + return encoding + + +class ContentMetaAttributeValue(AttributeValueWithCharsetSubstitution): + """A generic stand-in for the value of a meta tag's 'content' attribute. + + When Beautiful Soup parses the markup: + + + The value of the 'content' attribute will be one of these objects. + """ + + CHARSET_RE = re.compile("((^|;)\s*charset=)([^;]*)", re.M) + + def __new__(cls, original_value): + match = cls.CHARSET_RE.search(original_value) + if match is None: + # No substitution necessary. + return str.__new__(str, original_value) + + obj = str.__new__(cls, original_value) + obj.original_value = original_value + return obj + + def encode(self, encoding): + def rewrite(match): + return match.group(1) + encoding + return self.CHARSET_RE.sub(rewrite, self.original_value) + +class HTMLAwareEntitySubstitution(EntitySubstitution): + + """Entity substitution rules that are aware of some HTML quirks. + + Specifically, the contents of + +Hello, world! + + +''' + soup = self.soup(html) + self.assertEqual("text/javascript", soup.find('script')['type']) + + def test_comment(self): + # Comments are represented as Comment objects. + markup = "

foobaz

" + self.assertSoupEquals(markup) + + soup = self.soup(markup) + comment = soup.find(text="foobar") + self.assertEqual(comment.__class__, Comment) + + # The comment is properly integrated into the tree. + foo = soup.find(text="foo") + self.assertEqual(comment, foo.next_element) + baz = soup.find(text="baz") + self.assertEqual(comment, baz.previous_element) + + def test_preserved_whitespace_in_pre_and_textarea(self): + """Whitespace must be preserved in 
 and "
+        self.assertSoupEquals(pre_markup)
+        self.assertSoupEquals(textarea_markup)
+
+        soup = self.soup(pre_markup)
+        self.assertEqual(soup.pre.prettify(), pre_markup)
+
+        soup = self.soup(textarea_markup)
+        self.assertEqual(soup.textarea.prettify(), textarea_markup)
+
+        soup = self.soup("")
+        self.assertEqual(soup.textarea.prettify(), "")
+
+    def test_nested_inline_elements(self):
+        """Inline elements can be nested indefinitely."""
+        b_tag = "Inside a B tag"
+        self.assertSoupEquals(b_tag)
+
+        nested_b_tag = "
A nested tag
"
+        self.assertSoupEquals(nested_b_tag)
+
+        double_nested_b_tag = "
A doubly nested tag
"
+        self.assertSoupEquals(nested_b_tag)
+
+    def test_nested_block_level_elements(self):
+        """Block elements can be nested."""
+        soup = self.soup('
Foo
')
+        blockquote = soup.blockquote
+        self.assertEqual(blockquote.p.b.string, 'Foo')
+        self.assertEqual(blockquote.b.string, 'Foo')
+
+    def test_correctly_nested_tables(self):
+        """One table can go inside another one."""
+        markup = (''
+                  ''
+                  "')
+
+        self.assertSoupEquals(
+            markup,
+            '
Here's another table:"
+                  ''
+                  ''
+                  '
foo
Here\'s another table:'
+            '
foo
'
+            '
')
+
+        self.assertSoupEquals(
+            ""
+            ""
+            "
Foo
Bar
Baz
")
+
+    def test_deeply_nested_multivalued_attribute(self):
+        # html5lib can set the attributes of the same tag many times
+        # as it rearranges the tree. This has caused problems with
+        # multivalued attributes.
+        markup = '
'
+        soup = self.soup(markup)
+        self.assertEqual(["css"], soup.div.div['class'])
+
+    def test_multivalued_attribute_on_html(self):
+        # html5lib uses a different API to set the attributes ot the
+        #  tag. This has caused problems with multivalued
+        # attributes.
