Tokeniser updates to match spec: missing attribute values (r2122), lowercase doctype names (r2502)

philiptaylor · philiptaylor · commit ddfddb93c50e · 2008-12-22T01:13:38.000Z
--HG--
extra : convert_revision : svn%3Aacbfec75-9323-0410-a652-858a13e371e0/trunk%401247
diff --git a/src/html5lib/constants.py b/src/html5lib/constants.py
@@ -64,6 +64,8 @@
        _(u"Unexpected end of file. Expected = or end of tag."),
     "expected-attribute-value-but-got-eof":
        _(u"Unexpected end of file. Expected attribute value."),
+    "expected-attribute-value-but-got-right-bracket":
+       _(u"Expected attribute value. Got '>' instead."),
     "eof-in-attribute-value-double-quote":
        _(u"Unexpected end of file in attribute value (\")."),
     "eof-in-attribute-value-single-quote":
diff --git a/src/html5lib/tokenizer.py b/src/html5lib/tokenizer.py
@@ -615,6 +615,8 @@ def beforeAttributeValueState(self):
         elif data == u"'":
             self.state = self.states["attributeValueSingleQuoted"]
         elif data == u">":
+            self.tokenQueue.append({"type": "ParseError", "data":
+              "expected-attribute-value-but-got-right-bracket"})
             self.emitCurrentToken()
         elif data == u"=":
             self.tokenQueue.append({"type": "ParseError", "data":
@@ -888,14 +890,17 @@ def beforeDoctypeNameState(self):
     def doctypeNameState(self):
         data = self.stream.char()
         if data in spaceCharacters:
+            self.currentToken["name"] = self.currentToken["name"].translate(asciiUpper2Lower)
             self.state = self.states["afterDoctypeName"]
         elif data == u">":
+            self.currentToken["name"] = self.currentToken["name"].translate(asciiUpper2Lower)
             self.tokenQueue.append(self.currentToken)
             self.state = self.states["data"]
         elif data is EOF:
             self.tokenQueue.append({"type": "ParseError", "data":
               "eof-in-doctype-name"})
             self.currentToken["correct"] = False
+            self.currentToken["name"] = self.currentToken["name"].translate(asciiUpper2Lower)
             self.tokenQueue.append(self.currentToken)
             self.state = self.states["data"]
         else:
