File 0002-CVE-2020-27783.patch of Package python-lxml

Overview Repositories Revisions Requests Users Attributes Meta

File 0002-CVE-2020-27783.patch of Package python-lxml

From a105ab8dc262ec6735977c25c13f0bdfcdec72a7 Mon Sep 17 00:00:00 2001
From: Stefan Behnel <stefan_ml@behnel.de>
Date: Thu, 26 Nov 2020 09:20:52 +0100
Subject: [PATCH] Prevent combinations of <math/svg> and <style> to sneak
 JavaScript through the HTML cleaner.

---
 CHANGES.txt                        | 11 +++++++++++
 src/lxml/html/clean.py             | 22 ++++++++++++++--------
 src/lxml/html/tests/test_clean.py  | 10 ++++++++++
 src/lxml/html/tests/test_clean.txt | 18 +++++++++++++++---
 4 files changed, 50 insertions(+), 11 deletions(-)

diff --git a/CHANGES.txt b/CHANGES.txt
index 7afec7e28..e3b771401 100644
--- a/CHANGES.txt
+++ b/CHANGES.txt
@@ -2,6 +2,17 @@
 lxml changelog
 ==============
 
+4.6.2 (2020-11-26)
+==================
+
+Bugs fixed
+----------
+
+* A vulnerability (CVE-2020-27783) was discovered in the HTML Cleaner by Yaniv Nizry,
+  which allowed JavaScript to pass through.  The cleaner now removes more sneaky
+  "style" content.
+
+
 4.6.1 (2020-10-18)
 ==================
 
diff --git a/src/lxml/html/clean.py b/src/lxml/html/clean.py
index 7b51981d7..0fa1544c4 100644
--- a/src/lxml/html/clean.py
+++ b/src/lxml/html/clean.py
@@ -61,12 +61,15 @@
 
 # This is an IE-specific construct you can have in a stylesheet to
 # run some Javascript:
-_css_javascript_re = re.compile(
-    r'expression\s*\(.*?\)', re.S|re.I)
+_replace_css_javascript = re.compile(
+    r'expression\s*\(.*?\)', re.S|re.I).sub
 
 # Do I have to worry about @\nimport?
-_css_import_re = re.compile(
-    r'@\s*import', re.I)
+_replace_css_import = re.compile(
+    r'@\s*import', re.I).sub
+
+_looks_like_tag_content = re.compile(
+    r'</?[a-zA-Z]+|\son[a-zA-Z]+\s*=', ).search
 
 # All kinds of schemes besides just javascript: that can cause
 # execution:
@@ -304,8 +307,8 @@ def __call__(self, doc):
             if not self.inline_style:
                 for el in _find_styled_elements(doc):
                     old = el.get('style')
-                    new = _css_javascript_re.sub('', old)
-                    new = _css_import_re.sub('', new)
+                    new = _replace_css_javascript('', old)
+                    new = _replace_css_import('', new)
                     if self._has_sneaky_javascript(new):
                         # Something tricky is going on...
                         del el.attrib['style']
@@ -317,9 +320,9 @@ def __call__(self, doc):
                         el.drop_tree()
                         continue
                     old = el.text or ''
-                    new = _css_javascript_re.sub('', old)
+                    new = _replace_css_javascript('', old)
                     # The imported CSS can do anything; we just can't allow:
-                    new = _css_import_re.sub('', old)
+                    new = _replace_css_import('', new)
                     if self._has_sneaky_javascript(new):
                         # Something tricky is going on...
                         el.text = '/* deleted */'
@@ -539,6 +542,9 @@ def _has_sneaky_javascript(self, style):
         if '</noscript' in style:
             # e.g. '<noscript><style><a title="</noscript><img src=x onerror=alert(1)>">'
             return True
+        if _looks_like_tag_content(style):
+            # e.g. '<math><style><img src=x onerror=alert(1)></style></math>'
+            return True
         return False
 
     def clean_html(self, html):
diff --git a/src/lxml/html/tests/test_clean.py b/src/lxml/html/tests/test_clean.py
index 3c8ee252f..0e669f98d 100644
--- a/src/lxml/html/tests/test_clean.py
+++ b/src/lxml/html/tests/test_clean.py
@@ -113,6 +113,16 @@ def test_sneaky_noscript_in_style(self):
             b'<noscript><style>/* deleted */</style></noscript>',
             lxml.html.tostring(clean_html(s)))
 
+    def test_sneaky_js_in_math_style(self):
+        # This gets parsed as <math> -> <style>"..."</style>
+        # thus passing any tag/script/whatever content through into the output.
+        html = '<math><style><img src=x onerror=alert(1)></style></math>'
+        s = lxml.html.fragment_fromstring(html)
+
+        self.assertEqual(
+            b'<math><style>/* deleted */</style></math>',
+            lxml.html.tostring(clean_html(s)))
+
 
 def test_suite():
     suite = unittest.TestSuite()
diff --git a/src/lxml/html/tests/test_clean.txt b/src/lxml/html/tests/test_clean.txt
index 275be07c6..18e6c7e61 100644
--- a/src/lxml/html/tests/test_clean.txt
+++ b/src/lxml/html/tests/test_clean.txt
@@ -104,7 +104,11 @@
 >>> print(Cleaner(page_structure=False, comments=False).clean_html(doc))
 <html>
   <head>
-    <style>/* deleted */</style>
+    <style>
+      body {background-image: url()};
+      div {background-image: url()};
+      div {color: };
+    </style>
   </head>
   <body>
     
@@ -126,7 +130,11 @@
 >>> print(Cleaner(page_structure=False, safe_attrs_only=False).clean_html(doc))
 <html>
   <head>
-    <style>/* deleted */</style>
+    <style>
+      body {background-image: url()};
+      div {background-image: url()};
+      div {color: };
+    </style>
   </head>
   <body>
     <a href="">a link</a>

Places

File 0002-CVE-2020-27783.patch of Package python-lxml

Places