Fail on namespaces with BS.

gsnedders · gsnedders · commit 89114967cf99 · 2009-10-22T18:43:21.000+02:00
diff --git a/src/html5lib/treebuilders/soup.py b/src/html5lib/treebuilders/soup.py
@@ -134,6 +134,11 @@ def cloneNode(self):
         raise NotImplementedError
 
 class TreeBuilder(_base.TreeBuilder):
+    def __init__(self, namespaceHTMLElements):
+        if namespaceHTMLElements:
+            warnings.warn("BeautifulSoup cannot represent elements in any namespace", DataLossWarning)
+        _base.TreeBuilder.__init__(self, namespaceHTMLElements)
+        
     def documentClass(self):
         self.soup = BeautifulSoup("")
         return Element(self.soup, self.soup, None)
@@ -152,8 +157,8 @@ def insertDoctype(self, token):
             self.soup.insert(0, Declaration(name))
     
     def elementClass(self, name, namespace):
-        if namespace not in (None, namespaces["html"]):
-            warnings.warn("BeautifulSoup cannot represent elemens in nn-html namespace", DataLossWarning)
+        if namespace is not None:
+            warnings.warn("BeautifulSoup cannot represent elements in any namespace", DataLossWarning)
         return Element(Tag(self.soup, name), self.soup, namespace)
         
     def commentClass(self, data):
diff --git a/tests/test_parser.py b/tests/test_parser.py
@@ -80,8 +80,11 @@ def runParserTest(self, innerHTML, input, expected, errors, treeClass,
         namespaceHTMLElements):
         #XXX - move this out into the setup function
         #concatenate all consecutive character tokens into a single token
-        p = html5parser.HTMLParser(tree = treeClass,
-                                   namespaceHTMLElements=namespaceHTMLElements)
+        try:
+            p = html5parser.HTMLParser(tree = treeClass,
+                                       namespaceHTMLElements=namespaceHTMLElements)
+        except constants.DataLossWarning:
+            return 
 
         errors = [item.decode("utf-8") for item in errors]
         
@@ -92,7 +95,6 @@ def runParserTest(self, innerHTML, input, expected, errors, treeClass,
                 try:
                     document = p.parse(StringIO.StringIO(input))
                 except constants.DataLossWarning:
-                    sys.stderr.write("Test input causes known dataloss, skipping")
                     return 
         except:
             errorMsg = "\n".join(["\n\nInput:", input, "\nExpected:", expected,
diff --git a/tests/test_treewalkers.py b/tests/test_treewalkers.py
@@ -225,8 +225,8 @@ def sortattrs(x):
 
 class TestCase(unittest.TestCase):
     def runTest(self, innerHTML, input, expected, errors, treeClass):
-        p = html5parser.HTMLParser(tree = treeClass["builder"])
         try:
+            p = html5parser.HTMLParser(tree = treeClass["builder"])
             if innerHTML:
                 document = p.parseFragment(StringIO.StringIO(input), innerHTML)
             else:
