From febd6cea8cdd665a6f13c7040c3ff7d60d350e41 Mon Sep 17 00:00:00 2001
From: Leonard Richardson <leonardr@segfault.org>
Date: Sat, 21 Jul 2018 12:18:17 -0400
Subject: Fixed a problem where the html.parser tree builder interpreted   a
 string like '&foo ' as the character entity '&foo;'  [bug=1728706]

---
 bs4/testing.py | 8 ++++++++
 1 file changed, 8 insertions(+)

(limited to 'bs4/testing.py')

diff --git a/bs4/testing.py b/bs4/testing.py
index 5b0eb8f..bbcc271 100644
--- a/bs4/testing.py
+++ b/bs4/testing.py
@@ -319,6 +319,14 @@ Hello, world!
     def test_angle_brackets_in_attribute_values_are_escaped(self):
         self.assertSoupEquals('<a b="<a>"></a>', '<a b="&lt;a&gt;"></a>')
 
+    def test_strings_resembling_character_entity_references(self):
+        # "&T" and "&p" look like incomplete character entities, but they are
+        # not.
+        self.assertSoupEquals(
+            u"<p>&bull; AT&T is in the s&p 500</p>",
+            u"<p>\u2022 AT&amp;T is in the s&amp;p 500</p>"
+        )
+        
     def test_entities_in_attributes_converted_to_unicode(self):
         expect = u'<p id="pi\N{LATIN SMALL LETTER N WITH TILDE}ata"></p>'
         self.assertSoupEquals('<p id="pi&#241;ata"></p>', expect)
-- 
cgit v1.2.3