From 3ae64f165eae7500106324c9521a5fd6ad52fa0e Mon Sep 17 00:00:00 2001
From: Richard Harding <rharding@mitechie.com>
Date: Fri, 15 Jun 2012 20:15:37 -0400
Subject: [PATCH] Update and merge

---
 src/breadability/__init__.py |  1 -
 src/breadability/readable.py | 12 ++++++++----
 2 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/src/breadability/__init__.py b/src/breadability/__init__.py
index 5754926..856ce1d 100644
--- a/src/breadability/__init__.py
+++ b/src/breadability/__init__.py
@@ -1,2 +1 @@
 VERSION = '0.1.2'
-import client
diff --git a/src/breadability/readable.py b/src/breadability/readable.py
index 648cfa9..80e7208 100644
--- a/src/breadability/readable.py
+++ b/src/breadability/readable.py
@@ -1,5 +1,4 @@
 import re
-from lxml.etree import Element
 from lxml.etree import tounicode
 from lxml.etree import tostring
 from lxml.html.clean import Cleaner
@@ -8,6 +7,7 @@ from lxml.html import fromstring
 from operator import attrgetter
 from pprint import PrettyPrinter
 
+from breadability.document import build_doc
 from breadability.document import OriginalDocument
 from breadability.logconfig import LOG
 from breadability.logconfig import LNODE
@@ -202,13 +202,17 @@ def prep_article(doc):
         LOG.debug('Cleaning document')
         clean_list = ['object', 'h1']
 
+        # To start out, take our node and reload it so that our iterator is
+        # reset and we can process it completely.
+        re_node = build_doc(tounicode(node))
+
         # If there is only one h2, they are probably using it as a header and
         # not a subheader, so remove it since we already have a header.
-        if len(node.findall('.//h2')) == 1:
+        if len(re_node.findall('.//h2')) == 1:
             LOG.debug('Adding H2 to list of nodes to clean.')
             clean_list.append('h2')
 
-        for n in node.iter(tag=Element):
+        for n in re_node.iter():
             LNODE.log(n, 2, "Cleaning iter node")
             # clean out any incline style properties
             if 'style' in n.attrib:
@@ -262,7 +266,7 @@ def prep_article(doc):
                 if n.getparent() is not None:
                     n.drop_tree()
 
-        return node
+        return re_node
 
     def clean_conditionally(node):
         """Remove the clean_el if it looks like bad content based on rules."""