Merge pull request #9024 from hvitved/dataflow/content-flow-lib

Data flow: Introduce `ContentDataFlow.qll`
This commit is contained in:
Tom Hvitved
2022-05-24 15:09:16 +02:00
committed by GitHub
12 changed files with 6135 additions and 91 deletions

View File

@@ -22,6 +22,7 @@
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowImpl3.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowImpl4.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowImpl5.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowImplForContentDataFlow.qll",
"python/ql/lib/semmle/python/dataflow/new/internal/DataFlowImpl.qll",
"python/ql/lib/semmle/python/dataflow/new/internal/DataFlowImpl2.qll",
"python/ql/lib/semmle/python/dataflow/new/internal/DataFlowImpl3.qll",
@@ -559,4 +560,4 @@
"javascript/ql/src/Expressions/TypoDatabase.qll",
"ql/ql/src/codeql_ql/style/TypoDatabase.qll"
]
}
}

View File

@@ -788,24 +788,31 @@ private module Cached {
cached
predicate readSet(Node node1, ContentSet c, Node node2) { readStep(node1, c, node2) }
cached
predicate storeSet(
Node node1, ContentSet c, Node node2, DataFlowType contentType, DataFlowType containerType
) {
storeStep(node1, c, node2) and
contentType = getNodeDataFlowType(node1) and
containerType = getNodeDataFlowType(node2)
or
exists(Node n1, Node n2 |
n1 = node1.(PostUpdateNode).getPreUpdateNode() and
n2 = node2.(PostUpdateNode).getPreUpdateNode()
|
argumentValueFlowsThrough(n2, TReadStepTypesSome(containerType, c, contentType), n1)
or
readSet(n2, c, n1) and
contentType = getNodeDataFlowType(n1) and
containerType = getNodeDataFlowType(n2)
)
}
private predicate store(
Node node1, Content c, Node node2, DataFlowType contentType, DataFlowType containerType
) {
exists(ContentSet cs | c = cs.getAStoreContent() |
storeStep(node1, cs, node2) and
contentType = getNodeDataFlowType(node1) and
containerType = getNodeDataFlowType(node2)
or
exists(Node n1, Node n2 |
n1 = node1.(PostUpdateNode).getPreUpdateNode() and
n2 = node2.(PostUpdateNode).getPreUpdateNode()
|
argumentValueFlowsThrough(n2, TReadStepTypesSome(containerType, cs, contentType), n1)
or
readSet(n2, cs, n1) and
contentType = getNodeDataFlowType(n1) and
containerType = getNodeDataFlowType(n2)
)
exists(ContentSet cs |
c = cs.getAStoreContent() and storeSet(node1, cs, node2, contentType, containerType)
)
}

View File

@@ -788,24 +788,31 @@ private module Cached {
cached
predicate readSet(Node node1, ContentSet c, Node node2) { readStep(node1, c, node2) }
cached
predicate storeSet(
Node node1, ContentSet c, Node node2, DataFlowType contentType, DataFlowType containerType
) {
storeStep(node1, c, node2) and
contentType = getNodeDataFlowType(node1) and
containerType = getNodeDataFlowType(node2)
or
exists(Node n1, Node n2 |
n1 = node1.(PostUpdateNode).getPreUpdateNode() and
n2 = node2.(PostUpdateNode).getPreUpdateNode()
|
argumentValueFlowsThrough(n2, TReadStepTypesSome(containerType, c, contentType), n1)
or
readSet(n2, c, n1) and
contentType = getNodeDataFlowType(n1) and
containerType = getNodeDataFlowType(n2)
)
}
private predicate store(
Node node1, Content c, Node node2, DataFlowType contentType, DataFlowType containerType
) {
exists(ContentSet cs | c = cs.getAStoreContent() |
storeStep(node1, cs, node2) and
contentType = getNodeDataFlowType(node1) and
containerType = getNodeDataFlowType(node2)
or
exists(Node n1, Node n2 |
n1 = node1.(PostUpdateNode).getPreUpdateNode() and
n2 = node2.(PostUpdateNode).getPreUpdateNode()
|
argumentValueFlowsThrough(n2, TReadStepTypesSome(containerType, cs, contentType), n1)
or
readSet(n2, cs, n1) and
contentType = getNodeDataFlowType(n1) and
containerType = getNodeDataFlowType(n2)
)
exists(ContentSet cs |
c = cs.getAStoreContent() and storeSet(node1, cs, node2, contentType, containerType)
)
}

