From: Victor Stinner <victor.stinner@haypocalc.com>
Date: Mon, 24 May 2010 21:46:25 +0000 (+0000)
Subject: Recorded merge of revisions 81500-81501 via svnmerge from
X-Git-Tag: v3.2a1~703
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=e021f4b20622d272349ce9d05b9b1b4f5d183f8a;p=python

Recorded merge of revisions 81500-81501 via svnmerge from
svn+ssh://pythondev@svn.python.org/python/trunk

........
  r81500 | victor.stinner | 2010-05-24 23:33:24 +0200 (lun., 24 mai 2010) | 2 lines

  Issue #6662: Fix parsing of malformatted charref (&#bad;)
........
  r81501 | victor.stinner | 2010-05-24 23:37:28 +0200 (lun., 24 mai 2010) | 2 lines

  Add the author of the last fix (Issue #6662)
........
---

diff --git a/Lib/html/parser.py b/Lib/html/parser.py
index 83a58258c7..c2c7f6bf5d 100644
--- a/Lib/html/parser.py
+++ b/Lib/html/parser.py
@@ -175,6 +175,9 @@ class HTMLParser(_markupbase.ParserBase):
                     i = self.updatepos(i, k)
                     continue
                 else:
+                    if ";" in rawdata[i:]: #bail by consuming &#
+                        self.handle_data(rawdata[0:2])
+                        i = self.updatepos(i, 2)
                     break
             elif startswith('&', i):
                 match = entityref.match(rawdata, i)
diff --git a/Lib/test/test_htmlparser.py b/Lib/test/test_htmlparser.py
index dd74aac09b..e982218dba 100644
--- a/Lib/test/test_htmlparser.py
+++ b/Lib/test/test_htmlparser.py
@@ -136,6 +136,13 @@ text
     ("data", "\n"),
     ])
 
+    def test_malformatted_charref(self):
+        self._run_check("<p>&#bad;</p>", [
+            ("starttag", "p", []),
+            ("data", "&#bad;"),
+            ("endtag", "p"),
+        ])
+
     def test_unclosed_entityref(self):
         self._run_check("&entityref foo", [
             ("entityref", "entityref"),
diff --git a/Misc/ACKS b/Misc/ACKS
index b35acaf555..ee625420e7 100644
--- a/Misc/ACKS
+++ b/Misc/ACKS
@@ -871,3 +871,4 @@ Siebren van der Zee
 Uwe Zessin
 Tarek Ziadé
 Peter Åstrand
+Fredrik Håård
diff --git a/Misc/NEWS b/Misc/NEWS
index 36f374bdc9..df363451ac 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -392,6 +392,9 @@ C-API
 Library
 -------
 
+- Issue #6662: Fix parsing of malformatted charref (&#bad;), patch written by
+  Fredrik Håård
+
 - Issue #8540: Decimal module: rename the Context._clamp attribute to
   Context.clamp and make it public.  This is useful in creating
   contexts that correspond to the decimal interchange formats