First pass at changing parser test runner to test both namespaced and non-namespaced modes. This also removes the checks to stop namespaced HTML elements having their prefixes removed during test serialization.

gsnedders · gsnedders · commit 2db686153f7a · 2009-10-21T16:32:27.000+02:00
diff --git a/src/html5lib/treebuilders/dom.py b/src/html5lib/treebuilders/dom.py
@@ -192,8 +192,7 @@ def serializeElement(element, indent=0):
                 rv.append("|%s\"%s\"" %(' '*indent, element.nodeValue))
             else:
                 if (hasattr(element, "namespaceURI") and
-                    element.namespaceURI not in (None,
-                                              constants.namespaces["html"])):
+                    element.namespaceURI != None):
                     name = "%s %s"%(constants.prefixes[element.namespaceURI],
                                     element.nodeName)
                 else:
diff --git a/src/html5lib/treebuilders/etree.py b/src/html5lib/treebuilders/etree.py
@@ -227,8 +227,7 @@ def serializeElement(element, indent=0):
                 else:
                     ns, name = nsmatch.groups()
                     prefix = constants.prefixes[ns]
-                    if prefix != "html":
-                        name = "%s %s"%(prefix, name)
+                    name = "%s %s"%(prefix, name)
                 rv.append("|%s<%s>"%(' '*indent, name))
 
                 if hasattr(element, "attrib"):
diff --git a/src/html5lib/treebuilders/etree_lxml.py b/src/html5lib/treebuilders/etree_lxml.py
@@ -86,12 +86,8 @@ def serializeElement(element, indent=0):
                 ns = nsmatch.group(1)
                 tag = nsmatch.group(2)
                 prefix = constants.prefixes[ns]
-                if prefix != "html":
-                    rv.append("|%s<%s %s>"%(' '*indent, prefix,
-                                            filter.fromXmlName(tag)))
-                else:
-                    rv.append("|%s<%s>"%(' '*indent,
-                                         filter.fromXmlName(tag)))
+                rv.append("|%s<%s %s>"%(' '*indent, prefix,
+                                        filter.fromXmlName(tag)))
             else:
                 rv.append("|%s<%s>"%(' '*indent,
                                      filter.fromXmlName(element.tag)))
diff --git a/src/html5lib/treebuilders/simpletree.py b/src/html5lib/treebuilders/simpletree.py
@@ -162,7 +162,7 @@ def __init__(self, name, namespace=None):
         self.attributes = {}
 
     def __unicode__(self):
-        if self.namespace in (None, namespaces["html"]):
+        if self.namespace == None:
             return u"<%s>" % self.name
         else:
             return u"<%s %s>"%(prefixes[self.namespace], self.name)
diff --git a/tests/test_parser.py b/tests/test_parser.py
@@ -73,12 +73,15 @@ def sortattrs(x):
   lines = x.group(0).split("\n")
   lines.sort()
   return "\n".join(lines)
+namespaceExpected = re.compile(r"^(\s*)<(\w+)>", re.M).sub
 
 class TestCase(unittest.TestCase):
-    def runParserTest(self, innerHTML, input, expected, errors, treeClass):
+    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)
+        p = html5parser.HTMLParser(tree = treeClass,
+                                   namespaceHTMLElements=namespaceHTMLElements)
 
         errors = [item.decode("utf-8") for item in errors]
         
@@ -101,6 +104,9 @@ def runParserTest(self, innerHTML, input, expected, errors, treeClass):
         
         expected = convertExpected(expected)
         expected = attrlist.sub(sortattrs, expected)
+        if namespaceHTMLElements:
+            expected = namespaceExpected(r"\1<html \2>", expected)
+        
         errorMsg = "\n".join(["\n\nInput:", input, "\nExpected:", expected,
                               "\nReceived:", output])
         self.assertEquals(expected, output, errorMsg.encode("utf-8"))
@@ -133,12 +139,17 @@ def buildTestSuite():
                                                       'document']
                 if errors:
                     errors = errors.split("\n")
-                def testFunc(self, innerHTML=innerHTML, input=input,
-                    expected=expected, errors=errors, treeCls=treeCls): 
-                    return self.runParserTest(innerHTML, input, expected, errors, treeCls)
-                testFunc.__name__ = "test_%s_%d_%s" % (testName,index+1,treeName)
-                setattr(TestCase, testFunc.__name__,
-                     testFunc)
+                
+                for namespaceHTMLElements in (True, False):
+                    def testFunc(self, innerHTML=innerHTML, input=input,
+                        expected=expected, errors=errors, treeCls=treeCls,
+                        namespaceHTMLElements=namespaceHTMLElements): 
+                        return self.runParserTest(innerHTML, input, expected,
+                                                  errors, treeCls,
+                                                  namespaceHTMLElements)
+                    testFunc.__name__ = "test_%s_%d_%s_%s" % (testName,index+1,treeName, namespaceHTMLElements and "namespaced" or "no_html_namespace")
+                    setattr(TestCase, testFunc.__name__,
+                         testFunc)
 
     return unittest.TestLoader().loadTestsFromTestCase(TestCase)
 
