Deal more appropriately with bare ampersands and pointy brackets; this
module has to deal with "class" HTML-as-deployed as well as XHTML, so we
cannot be as strict as XHTML allows.
This closes SF bug #453059, but uses a different fix than suggested in
the bug comments.
diff --git a/Lib/test/test_htmlparser.py b/Lib/test/test_htmlparser.py
index e0e212c..bb6e0b0 100755
--- a/Lib/test/test_htmlparser.py
+++ b/Lib/test/test_htmlparser.py
@@ -1,6 +1,7 @@
"""Tests for HTMLParser.py."""
import HTMLParser
+import pprint
import sys
import test_support
import unittest
@@ -83,9 +84,10 @@
for c in self.epilogue:
parser.feed(c)
parser.close()
- self.assert_(parser.get_events() ==
- self.initial_events + events + self.final_events,
- parser.get_events())
+ events = parser.get_events()
+ self.assertEqual(events,
+ self.initial_events + events + self.final_events,
+ "got events:\n" + pprint.pformat(events))
def _run_check_extra(self, source, events):
self._run_check(source, events, EventCollectorExtra)
@@ -137,6 +139,18 @@
("data", "\n"),
])
+ def test_doctype_decl(self):
+ inside = """\
+DOCTYPE html [
+ <!ELEMENT html - O EMPTY>
+ <!ATTLIST html
+ version CDATA #IMPLIED '4.0'>
+ <!-- comment -->
+]"""
+ self._run_check("<!%s>" % inside, [
+ ("decl", inside),
+ ])
+
def test_bad_nesting(self):
# Strangely, this *is* supposed to test that overlapping
# elements are allowed. HTMLParser is more geared toward
@@ -148,6 +162,16 @@
("endtag", "b"),
])
+ def test_bare_ampersands(self):
+ self._run_check("this text & contains & ampersands &", [
+ ("data", "this text & contains & ampersands &"),
+ ])
+
+ def test_bare_pointy_brackets(self):
+ self._run_check("this < text > contains < bare>pointy< brackets", [
+ ("data", "this < text > contains < bare>pointy< brackets"),
+ ])
+
def test_attr_syntax(self):
output = [
("starttag", "a", [("b", "v"), ("c", "v"), ("d", "v"), ("e", None)])
@@ -199,16 +223,12 @@
self._run_check(["<a b='>'", ">"], output)
def test_starttag_junk_chars(self):
- self._parse_error("<")
- self._parse_error("<>")
self._parse_error("</>")
self._parse_error("</$>")
self._parse_error("</")
self._parse_error("</a")
self._parse_error("<a<a>")
self._parse_error("</a<a>")
- self._parse_error("<$")
- self._parse_error("<$>")
self._parse_error("<!")
self._parse_error("<a $>")
self._parse_error("<a")