Merge pull request #734 from LuisFigueroaG/fix/empty-list-items

Fix empty list items being parsed as nested lists

Author: nicholasserra

CHANGES.md

@@ -19,6 +19,7 @@
 - [pull #720] Add `wiki-links` extra for `[[Page Name]]` style links (#221)
 - [pull #722] Harden URL safety checks and sanitization in safe mode (#721)
 - [pull #728] Fix XSS from zero-padded colon entities in link URLs (#726)
+- [pull #734] Fix empty list items being parsed as nested lists (#167)
 
 
 ## python-markdown2 2.5.5

lib/markdown2.py

@@ -1883,7 +1883,7 @@ def _do_lists(self, text: str) -> str:
         (\n)?                   # leading line = \1
         (^[ \t]*)               # leading whitespace = \2
         (?P<marker>{}) [ \t]+   # list marker = \3
-        ((?:.+?)                # list item text = \4
+        ((?:.*?)                # list item text = \4 (may be empty)
         (\n{{1,2}}))              # eols = \5
         (?= \n* (\Z | \2 (?P<next_marker>{}) [ \t]+))
         '''.format(_marker_any, _marker_any),

test/tm-cases/empty_list_items_issue167.html

@@ -0,0 +1,38 @@
+<ol>
+<li></li>
+<li></li>
+</ol>
+
+<p>Empty bullet items:</p>
+
+<ul>
+<li></li>
+<li></li>
+</ul>
+
+<p>An empty item between others:</p>
+
+<ul>
+<li>foo</li>
+<li></li>
+<li>bar</li>
+</ul>
+
+<p>An empty numbered item:</p>
+
+<ol>
+<li>foo</li>
+<li></li>
+<li>bar</li>
+</ol>
+
+<p>Empty items in a sub-list:</p>
+
+<ul>
+<li>foo
+<ul>
+<li></li>
+<li></li>
+</ul></li>
+<li>bar</li>
+</ul>

test/tm-cases/empty_list_items_issue167.text

@@ -0,0 +1,26 @@
+ 1. 
+ 2. 
+
+Empty bullet items:
+
+- 
+- 
+
+An empty item between others:
+
+- foo
+- 
+- bar
+
+An empty numbered item:
+
+1. foo
+2. 
+3. bar
+
+Empty items in a sub-list:
+
+- foo
+    - 
+    - 
+- bar

test/tm-cases/xss_smuggling_spans_in_image_attrs.html

@@ -9,10 +9,8 @@
 <p><a href="#">x</a></p>
 
 <ul>
-<li>
-<ul>
+<li></li>
 <li><img src="x" alt="" title="&lt;code&gt;![](&lt;/code&gt;" /> onerror=alert(origin) )</li>
-</ul></li>
 </ul>
 
 <p><img src="code&gt;&lt;A B=&quot;