+        markup = ''
+        soup = self.soup(markup)
+        self.assertEqual(["a", "b"], soup.html['class'])
+
+    def test_angle_brackets_in_attribute_values_are_escaped(self):
+        self.assertSoupEquals('', '')
+
+    def test_entities_in_attributes_converted_to_unicode(self):
+        expect = '
'
+        self.assertSoupEquals('
', expect)
+        self.assertSoupEquals('
', expect)
+        self.assertSoupEquals('
', expect)
+        self.assertSoupEquals('
', expect)
+
+    def test_entities_in_text_converted_to_unicode(self):
+        expect = '
pi\N{LATIN SMALL LETTER N WITH TILDE}ata
'
+        self.assertSoupEquals("
piñata
", expect)
+        self.assertSoupEquals("
piñata
", expect)
+        self.assertSoupEquals("
piñata
", expect)
+        self.assertSoupEquals("
piñata
", expect)
+
+    def test_quot_entity_converted_to_quotation_mark(self):
+        self.assertSoupEquals("
I said "good day!"
",
+                              '
I said "good day!"
')
+
+    def test_out_of_range_entity(self):
+        expect = "\N{REPLACEMENT CHARACTER}"
+        self.assertSoupEquals("�", expect)
+        self.assertSoupEquals("�", expect)
+        self.assertSoupEquals("�", expect)
+
+    def test_multipart_strings(self):
+        "Mostly to prevent a recurrence of a bug in the html5lib treebuilder."
+        soup = self.soup("
\nfoo
")
+        self.assertEqual("p", soup.h2.string.next_element.name)
+        self.assertEqual("p", soup.p.name)
+        self.assertConnectedness(soup)
+
+    def test_empty_element_tags(self):
+        """Verify consistent handling of empty-element tags,
+        no matter how they come in through the markup.
+        """
+        self.assertSoupEquals('


', "


")
+        self.assertSoupEquals('


', "


")
+        
+    def test_head_tag_between_head_and_body(self):
+        "Prevent recurrence of a bug in the html5lib treebuilder."
+        content = """
+  
+  foo
+
+"""
+        soup = self.soup(content)
+        self.assertNotEqual(None, soup.html.body)
+        self.assertConnectedness(soup)
+
+    def test_multiple_copies_of_a_tag(self):
+        "Prevent recurrence of a bug in the html5lib treebuilder."
+        content = """
+
+ 
+   
+  
+
+"""
+        soup = self.soup(content)
+        self.assertConnectedness(soup.article)
+
+    def test_basic_namespaces(self):
+        """Parsers don't need to *understand* namespaces, but at the
+        very least they should not choke on namespaces or lose
+        data."""
+
+        markup = b'4'
+        soup = self.soup(markup)
+        self.assertEqual(markup, soup.encode())
+        html = soup.html
+        self.assertEqual('http://www.w3.org/1999/xhtml', soup.html['xmlns'])
+        self.assertEqual(
+            'http://www.w3.org/1998/Math/MathML', soup.html['xmlns:mathml'])
+        self.assertEqual(
+            'http://www.w3.org/2000/svg', soup.html['xmlns:svg'])
+
+    def test_multivalued_attribute_value_becomes_list(self):
+        markup = b''
+        soup = self.soup(markup)
+        self.assertEqual(['foo', 'bar'], soup.a['class'])
+
+    #
+    # Generally speaking, tests below this point are more tests of
+    # Beautiful Soup than tests of the tree builders. But parsers are
+    # weird, so we run these tests separately for every tree builder
+    # to detect any differences between them.
+    #
+
+    def test_can_parse_unicode_document(self):
+        # A seemingly innocuous document... but it's in Unicode! And
+        # it contains characters that can't be represented in the
+        # encoding found in the  declaration! The horror!
+        markup = 'Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!'
+        soup = self.soup(markup)
+        self.assertEqual('Sacr\xe9 bleu!', soup.body.string)
+
+    def test_soupstrainer(self):
+        """Parsers should be able to work with SoupStrainers."""
