]> git.ipfire.org Git - thirdparty/Python/cpython.git/commitdiff
projects / thirdparty / Python / cpython.git / commitdiff
? search:
summary | shortlog | log | commit | commitdiff | tree
raw | patch | inline | side by side (parent: 5211ffe)
Fix htmlparser tests to always use the right collector.
authorEzio Melotti <ezio.melotti@gmail.com>
Mon, 13 Feb 2012 12:11:27 +0000 (14:11 +0200)
committerEzio Melotti <ezio.melotti@gmail.com>
Mon, 13 Feb 2012 12:11:27 +0000 (14:11 +0200)
Lib/test/test_htmlparser.py patch | blob | blame | history

diff --git a/Lib/test/test_htmlparser.py b/Lib/test/test_htmlparser.py
index 2e7277e573db03d0cb209423a96c4cdd6c3b3577..3d3f8593b556a6884d61297549e1d317b5f8908c 100644 (file)
--- a/Lib/test/test_htmlparser.py
+++ b/Lib/test/test_htmlparser.py
@@ -93,7 +93,7 @@ class TestCaseBase(unittest.TestCase):
 
     def _parse_error(self, source):
         def parse(source=source):
-            parser = html.parser.HTMLParser()
+            parser = self.get_collector()
             parser.feed(source)
             parser.close()
         self.assertRaises(html.parser.HTMLParseError, parse)
@@ -368,6 +368,30 @@ class HTMLParserTolerantTestCase(HTMLParserStrictTestCase):
                             ('comment', '/img'),
                             ('endtag', 'html<')])
 
+    def test_starttag_junk_chars(self):
+        self._run_check("</>", [])
+        self._run_check("</$>", [('comment', '$')])
+        self._run_check("</", [('data', '</')])
+        self._run_check("</a", [('data', '</a')])
+        # XXX this might be wrong
+        self._run_check("<a<a>", [('data', '<a'), ('starttag', 'a', [])])
+        self._run_check("</a<a>", [('endtag', 'a<a')])
+        self._run_check("<!", [('data', '<!')])
+        self._run_check("<a", [('data', '<a')])
+        self._run_check("<a foo='bar'", [('data', "<a foo='bar'")])
+        self._run_check("<a foo='bar", [('data', "<a foo='bar")])
+        self._run_check("<a foo='>'", [('data', "<a foo='>'")])
+        self._run_check("<a foo='>", [('data', "<a foo='>")])
+
+    def test_declaration_junk_chars(self):
+        # XXX this is wrong
+        self._run_check("<!DOCTYPE foo $ >", [('comment', 'DOCTYPE foo $ ')])
+
+    def test_illegal_declarations(self):
+        # XXX this might be wrong
+        self._run_check('<!spacer type="block" height="25">',
+                        [('comment', 'spacer type="block" height="25"')])
+
     def test_with_unquoted_attributes(self):
         # see #12008
         html = ("<html><body bgcolor=d0ca90 text='181008'>"
@@ -476,7 +500,7 @@ class HTMLParserTolerantTestCase(HTMLParserStrictTestCase):
         self._run_check(html, expected)
 
     def test_unescape_function(self):
-        p = html.parser.HTMLParser()
+        p = self.get_collector()
         self.assertEqual(p.unescape('&#bad;'),'&#bad;')
         self.assertEqual(p.unescape('&#0038;'),'&')
         # see #12888
Mirror of https://github.com/python/cpython.git