Merge branch 'kaedroho-html5lib'

CHANGELOG.txt

@@ -4,6 +4,7 @@ Changelog
 0.5 (xx.xx.20xx)
 ~~~~~~~~~~~~~~~~
  * Added decorator syntax for hooks
+ * Replaced lxml dependency with html5lib, to simplify installation
 
 0.4.1 (14.07.2014)
 ~~~~~~~~~~~~~~~~~~

docs/releases/0.5.rst

@@ -7,8 +7,8 @@ Wagtail 0.5 release notes - IN DEVELOPMENT
     :depth: 1
 
 
-Whats new
-=========
+What's new
+==========
 
 
 Minor features
@@ -27,6 +27,8 @@ Core
             MenuItem('Kittens!', '/kittens/', classnames='icon icon-folder-inverse', order=1000)
         )
 
+ * The lxml library (used for whitelisting and rewriting of rich text fields) has been replaced with the pure-python html5lib library, to simplify installation.
+
 
 Bug fixes
 ~~~~~~~~~

setup.py

@@ -32,7 +32,7 @@ install_requires = [
     "django-treebeard==2.0",
     "Pillow>=2.3.0",
     "beautifulsoup4>=4.3.2",
-    "lxml>=3.3.0",
+    "html5lib==0.999",
     "Unidecode>=0.04.14",
     "six==1.7.3",
     'requests==2.3.0',

wagtail/wagtailcore/tests/test_dbwhitelister.py

@@ -25,9 +25,9 @@ class TestDbWhitelister(TestCase):
         self.assertHtmlEqual(expected, output_html)
 
     def test_image_embed_is_rewritten(self):
-        input_html = '<p>OMG look at this picture of a kitten: <figure data-embedtype="image" data-id="5" data-format="image-with-caption" data-alt="A cute kitten" class="fancy-image"><img src="/media/images/kitten.jpg" width="320" height="200" alt="A cute kitten" /><figcaption>A kitten, yesterday.</figcaption></figure></p>'
+        input_html = '<p>OMG look at this picture of a kitten:</p><figure data-embedtype="image" data-id="5" data-format="image-with-caption" data-alt="A cute kitten" class="fancy-image"><img src="/media/images/kitten.jpg" width="320" height="200" alt="A cute kitten" /><figcaption>A kitten, yesterday.</figcaption></figure>'
         output_html = DbWhitelister.clean(input_html)
-        expected = '<p>OMG look at this picture of a kitten: <embed embedtype="image" id="5" format="image-with-caption" alt="A cute kitten" /></p>'
+        expected = '<p>OMG look at this picture of a kitten:</p><embed embedtype="image" id="5" format="image-with-caption" alt="A cute kitten" />'
         self.assertHtmlEqual(expected, output_html)
 
     def test_media_embed_is_rewritten(self):

wagtail/wagtailcore/whitelist.py

@@ -81,7 +81,7 @@ class Whitelister(object):
     def clean(cls, html):
         """Clean up an HTML string to contain just the allowed elements /
         attributes"""
-        doc = BeautifulSoup(html, 'lxml')
+        doc = BeautifulSoup(html, 'html5lib')
         cls.clean_node(doc, doc)
         return doc.decode()