hohn/codeql
1
0
Fork 0
mirror of https://github.com/github/codeql.git synced 2026-04-29 18:55:14 +02:00
Code Issues Packages Projects Releases Wiki Activity

Python: Change XmlBomb vulnerability kind

Browse Source
This commit is contained in:
Rasmus Wriedt Larsen
2022-04-07 10:45:43 +02:00
parent f2f0873d91
commit 7728b6cf1b
12 changed files with 73 additions and 78 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
python/ql/lib/semmle/python/Concepts.qll
View File

@@ -507,15 +507,18 @@ module XML {
* See PoC at `python/PoCs/XmlParsing/PoC.py` for some tests of vulnerable XML parsing.
*/
class XmlParsingVulnerabilityKind extends string {
XmlParsingVulnerabilityKind() {
this in ["Billion Laughs", "Quadratic Blowup", "XXE", "DTD retrieval"]
}
XmlParsingVulnerabilityKind() { this in ["XML bomb", "XXE", "DTD retrieval"] }
/** Holds for Billion Laughs vulnerability kind. */
predicate isBillionLaughs() { this = "Billion Laughs" }
/** Holds for Quadratic Blowup vulnerability kind. */
predicate isQuadraticBlowup() { this = "Quadratic Blowup" }
/**
* Holds for XML bomb vulnerability kind, such as 'Billion Laughs' and 'Quadratic
* Blowup'.
*
* While a parser could technically be vulnerable to one and not the other, from our
* point of view the interesting part is that it IS vulnerable to these types of
* attacks, and not so much which one specifically works. In practice I haven't seen
* a parser that is vulnerable to one and not the other.
*/
predicate isXmlBomb() { this = "XML bomb" }
/** Holds for XXE vulnerability kind. */
predicate isXxe() { this = "XXE" }

4
python/ql/lib/semmle/python/frameworks/Lxml.qll
View File

@@ -144,7 +144,7 @@ private module Lxml {
this.getKeywordParameter("resolve_entities").getAValueReachingRhs().asExpr() = any(True t)
)
or
(kind.isBillionLaughs() or kind.isQuadraticBlowup()) and
kind.isXmlBomb() and
this.getKeywordParameter("huge_tree").getAValueReachingRhs().asExpr() = any(True t) and
not this.getKeywordParameter("resolve_entities").getAValueReachingRhs().asExpr() =
any(False t)
@@ -318,7 +318,7 @@ private module Lxml {
// note that there is no `resolve_entities` argument, so it's not possible to turn off XXE :O
kind.isXxe()
or
(kind.isBillionLaughs() or kind.isQuadraticBlowup()) and
kind.isXmlBomb() and
this.getKeywordParameter("huge_tree").getAValueReachingRhs().asExpr() = any(True t)
or
kind.isDtdRetrieval() and

12
python/ql/lib/semmle/python/frameworks/Stdlib.qll
View File

@@ -3241,9 +3241,7 @@ private module StdlibPrivate {
override DataFlow::Node getAnInput() { result in [this.getArg(0), this.getArgByName("data")] }
override predicate vulnerableTo(XML::XmlParsingVulnerabilityKind kind) {
kind.isBillionLaughs() or kind.isQuadraticBlowup()
}
override predicate vulnerableTo(XML::XmlParsingVulnerabilityKind kind) { kind.isXmlBomb() }
override predicate mayExecuteInput() { none() }
@@ -3301,7 +3299,7 @@ private module StdlibPrivate {
override predicate vulnerableTo(XML::XmlParsingVulnerabilityKind kind) {
// note: it does not matter what `xml.etree` parser you are using, you cannot
// change the security features anyway :|
kind.isBillionLaughs() or kind.isQuadraticBlowup()
kind.isXmlBomb()
}
override predicate mayExecuteInput() { none() }
@@ -3461,7 +3459,7 @@ private module StdlibPrivate {
override predicate vulnerableTo(XML::XmlParsingVulnerabilityKind kind) {
// always vuln to these
(kind.isBillionLaughs() or kind.isQuadraticBlowup())
kind.isXmlBomb()
or
// can be vuln to other things if features has been turned on
this.getObject() = saxParserWithFeatureExternalGesTurnedOn() and
@@ -3514,7 +3512,7 @@ private module StdlibPrivate {
override predicate vulnerableTo(XML::XmlParsingVulnerabilityKind kind) {
// always vuln to these
(kind.isBillionLaughs() or kind.isQuadraticBlowup())
kind.isXmlBomb()
}
override predicate mayExecuteInput() { none() }
@@ -3590,7 +3588,7 @@ private module StdlibPrivate {
this.getParserArg() = saxParserWithFeatureExternalGesTurnedOn() and
(kind.isXxe() or kind.isDtdRetrieval())
or
(kind.isBillionLaughs() or kind.isQuadraticBlowup())
kind.isXmlBomb()
}
override predicate mayExecuteInput() { none() }

