mirror of
https://github.com/github/codeql.git
synced 2025-12-18 01:33:15 +01:00
Merge branch 'main' into stdlib-optparse
This commit is contained in:
@@ -1,3 +1,13 @@
|
||||
## 2.1.0
|
||||
|
||||
### New Features
|
||||
|
||||
* Added support for custom threat-models, which can be used in most of our taint-tracking queries, see our [documentation](https://docs.github.com/en/code-security/code-scanning/creating-an-advanced-setup-for-code-scanning/customizing-your-advanced-setup-for-code-scanning#extending-codeql-coverage-with-threat-models) for more details.
|
||||
|
||||
### Minor Analysis Improvements
|
||||
|
||||
* The common sanitizer guard `StringConstCompareBarrier` has been renamed to `ConstCompareBarrier` and expanded to cover comparisons with other constant values such as `None`. This may result in fewer false positive results for several queries.
|
||||
|
||||
## 2.0.0
|
||||
|
||||
### Breaking Changes
|
||||
|
||||
@@ -1,4 +0,0 @@
|
||||
---
|
||||
category: minorAnalysis
|
||||
---
|
||||
* The common sanitizer guard `StringConstCompareBarrier` has been renamed to `ConstCompareBarrier` and expanded to cover comparisons with other constant values such as `None`. This may result in fewer false positive results for several queries.
|
||||
9
python/ql/lib/change-notes/released/2.1.0.md
Normal file
9
python/ql/lib/change-notes/released/2.1.0.md
Normal file
@@ -0,0 +1,9 @@
|
||||
## 2.1.0
|
||||
|
||||
### New Features
|
||||
|
||||
* Added support for custom threat-models, which can be used in most of our taint-tracking queries, see our [documentation](https://docs.github.com/en/code-security/code-scanning/creating-an-advanced-setup-for-code-scanning/customizing-your-advanced-setup-for-code-scanning#extending-codeql-coverage-with-threat-models) for more details.
|
||||
|
||||
### Minor Analysis Improvements
|
||||
|
||||
* The common sanitizer guard `StringConstCompareBarrier` has been renamed to `ConstCompareBarrier` and expanded to cover comparisons with other constant values such as `None`. This may result in fewer false positive results for several queries.
|
||||
@@ -1,2 +1,2 @@
|
||||
---
|
||||
lastReleaseVersion: 2.0.0
|
||||
lastReleaseVersion: 2.1.0
|
||||
|
||||
8
python/ql/lib/ext/default-threat-models-fixup.model.yml
Normal file
8
python/ql/lib/ext/default-threat-models-fixup.model.yml
Normal file
@@ -0,0 +1,8 @@
|
||||
extensions:
|
||||
- addsTo:
|
||||
pack: codeql/threat-models
|
||||
extensible: threatModelConfiguration
|
||||
data:
|
||||
# Since responses are enabled by default in the shared threat-models configuration,
|
||||
# we need to disable it here to keep existing behavior for the python analysis.
|
||||
- ["response", false, -2147483647]
|
||||
@@ -1,5 +1,5 @@
|
||||
name: codeql/python-all
|
||||
version: 2.0.1-dev
|
||||
version: 2.1.1-dev
|
||||
groups: python
|
||||
dbscheme: semmlecode.python.dbscheme
|
||||
extractor: python
|
||||
@@ -9,10 +9,12 @@ dependencies:
|
||||
codeql/dataflow: ${workspace}
|
||||
codeql/mad: ${workspace}
|
||||
codeql/regex: ${workspace}
|
||||
codeql/threat-models: ${workspace}
|
||||
codeql/tutorial: ${workspace}
|
||||
codeql/util: ${workspace}
|
||||
codeql/xml: ${workspace}
|
||||
codeql/yaml: ${workspace}
|
||||
dataExtensions:
|
||||
- semmle/python/frameworks/**/*.model.yml
|
||||
- ext/*.model.yml
|
||||
warnOnImplicitThis: true
|
||||
|
||||
@@ -10,6 +10,62 @@ private import semmle.python.dataflow.new.RemoteFlowSources
|
||||
private import semmle.python.dataflow.new.TaintTracking
|
||||
private import semmle.python.Frameworks
|
||||
private import semmle.python.security.internal.EncryptionKeySizes
|
||||
private import codeql.threatmodels.ThreatModels
|
||||
|
||||
/**
|
||||
* A data flow source, for a specific threat-model.
|
||||
*
|
||||
* Extend this class to refine existing API models. If you want to model new APIs,
|
||||
* extend `ThreatModelSource::Range` instead.
|
||||
*/
|
||||
class ThreatModelSource extends DataFlow::Node instanceof ThreatModelSource::Range {
|
||||
/**
|
||||
* Gets a string that represents the source kind with respect to threat modeling.
