Reintroduce the old sanitizer testsuite from html5lib-tests

This is imported into this repo as its expectations are very much
implementation dependent

Author: gsnedders

html5lib/serializer/htmlserializer.py

@@ -184,6 +184,12 @@ def serialize(self, treewalker, encoding=None):
         if encoding and self.inject_meta_charset:
             from ..filters.inject_meta_charset import Filter
             treewalker = Filter(treewalker, encoding)
+        # Alphabetical attributes is here under the assumption that none of
+        # the later filters add or change order of attributes; it needs to be
+        # before the sanitizer so escaped elements come out correctly
+        if self.alphabetical_attributes:
+            from ..filters.alphabeticalattributes import Filter
+            treewalker = Filter(treewalker)
         # WhitespaceFilter should be used before OptionalTagFilter
         # for maximum efficiently of this latter filter
         if self.strip_whitespace:
@@ -195,11 +201,6 @@ def serialize(self, treewalker, encoding=None):
         if self.omit_optional_tags:
             from ..filters.optionaltags import Filter
             treewalker = Filter(treewalker)
-        # Alphabetical attributes must be last, as other filters
-        # could add attributes and alter the order
-        if self.alphabetical_attributes:
-            from ..filters.alphabeticalattributes import Filter
-            treewalker = Filter(treewalker)
 
         for token in treewalker:
             type = token["type"]

html5lib/tests/conftest.py

@@ -2,11 +2,13 @@
 
 from .tree_construction import TreeConstructionFile
 from .tokenizer import TokenizerFile
+from .sanitizer import SanitizerFile
 
 _dir = os.path.abspath(os.path.dirname(__file__))
 _testdata = os.path.join(_dir, "testdata")
 _tree_construction = os.path.join(_testdata, "tree-construction")
 _tokenizer = os.path.join(_testdata, "tokenizer")
+_sanitizer_testdata = os.path.join(_dir, "sanitizer-testdata")
 
 
 def pytest_collectstart():
@@ -24,3 +26,6 @@ def pytest_collect_file(path, parent):
     elif dir == _tokenizer:
         if path.ext == ".test":
             return TokenizerFile(path, parent)
+    elif dir == _sanitizer_testdata:
+        if path.ext == ".dat":
+            return SanitizerFile(path, parent)