2
python/ql/lib/semmle/python/frameworks/Xmltodict.qll
View File

@@ -28,7 +28,7 @@ private module Xmltodict {
}
override predicate vulnerableTo(XML::XmlParsingVulnerabilityKind kind) {
(kind.isBillionLaughs() or kind.isQuadraticBlowup()) and
kind.isXmlBomb() and
this.getKeywordParameter("disable_entities").getAValueReachingRhs().asExpr() = any(False f)
}

12
python/ql/src/experimental/Security/CWE-611/SimpleXmlRpcServer.ql
View File

@@ -13,13 +13,7 @@ private import python
private import semmle.python.Concepts
private import semmle.python.ApiGraphs
from DataFlow::CallCfgNode call, string kinds
from DataFlow::CallCfgNode call
where
call = API::moduleImport("xmlrpc").getMember("server").getMember("SimpleXMLRPCServer").getACall() and
kinds =
strictconcat(XML::XmlParsingVulnerabilityKind kind |
kind.isBillionLaughs() or kind.isQuadraticBlowup()
|
kind, ", "
)
select call, "SimpleXMLRPCServer is vulnerable to: " + kinds + "."
call = API::moduleImport("xmlrpc").getMember("server").getMember("SimpleXMLRPCServer").getACall()
select call, "SimpleXMLRPCServer is vulnerable to XML bombs"

2
python/ql/src/experimental/semmle/python/security/dataflow/XmlBombCustomizations.qll
View File

@@ -41,7 +41,7 @@ module XmlBomb {
class XmlParsingWithEntityResolution extends Sink {
XmlParsingWithEntityResolution() {
exists(XML::XmlParsing parsing, XML::XmlParsingVulnerabilityKind kind |
(kind.isBillionLaughs() or kind.isQuadraticBlowup()) and
kind.isXmlBomb() and
parsing.vulnerableTo(kind) and
this = parsing.getAnInput()
)

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='Billion Laughs' xmlVuln='Quadratic Blowup' xmlVuln='XXE' decodeOutput=lxml.etree.fromstring(..)
lxml.etree.fromstring(x, parser=parser) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb' 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='Billion Laughs' xmlVuln='Quadratic Blowup' xmlVuln='XXE' decodeOutput=lxml.etree.iterparse(..) getAPathArgument=xml_file
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, load_dtd=True, no_network=False) # $ decodeFormat=XML decodeInput=xml_file xmlVuln='DTD retrieval' xmlVuln='XXE' decodeOutput=lxml.etree.iterparse(..) getAPathArgument=xml_file