View File

@@ -0,0 +1,529 @@
/**
* Provides classes for performing global (inter-procedural)
* content-sensitive data flow analyses.
*/
private import DataFlowImplCommon
module ContentDataFlow {
private import DataFlowImplSpecific::Private
private import DataFlowImplSpecific::Private as DataFlowPrivate
private import DataFlowImplForContentDataFlow as DF
class Node = DF::Node;
class FlowFeature = DF::FlowFeature;
class ContentSet = DF::ContentSet;
predicate stageStats = DF::stageStats/8;
/**
* A configuration of interprocedural data flow analysis. This defines
* sources, sinks, and any other configurable aspect of the analysis. Each
* use of the global data flow library must define its own unique extension
* of this abstract class. To create a configuration, extend this class with
* a subclass whose characteristic predicate is a unique singleton string.
* For example, write
*
* ```ql
* class MyAnalysisConfiguration extends ContentDataFlowConfiguration {
* MyAnalysisConfiguration() { this = "MyAnalysisConfiguration" }
* // Override `isSource` and `isSink`.
* // Optionally override `isBarrier`.
* // Optionally override `isAdditionalFlowStep`.
* // Optionally override `getAFeature`.
* // Optionally override `accessPathLimit`.
* // Optionally override `isRelevantContent`.
* }
* ```
*
* Unlike `DataFlow::Configuration` (on which this class is based), we allow
* for data to be stored (possibly nested) inside contents of sources and sinks.
* We track flow paths of the form
*
* ```
* source --value-->* node
* (--read--> node --value-->* node)*
* --(non-value|value)-->* node
* (--store--> node --value-->* node)*
* --value-->* sink
* ```
*
* where `--value-->` is a value-preserving flow step, `--read-->` is a read
* step, `--store-->` is a store step, and `--(non-value)-->` is a
* non-value-preserving flow step.
*
* That is, first a sequence of 0 or more reads, followed by 0 or more additional
* steps, followed by 0 or more stores, with value-preserving steps allowed in
* between all other steps.
*/
abstract class Configuration extends string {
bindingset[this]
Configuration() { any() }
/**
* Holds if `source` is a relevant data flow source.
*/
abstract predicate isSource(Node source);
/**
* Holds if `sink` is a relevant data flow sink.
*/
abstract predicate isSink(Node sink);
/**
* Holds if data may flow from `node1` to `node2` in addition to the normal data-flow steps.
*/
predicate isAdditionalFlowStep(Node node1, Node node2) { none() }
/** Holds if data flow into `node` is prohibited. */
predicate isBarrier(Node node) { none() }
/**
* Gets a data flow configuration feature to add restrictions to the set of
* valid flow paths.
*
* - `FeatureHasSourceCallContext`:
* Assume that sources have some existing call context to disallow
* conflicting return-flow directly following the source.
* - `FeatureHasSinkCallContext`:
* Assume that sinks have some existing call context to disallow
* conflicting argument-to-parameter flow directly preceding the sink.
* - `FeatureEqualSourceSinkCallContext`:
* Implies both of the above and additionally ensures that the entire flow
* path preserves the call context.
*/
FlowFeature getAFeature() { none() }
/** Gets a limit on the number of reads out of sources and number of stores into sinks. */
int accessPathLimit() { result = DataFlowPrivate::accessPathLimit() }
/** Holds if `c` is relevant for reads out of sources or stores into sinks. */
predicate isRelevantContent(ContentSet c) { any() }
/**
* Holds if data stored inside `sourceAp` on `source` flows to `sinkAp` inside `sink`
* for this configuration. `preservesValue` indicates whether any of the additional
* flow steps defined by `isAdditionalFlowStep` are needed.
*
* For the source access path, `sourceAp`, the top of the stack represents the content
* that was last read from. That is, if `sourceAp` is `Field1.Field2` (with `Field1`
* being the top of the stack), then there is flow from `source.Field2.Field1`.