|
||||
*
|
||||
* See
|
||||
* - https://github.com/github/codeql/blob/main/docs/codeql/reusables/threat-model-description.rst
|
||||
* - https://github.com/github/codeql/blob/main/shared/threat-models/ext/threat-model-grouping.model.yml
|
||||
*/
|
||||
string getThreatModel() { result = super.getThreatModel() }
|
||||
|
||||
/** Gets a string that describes the type of this threat-model source. */
|
||||
string getSourceType() { result = super.getSourceType() }
|
||||
}
|
||||
|
||||
/** Provides a class for modeling new sources for specific threat-models. */
|
||||
module ThreatModelSource {
|
||||
/**
|
||||
* A data flow source, for a specific threat-model.
|
||||
*
|
||||
* Extend this class to model new APIs. If you want to refine existing API models,
|
||||
* extend `ThreatModelSource` instead.
|
||||
*/
|
||||
abstract class Range extends DataFlow::Node {
|
||||
/**
|
||||
* Gets a string that represents the source kind with respect to threat modeling.
|
||||
*
|
||||
* See
|
||||
* - https://github.com/github/codeql/blob/main/docs/codeql/reusables/threat-model-description.rst
|
||||
* - https://github.com/github/codeql/blob/main/shared/threat-models/ext/threat-model-grouping.model.yml
|
||||
*/
|
||||
abstract string getThreatModel();
|
||||
|
||||
/** Gets a string that describes the type of this threat-model source. */
|
||||
abstract string getSourceType();
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* A data flow source that is enabled in the current threat model configuration.
|
||||
*/
|
||||
class ActiveThreatModelSource extends ThreatModelSource {
|
||||
ActiveThreatModelSource() {
|
||||
exists(string kind |
|
||||
currentThreatModel(kind) and
|
||||
this.getThreatModel() = kind
|
||||
)
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* A data-flow node that executes an operating system command,
|
||||
|
||||
@@ -15,10 +15,7 @@ private import semmle.python.Concepts
|
||||
* Extend this class to refine existing API models. If you want to model new APIs,
|
||||
* extend `RemoteFlowSource::Range` instead.
|
||||
*/
|
||||
class RemoteFlowSource extends DataFlow::Node instanceof RemoteFlowSource::Range {
|
||||
/** Gets a string that describes the type of this remote flow source. */
|
||||
string getSourceType() { result = super.getSourceType() }
|
||||
}
|
||||
class RemoteFlowSource extends ThreatModelSource instanceof RemoteFlowSource::Range { }
|
||||
|
||||
/** Provides a class for modeling new sources of remote user input. */
|
||||
module RemoteFlowSource {
|
||||
@@ -28,8 +25,7 @@ module RemoteFlowSource {
|
||||
* Extend this class to model new APIs. If you want to refine existing API models,
|
||||
* extend `RemoteFlowSource` instead.
|
||||
*/
|
||||
abstract class Range extends DataFlow::Node {
|
||||
/** Gets a string that describes the type of this remote flow source. */
|
||||
abstract string getSourceType();
|
||||
abstract class Range extends ThreatModelSource::Range {
|
||||
override string getThreatModel() { result = "remote" }
|
||||
}
|
||||
}
|
||||
|
||||
@@ -81,6 +81,24 @@ module PEP249 {
|
||||
}
|
||||
}
|
||||
|
||||
/** A call to a method that fetches rows from a previous execution. */
|
||||
private class FetchMethodCall extends ThreatModelSource::Range, API::CallNode {
|
||||
FetchMethodCall() {
|
||||
exists(API::Node start |
|
||||
start instanceof DatabaseCursor or start instanceof DatabaseConnection
|
||||
|
|
||||
// note: since we can't currently provide accesspaths for sources, these are all
|
||||
// lumped together, although clearly the fetchmany/fetchall returns a
|
||||
// list/iterable with rows.
