hohn/codeql
1
0
Fork 0
mirror of https://github.com/github/codeql.git synced 2026-04-27 01:35:13 +02:00
Code Issues Packages Projects Releases Wiki Activity

Removed lxml.etree.XMLParser from xml bomb sinks

Browse Source
This commit is contained in:
Napalys Klicius
2025-07-15 09:54:55 +02:00
parent 887d80f49f
commit 638f6498f0
3 changed files with 4 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
python/ql/test/library-tests/frameworks/lxml/parsing.py
View File

@@ -50,7 +50,7 @@ lxml.etree.fromstring(x, parser=parser) # $ decodeFormat=XML decodeInput=x xmlVu
# Billion laughs vuln (also XXE)
parser = lxml.etree.XMLParser(huge_tree=True)
lxml.etree.fromstring(x, parser=parser) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb' xmlVuln='XXE' decodeOutput=lxml.etree.fromstring(..)
lxml.etree.fromstring(x, parser=parser) # $ decodeFormat=XML decodeInput=x xmlVuln='XXE' decodeOutput=lxml.etree.fromstring(..)
# Safe for both Billion laughs and XXE
parser = lxml.etree.XMLParser(resolve_entities=False, huge_tree=True)
@@ -63,5 +63,5 @@ lxml.etree.fromstring(x, parser=parser) # $ decodeFormat=XML decodeInput=x xmlVu
# iterparse configurations ... this doesn't use a parser argument but takes MOST (!) of
# the normal XMLParser arguments. Specifically, it doesn't allow disabling XXE :O
lxml.etree.iterparse(xml_file, huge_tree=True) # $ decodeFormat=XML decodeInput=xml_file xmlVuln='XML bomb' xmlVuln='XXE' decodeOutput=lxml.etree.iterparse(..) getAPathArgument=xml_file
lxml.etree.iterparse(xml_file, huge_tree=True) # $ decodeFormat=XML decodeInput=xml_file xmlVuln='XXE' decodeOutput=lxml.etree.iterparse(..) getAPathArgument=xml_file
lxml.etree.iterparse(xml_file, load_dtd=True, no_network=False) # $ decodeFormat=XML decodeInput=xml_file xmlVuln='DTD retrieval' xmlVuln='XXE' decodeOutput=lxml.etree.iterparse(..) getAPathArgument=xml_file

10
python/ql/test/query-tests/Security/CWE-776-XmlBomb/XmlBomb.expected
View File

@@ -1,14 +1,4 @@
edges
| test.py:1:26:1:32 | ControlFlowNode for ImportMember | test.py:1:26:1:32 | ControlFlowNode for request | provenance | |
| test.py:1:26:1:32 | ControlFlowNode for request | test.py:19:19:19:25 | ControlFlowNode for request | provenance | |
| test.py:19:5:19:15 | ControlFlowNode for xml_content | test.py:30:34:30:44 | ControlFlowNode for xml_content | provenance | |
| test.py:19:19:19:25 | ControlFlowNode for request | test.py:19:5:19:15 | ControlFlowNode for xml_content | provenance | AdditionalTaintStep |
nodes
| test.py:1:26:1:32 | ControlFlowNode for ImportMember | semmle.label | ControlFlowNode for ImportMember |
| test.py:1:26:1:32 | ControlFlowNode for request | semmle.label | ControlFlowNode for request |
| test.py:19:5:19:15 | ControlFlowNode for xml_content | semmle.label | ControlFlowNode for xml_content |
| test.py:19:19:19:25 | ControlFlowNode for request | semmle.label | ControlFlowNode for request |
| test.py:30:34:30:44 | ControlFlowNode for xml_content | semmle.label | ControlFlowNode for xml_content |
subpaths
#select
| test.py:30:34:30:44 | ControlFlowNode for xml_content | test.py:1:26:1:32 | ControlFlowNode for ImportMember | test.py:30:34:30:44 | ControlFlowNode for xml_content | XML parsing depends on a $@ without guarding against uncontrolled entity expansion. | test.py:1:26:1:32 | ControlFlowNode for ImportMember | user-provided value |