*
* For the sink access path, `sinkAp`, the top of the stack represents the content
* that was last stored into. That is, if `sinkAp` is `Field1.Field2` (with `Field1`
* being the top of the stack), then there is flow into `sink.Field1.Field2`.
*/
final predicate hasFlow(
Node source, AccessPath sourceAp, Node sink, AccessPath sinkAp, boolean preservesValue
) {
exists(DF::PathNode pathSource, DF::PathNode pathSink |
this.(ConfigurationAdapter).hasFlowPath(pathSource, pathSink) and
nodeReaches(pathSource, TAccessPathNil(), TAccessPathNil(), pathSink, sourceAp, sinkAp) and
source = pathSource.getNode() and
sink = pathSink.getNode()
|
pathSink.getState().(InitState).decode(preservesValue)
or
pathSink.getState().(ReadState).decode(_, preservesValue)
or
pathSink.getState().(StoreState).decode(_, preservesValue)
)
}
}
/** A flow state representing no reads or stores. */
private class InitState extends DF::FlowState {
private boolean preservesValue_;
InitState() { this = "Init(" + preservesValue_ + ")" and preservesValue_ in [false, true] }
predicate decode(boolean preservesValue) { preservesValue = preservesValue_ }
}
/** A flow state representing that content has been stored into. */
private class StoreState extends DF::FlowState {
private boolean preservesValue_;
private int size_;
StoreState() {
preservesValue_ in [false, true] and
size_ in [1 .. any(Configuration c).accessPathLimit()] and
this = "StoreState(" + size_ + "," + preservesValue_ + ")"
}
predicate decode(int size, boolean preservesValue) {
size = size_ and preservesValue = preservesValue_
}
}
/** A flow state representing that content has been read from. */
private class ReadState extends DF::FlowState {
private boolean preservesValue_;
private int size_;
ReadState() {
preservesValue_ in [false, true] and
size_ in [1 .. any(Configuration c).accessPathLimit()] and
this = "ReadState(" + size_ + "," + preservesValue_ + ")"
}
predicate decode(int size, boolean preservesValue) {
size = size_ and preservesValue = preservesValue_
}
}
private predicate storeStep(
Node node1, DF::FlowState state1, ContentSet c, Node node2, StoreState state2,
Configuration config
) {
exists(boolean preservesValue, int size |
storeSet(node1, c, node2, _, _) and
config.isRelevantContent(c) and
state2.decode(size + 1, preservesValue)
|
state1.(InitState).decode(preservesValue) and size = 0
or
state1.(ReadState).decode(_, preservesValue) and size = 0
or
state1.(StoreState).decode(size, preservesValue)
)
}
private predicate readStep(
Node node1, DF::FlowState state1, ContentSet c, Node node2, ReadState state2,
Configuration config
) {
exists(int size |
readSet(node1, c, node2) and
config.isRelevantContent(c) and
state2.decode(size + 1, true)
|
state1.(InitState).decode(true) and
size = 0
or
state1.(ReadState).decode(size, true)
)
}
private predicate additionalStep(
Node node1, DF::FlowState state1, Node node2, DF::FlowState state2, Configuration config
) {
config.isAdditionalFlowStep(node1, node2) and
(
state1 instanceof InitState and
state2.(InitState).decode(false)
or
exists(int size |
state1.(ReadState).decode(size, _) and
state2.(ReadState).decode(size, false)
)
)
}
private class ConfigurationAdapter extends DF::Configuration {
private Configuration c;
ConfigurationAdapter() { this = c }
final override predicate isSource(Node source, DF::FlowState state) {
c.isSource(source) and
state.(InitState).decode(true)
}
final override predicate isSink(Node sink, DF::FlowState state) {
c.isSink(sink) and
(
state instanceof InitState or
state instanceof StoreState or
state instanceof ReadState
)
}
final override predicate isAdditionalFlowStep(
Node node1, DF::FlowState state1, Node node2, DF::FlowState state2
) {
storeStep(node1, state1, _, node2, state2, this) or
readStep(node1, state1, _, node2, state2, this) or
additionalStep(node1, state1, node2, state2, this)
}
final override predicate isBarrier(Node node) { c.isBarrier(node) }