+        strainer = SoupStrainer("b")
+        soup = self.soup("A bold  statement",
+                         parse_only=strainer)
+        self.assertEqual(soup.decode(), "bold")
+
+    def test_single_quote_attribute_values_become_double_quotes(self):
+        self.assertSoupEquals("",
+                              '')
+
+    def test_attribute_values_with_nested_quotes_are_left_alone(self):
+        text = """a"""
+        self.assertSoupEquals(text)
+
+    def test_attribute_values_with_double_nested_quotes_get_quoted(self):
+        text = """a"""
+        soup = self.soup(text)
+        soup.foo['attr'] = 'Brawls happen at "Bob\'s Bar"'
+        self.assertSoupEquals(
+            soup.foo.decode(),
+            """a""")
+
+    def test_ampersand_in_attribute_value_gets_escaped(self):
+        self.assertSoupEquals('',
+                              '')
+
+        self.assertSoupEquals(
+            'foo',
+            'foo')
+
+    def test_escaped_ampersand_in_attribute_value_is_left_alone(self):
+        self.assertSoupEquals('')
+
+    def test_entities_in_strings_converted_during_parsing(self):
+        # Both XML and HTML entities are converted to Unicode characters
+        # during parsing.
+        text = "
<<sacré bleu!>>
"
+        expected = "
<<sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>>
"
+        self.assertSoupEquals(text, expected)
+
+    def test_smart_quotes_converted_on_the_way_in(self):
+        # Microsoft smart quotes are converted to Unicode characters during
+        # parsing.
+        quote = b"
\x91Foo\x92
"
+        soup = self.soup(quote)
+        self.assertEqual(
+            soup.p.string,
+            "\N{LEFT SINGLE QUOTATION MARK}Foo\N{RIGHT SINGLE QUOTATION MARK}")
+
+    def test_non_breaking_spaces_converted_on_the_way_in(self):
+        soup = self.soup("  ")
+        self.assertEqual(soup.a.string, "\N{NO-BREAK SPACE}" * 2)
+
+    def test_entities_converted_on_the_way_out(self):
+        text = "
<<sacré bleu!>>
"
+        expected = "
<<sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!>>
".encode("utf-8")
+        soup = self.soup(text)
+        self.assertEqual(soup.p.encode("utf-8"), expected)
+
+    def test_real_iso_latin_document(self):
+        # Smoke test of interrelated functionality, using an
+        # easy-to-understand document.
+
+        # Here it is in Unicode. Note that it claims to be in ISO-Latin-1.
+        unicode_html = '
Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!
'
+
+        # That's because we're going to encode it into ISO-Latin-1, and use
+        # that to test.
+        iso_latin_html = unicode_html.encode("iso-8859-1")
+
+        # Parse the ISO-Latin-1 HTML.
+        soup = self.soup(iso_latin_html)
+        # Encode it to UTF-8.
+        result = soup.encode("utf-8")
+
+        # What do we expect the result to look like? Well, it would
+        # look like unicode_html, except that the META tag would say
+        # UTF-8 instead of ISO-Latin-1.
+        expected = unicode_html.replace("ISO-Latin-1", "utf-8")
+
+        # And, of course, it would be in UTF-8, not Unicode.
+        expected = expected.encode("utf-8")
+
+        # Ta-da!
+        self.assertEqual(result, expected)
+
+    def test_real_shift_jis_document(self):
+        # Smoke test to make sure the parser can handle a document in
+        # Shift-JIS encoding, without choking.
+        shift_jis_html = (
+            b'
'
+            b'\x82\xb1\x82\xea\x82\xcdShift-JIS\x82\xc5\x83R\x81[\x83f'
+            b'\x83B\x83\x93\x83O\x82\xb3\x82\xea\x82\xbd\x93\xfa\x96{\x8c'
+            b'\xea\x82\xcc\x83t\x83@\x83C\x83\x8b\x82\xc5\x82\xb7\x81B'
+            b'
')
+        unicode_html = shift_jis_html.decode("shift-jis")
+        soup = self.soup(unicode_html)
+
+        # Make sure the parse tree is correctly encoded to various
+        # encodings.