6
python/ql/test/library-tests/frameworks/stdlib/XPathExecution.py
View File

@@ -2,7 +2,7 @@ match = "dc:title"
ns = {'dc': 'http://purl.org/dc/elements/1.1/'}
import xml.etree.ElementTree as ET
tree = ET.parse('country_data.xml') # $ decodeFormat=XML decodeInput='country_data.xml' decodeOutput=ET.parse(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' getAPathArgument='country_data.xml'
tree = ET.parse('country_data.xml') # $ decodeFormat=XML decodeInput='country_data.xml' decodeOutput=ET.parse(..) xmlVuln='XML bomb' getAPathArgument='country_data.xml'
root = tree.getroot()
root.find(match, namespaces=ns) # $ getXPath=match
@@ -10,13 +10,13 @@ root.findall(match, namespaces=ns) # $ getXPath=match
root.findtext(match, default=None, namespaces=ns) # $ getXPath=match
tree = ET.ElementTree()
tree.parse("index.xhtml") # $ decodeFormat=XML decodeInput="index.xhtml" decodeOutput=tree.parse(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' getAPathArgument="index.xhtml"
tree.parse("index.xhtml") # $ decodeFormat=XML decodeInput="index.xhtml" decodeOutput=tree.parse(..) xmlVuln='XML bomb' getAPathArgument="index.xhtml"
tree.find(match, namespaces=ns) # $ getXPath=match
tree.findall(match, namespaces=ns) # $ getXPath=match
tree.findtext(match, default=None, namespaces=ns) # $ getXPath=match
parser = ET.XMLParser()
parser.feed("<foo>bar</foo>") # $ decodeFormat=XML decodeInput="<foo>bar</foo>" xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup'
parser.feed("<foo>bar</foo>") # $ decodeFormat=XML decodeInput="<foo>bar</foo>" xmlVuln='XML bomb'
tree = parser.close() # $ decodeOutput=parser.close()
tree.find(match, namespaces=ns) # $ getXPath=match

24
python/ql/test/library-tests/frameworks/stdlib/xml_dom.py
View File

@@ -6,26 +6,26 @@ import xml.sax
x = "some xml"
# minidom
xml.dom.minidom.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.dom.minidom.parse(..) getAPathArgument=StringIO(..)
xml.dom.minidom.parse(file=StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.dom.minidom.parse(..) getAPathArgument=StringIO(..)
xml.dom.minidom.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' decodeOutput=xml.dom.minidom.parse(..) getAPathArgument=StringIO(..)
xml.dom.minidom.parse(file=StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' decodeOutput=xml.dom.minidom.parse(..) getAPathArgument=StringIO(..)
xml.dom.minidom.parseString(x) # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.dom.minidom.parseString(..)
xml.dom.minidom.parseString(string=x) # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.dom.minidom.parseString(..)
xml.dom.minidom.parseString(x) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb' decodeOutput=xml.dom.minidom.parseString(..)
xml.dom.minidom.parseString(string=x) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb' decodeOutput=xml.dom.minidom.parseString(..)
# pulldom
xml.dom.pulldom.parse(StringIO(x))['START_DOCUMENT'][1] # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.dom.pulldom.parse(..) getAPathArgument=StringIO(..)