final override FlowFeature getAFeature() { result = c.getAFeature() }
// needed to record reads/stores inside summarized callables
final override predicate includeHiddenNodes() { any() }
}
private newtype TAccessPath =
TAccessPathNil() or
TAccessPathCons(ContentSet head, AccessPath tail) {
nodeReachesStore(_, _, _, _, head, _, tail)
or
nodeReachesRead(_, _, _, _, head, tail, _)
}
/** An access path. */
class AccessPath extends TAccessPath {
/** Gets the head of this access path, if any. */
ContentSet getHead() { this = TAccessPathCons(result, _) }
/** Gets the tail of this access path, if any. */
AccessPath getTail() { this = TAccessPathCons(_, result) }
/**
* Gets a textual representation of this access path.
*
* Elements are dot-separated, and the head of the stack is
* rendered first.
*/
string toString() {
this = TAccessPathNil() and
result = ""
or
exists(ContentSet head, AccessPath tail |
this = TAccessPathCons(head, tail) and
result = head + "." + tail
)
}
}
// important to use `edges` and not `PathNode::getASuccessor()`, as the latter
// is not pruned for reachability
private predicate pathSucc = DF::PathGraph::edges/2;
/**
* Provides a big-step flow relation, where flow stops at read/store steps that
* must be recorded, and flow via `subpaths` such that reads/stores inside
* summarized callables can be recorded as well.
*/
private module BigStepFlow {
private predicate reachesSink(DF::PathNode node) {
any(ConfigurationAdapter config).isSink(node.getNode(), node.getState())
or
exists(DF::PathNode mid |
pathSucc(node, mid) and
reachesSink(mid)
)
}
/**
* Holds if the flow step `pred -> succ` should not be allowed to be included
* in the big-step relation.
*/
pragma[nomagic]
private predicate excludeStep(DF::PathNode pred, DF::PathNode succ) {
pathSucc(pred, succ) and
(
// we need to record reads/stores inside summarized callables
DF::PathGraph::subpaths(pred, _, _, succ)
or
// only allow flow into a summarized callable, as part of the big-step
// relation, when flow can reach a sink without going back out
DF::PathGraph::subpaths(pred, succ, _, _) and
not reachesSink(succ)
or
// needed to record store steps
storeStep(pred.getNode(), pred.getState(), _, succ.getNode(), succ.getState(),
pred.getConfiguration())
or
// needed to record read steps
readStep(pred.getNode(), pred.getState(), _, succ.getNode(), succ.getState(),
pred.getConfiguration())
)
}
pragma[nomagic]
private DataFlowCallable getEnclosingCallableImpl(DF::PathNode node) {
result = getNodeEnclosingCallable(node.getNode())
}
pragma[inline]
private DataFlowCallable getEnclosingCallable(DF::PathNode node) {
pragma[only_bind_into](result) = getEnclosingCallableImpl(pragma[only_bind_out](node))
}
pragma[nomagic]
private predicate bigStepEntry(DF::PathNode node) {
node.getConfiguration() instanceof Configuration and
(
any(ConfigurationAdapter config).isSource(node.getNode(), node.getState())
or
excludeStep(_, node)
or
DF::PathGraph::subpaths(_, node, _, _)
)
}
pragma[nomagic]
private predicate bigStepExit(DF::PathNode node) {
node.getConfiguration() instanceof Configuration and
(
bigStepEntry(node)
or
any(ConfigurationAdapter config).isSink(node.getNode(), node.getState())
or
excludeStep(node, _)
or
DF::PathGraph::subpaths(_, _, node, _)
)
}
pragma[nomagic]
private predicate step(DF::PathNode pred, DF::PathNode succ) {
pathSucc(pred, succ) and
not excludeStep(pred, succ)
}
pragma[nomagic]
private predicate stepRec(DF::PathNode pred, DF::PathNode succ) {
step(pred, succ) and
not bigStepEntry(pred)
}
private predicate stepRecPlus(DF::PathNode n1, DF::PathNode n2) = fastTC(stepRec/2)(n1, n2)
/**
* Holds if there is flow `pathSucc+(pred) = succ`, and such a flow path does
* not go through any reads/stores that need to be recorded, or summarized
* steps.
*/
pragma[nomagic]
private predicate bigStep(DF::PathNode pred, DF::PathNode succ) {
exists(DF::PathNode mid |
bigStepEntry(pred) and
step(pred, mid)
|
succ = mid
or