+        self.assertEqual(soup.encode("utf-8"), unicode_html.encode("utf-8"))
+        self.assertEqual(soup.encode("euc_jp"), unicode_html.encode("euc_jp"))
+
+    def test_real_hebrew_document(self):
+        # A real-world test to make sure we can convert ISO-8859-9 (a
+        # Hebrew encoding) to UTF-8.
+        hebrew_document = b'Hebrew (ISO 8859-8) in Visual Directionality
Hebrew (ISO 8859-8) in Visual Directionality
\xed\xe5\xec\xf9'
+        soup = self.soup(
+            hebrew_document, from_encoding="iso8859-8")
+        # Some tree builders call it iso8859-8, others call it iso-8859-9.
+        # That's not a difference we really care about.
+        assert soup.original_encoding in ('iso8859-8', 'iso-8859-8')
+        self.assertEqual(
+            soup.encode('utf-8'),
+            hebrew_document.decode("iso8859-8").encode("utf-8"))
+
+    def test_meta_tag_reflects_current_encoding(self):
+        # Here's the  tag saying that a document is
+        # encoded in Shift-JIS.
+        meta_tag = ('')
+
+        # Here's a document incorporating that meta tag.
+        shift_jis_html = (
+            '\n%s\n'
+            ''
+            'Shift-JIS markup goes here.') % meta_tag
+        soup = self.soup(shift_jis_html)
+
+        # Parse the document, and the charset is seemingly unaffected.
+        parsed_meta = soup.find('meta', {'http-equiv': 'Content-type'})
+        content = parsed_meta['content']
+        self.assertEqual('text/html; charset=x-sjis', content)
+
+        # But that value is actually a ContentMetaAttributeValue object.
+        self.assertTrue(isinstance(content, ContentMetaAttributeValue))
+
+        # And it will take on a value that reflects its current
+        # encoding.
+        self.assertEqual('text/html; charset=utf8', content.encode("utf8"))
+
+        # For the rest of the story, see TestSubstitutions in
+        # test_tree.py.
+
+    def test_html5_style_meta_tag_reflects_current_encoding(self):
+        # Here's the  tag saying that a document is
+        # encoded in Shift-JIS.
+        meta_tag = ('')
+
+        # Here's a document incorporating that meta tag.
+        shift_jis_html = (
+            '\n%s\n'
+            ''
+            'Shift-JIS markup goes here.') % meta_tag
+        soup = self.soup(shift_jis_html)
+
+        # Parse the document, and the charset is seemingly unaffected.
+        parsed_meta = soup.find('meta', id="encoding")
+        charset = parsed_meta['charset']
+        self.assertEqual('x-sjis', charset)
+
+        # But that value is actually a CharsetMetaAttributeValue object.
+        self.assertTrue(isinstance(charset, CharsetMetaAttributeValue))
+
+        # And it will take on a value that reflects its current
+        # encoding.
+        self.assertEqual('utf8', charset.encode("utf8"))
+
+    def test_tag_with_no_attributes_can_have_attributes_added(self):
+        data = self.soup("text")
+        data.a['foo'] = 'bar'
+        self.assertEqual('text', data.a.decode())
+
+class XMLTreeBuilderSmokeTest(object):
+
+    def test_pickle_and_unpickle_identity(self):
+        # Pickling a tree, then unpickling it, yields a tree identical
+        # to the original.