|
||||
this = start.getMember(["fetchone", "fetchmany", "fetchall"]).getACall()
|
||||
)
|
||||
}
|
||||
|
||||
override string getThreatModel() { result = "database" }
|
||||
|
||||
override string getSourceType() { result = "cursor.fetch*()" }
|
||||
}
|
||||
|
||||
// ---------------------------------------------------------------------------
|
||||
// asyncio implementations
|
||||
// ---------------------------------------------------------------------------
|
||||
|
||||
29
python/ql/lib/semmle/python/frameworks/Stdlib.model.yml
Normal file
29
python/ql/lib/semmle/python/frameworks/Stdlib.model.yml
Normal file
@@ -0,0 +1,29 @@
|
||||
extensions:
|
||||
- addsTo:
|
||||
pack: codeql/python-all
|
||||
extensible: sourceModel
|
||||
data:
|
||||
- ['os', 'Member[getenv].ReturnValue', 'environment']
|
||||
- ['os', 'Member[getenvb].ReturnValue', 'environment']
|
||||
- ['os', 'Member[environ]', 'environment']
|
||||
- ['os', 'Member[environb]', 'environment']
|
||||
- ['posix', 'Member[environ]', 'environment']
|
||||
|
||||
- ['sys', 'Member[argv]', 'commandargs']
|
||||
- ['sys', 'Member[orig_argv]', 'commandargs']
|
||||
|
||||
- ['sys', 'Member[stdin]', 'stdin']
|
||||
- ['builtins', 'Member[input].ReturnValue', 'stdin']
|
||||
- ['builtins', 'Member[raw_input].ReturnValue', 'stdin'] # python 2 only
|
||||
|
||||
|
||||
# if no argument is given, the default is to use sys.argv[1:]
|
||||
- ['argparse.ArgumentParser', 'Member[parse_args,parse_known_args].WithArity[0].ReturnValue', 'commandargs']
|
||||
|
||||
- ['os', 'Member[read].ReturnValue', 'file']
|
||||
- addsTo:
|
||||
pack: codeql/python-all
|
||||
extensible: summaryModel
|
||||
data:
|
||||
- ['argparse.ArgumentParser', 'Member[parse_args,parse_known_args]', 'Argument[0,args:]', 'ReturnValue', 'taint']
|
||||
# note: taint of attribute lookups is handled in QL
|
||||
@@ -349,7 +349,7 @@ module StdlibPrivate {
|
||||
* Modeling of path related functions in the `os` module.
|
||||
* Wrapped in QL module to make it easy to fold/unfold.
|
||||
*/
|
||||
private module OsFileSystemAccessModeling {
|
||||
module OsFileSystemAccessModeling {
|
||||
/**
|
||||
* A call to the `os.fsencode` function.
|
||||
*
|
||||
@@ -406,7 +406,7 @@ module StdlibPrivate {
|
||||
*
|
||||
* See https://docs.python.org/3/library/os.html#os.open
|
||||
*/
|
||||
private class OsOpenCall extends FileSystemAccess::Range, DataFlow::CallCfgNode {
|
||||
class OsOpenCall extends FileSystemAccess::Range, DataFlow::CallCfgNode {
|
||||
OsOpenCall() { this = os().getMember("open").getACall() }
|
||||
|
||||
override DataFlow::Node getAPathArgument() {
|
||||
@@ -1513,13 +1513,22 @@ module StdlibPrivate {
|
||||
* See https://docs.python.org/3/library/functions.html#open
|
||||
*/
|
||||
private class OpenCall extends FileSystemAccess::Range, Stdlib::FileLikeObject::InstanceSource,
|
||||
DataFlow::CallCfgNode
|
||||
ThreatModelSource::Range, DataFlow::CallCfgNode
|
||||
{
|
||||
OpenCall() { this = getOpenFunctionRef().getACall() }
|
||||
OpenCall() {
|
||||
this = getOpenFunctionRef().getACall() and
|
||||
// when analyzing stdlib code for os.py we wrongly assume that `os.open` is an
|
||||
// alias of the builtins `open` function
|
||||
not this instanceof OsFileSystemAccessModeling::OsOpenCall
|
||||
}
|
||||
|
||||
override DataFlow::Node getAPathArgument() {
|
||||
result in [this.getArg(0), this.getArgByName("file")]
|
||||
}
|
||||
|
||||
override string getThreatModel() { result = "file" }
|
||||
|
||||
override string getSourceType() { result = "open()" }
|
||||
}
|
||||
|
||||
/**
|
||||
@@ -5008,6 +5017,39 @@ module StdlibPrivate {
|
||||
|
||||
override string getKind() { result = Escaping::getHtmlKind() }
|
||||
}
|
||||
|
||||
// ---------------------------------------------------------------------------
|
||||
// argparse
|
||||
// ---------------------------------------------------------------------------
|
||||
/**
|
||||
* if result of `parse_args` is tainted (because it uses command-line arguments),
|
||||
* then the parsed values accesssed on any attribute lookup is also tainted.
|
||||
*/
|
||||
private class ArgumentParserAnyAttributeStep extends TaintTracking::AdditionalTaintStep {
|
||||
override predicate step(DataFlow::Node nodeFrom, DataFlow::Node nodeTo) {
|
||||
nodeFrom =
|
||||
API::moduleImport("argparse")
|
||||
.getMember("ArgumentParser")
|
||||
.getReturn()
|
||||
.getMember("parse_args")
|
||||
.getReturn()
|
||||
.getAValueReachableFromSource() and
|
||||
nodeTo.(DataFlow::AttrRead).getObject() = nodeFrom
|
||||
}
|
||||
}
|
||||
|
||||
// ---------------------------------------------------------------------------
|
||||
// sys
|
||||
// ---------------------------------------------------------------------------
|
||||
/**
|
||||
* An access of `sys.stdin`/`sys.stdout`/`sys.stderr`, to get additional FileLike
|
||||
* modeling.
