soup_adapter.py: migrate to BeautifulSoup4

Author: lanodan

python/gumbo/soup_adapter.py

@@ -13,15 +13,15 @@
 # limitations under the License.
 #
 
-"""Adapter between Gumbo and BeautifulSoup.
+"""Adapter between Gumbo and BeautifulSoup4.
 
 This parses an HTML document and gives back a BeautifulSoup object, which you
 can then manipulate like a normal BeautifulSoup parse tree.
 """
 
 __author__ = '[email protected] (Jonathan Tang)'
 
-import BeautifulSoup
+import bs4
 
 import gumboc
 
@@ -45,7 +45,7 @@ def _convert_attrs(attrs):
   # TODO(jdtang): Ideally attributes would pass along their positions as well,
   # but I can't extend the built in str objects with new attributes.  Maybe work
   # around this with a subclass in some way...
-  return [(_utf8(attr.name), _utf8(attr.value)) for attr in attrs]
+  return {_utf8(attr.name): _utf8(attr.value) for attr in attrs}
 
 
 def _add_document(soup, element):
@@ -57,8 +57,7 @@ def _add_document(soup, element):
 def _add_element(soup, element):
   # TODO(jdtang): Expose next/previous in gumbo so they can be passed along to
   # BeautifulSoup.
-  tag = BeautifulSoup.Tag(
-      soup, _utf8(element.tag_name), _convert_attrs(element.attributes))
+  tag = bs4.Tag(parser = soup, name = _utf8(element.tag_name), attrs = _convert_attrs(element.attributes))
   for child in element.children:
     tag.append(_add_node(soup, child))
   _add_source_info(
@@ -78,10 +77,10 @@ def add_text_internal(soup, element):
 _HANDLERS = [
     _add_document,
     _add_element,
-    _add_text(BeautifulSoup.NavigableString),
-    _add_text(BeautifulSoup.CData),
-    _add_text(BeautifulSoup.Comment),
-    _add_text(BeautifulSoup.NavigableString),
+    _add_text(bs4.NavigableString),
+    _add_text(bs4.CData),
+    _add_text(bs4.Comment),
+    _add_text(bs4.NavigableString),
     _add_element,
     ]
 
@@ -90,32 +89,8 @@ def _add_node(soup, node):
   return _HANDLERS[node.type.value](soup, node.contents)
 
 
-def _add_next_prev_pointers(soup):
-  def _traverse(node):
-    # .findAll requires the .next pointer, which is what we're trying to add
-    # when we call this, and so we manually supply a generator to yield the
-    # nodes in DOM order.
-    yield node
-    try:
-      for child in node.contents:
-        for descendant in _traverse(child):
-          yield descendant
-    except AttributeError:
-      # Not an element.
-      return
-
-  nodes = sorted(_traverse(soup), key=lambda node: node.offset)
-  if nodes:
-    nodes[0].previous = None
-    nodes[-1].next = None
-  for i, node in enumerate(nodes[1:-1], 1):
-    nodes[i-1].next = node
-    node.previous = nodes[i-1]
-
-
 def parse(text, **kwargs):
   with gumboc.parse(text, **kwargs) as output:
-    soup = BeautifulSoup.BeautifulSoup()
+    soup = bs4.BeautifulSoup()
     soup.append(_add_node(soup, output.contents.root.contents))
-    _add_next_prev_pointers(soup)
     return soup