ox.html.sanitize_fragment: documentation, tests

ox/html.py

@@ -393,6 +393,21 @@ def split_tags(string):
     return [item for sublist in zip(strings, tags) for item in sublist][:-1]
 
 def sanitize_fragment(html):
+    '''
+    Ensures that tags are closed (or not, as appropriate), attributes
+    are quoted, etc. Does not strip potentially-malicious HTML: use
+    sanitize_html() for that.
+
+    >>> sanitize_fragment(u'<span lang="en">')
+    u'<span lang="en"></span>'
+    >>> sanitize_fragment(u'<span lang=en></span>')
+    u'<span lang="en"></span>'
+    >>> sanitize_fragment(u'<br><br/></br>')
+    u'<br><br>'
+    >>> sanitize_fragment(u'<a href="javascript:alert()">foo</a>')
+    u'<a href="javascript:alert()">foo</a>'
+    '''
+
     '''
     #html5lib reorders arguments, so not usable
     import html5lib