rutracker revision

- Now uses requests with more logging
- Update to latest BeautifulSoup and html5lib libs

lib/bs4/builder/__init__.py

@@ -80,9 +80,12 @@ 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.

lib/bs4/builder/_html5lib.py

@@ -2,6 +2,7 @@ __all__ = [
     'HTML5TreeBuilder',
     ]
 
+from pdb import set_trace
 import warnings
 from bs4.builder import (
     PERMISSIVE,
@@ -9,7 +10,10 @@ from bs4.builder import (
     HTML_5,
     HTMLTreeBuilder,
     )
-from bs4.element import NamespacedAttribute
+from bs4.element import (
+    NamespacedAttribute,
+    whitespace_re,
+)
 import html5lib
 from html5lib.constants import namespaces
 from bs4.element import (
@@ -22,11 +26,20 @@ from bs4.element import (
 class HTML5TreeBuilder(HTMLTreeBuilder):
     """Use html5lib to build a tree."""
 
-    features = ['html5lib', PERMISSIVE, HTML_5, HTML]
+    NAME = "html5lib"
 
-    def prepare_markup(self, markup, user_specified_encoding):
+    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.
@@ -101,7 +114,13 @@ class AttrList(object):
     def __iter__(self):
         return list(self.attrs.items()).__iter__()
     def __setitem__(self, name, value):
-        "set attr", 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])):
+            value = whitespace_re.split(value)
         self.element[name] = value
     def items(self):
         return list(self.attrs.items())
@@ -161,6 +180,12 @@ class Element(html5lib.treebuilders._base.Node):
             # 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
 
@@ -172,6 +197,7 @@ class Element(html5lib.treebuilders._base.Node):
         return AttrList(self.element)
 
     def setAttributes(self, attributes):
+
         if attributes is not None and len(attributes) > 0:
 
             converted_attributes = []
@@ -218,6 +244,9 @@ class Element(html5lib.treebuilders._base.Node):
 
     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
@@ -236,17 +265,28 @@ class Element(html5lib.treebuilders._base.Node):
             new_parents_last_descendant_next_element = new_parent_element.next_element
 
         to_append = element.contents
-        append_after = new_parent.element.contents
+        append_after = new_parent_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]
-            first_child.previous_element = new_parents_last_descendant
+            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
 
             # Fix the last child's next_element and next_sibling
             last_child = to_append[-1]
             last_child.next_element = new_parents_last_descendant_next_element
+            if new_parents_last_descendant_next_element:
+                new_parents_last_descendant_next_element.previous_element = last_child
             last_child.next_sibling = None
 
         for child in to_append:
@@ -257,6 +297,10 @@ class Element(html5lib.treebuilders._base.Node):
         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)
@@ -268,7 +312,7 @@ class Element(html5lib.treebuilders._base.Node):
         return self.element.contents
 
     def getNameTuple(self):
-        if self.namespace is None:
+        if self.namespace == None:
             return namespaces["html"], self.name
         else:
             return self.namespace, self.name

lib/bs4/builder/_htmlparser.py

@@ -4,10 +4,16 @@ __all__ = [
     'HTMLParserTreeBuilder',
     ]
 
-from HTMLParser import (
-    HTMLParser,
-    HTMLParseError,
-    )
+from HTMLParser import HTMLParser
+
+try:
+    from HTMLParser import HTMLParseError
+except ImportError, 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
 
@@ -19,10 +25,10 @@ import warnings
 # 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
-    or (major == 3 and minor > 2)
-    or (major == 3 and minor == 2 and release >= 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,
@@ -63,7 +69,8 @@ class BeautifulSoupHTMLParser(HTMLParser):
 
     def handle_charref(self, name):
         # XXX workaround for a bug in HTMLParser. Remove this once
-        # it's fixed.
+        # 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'):
@@ -113,14 +120,6 @@ class BeautifulSoupHTMLParser(HTMLParser):
 
     def handle_pi(self, data):
         self.soup.endData()
-        if data.endswith("?") and data.lower().startswith("xml"):
-            # "An XHTML processing instruction using the trailing '?'
-            # will cause the '?' to be included in data." - HTMLParser
-            # docs.
-            #
-            # Strip the question mark so we don't end up with two
-            # question marks.
-            data = data[:-1]
         self.soup.handle_data(data)
         self.soup.endData(ProcessingInstruction)
 