+        tree = self.soup("foo")
+        dumped = pickle.dumps(tree, 2)
+        loaded = pickle.loads(dumped)
+        self.assertEqual(loaded.__class__, BeautifulSoup)
+        self.assertEqual(loaded.decode(), tree.decode())
+
+    def test_docstring_generated(self):
+        soup = self.soup("")
+        self.assertEqual(
+            soup.encode(), b'\n')
+
+    def test_xml_declaration(self):
+        markup = b"""\n"""
+        soup = self.soup(markup)
+        self.assertEqual(markup, soup.encode("utf8"))
+
+    def test_processing_instruction(self):
+        markup = b"""\n"""
+        soup = self.soup(markup)
+        self.assertEqual(markup, soup.encode("utf8"))
+
+    def test_real_xhtml_document(self):
+        """A real XHTML document should come out *exactly* the same as it went in."""
+        markup = b"""
+
+
+Hello.
+Goodbye.
+"""
+        soup = self.soup(markup)
+        self.assertEqual(
+            soup.encode("utf-8"), markup)
+
+    def test_formatter_processes_script_tag_for_xml_documents(self):
+        doc = """
+  
+"""
+        soup = BeautifulSoup(doc, "lxml-xml")
+        # lxml would have stripped this while parsing, but we can add
+        # it later.
+        soup.script.string = 'console.log("< < hey > > ");'
+        encoded = soup.encode()
+        self.assertTrue(b"< < hey > >" in encoded)
+
+    def test_can_parse_unicode_document(self):
+        markup = 'Sacr\N{LATIN SMALL LETTER E WITH ACUTE} bleu!'
+        soup = self.soup(markup)
+        self.assertEqual('Sacr\xe9 bleu!', soup.root.string)
+
+    def test_popping_namespaced_tag(self):
+        markup = 'b2012-07-02T20:33:42Zcd'
+        soup = self.soup(markup)
+        self.assertEqual(
+            str(soup.rss), markup)
+
+    def test_docstring_includes_correct_encoding(self):
+        soup = self.soup("")
+        self.assertEqual(
+            soup.encode("latin1"),
+            b'\n')
+
+    def test_large_xml_document(self):
+        """A large XML document should come out the same as it went in."""
+        markup = (b'\n'
+                  + b'0' * (2**12)
+                  + b'')
+        soup = self.soup(markup)
+        self.assertEqual(soup.encode("utf-8"), markup)
+
+
+    def test_tags_are_empty_element_if_and_only_if_they_are_empty(self):
+        self.assertSoupEquals("
", "
")
+        self.assertSoupEquals("
foo
")
+
+    def test_namespaces_are_preserved(self):
+        markup = 'This tag is in the a namespaceThis tag is in the b namespace'
+        soup = self.soup(markup)
+        root = soup.root
+        self.assertEqual("http://example.com/", root['xmlns:a'])
+        self.assertEqual("http://example.net/", root['xmlns:b'])
+
+    def test_closing_namespaced_tag(self):
+        markup = '
20010504
'
+        soup = self.soup(markup)
+        self.assertEqual(str(soup.p), markup)
+
+    def test_namespaced_attributes(self):
+        markup = ''
+        soup = self.soup(markup)
+        self.assertEqual(str(soup.foo), markup)
+
+    def test_namespaced_attributes_xml_namespace(self):
+        markup = 'bar'
+        soup = self.soup(markup)
+        self.assertEqual(str(soup.foo), markup)
+
+    def test_find_by_prefixed_name(self):
+        doc = """
+foo
+    bar
+    baz
+
+"""
+        soup = self.soup(doc)
+
+        # There are three  tags.
+        self.assertEqual(3, len(soup.find_all('tag')))
+
+        # But two of them are ns1:tag and one of them is ns2:tag.
+        self.assertEqual(2, len(soup.find_all('ns1:tag')))
+        self.assertEqual(1, len(soup.find_all('ns2:tag')))
+        
+        self.assertEqual(1, len(soup.find_all('ns2:tag', key='value')))
+        self.assertEqual(3, len(soup.find_all(['ns1:tag', 'ns2:tag'])))
+        
+    def test_copy_tag_preserves_namespace(self):
+        xml = """
+"""
+    
+        soup = self.soup(xml)
+        tag = soup.document
+        duplicate = copy.copy(tag)
+
+        # The two tags have the same namespace prefix.
