Introduce type hierarchy for tag-related tokens

jayaddison · jayaddison · commit 67262f8e6f0a · 2020-12-30T16:17:46.000Z
diff --git a/html5lib/_tokenizer.py b/html5lib/_tokenizer.py
@@ -41,21 +41,25 @@ class Characters(Token):
 class SpaceCharacters(Token):
     pass
 
-class StartTag(Token):
-    def __init__(self, name, data, self_closing, self_closing_acknowledged=False):
+
+class Tag(Token):
+    def __init__(self, name, data, self_closing):
         self.name = name
         self.data = data
         self.self_closing = self_closing
+
+class StartTag(Tag):
+    def __init__(self, name, data, self_closing, self_closing_acknowledged=False):
+        super(StartTag, self).__init__(name, data, self_closing)
         self.self_closing_acknowledged = self_closing_acknowledged
 
-class EndTag(Token):
+class EndTag(Tag):
     def __init__(self, name, data, self_closing):
-        self.name = name
-        self.data = data
-        self.self_closing = self_closing
+        super(EndTag, self).__init__(name, data, self_closing)
 
-class EmptyTag(Token):
-    pass
+class EmptyTag(Tag):
+    def __init__(self, name, data):
+        super(EmptyTag, self).__init__(name, data, self_closing)
 
 class Comment(Token):
     pass
@@ -66,9 +70,6 @@ def __init__(self, data, datavars=None):
         self.datavars = datavars or {}
 
 
-tagTokenTypes = frozenset([StartTag, EndTag, EmptyTag])
-
-
 class HTMLTokenizer(object):
     """ This class takes care of tokenizing HTML.
 
@@ -266,7 +267,7 @@ def emitCurrentToken(self):
         """
         token = self.currentToken
         # Add token to the queue to be yielded
-        if (type(token) in tagTokenTypes):
+        if isinstance(token, Tag):
             token.name = token.name.translate(asciiUpper2Lower)
             if isinstance(token, StartTag):
                 raw = token.data
diff --git a/html5lib/html5parser.py b/html5lib/html5parser.py
@@ -13,7 +13,7 @@
     Comment,
     Doctype,
     ParseError as TokenizerParseError,
-    tagTokenTypes,
+    Tag,
 )
 
 from . import treebuilders
@@ -400,9 +400,8 @@ def log(function):
         def wrapped(self, *args, **kwargs):
             if function.__name__.startswith("process") and len(args) > 0:
                 token = args[0]
-                token_type = type(token)
-                info = {"type": token_type.__name__}
-                if token_type in tagTokenTypes:
+                info = {"type": token.__class__.__name__}
+                if isinstance(token, Tag):
                     info["name"] = token.name
 
                 self.parser.log.append((self.parser.tokenizer.state.__name__,