|
||||
*/
|
||||
private class SysStandardStreams extends Stdlib::FileLikeObject::InstanceSource, DataFlow::Node {
|
||||
SysStandardStreams() {
|
||||
this = API::moduleImport("sys").getMember(["stdin", "stdout", "stderr"]).asSource()
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// ---------------------------------------------------------------------------
|
||||
|
||||
@@ -18,14 +18,19 @@ private import semmle.python.dataflow.new.RemoteFlowSources
|
||||
private import semmle.python.dataflow.new.DataFlow
|
||||
private import semmle.python.ApiGraphs
|
||||
private import semmle.python.dataflow.new.FlowSummary
|
||||
private import semmle.python.Concepts
|
||||
|
||||
/**
|
||||
* A remote flow source originating from a CSV source row.
|
||||
* A threat-model flow source originating from a data extension.
|
||||
*/
|
||||
private class RemoteFlowSourceFromCsv extends RemoteFlowSource::Range {
|
||||
RemoteFlowSourceFromCsv() { this = ModelOutput::getASourceNode("remote").asSource() }
|
||||
private class ThreatModelSourceFromDataExtension extends ThreatModelSource::Range {
|
||||
ThreatModelSourceFromDataExtension() { this = ModelOutput::getASourceNode(_).asSource() }
|
||||
|
||||
override string getSourceType() { result = "Remote flow (from model)" }
|
||||
override string getThreatModel() { this = ModelOutput::getASourceNode(result).asSource() }
|
||||
|
||||
override string getSourceType() {
|
||||
result = "Source node (" + this.getThreatModel() + ") [from data-extension]"
|
||||
}
|
||||
}
|
||||
|
||||
private class SummarizedCallableFromModel extends SummarizedCallable {
|
||||
|
||||
@@ -33,9 +33,14 @@ module CodeInjection {
|
||||
abstract class Sanitizer extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* A code execution, considered as a flow sink.
|
||||
|
||||
@@ -33,9 +33,14 @@ module CommandInjection {
|
||||
abstract class Sanitizer extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* A command execution, considered as a flow sink.
|
||||
|
||||
@@ -31,9 +31,14 @@ module CookieInjection {
|
||||
abstract class Sanitizer extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* A write to a cookie, considered as a sink.
|
||||
|
||||
@@ -32,9 +32,14 @@ module HttpHeaderInjection {
|
||||
abstract class Sanitizer extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* A HTTP header write, considered as a flow sink.
|
||||
|
||||
@@ -42,9 +42,14 @@ module LdapInjection {
|
||||
abstract class FilterSanitizer extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* A logging operation, considered as a flow sink.
|
||||
|
||||
@@ -33,9 +33,14 @@ module LogInjection {
|
||||
abstract class Sanitizer extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* A logging operation, considered as a flow sink.
|
||||
|
||||
@@ -7,6 +7,7 @@ import python
|
||||
import semmle.python.ApiGraphs
|
||||
import semmle.python.dataflow.new.TaintTracking
|
||||
import semmle.python.dataflow.new.RemoteFlowSources
|
||||
import semmle.python.Concepts
|
||||
|
||||
/**
|
||||
* Provides default sources, sinks and sanitizers for detecting
|
||||
@@ -39,9 +40,14 @@ module PamAuthorizationCustomizations {
|
||||
abstract class Sink extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* A vulnerable `pam_authenticate` call considered as a flow sink.
|
||||
|
||||
@@ -43,9 +43,14 @@ module PathInjection {
|
||||
abstract class Sanitizer extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* A file system access, considered as a flow sink.
|
||||
|
||||
@@ -47,9 +47,14 @@ module PolynomialReDoS {
|
||||
abstract class Sanitizer extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* A regex execution, considered as a flow sink.
|
||||
|
||||
@@ -33,9 +33,14 @@ module ReflectedXss {
|
||||
abstract class Sanitizer extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* A data flow sink for "reflected cross-site scripting" vulnerabilities.
|
||||
|
||||
@@ -40,9 +40,14 @@ module RegexInjection {
|
||||
abstract class Sanitizer extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* A regex escaping, considered as a sanitizer.
|
||||
|
||||
@@ -45,9 +45,14 @@ module ServerSideRequestForgery {
|
||||
abstract class FullUrlControlSanitizer extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/** The URL of an HTTP request, considered as a sink. */
|
||||
class HttpRequestUrlAsSink extends Sink {
|
||||
|
||||
@@ -32,9 +32,14 @@ module SqlInjection {
|
||||
abstract class Sanitizer extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* A SQL statement of a SQL construction, considered as a flow sink.