+        self.assertEqual(tag.prefix, duplicate.prefix)
+
+
+class HTML5TreeBuilderSmokeTest(HTMLTreeBuilderSmokeTest):
+    """Smoke test for a tree builder that supports HTML5."""
+
+    def test_real_xhtml_document(self):
+        # Since XHTML is not HTML5, HTML5 parsers are not tested to handle
+        # XHTML documents in any particular way.
+        pass
+
+    def test_html_tags_have_namespace(self):
+        markup = ""
+        soup = self.soup(markup)
+        self.assertEqual("http://www.w3.org/1999/xhtml", soup.a.namespace)
+
+    def test_svg_tags_have_namespace(self):
+        markup = ''
+        soup = self.soup(markup)
+        namespace = "http://www.w3.org/2000/svg"
+        self.assertEqual(namespace, soup.svg.namespace)
+        self.assertEqual(namespace, soup.circle.namespace)
+
+
+    def test_mathml_tags_have_namespace(self):
+        markup = '5'
+        soup = self.soup(markup)
+        namespace = 'http://www.w3.org/1998/Math/MathML'
+        self.assertEqual(namespace, soup.math.namespace)
+        self.assertEqual(namespace, soup.msqrt.namespace)
+
+    def test_xml_declaration_becomes_comment(self):
+        markup = ''
+        soup = self.soup(markup)
+        self.assertTrue(isinstance(soup.contents[0], Comment))
+        self.assertEqual(soup.contents[0], '?xml version="1.0" encoding="utf-8"?')
+        self.assertEqual("html", soup.contents[0].next_element.name)
+
+def skipIf(condition, reason):
+   def nothing(test, *args, **kwargs):
+       return None
+
+   def decorator(test_item):
+       if condition:
+           return nothing
+       else:
+           return test_item
+
+   return decorator
diff --git a/cleanerupper/plugin.py b/cleanerupper/plugin.py
new file mode 100644
index 0000000..d38e400
--- /dev/null
+++ b/cleanerupper/plugin.py
@@ -0,0 +1,286 @@
+import re
+import sys
+import bs4
+import os
+
+html_cleaners = []
+soup_cleaners = []
+global_footnotes = {}
+
+def html_cleaner(function):
+    html_cleaners.append(function)
+
+def soup_cleaner(function):
+    soup_cleaners.append(function)
+
+def raise_children_and_delete(element):
+    children = list(element.children)
+    while children:
+        element.insert_after(children.pop(-1))
+    element.decompose()
+
+def remove_class(element, cls):
+    if not hasattr(element, 'class') or element['class'] is None:
+        return
+    if isinstance(element['class'], str):
+        if element['class'] == cls:
+            del element['class']
+            return
+        else:
+            element['class'] = element['class'].split()
+    try:
+        element['class'].remove(cls)
+    except IndexError:
+        pass
+    if len(element['class']) == 0:
+        del element['class']
+
+@html_cleaner
+def remove_unwanted_stylesheets(html):
+    html = re.sub(r'', '', html)
+    return html
+
+@html_cleaner
+def merge_neighboring_sametag(html):
+    html = re.sub(r'\s*', ' ', html)
+    html = re.sub(r'\s*', ' ', html)
+    return html
+
+@html_cleaner
+def bring_punctuation_into_italics(html):
+    for tag in ['i', 'b']:
+        for punct in ['.', ',', '-', '—']:
+            html = re.sub('\\{punct}<{tag}>'.format(**locals()), '<{tag}>{punct}'.format(**locals()), html)
+            html = re.sub('\\{punct}'.format(**locals()), '{punct}'.format(**locals()), html)
+    return html
+
+@html_cleaner
+def remove_space_around_br(html):
+    html = re.sub(r'\s*
\s*', '
', html)
+    return html
+
+@html_cleaner
+def replace_smart_quotes(html):
+    html = re.sub(r'”|“', '"', html)
+    html = re.sub(r'‘|’|ʹ', "'", html)
+    return html
+
+@html_cleaner
+def remove_empty_elements(html):
+    html = re.sub(r'(?s)<(\w+)>(&(nbsp|emsp|ensp|thinsp|#160);|\s|
)*', '', html)
+    return html
+
+@soup_cleaner
+def inject_footnotes(soup):
+    footnotes = soup.find_all('blockquote', {'class': 'gcufootnote_content'})
+    for footnote in footnotes:
+        try:
+            footnote_id = next(footnote.stripped_strings)
+        except StopIteration:
+            print(footnote, 'is malformed. No string contents.')