@@ -128,15 +127,19 @@ class BeautifulSoupHTMLParser(HTMLParser):
 class HTMLParserTreeBuilder(HTMLTreeBuilder):
 
     is_xml = False
-    features = [HTML, STRICT, HTMLPARSER]
+    picklable = True
+    NAME = HTMLPARSER
+    features = [NAME, HTML, STRICT]
 
     def __init__(self, *args, **kwargs):
-        if CONSTRUCTOR_TAKES_STRICT:
+        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):
+                       document_declared_encoding=None, exclude_encodings=None):
         """
         :return: A 4-tuple (markup, original encoding, encoding
         declared within markup, whether any characters had to be
@@ -147,7 +150,8 @@ class HTMLParserTreeBuilder(HTMLTreeBuilder):
             return
 
         try_encodings = [user_specified_encoding, document_declared_encoding]
-        dammit = UnicodeDammit(markup, try_encodings, is_html=True)
+        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)

lib/bs4/builder/_lxml.py

@@ -7,7 +7,12 @@ from io import BytesIO
 from StringIO import StringIO
 import collections
 from lxml import etree
-from bs4.element import Comment, Doctype, NamespacedAttribute
+from bs4.element import (
+    Comment,
+    Doctype,
+    NamespacedAttribute,
+    ProcessingInstruction,
+)
 from bs4.builder import (
     FAST,
     HTML,
@@ -25,8 +30,11 @@ class LXMLTreeBuilderForXML(TreeBuilder):
 
     is_xml = True
 
+    NAME = "lxml-xml"
+    ALTERNATE_NAMES = ["xml"]
+
     # Well, it's permissive by XML parser standards.
-    features = [LXML, XML, FAST, PERMISSIVE]
+    features = [NAME, LXML, XML, FAST, PERMISSIVE]
 
     CHUNK_SIZE = 512
 
@@ -70,6 +78,7 @@ class LXMLTreeBuilderForXML(TreeBuilder):
             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.
@@ -95,7 +104,8 @@ class LXMLTreeBuilderForXML(TreeBuilder):
         # the document as each one in turn.
         is_html = not self.is_xml
         try_encodings = [user_specified_encoding, document_declared_encoding]
-        detector = EncodingDetector(markup, try_encodings, is_html)
+        detector = EncodingDetector(
+            markup, try_encodings, is_html, exclude_encodings)
         for encoding in detector.encodings:
             yield (detector.markup, encoding, document_declared_encoding, False)
 
@@ -189,7 +199,9 @@ class LXMLTreeBuilderForXML(TreeBuilder):
             self.nsmaps.pop()
 
     def pi(self, target, data):
-        pass
+        self.soup.endData()
+        self.soup.handle_data(target + ' ' + data)
+        self.soup.endData(ProcessingInstruction)
 
     def data(self, content):
         self.soup.handle_data(content)
@@ -212,7 +224,10 @@ class LXMLTreeBuilderForXML(TreeBuilder):
 
 class LXMLTreeBuilder(HTMLTreeBuilder, LXMLTreeBuilderForXML):
 
-    features = [LXML, HTML, FAST, PERMISSIVE]
+    NAME = LXML
+    ALTERNATE_NAMES = ["lxml-html"]
+
+    features = ALTERNATE_NAMES + [NAME, HTML, FAST, PERMISSIVE]
     is_xml = False
 
     def default_parser(self, encoding):