From a8e06b7cef5c5b0b372b9a0db0f56c4e7094e2bd Mon Sep 17 00:00:00 2001
From: Waylan Limberg <waylan@gmail.com>
Date: Thu, 13 Dec 2012 14:05:33 -0500
Subject: Fixed #164. attr_list extension attribute names are now sanitized and
 won't crash the serealizer.

---
 markdown/extensions/attr_list.py | 15 +++++++++++++--
 tests/extensions/attr_list.html  |  3 ++-
 tests/extensions/attr_list.txt   |  1 +
 3 files changed, 16 insertions(+), 3 deletions(-)

diff --git a/markdown/extensions/attr_list.py b/markdown/extensions/attr_list.py
index 0aa18e0..36f3e3a 100644
--- a/markdown/extensions/attr_list.py
+++ b/markdown/extensions/attr_list.py
@@ -67,6 +67,10 @@ class AttrListTreeprocessor(markdown.treeprocessors.Treeprocessor):
     HEADER_RE = re.compile(r'[ ]*%s[ ]*$' % BASE_RE)
     BLOCK_RE = re.compile(r'\n[ ]*%s[ ]*$' % BASE_RE)
     INLINE_RE = re.compile(r'^%s' % BASE_RE)
+    NAME_RE = re.compile(r'[^A-Z_a-z\xc0-\xd6\xd8-\xf6\u00f8-\u02ff\u0370-\u037d'
+                          '\u037f-\u1fff\u200c-\u200d\u2070-\u218f\u2c00-\u2fef'
+                          '\u3001-\ud7ff\uf900-\ufdcf\ufdf0-\ufffd\u10000-\ueffff'
+                          '\-\.0-9\xb7\u0300-\u036f\u203f-\u2040]+')
 
     def run(self, doc):
         for elem in doc.getiterator():
@@ -114,8 +118,15 @@ class AttrListTreeprocessor(markdown.treeprocessors.Treeprocessor):
                 else:
                     elem.set('class', v)
             else:
-                # assing attr k with v
-                elem.set(k, v)
+                # assign attr k with v
+                elem.set(self.sanitize_name(k), v)
+
+    def sanitize_name(self, name):
+        """
+        Sanitize name as 'an XML Name, minus the ":"'.
+        See http://www.w3.org/TR/REC-xml-names/#NT-NCName
+        """
+        return self.NAME_RE.sub('_', name)
 
 
 class AttrListExtension(markdown.extensions.Extension):
diff --git a/tests/extensions/attr_list.html b/tests/extensions/attr_list.html
index 1e9c182..f50cd6a 100644
--- a/tests/extensions/attr_list.html
+++ b/tests/extensions/attr_list.html
@@ -14,4 +14,5 @@ And a <strong class="nest">nested <a class="linky2" href="http://example.com" ti
 {: #someid .someclass }
 </code></pre>
 <h3 id="hash3">No colon for compatability with Headerid ext</h3>
-<p id="the_end">Also a codespan: <code class="foo">{: .someclass}</code>.</p>
\ No newline at end of file
+<p id="the_end">Also a codespan: <code class="foo">{: .someclass}</code>.</p>
+<h3 _:="{:" id="hash5">Bad Syntax</h3>
\ No newline at end of file
diff --git a/tests/extensions/attr_list.txt b/tests/extensions/attr_list.txt
index d7ed274..cd7f398 100644
--- a/tests/extensions/attr_list.txt
+++ b/tests/extensions/attr_list.txt
@@ -33,3 +33,4 @@ Now test overrides
 Also a codespan: `{: .someclass}`{: .foo}.
 {: #the_end}
 
+### Bad Syntax { {: #hash5 }
-- 
cgit v1.2.3