+            continue
+        if not footnote_id.startswith('['):
+            print(footnote, 'is malformed. Should start with [id].')
+            continue
+        footnote_id = footnote_id.split('[', 1)[-1].split(']', 1)[0]
+
+        global_footnotes[footnote_id] = footnote
+
+    footnote_links = soup.find_all('span', {'class': 'gcufootnote_link'})
+    for footnote_link in reversed(footnote_links):
+        if len(footnote_link.contents) != 1:
+            print(footnote_link, 'is malformed. Should just be >[id<.')
+        footnote_id = footnote_link.contents[0]
+        if not footnote_id.startswith('['):
+            print(footnote_link, 'is malformed. Should start with [id].')
+            continue
+        footnote_id = footnote_id.split('[', 1)[-1].split(']', 1)[0]
+
+        if footnote_id not in global_footnotes:
+            continue
+        footnote = global_footnotes[footnote_id]
+        footnote_link.parent.insert_after(footnote)
+        footnote_link.insert_before(footnote_link.contents[0])
+        footnote_link.decompose()
+        remove_class(footnote, 'gcufootnote_content')
+
+@soup_cleaner
+def convert_textdivs_p(soup):
+    divs = soup.find_all('div')
+    for div in divs:
+        children = list(div.children)
+        if len(children) == 1 and isinstance(children[0], (str, bs4.element.NavigableString)):
+            div.name = 'p'
+
+@soup_cleaner
+def remove_body_br(soup):
+    for br in soup.find_all('br'):
+        if br.parent.name == 'body':
+            br.decompose()
+
+@soup_cleaner
+def remove_empty_paragraphs(soup):
+    brs = soup.find_all('br')
+    br_parents = set(br.parent for br in brs)
+    for br_parent in br_parents:
+        if all(child.name == 'br' for child in br_parent.contents):
+            br_parent.decompose()
+
+@soup_cleaner
+def remove_calibre_classes(soup):
+    PATTERNS = [
+        r'calibre\d*',
+        r'mbppagebreak',
+        r'calibre_pb_\d+',
+        r'filepos\d*',
+    ]
+    for tag in soup.descendants:
+        try:
+            tag['class']
+        except (TypeError, KeyError):
+            pass
+        else:
+            if isinstance(tag['class'], str):
+                tag['class'] = tag['class'].split()
+
+            for cls in list(tag['class']):
+                if any(re.match(pattern, cls) for pattern in PATTERNS):
+                    tag['class'].remove(cls)
+
+            if len(tag['class']) == 0 or tag['class'][0] == '':
+                del tag['class']
+
+        try:
+            tag['id']
+        except (TypeError, KeyError):
+            pass
+        else:
+            if any(re.match(pattern, tag['id']) for pattern in PATTERNS):
+                del tag['id']
+                continue
+
+@soup_cleaner
+def remove_header_italic_bold(soup):
+    headers = [h for tag in ['h1', 'h2', 'h3', 'h4', 'h5', 'h6'] for h in soup.find_all(tag)]
+    for header in headers:
+        children = list(header.children)
+        if len(children) > 1:
+            continue
+        child = children[0]
+        if isinstance(child, str):
+            continue
+        if child.name in ['i', 'b']:
+            raise_children_and_delete(child)
+
+@soup_cleaner
+def remove_useless_divs(soup):
+    divs = soup.find_all('div')
+    for div in divs:
+        if not div.attrs:
+            if all(isinstance(child, bs4.element.Tag) or child.isspace() for child in div.contents):
+                raise_children_and_delete(div)
+
+@soup_cleaner