|
||||
|
||||
@@ -33,9 +33,14 @@ module UnsafeDeserialization {
|
||||
abstract class Sanitizer extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* An insecure decoding, considered as a flow sink.
|
||||
|
||||
@@ -77,9 +77,14 @@ module UrlRedirect {
|
||||
}
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* A HTTP redirect response, considered as a flow sink.
|
||||
|
||||
@@ -30,9 +30,14 @@ module XpathInjection {
|
||||
abstract class Sanitizer extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* A construction of an XPath expression, considered as a sink.
|
||||
|
||||
@@ -1,3 +1,9 @@
|
||||
## 1.3.0
|
||||
|
||||
### New Queries
|
||||
|
||||
* The `py/cors-misconfiguration-with-credentials` query, which finds insecure CORS middleware configurations.
|
||||
|
||||
## 1.2.2
|
||||
|
||||
### Minor Analysis Improvements
|
||||
|
||||
@@ -1,4 +0,0 @@
|
||||
---
|
||||
category: newQuery
|
||||
---
|
||||
* The `py/cors-misconfiguration-with-credentials` query, which finds insecure CORS middleware configurations.
|
||||
5
python/ql/src/change-notes/released/1.3.0.md
Normal file
5
python/ql/src/change-notes/released/1.3.0.md
Normal file
@@ -0,0 +1,5 @@
|
||||
## 1.3.0
|
||||
|
||||
### New Queries
|
||||
|
||||
* The `py/cors-misconfiguration-with-credentials` query, which finds insecure CORS middleware configurations.
|
||||
@@ -1,2 +1,2 @@
|
||||
---
|
||||
lastReleaseVersion: 1.2.2
|
||||
lastReleaseVersion: 1.3.0
|
||||
|
||||
@@ -33,9 +33,14 @@ module TemplateInjection {
|
||||
abstract class Sanitizer extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* A SQL statement of a SQL construction, considered as a flow sink.
|
||||
|
||||
@@ -33,9 +33,14 @@ module XsltInjection {
|
||||
abstract class Sanitizer extends DataFlow::Node { }
|
||||
|
||||
/**
|
||||
* A source of remote user input, considered as a flow source.
|
||||
* DEPRECATED: Use `ActiveThreatModelSource` from Concepts instead!
|
||||
*/
|
||||
class RemoteFlowSourceAsSource extends Source, RemoteFlowSource { }
|
||||
deprecated class RemoteFlowSourceAsSource = ActiveThreatModelSourceAsSource;
|
||||
|
||||
/**
|
||||
* An active threat-model source, considered as a flow source.
|
||||
*/
|
||||
private class ActiveThreatModelSourceAsSource extends Source, ActiveThreatModelSource { }
|
||||
|
||||
/**
|
||||
* An XSLT construction, considered as a flow sink.
|
||||
|
||||
@@ -18,7 +18,7 @@ import semmle.python.dataflow.new.RemoteFlowSources
|
||||
import semmle.python.Concepts
|
||||
|
||||
module Js2PyFlowConfig implements DataFlow::ConfigSig {
|
||||
predicate isSource(DataFlow::Node node) { node instanceof RemoteFlowSource }
|
||||
predicate isSource(DataFlow::Node node) { node instanceof ActiveThreatModelSource }
|
||||
|
||||
predicate isSink(DataFlow::Node node) {
|
||||
API::moduleImport("js2py").getMember(["eval_js", "eval_js6", "EvalJs"]).getACall().getArg(_) =
|
||||
|
||||
@@ -1,5 +1,5 @@
|
||||
name: codeql/python-queries
|
||||
version: 1.2.3-dev
|
||||
version: 1.3.1-dev
|
||||
groups:
|
||||
- python
|
||||
- queries
|
||||
|
||||
@@ -3,6 +3,7 @@ import semmle.python.dataflow.new.DataFlow
|
||||
import semmle.python.Concepts
|
||||
import TestUtilities.InlineExpectationsTest
|
||||
private import semmle.python.dataflow.new.internal.PrintNode
|
||||
private import codeql.threatmodels.ThreatModels
|
||||
|
||||
module SystemCommandExecutionTest implements TestSig {
|
||||
string getARelevantTag() { result = "getCommand" }
|
||||
@@ -632,6 +633,22 @@ module XmlParsingTest implements TestSig {
|
||||
}
|
||||
}
|
||||
|
||||
module ThreatModelSourceTest implements TestSig {
|
||||
string getARelevantTag() {
|
||||
exists(string kind | knownThreatModel(kind) | result = "threatModelSource" + "[" + kind + "]")
|
||||
}
|
||||
|
||||
predicate hasActualResult(Location location, string element, string tag, string value) {
|
||||
exists(location.getFile().getRelativePath()) and
|
||||
exists(ThreatModelSource src | not src.getThreatModel() = "remote" |
|
||||
location = src.getLocation() and
|
||||
element = src.toString() and
|
||||
