Turn on chardet by default, if installed

jgraham · jgraham · commit 1523af2e5025 · 2007-03-11T19:20:34.000Z
--HG--
extra : convert_revision : svn%3Aacbfec75-9323-0410-a652-858a13e371e0/trunk%40524
diff --git a/src/__init__.py b/src/__init__.py
@@ -10,25 +10,6 @@
 f = open("my_document.html")
 p = html5lib.HTMLParser()
 tree = p.parse(f)
-
-By default the returned treeformat is a custom "simpletree", similar
-to a DOM tree; each element has attributes childNodes and parent
-holding the parents and children respectively, a name attribute
-holding the Element name, a data attribute holding the element data
-(for text and comment nodes) and an attributes dictionary holding the
-element's attributes (for Element nodes).
-
-To get output in ElementTree format:
-
-import html5lib
-from html5lib.treebuilders import etree
-p = html5lib.HTMLParser(tree=etree.TreeBuilder)
-elementtree = p.parse(f)
-
-Note: Because HTML documents support various features not in the
-default ElementTree (e.g. doctypes), we suppy our own simple
-serializer; html5lib.treebuilders.etree.tostring At present this does not
-have the encoding support offered by the elementtree serializer.
-
 """
 from html5parser import HTMLParser
+from liberalxmlparser import XMLParser, XHTMLParser
diff --git a/src/inputstream.py b/src/inputstream.py
@@ -14,7 +14,7 @@ class HTMLInputStream(object):
 
     """
 
-    def __init__(self, source, encoding=None, chardet=False):
+    def __init__(self, source, encoding=None, chardet=True):
         """Initialises the HTMLInputStream.
 
         HTMLInputStream(source, [encoding]) -> Normalized stream from source
diff --git a/tests/test_encoding.py b/tests/test_encoding.py
@@ -42,7 +42,7 @@ class TestCase(unittest.TestCase):
     def runEncodingTest(self, input, encoding):
         #XXX - move this out into the setup function
         #concatenate all consecutive character tokens into a single token
-        stream = inputstream.HTMLInputStream(input)
+        stream = inputstream.HTMLInputStream(input, chardet=False)
         
         errorMsg = "\n".join(["\n\nInput", input,"\nExpected:", encoding,
                               "\nRecieved:", stream.charEncoding])