+def remove_useless_spans(soup):
+    spans = soup.find_all('span')
+    for span in spans:
+        if span.attrs:
+            continue
+        raise_children_and_delete(span)
+
+@soup_cleaner
+def remove_nested_italic(soup):
+    elements = [element for tag in ['b', 'i'] for element in soup.find_all(tag)]
+    for element in elements:
+        if element.parent.name == element.name:
+            raise_children_and_delete(element)
+
+@soup_cleaner
+def replace_italic_bold_span(soup):
+    tags = {'italic': 'i', 'italics': 'i', 'bold': 'b'}
+    spans = set(span for cls in tags for span in soup.find_all('span', {'class': cls}))
+    for span in spans:
+        if isinstance(span['class'], str):
+            span['class'] = span['class'].split()
+
+        if len(span['class']) == 1:
+            new_name = tags[span['class'][0]]
+            del span['class']
+            span.name = new_name
+
+        elif all(cls in tags for cls in span['class']):
+            b = soup.new_tag('b')
+            del span['class']
+            span.name = 'i'
+            span.insert_before(b)
+            b.insert(0, span)
+
+@soup_cleaner
+def replace_pblock_blockquote(soup):
+    classes = ['block', 'block1', 'blockquote']
+    ptags = set(ptag for cls in classes for ptag in soup.find_all('p', {'class': cls}))
+    for ptag in ptags:
+        if isinstance(ptag['class'], str):
+            span['class'] = span['class'].split()
+        if len(ptag['class']) == 1:
+            ptag.name = 'blockquote'
+            ptag['class'] = []
+
+@soup_cleaner
+def strip_ptag_whitespace(soup):
+    ps = soup.find_all('p') + soup.find_all('blockquote')
+
+    for p in ps:
+        descendants = list(p.descendants)
+        while descendants and not isinstance(descendants[0], bs4.element.NavigableString):
+            if descendants[0].name == 'br':
+                descendants[0].decompose()
+            descendants.pop(0)
+        while descendants and not isinstance(descendants[-1], bs4.element.NavigableString):
+            if descendants[-1].name == 'br':
+                descendants[-1].decompose()
+            descendants.pop(-1)
+
+        if not descendants:
+            continue
+
+        if len(descendants) == 1:
+            descendants[0].replace_with(descendants[0].strip())
+            continue
+
+        descendants[0].replace_with(descendants[0].lstrip())
+        descendants[-1].replace_with(descendants[-1].rstrip())
+
+def cleanup_page(html):
+    previous_html = None
+    while previous_html != html:
+        previous_html = html
+
+        for cleaner in html_cleaners:
+            html = cleaner(html)
+
+        soup = bs4.BeautifulSoup(html, 'html.parser')
+
+        for cleaner in soup_cleaners:
+            cleaner(soup)
+
+        html = str(soup)
+
+    return html
+
+def run_once(book):
+    for (id, href) in book.text_iter():
+        if id in ('navid', 'nav.xhtml', 'nav.html'):
+            continue
+        print('Cleaning', id)
+        html = book.readfile(id)
+        html = cleanup_page(html)
+        book.writefile(id, html)
+
+def run(book):
+    run_once(book)
+
+    if global_footnotes:
+        run_once(book)
+
+    return 0
diff --git a/cleanerupper/plugin.xml b/cleanerupper/plugin.xml
new file mode 100644
index 0000000..dde989b
--- /dev/null
+++ b/cleanerupper/plugin.xml
@@ -0,0 +1,11 @@
+
+
+    cleanerupper
+    edit
+    voussoir
+    Behold immaculate html
+    python3.4
+    1.0
+    true
+    osx,win,unx
+