Add a documentEncoding property to HTML5Parser, fix #121.

Author: SimonSapin

html5lib/html5parser.py

@@ -129,6 +129,17 @@ def reset(self):
 
         self.framesetOK = True
 
+    @property
+    def documentEncoding(self):
+        """The name of the character encoding
+        that was used to decode the input stream,
+        or :obj:`None` if that is not determined yet.
+
+        """
+        if not hasattr(self, 'tokenizer'):
+            return None
+        return self.tokenizer.stream.charEncoding[0]
+
     def isHTMLIntegrationPoint(self, element):
         if (element.name == "annotation-xml" and
                 element.namespace == namespaces["mathml"]):

html5lib/tests/test_encoding.py

@@ -28,10 +28,11 @@ def test_codec_name_d(self):
 
 def runParserEncodingTest(data, encoding):
     p = HTMLParser()
+    assert p.documentEncoding is None
     p.parse(data, useChardet=False)
     encoding = encoding.lower().decode("ascii")
 
-    assert encoding == p.tokenizer.stream.charEncoding[0], errorMessage(data, encoding, p.tokenizer.stream.charEncoding[0])
+    assert encoding == p.documentEncoding, errorMessage(data, encoding, p.documentEncoding)
 
 
 def runPreScanEncodingTest(data, encoding):