value = prettyNodeForInlineTest(src) and
|
||||
tag = "threatModelSource[" + src.getThreatModel() + "]"
|
||||
)
|
||||
}
|
||||
}
|
||||
|
||||
module CorsMiddlewareTest implements TestSig {
|
||||
string getARelevantTag() { result = "CorsMiddleware" }
|
||||
|
||||
@@ -656,4 +673,4 @@ import MakeTest<MergeTests5<MergeTests5<SystemCommandExecutionTest, DecodingTest
|
||||
MergeTests5<FileSystemAccessTest, FileSystemWriteAccessTest, PathNormalizationTest,
|
||||
SafeAccessCheckTest, PublicKeyGenerationTest>,
|
||||
MergeTests5<CryptographicOperationTest, HttpClientRequestTest, CsrfProtectionSettingTest,
|
||||
CsrfLocalProtectionSettingTest, XmlParsingTest>>>
|
||||
CsrfLocalProtectionSettingTest, MergeTests<XmlParsingTest, ThreatModelSourceTest>>>>
|
||||
|
||||
@@ -15,6 +15,7 @@ import semmle.python.dataflow.new.TaintTracking
|
||||
import semmle.python.dataflow.new.RemoteFlowSources
|
||||
import TestUtilities.InlineExpectationsTest
|
||||
private import semmle.python.dataflow.new.internal.PrintNode
|
||||
private import semmle.python.Concepts
|
||||
|
||||
DataFlow::Node shouldBeTainted() {
|
||||
exists(DataFlow::CallCfgNode call |
|
||||
@@ -45,7 +46,7 @@ module Conf {
|
||||
source.(DataFlow::CfgNode).getNode() = call.getAnArg()
|
||||
)
|
||||
or
|
||||
source instanceof RemoteFlowSource
|
||||
source instanceof ThreatModelSource
|
||||
}
|
||||
|
||||
predicate isSink(DataFlow::Node sink) {
|
||||
|
||||
@@ -6,7 +6,7 @@ import sys
|
||||
|
||||
def main():
|
||||
"""Run administrative tasks."""
|
||||
os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'testproj.settings')
|
||||
os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'testproj.settings') # $ threatModelSource[environment]=os.environ
|
||||
try:
|
||||
from django.core.management import execute_from_command_line
|
||||
except ImportError as exc:
|
||||
@@ -15,7 +15,7 @@ def main():
|
||||
"available on your PYTHONPATH environment variable? Did you "
|
||||
"forget to activate a virtual environment?"
|
||||
) from exc
|
||||
execute_from_command_line(sys.argv)
|
||||
execute_from_command_line(sys.argv) # $ threatModelSource[commandargs]=sys.argv
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
|
||||
@@ -11,6 +11,6 @@ import os
|
||||
|
||||
from django.core.asgi import get_asgi_application
|
||||
|
||||
os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'testproj.settings')
|
||||
os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'testproj.settings') # $ threatModelSource[environment]=os.environ
|
||||
|
||||
application = get_asgi_application()
|
||||
|
||||
@@ -11,6 +11,6 @@ import os
|
||||
|
||||
from django.core.wsgi import get_wsgi_application
|
||||
|
||||
os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'testproj.settings')
|
||||
os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'testproj.settings') # $ threatModelSource[environment]=os.environ
|
||||
|
||||
application = get_wsgi_application()
|
||||
|
||||
@@ -12,3 +12,24 @@ with psycopg.connect(...) as conn:
|
||||
with conn.cursor() as cursor:
|
||||
cursor.execute("some sql", (42,)) # $ getSql="some sql"
|
||||
cursor.executemany("some sql", [(42,)]) # $ getSql="some sql"
|
||||
|
||||
|
||||
### test of threat-model sources
|
||||
row = cursor.fetchone() # $ threatModelSource[database]=cursor.fetchone()
|
||||
rows_many = cursor.fetchmany(10) # $ threatModelSource[database]=cursor.fetchmany(..)
|
||||
rows_all = cursor.fetchall() # $ threatModelSource[database]=cursor.fetchall()
|
||||
|
||||
ensure_tainted(
|
||||
row[0], # $ tainted
|
||||
rows_many[0][0], # $ tainted
|
||||
rows_all[0][0], # $ tainted
|
||||
|
||||
# pretending we created cursor to return dictionary results
|
||||
row["column"], # $ tainted
|
||||
rows_many[0]["column"], # $ tainted
|
||||
rows_all[0]["column"], # $ tainted
|
||||
)
|
||||
for row in rows_many:
|
||||
ensure_tainted(row[0], row["column"]) # $ tainted
|
||||
for row in rows_all:
|
||||
ensure_tainted(row[0], row["column"]) # tainted
|
||||
|
||||
@@ -6,7 +6,7 @@ import sys
|
||||
|
||||
def main():
|
||||
"""Run administrative tasks."""