xml.dom.pulldom.parse(stream_or_string=StringIO(x))['START_DOCUMENT'][1] # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.dom.pulldom.parse(..) getAPathArgument=StringIO(..)
xml.dom.pulldom.parse(StringIO(x))['START_DOCUMENT'][1] # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' decodeOutput=xml.dom.pulldom.parse(..) getAPathArgument=StringIO(..)
xml.dom.pulldom.parse(stream_or_string=StringIO(x))['START_DOCUMENT'][1] # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' decodeOutput=xml.dom.pulldom.parse(..) getAPathArgument=StringIO(..)
xml.dom.pulldom.parseString(x)['START_DOCUMENT'][1] # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.dom.pulldom.parseString(..)
xml.dom.pulldom.parseString(string=x)['START_DOCUMENT'][1] # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.dom.pulldom.parseString(..)
xml.dom.pulldom.parseString(x)['START_DOCUMENT'][1] # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb' decodeOutput=xml.dom.pulldom.parseString(..)
xml.dom.pulldom.parseString(string=x)['START_DOCUMENT'][1] # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb' decodeOutput=xml.dom.pulldom.parseString(..)
# These are based on SAX parses, and you can specify your own, so you can expose yourself to XXE (yay/)
parser = xml.sax.make_parser()
parser.setFeature(xml.sax.handler.feature_external_ges, True)
xml.dom.minidom.parse(StringIO(x), parser) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='DTD retrieval' xmlVuln='Quadratic Blowup' xmlVuln='XXE' decodeOutput=xml.dom.minidom.parse(..) getAPathArgument=StringIO(..)
xml.dom.minidom.parse(StringIO(x), parser=parser) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='DTD retrieval' xmlVuln='Quadratic Blowup' xmlVuln='XXE' decodeOutput=xml.dom.minidom.parse(..) getAPathArgument=StringIO(..)
xml.dom.minidom.parse(StringIO(x), parser) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' xmlVuln='DTD retrieval' xmlVuln='XXE' decodeOutput=xml.dom.minidom.parse(..) getAPathArgument=StringIO(..)
xml.dom.minidom.parse(StringIO(x), parser=parser) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' xmlVuln='DTD retrieval' xmlVuln='XXE' decodeOutput=xml.dom.minidom.parse(..) getAPathArgument=StringIO(..)
xml.dom.pulldom.parse(StringIO(x), parser) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='DTD retrieval' xmlVuln='Quadratic Blowup' xmlVuln='XXE' decodeOutput=xml.dom.pulldom.parse(..) getAPathArgument=StringIO(..)
xml.dom.pulldom.parse(StringIO(x), parser=parser) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='DTD retrieval' xmlVuln='Quadratic Blowup' xmlVuln='XXE' decodeOutput=xml.dom.pulldom.parse(..) getAPathArgument=StringIO(..)
xml.dom.pulldom.parse(StringIO(x), parser) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' xmlVuln='DTD retrieval' xmlVuln='XXE' decodeOutput=xml.dom.pulldom.parse(..) getAPathArgument=StringIO(..)
xml.dom.pulldom.parse(StringIO(x), parser=parser) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' xmlVuln='DTD retrieval' xmlVuln='XXE' decodeOutput=xml.dom.pulldom.parse(..) getAPathArgument=StringIO(..)

