Fixed #19693 -- Made truncatewords_html handle self-closing tags

Thanks sneawo for the report and Jonathan Loy for the patch.

2 files changed, 20 insertions, 9 deletions

diff --git a/django/utils/text.py b/django/utils/text.py
index 6664b18249..02c3f10678 100644
--- a/django/utils/text.py
+++ b/django/utils/text.py
@@ -24,7 +24,7 @@ capfirst = allow_lazy(capfirst, six.text_type)
 
 # Set up regular expressions
 re_words = re.compile(r'&.*?;|<.*?>|(\w[\w-]*)', re.U|re.S)
-re_tag = re.compile(r'<(/)?([^ ]+?)(?: (/)| .*?)?>', re.S)
+re_tag = re.compile(r'<(/)?([^ ]+?)(?:(\s*/)| .*?)?>', re.S)
 
 
 def wrap(text, width):
diff --git a/tests/regressiontests/utils/text.py b/tests/regressiontests/utils/text.py
index 3465282c7f..c9dde6b1b3 100644
--- a/tests/regressiontests/utils/text.py
+++ b/tests/regressiontests/utils/text.py
@@ -55,22 +55,33 @@ class TestUtilsText(SimpleTestCase):
             truncator.words(4, '[snip]'))
 
     def test_truncate_html_words(self):
-        truncator = text.Truncator('<p><strong><em>The quick brown fox jumped '
-            'over the lazy dog.</em></strong></p>')
-        self.assertEqual('<p><strong><em>The quick brown fox jumped over the '
-            'lazy dog.</em></strong></p>', truncator.words(10, html=True))
-        self.assertEqual('<p><strong><em>The quick brown fox...</em>'
+        truncator = text.Truncator('<p id="par"><strong><em>The quick brown fox'
+            ' jumped over the lazy dog.</em></strong></p>')
+        self.assertEqual('<p id="par"><strong><em>The quick brown fox jumped over'
+            ' the lazy dog.</em></strong></p>', truncator.words(10, html=True))
+        self.assertEqual('<p id="par"><strong><em>The quick brown fox...</em>'
             '</strong></p>', truncator.words(4, html=True))
-        self.assertEqual('<p><strong><em>The quick brown fox....</em>'
+        self.assertEqual('<p id="par"><strong><em>The quick brown fox....</em>'
             '</strong></p>', truncator.words(4, '....', html=True))
-        self.assertEqual('<p><strong><em>The quick brown fox</em></strong>'
-            '</p>', truncator.words(4, '', html=True))
+        self.assertEqual('<p id="par"><strong><em>The quick brown fox</em>'
+            '</strong></p>', truncator.words(4, '', html=True))
+
         # Test with new line inside tag
         truncator = text.Truncator('<p>The quick <a href="xyz.html"\n'
             'id="mylink">brown fox</a> jumped over the lazy dog.</p>')
         self.assertEqual('<p>The quick <a href="xyz.html"\n'
             'id="mylink">brown...</a></p>', truncator.words(3, '...', html=True))
 
+        # Test self-closing tags
+        truncator = text.Truncator('<br/>The <hr />quick brown fox jumped over'
+            ' the lazy dog.')
+        self.assertEqual('<br/>The <hr />quick brown...',
+            truncator.words(3, '...', html=True ))
+        truncator = text.Truncator('<br>The <hr/>quick <em>brown fox</em> '
+            'jumped over the lazy dog.')
+        self.assertEqual('<br>The <hr/>quick <em>brown...</em>',
+            truncator.words(3, '...', html=True ))
+
     def test_wrap(self):
         digits = '1234 67 9'
         self.assertEqual(text.wrap(digits, 100), '1234 67 9')