|
||||
os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'testproj.settings')
|
||||
os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'testproj.settings') # $ threatModelSource[environment]=os.environ
|
||||
try:
|
||||
from django.core.management import execute_from_command_line
|
||||
except ImportError as exc:
|
||||
@@ -15,7 +15,7 @@ def main():
|
||||
"available on your PYTHONPATH environment variable? Did you "
|
||||
"forget to activate a virtual environment?"
|
||||
) from exc
|
||||
execute_from_command_line(sys.argv)
|
||||
execute_from_command_line(sys.argv) # $ threatModelSource[commandargs]=sys.argv
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
|
||||
@@ -17,7 +17,7 @@ p.open("wt").write("hello") # $ getAPathArgument=p fileWriteData="hello"
|
||||
|
||||
name = windows.parent.name
|
||||
o = open
|
||||
o(name) # $ getAPathArgument=name
|
||||
o(name) # $ getAPathArgument=name threatModelSource[file]=o(..)
|
||||
|
||||
wb = p.write_bytes
|
||||
wb(b"hello") # $ getAPathArgument=p fileWriteData=b"hello"
|
||||
|
||||
@@ -5,25 +5,25 @@ import stat
|
||||
import tempfile
|
||||
import shutil
|
||||
|
||||
open("file") # $ getAPathArgument="file"
|
||||
open(file="file") # $ getAPathArgument="file"
|
||||
open("file") # $ getAPathArgument="file" threatModelSource[file]=open(..)
|
||||
open(file="file") # $ getAPathArgument="file" threatModelSource[file]=open(..)
|
||||
|
||||
o = open
|
||||
|
||||
o("file") # $ getAPathArgument="file"
|
||||
o(file="file") # $ getAPathArgument="file"
|
||||
o("file") # $ getAPathArgument="file" threatModelSource[file]=o(..)
|
||||
o(file="file") # $ getAPathArgument="file" threatModelSource[file]=o(..)
|
||||
|
||||
|
||||
builtins.open("file") # $ getAPathArgument="file"
|
||||
builtins.open(file="file") # $ getAPathArgument="file"
|
||||
builtins.open("file") # $ getAPathArgument="file" threatModelSource[file]=builtins.open(..)
|
||||
builtins.open(file="file") # $ getAPathArgument="file" threatModelSource[file]=builtins.open(..)
|
||||
|
||||
|
||||
io.open("file") # $ getAPathArgument="file"
|
||||
io.open(file="file") # $ getAPathArgument="file"
|
||||
io.open("file") # $ getAPathArgument="file" threatModelSource[file]=io.open(..)
|
||||
io.open(file="file") # $ getAPathArgument="file" threatModelSource[file]=io.open(..)
|
||||
io.open_code("file") # $ getAPathArgument="file"
|
||||
io.FileIO("file") # $ getAPathArgument="file"
|
||||
|
||||
f = open("path") # $ getAPathArgument="path"
|
||||
f = open("path") # $ getAPathArgument="path" threatModelSource[file]=open(..)
|
||||
f.write("foo") # $ getAPathArgument="path" fileWriteData="foo"
|
||||
lines = ["foo"]
|
||||
f.writelines(lines) # $ getAPathArgument="path" fileWriteData=lines
|
||||
|
||||
@@ -0,0 +1,71 @@
|
||||
import os
|
||||
import sys
|
||||
import posix
|
||||
|
||||
ensure_tainted(
|
||||
os.getenv("foo"), # $ tainted threatModelSource[environment]=os.getenv(..)
|
||||
os.getenvb("bar"), # $ tainted threatModelSource[environment]=os.getenvb(..)