38
python/ql/test/library-tests/frameworks/stdlib/xml_etree.py
View File

@@ -4,43 +4,43 @@ import xml.etree.ElementTree
x = "some xml"
# Parsing in different ways
xml.etree.ElementTree.fromstring(x) # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.etree.ElementTree.fromstring(..)
xml.etree.ElementTree.fromstring(text=x) # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.etree.ElementTree.fromstring(..)
xml.etree.ElementTree.fromstring(x) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb' decodeOutput=xml.etree.ElementTree.fromstring(..)
xml.etree.ElementTree.fromstring(text=x) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb' decodeOutput=xml.etree.ElementTree.fromstring(..)
xml.etree.ElementTree.fromstringlist([x]) # $ decodeFormat=XML decodeInput=List xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.etree.ElementTree.fromstringlist(..)
xml.etree.ElementTree.fromstringlist(sequence=[x]) # $ decodeFormat=XML decodeInput=List xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.etree.ElementTree.fromstringlist(..)
xml.etree.ElementTree.fromstringlist([x]) # $ decodeFormat=XML decodeInput=List xmlVuln='XML bomb' decodeOutput=xml.etree.ElementTree.fromstringlist(..)
xml.etree.ElementTree.fromstringlist(sequence=[x]) # $ decodeFormat=XML decodeInput=List xmlVuln='XML bomb' decodeOutput=xml.etree.ElementTree.fromstringlist(..)
xml.etree.ElementTree.XML(x) # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.etree.ElementTree.XML(..)
xml.etree.ElementTree.XML(text=x) # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.etree.ElementTree.XML(..)
xml.etree.ElementTree.XML(x) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb' decodeOutput=xml.etree.ElementTree.XML(..)
xml.etree.ElementTree.XML(text=x) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb' decodeOutput=xml.etree.ElementTree.XML(..)
xml.etree.ElementTree.XMLID(x) # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.etree.ElementTree.XMLID(..)
xml.etree.ElementTree.XMLID(text=x) # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.etree.ElementTree.XMLID(..)
xml.etree.ElementTree.XMLID(x) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb' decodeOutput=xml.etree.ElementTree.XMLID(..)
xml.etree.ElementTree.XMLID(text=x) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb' decodeOutput=xml.etree.ElementTree.XMLID(..)
xml.etree.ElementTree.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.etree.ElementTree.parse(..) getAPathArgument=StringIO(..)
xml.etree.ElementTree.parse(source=StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.etree.ElementTree.parse(..) getAPathArgument=StringIO(..)
xml.etree.ElementTree.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' decodeOutput=xml.etree.ElementTree.parse(..) getAPathArgument=StringIO(..)
xml.etree.ElementTree.parse(source=StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' decodeOutput=xml.etree.ElementTree.parse(..) getAPathArgument=StringIO(..)
xml.etree.ElementTree.iterparse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.etree.ElementTree.iterparse(..) getAPathArgument=StringIO(..)
xml.etree.ElementTree.iterparse(source=StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.etree.ElementTree.iterparse(..) getAPathArgument=StringIO(..)
xml.etree.ElementTree.iterparse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' decodeOutput=xml.etree.ElementTree.iterparse(..) getAPathArgument=StringIO(..)
xml.etree.ElementTree.iterparse(source=StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' decodeOutput=xml.etree.ElementTree.iterparse(..) getAPathArgument=StringIO(..)
tree = xml.etree.ElementTree.ElementTree()
tree.parse("file.xml") # $ decodeFormat=XML decodeInput="file.xml" xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=tree.parse(..) getAPathArgument="file.xml"
tree.parse(source="file.xml") # $ decodeFormat=XML decodeInput="file.xml" xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=tree.parse(..) getAPathArgument="file.xml"
tree.parse("file.xml") # $ decodeFormat=XML decodeInput="file.xml" xmlVuln='XML bomb' decodeOutput=tree.parse(..) getAPathArgument="file.xml"
tree.parse(source="file.xml") # $ decodeFormat=XML decodeInput="file.xml" xmlVuln='XML bomb' decodeOutput=tree.parse(..) getAPathArgument="file.xml"
# With parsers (no options available to disable/enable security features)
parser = xml.etree.ElementTree.XMLParser()
xml.etree.ElementTree.fromstring(x, parser=parser) # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xml.etree.ElementTree.fromstring(..)
xml.etree.ElementTree.fromstring(x, parser=parser) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb' decodeOutput=xml.etree.ElementTree.fromstring(..)
# manual use of feed method
parser = xml.etree.ElementTree.XMLParser()
parser.feed(x) # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup'
parser.feed(data=x) # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup'
parser.feed(x) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb'
parser.feed(data=x) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb'
parser.close() # $ decodeOutput=parser.close()
# manual use of feed method on XMLPullParser
parser = xml.etree.ElementTree.XMLPullParser()
parser.feed(x) # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup'
parser.feed(data=x) # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup'
parser.feed(x) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb'
parser.feed(data=x) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb'
parser.close() # $ decodeOutput=parser.close()
# note: it's technically possible to use the thing wrapper func `fromstring` with an

