rutracker revision

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

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