|
||||
|
||||
os.environ["foo"], # $ tainted threatModelSource[environment]=os.environ
|
||||
os.environ.get("foo"), # $ tainted threatModelSource[environment]=os.environ
|
||||
|
||||
os.environb["bar"], # $ tainted threatModelSource[environment]=os.environb
|
||||
posix.environ[b"foo"], # $ tainted threatModelSource[environment]=posix.environ
|
||||
|
||||
|
||||
sys.argv[1], # $ tainted threatModelSource[commandargs]=sys.argv
|
||||
sys.orig_argv[1], # $ tainted threatModelSource[commandargs]=sys.orig_argv
|
||||
)
|
||||
|
||||
for k,v in os.environ.items(): # $ threatModelSource[environment]=os.environ
|
||||
ensure_tainted(k) # $ tainted
|
||||
ensure_tainted(v) # $ tainted
|
||||
|
||||
|
||||
########################################
|
||||
# argparse
|
||||
########################################
|
||||
|
||||
import argparse
|
||||
parser = argparse.ArgumentParser()
|
||||
parser.add_argument("foo")
|
||||
|
||||
args = parser.parse_args() # $ threatModelSource[commandargs]=parser.parse_args()
|
||||
ensure_tainted(args.foo) # $ tainted
|
||||
|
||||
explicit_argv_parsing = parser.parse_args(sys.argv) # $ threatModelSource[commandargs]=sys.argv
|
||||
ensure_tainted(explicit_argv_parsing.foo) # $ tainted
|
||||
|
||||
fake_args = parser.parse_args(["<foo>"])
|
||||
ensure_not_tainted(fake_args.foo) # $ SPURIOUS: tainted
|
||||
|
||||
########################################
|
||||
# reading input from stdin
|
||||
########################################
|
||||
|
||||
ensure_tainted(
|
||||
sys.stdin.readline(), # $ tainted threatModelSource[stdin]=sys.stdin
|
||||
input(), # $ tainted threatModelSource[stdin]=input()
|
||||
)
|
||||
|
||||
########################################
|
||||
# reading data from files
|
||||
########################################
|
||||
|
||||
ensure_tainted(
|
||||
open("foo"), # $ tainted threatModelSource[file]=open(..) getAPathArgument="foo"
|
||||
open("foo").read(), # $ tainted threatModelSource[file]=open(..) getAPathArgument="foo"
|
||||
open("foo").readline(), # $ tainted threatModelSource[file]=open(..) getAPathArgument="foo"
|
||||
open("foo").readlines(), # $ tainted threatModelSource[file]=open(..) getAPathArgument="foo"
|
||||
|
||||
os.read(os.open("foo"), 1024), # $ tainted threatModelSource[file]=os.read(..) getAPathArgument="foo"
|
||||
)
|
||||
|
||||
########################################
|
||||
# socket
|
||||
########################################
|
||||
|
||||
import socket
|
||||
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
|
||||
s.connect(("example.com", 1234))
|
||||
ensure_tainted(s.recv(1024)) # $ MISSING: tainted threatModelSource[socket]
|
||||
@@ -45,7 +45,7 @@ def func2(environ, start_response): # $ requestHandler
|
||||
start_response(status, headers) # $ headerWriteBulk=headers headerWriteBulkUnsanitized=name,value
|
||||
return [b"Hello"] # $ HttpResponse responseBody=List
|
||||
|
||||
case = sys.argv[1]
|
||||
case = sys.argv[1] # $ threatModelSource[commandargs]=sys.argv
|
||||
if case == "1":
|
||||
server = wsgiref.simple_server.WSGIServer(ADDRESS, wsgiref.simple_server.WSGIRequestHandler)
|
||||
server.set_app(func)
|
||||
|
||||
@@ -0,0 +1,3 @@
|
||||
| default |
|
||||
| remote |
|
||||
| request |
|
||||
@@ -0,0 +1,7 @@
|
||||
private import codeql.threatmodels.ThreatModels
|
||||
|
||||
from string kind
|
||||
where
|
||||
knownThreatModel(kind) and
|
||||
currentThreatModel(kind)
|
||||
select kind
|
||||
@@ -0,0 +1,8 @@
|
||||
edges
|
||||
| test.py:6:14:6:21 | ControlFlowNode for Attribute | test.py:6:14:6:24 | ControlFlowNode for Subscript | provenance | Src:MaD:17 |
|
||||
nodes
|
||||
| test.py:6:14:6:21 | ControlFlowNode for Attribute | semmle.label | ControlFlowNode for Attribute |
|
||||
| test.py:6:14:6:24 | ControlFlowNode for Subscript | semmle.label | ControlFlowNode for Subscript |
|
||||
subpaths
|
||||
#select
|
||||
| test.py:6:14:6:24 | ControlFlowNode for Subscript | test.py:6:14:6:21 | ControlFlowNode for Attribute | test.py:6:14:6:24 | ControlFlowNode for Subscript | This SQL query depends on a $@. | test.py:6:14:6:21 | ControlFlowNode for Attribute | user-provided value |
|
||||
@@ -0,0 +1,6 @@
|
||||
extensions:
|
||||
- addsTo:
|
||||
pack: codeql/threat-models
|
||||
extensible: threatModelConfiguration
|
||||
data:
|
||||
- ["local", true, 0]
|
||||
@@ -0,0 +1 @@
|
||||
Security/CWE-089/SqlInjection.ql
|
||||
@@ -0,0 +1,6 @@
|
||||
# test that enabling local threat-model works end-to-end
|
||||
import sys
|
||||
import psycopg
|
||||
|
||||
conn = psycopg.connect(...)
|
||||
conn.execute(sys.argv[1])
|
||||
Reference in New Issue
Block a user