26
python/ql/test/library-tests/frameworks/stdlib/xml_sax.py
View File

@@ -10,41 +10,41 @@ class MainHandler(xml.sax.ContentHandler):
def characters(self, data):
self._result.append(data)
xml.sax.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' getAPathArgument=StringIO(..)
xml.sax.parse(source=StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' getAPathArgument=StringIO(..)
xml.sax.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' getAPathArgument=StringIO(..)
xml.sax.parse(source=StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' getAPathArgument=StringIO(..)
xml.sax.parseString(x) # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup'
xml.sax.parseString(string=x) # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup'
xml.sax.parseString(x) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb'
xml.sax.parseString(string=x) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb'
parser = xml.sax.make_parser()
parser.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' getAPathArgument=StringIO(..)
parser.parse(source=StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' getAPathArgument=StringIO(..)
parser.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' getAPathArgument=StringIO(..)
parser.parse(source=StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' getAPathArgument=StringIO(..)
# You can make it vuln to both XXE and DTD retrieval by setting this flag
# see https://docs.python.org/3/library/xml.sax.handler.html#xml.sax.handler.feature_external_ges
parser = xml.sax.make_parser()
parser.setFeature(xml.sax.handler.feature_external_ges, True)
parser.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='DTD retrieval' xmlVuln='Quadratic Blowup' xmlVuln='XXE' getAPathArgument=StringIO(..)
parser.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' xmlVuln='DTD retrieval' xmlVuln='XXE' getAPathArgument=StringIO(..)
parser = xml.sax.make_parser()
parser.setFeature(xml.sax.handler.feature_external_ges, False)
parser.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' getAPathArgument=StringIO(..)
parser.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' getAPathArgument=StringIO(..)
# Forward Type Tracking test
def func(cond):
parser = xml.sax.make_parser()
if cond:
parser.setFeature(xml.sax.handler.feature_external_ges, True)
parser.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='DTD retrieval' xmlVuln='Quadratic Blowup' xmlVuln='XXE' getAPathArgument=StringIO(..)
parser.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' xmlVuln='DTD retrieval' xmlVuln='XXE' getAPathArgument=StringIO(..)
else:
parser.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' getAPathArgument=StringIO(..)
parser.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' getAPathArgument=StringIO(..)
# make it vuln, then making it safe
# a bit of an edge-case, but is nice to be able to handle.
parser = xml.sax.make_parser()
parser.setFeature(xml.sax.handler.feature_external_ges, True)
parser.setFeature(xml.sax.handler.feature_external_ges, False)
parser.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' getAPathArgument=StringIO(..)
parser.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' getAPathArgument=StringIO(..)
def check_conditional_assignment(cond):
parser = xml.sax.make_parser()
@@ -52,7 +52,7 @@ def check_conditional_assignment(cond):
parser.setFeature(xml.sax.handler.feature_external_ges, True)
else:
parser.setFeature(xml.sax.handler.feature_external_ges, False)
parser.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='DTD retrieval' xmlVuln='Quadratic Blowup' xmlVuln='XXE' getAPathArgument=StringIO(..)
parser.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' xmlVuln='DTD retrieval' xmlVuln='XXE' getAPathArgument=StringIO(..)
def check_conditional_assignment2(cond):
parser = xml.sax.make_parser()
@@ -61,4 +61,4 @@ def check_conditional_assignment2(cond):
else:
flag_value = False
parser.setFeature(xml.sax.handler.feature_external_ges, flag_value)
parser.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='Billion Laughs' xmlVuln='DTD retrieval' xmlVuln='Quadratic Blowup' xmlVuln='XXE' getAPathArgument=StringIO(..)
parser.parse(StringIO(x)) # $ decodeFormat=XML decodeInput=StringIO(..) xmlVuln='XML bomb' xmlVuln='DTD retrieval' xmlVuln='XXE' getAPathArgument=StringIO(..)

2
python/ql/test/library-tests/frameworks/xmltodict/test.py
View File

@@ -5,4 +5,4 @@ x = "some xml"
xmltodict.parse(x) # $ decodeFormat=XML decodeInput=x decodeOutput=xmltodict.parse(..)
xmltodict.parse(xml_input=x) # $ decodeFormat=XML decodeInput=x decodeOutput=xmltodict.parse(..)
xmltodict.parse(x, disable_entities=False) # $ decodeFormat=XML decodeInput=x xmlVuln='Billion Laughs' xmlVuln='Quadratic Blowup' decodeOutput=xmltodict.parse(..)
xmltodict.parse(x, disable_entities=False) # $ decodeFormat=XML decodeInput=x xmlVuln='XML bomb' decodeOutput=xmltodict.parse(..)