Fixed #4310 -- Fixed a regular expression bug in strip_entities function and added tests for several django.utils.html functions. Based on patch from Brian Harring.

gdub · gdub · commit b68f188628dd · 2007-07-15T05:03:28.000Z
git-svn-id: http://code.djangoproject.com/svn/django/trunk@5701 bcc190cf-cafb-0310-a4f2-bffc1f526a37
diff --git a/django/utils/html.py b/django/utils/html.py
@@ -50,7 +50,7 @@ def strip_spaces_between_tags(value):
 
 def strip_entities(value):
     "Returns the given HTML with all entities (&something;) stripped"
-    return re.sub(r'&(?:\w+|#\d);', '', force_unicode(value))
+    return re.sub(r'&(?:\w+|#\d+);', '', force_unicode(value))
 strip_entities = allow_lazy(strip_entities, unicode)
 
 def fix_ampersands(value):
diff --git a/tests/regressiontests/utils/__init__.py b/tests/regressiontests/utils/__init__.py
diff --git a/tests/regressiontests/utils/models.py b/tests/regressiontests/utils/models.py
@@ -0,0 +1 @@
+# Test runner needs a models.py file.
diff --git a/tests/regressiontests/utils/tests.py b/tests/regressiontests/utils/tests.py
@@ -0,0 +1,115 @@
+"""
+Tests for django.utils.
+"""
+
+from unittest import TestCase
+
+from django.utils import html
+
+class TestUtilsHtml(TestCase):
+
+    def check_output(self, function, value, output=None):
+        """
+        Check that function(value) equals output.  If output is None,
+        check that function(value) equals value.
+        """
+        if output is None:
+            output = value
+        self.assertEqual(function(value), output)
+
+    def test_escape(self):
+        f = html.escape
+        items = (
+            ('&','&amp;'),
+            ('<', '&lt;'),
+            ('>', '&gt;'),
+            ('"', '&quot;'),
+            ("'", '&#39;'),
+        )
+        # Substitution patterns for testing the above items.
+        patterns = ("%s", "asdf%sfdsa", "%s1", "1%sb")
+        for value, output in items:
+            for pattern in patterns:
+                self.check_output(f, pattern % value, pattern % output)
+            # Check repeated values.
+            self.check_output(f, value * 2, output * 2)
+        # Verify it doesn't double replace &.
+        self.check_output(f, '<&', '&lt;&amp;')
+
+    def test_linebreaks(self):
+        f = html.linebreaks
+        items = (
+            ("para1\n\npara2\r\rpara3", "<p>para1</p>\n\n<p>para2</p>\n\n<p>para3</p>"),
+            ("para1\nsub1\rsub2\n\npara2", "<p>para1<br />sub1<br />sub2</p>\n\n<p>para2</p>"),
+            ("para1\r\n\r\npara2\rsub1\r\rpara4", "<p>para1</p>\n\n<p>para2<br />sub1</p>\n\n<p>para4</p>"),
+            ("para1\tmore\n\npara2", "<p>para1\tmore</p>\n\n<p>para2</p>"),
+        )
+        for value, output in items:
+            self.check_output(f, value, output)
+
+    def test_strip_tags(self):
+        f = html.strip_tags
+        items = (
+            ('<adf>a', 'a'),
+            ('</adf>a', 'a'),
+            ('<asdf><asdf>e', 'e'),
+            ('<f', '<f'),
+            ('</fe', '</fe'),
+            ('<x>b<y>', 'b'),
+        )
+        for value, output in items:
+            self.check_output(f, value, output)
+
+    def test_strip_spaces_between_tags(self):
+        f = html.strip_spaces_between_tags
+        # Strings that should come out untouched.
+        items = (' <adf>', '<adf> ', ' </adf> ', ' <f> x</f>')
+        for value in items:
+            self.check_output(f, value)
+        # Strings that have spaces to strip.
+        items = (
+            ('<d> </d>', '<d></d>'),
+            ('<p>hello </p>\n<p> world</p>', '<p>hello </p><p> world</p>'),
+            ('\n<p>\t</p>\n<p> </p>\n', '\n<p></p><p></p>\n'),
+        )
+        for value, output in items:
+            self.check_output(f, value, output)
+
+    def test_strip_entities(self):
+        f = html.strip_entities
+        # Strings that should come out untouched.
+        values = ("&", "&a", "&a", "a&#a")
+        for value in values:
+            self.check_output(f, value)
+        # Valid entities that should be stripped from the patterns.
+        entities = ("&#1;", "&#12;", "&a;", "&fdasdfasdfasdf;")
+        patterns = (
+            ("asdf %(entity)s ", "asdf  "),
+            ("%(entity)s%(entity)s", ""),
+            ("&%(entity)s%(entity)s", "&"),
+            ("%(entity)s3", "3"),
+        )
+        for entity in entities:
+            for in_pattern, output in patterns:
+                self.check_output(f, in_pattern % {'entity': entity}, output)
+
+    def test_fix_ampersands(self):
+        f = html.fix_ampersands
+        # Strings without ampersands or with ampersands already encoded.
+        values = ("a&#1;", "b", "&a;", "&amp; &x; ", "asdf")
+        patterns = (
+            ("%s", "%s"),
+            ("&%s", "&amp;%s"),
+            ("&%s&", "&amp;%s&amp;"),
+        )
+        for value in values:
+            for in_pattern, out_pattern in patterns:
+                self.check_output(f, in_pattern % value, out_pattern % value)
+        # Strings with ampersands that need encoding.
+        items = (
+            ("&#;", "&amp;#;"),
+            ("&#875 ;", "&amp;#875 ;"),
+            ("&#4abc;", "&amp;#4abc;"),
+        )
+        for value, output in items:
+            self.check_output(f, value, output)