stepRecPlus(mid, succ)
) and
bigStepExit(succ)
}
pragma[nomagic]
predicate bigStepNotLocal(DF::PathNode pred, DF::PathNode succ) {
bigStep(pred, succ) and
not getEnclosingCallable(pred) = getEnclosingCallable(succ)
}
pragma[nomagic]
predicate bigStepMaybeLocal(DF::PathNode pred, DF::PathNode succ) {
bigStep(pred, succ) and
getEnclosingCallable(pred) = getEnclosingCallable(succ)
}
}
/**
* Holds if `source` can reach `node`, having read `reads` from the source and
* written `stores` into `node`.
*
* `source` is either a source from a configuration, in which case `scReads` and
* `scStores` are always empty, or it is the parameter of a summarized callable,
* in which case `scReads` and `scStores` record the reads/stores for a summary
* context, that is, the reads/stores for an argument that can reach the parameter.
*/
pragma[nomagic]
private predicate nodeReaches(
DF::PathNode source, AccessPath scReads, AccessPath scStores, DF::PathNode node,
AccessPath reads, AccessPath stores
) {
exists(ConfigurationAdapter config |
node = source and
reads = scReads and
stores = scStores
|
config.hasFlowPath(source, _) and
scReads = TAccessPathNil() and
scStores = TAccessPathNil()
or
// the argument in a sub path can be reached, so we start flow from the sub path
// parameter, while recording the read/store summary context
exists(DF::PathNode arg |
nodeReachesSubpathArg(_, _, _, arg, scReads, scStores) and
DF::PathGraph::subpaths(arg, source, _, _)
)
)
or
exists(DF::PathNode mid |
nodeReaches(source, scReads, scStores, mid, reads, stores) and
BigStepFlow::bigStepMaybeLocal(mid, node)
)
or
exists(DF::PathNode mid |
nodeReaches(source, scReads, scStores, mid, reads, stores) and
BigStepFlow::bigStepNotLocal(mid, node) and
// when flow is not local, we cannot flow back out, so we may stop
// flow early when computing summary flow
any(ConfigurationAdapter config).hasFlowPath(source, _) and
scReads = TAccessPathNil() and
scStores = TAccessPathNil()
)
or
// store step
exists(AccessPath storesMid, ContentSet c |
nodeReachesStore(source, scReads, scStores, node, c, reads, storesMid) and
stores = TAccessPathCons(c, storesMid)
)
or
// read step
exists(AccessPath readsMid, ContentSet c |
nodeReachesRead(source, scReads, scStores, node, c, readsMid, stores) and
reads = TAccessPathCons(c, readsMid)
)
or
// flow-through step; match outer stores/reads with inner store/read summary contexts
exists(DF::PathNode mid, AccessPath innerScReads, AccessPath innerScStores |
nodeReachesSubpathArg(source, scReads, scStores, mid, innerScReads, innerScStores) and
subpathArgReachesOut(mid, innerScReads, innerScStores, node, reads, stores)
)
}
pragma[nomagic]
private predicate nodeReachesStore(
DF::PathNode source, AccessPath scReads, AccessPath scStores, DF::PathNode node, ContentSet c,
AccessPath reads, AccessPath stores
) {
exists(DF::PathNode mid |
nodeReaches(source, scReads, scStores, mid, reads, stores) and
storeStep(mid.getNode(), mid.getState(), c, node.getNode(), node.getState(),
node.getConfiguration()) and
pathSucc(mid, node)
)
}
pragma[nomagic]
private predicate nodeReachesRead(
DF::PathNode source, AccessPath scReads, AccessPath scStores, DF::PathNode node, ContentSet c,
AccessPath reads, AccessPath stores
) {
exists(DF::PathNode mid |
nodeReaches(source, scReads, scStores, mid, reads, stores) and
readStep(mid.getNode(), mid.getState(), c, node.getNode(), node.getState(),
node.getConfiguration()) and
pathSucc(mid, node)
)
}
pragma[nomagic]
private predicate nodeReachesSubpathArg(
DF::PathNode source, AccessPath scReads, AccessPath scStores, DF::PathNode arg,
AccessPath reads, AccessPath stores
) {
nodeReaches(source, scReads, scStores, arg, reads, stores) and
DF::PathGraph::subpaths(arg, _, _, _)
}
pragma[nomagic]
private predicate subpathArgReachesOut(
DF::PathNode arg, AccessPath scReads, AccessPath scStores, DF::PathNode out, AccessPath reads,
AccessPath stores
) {
exists(DF::PathNode source, DF::PathNode ret |
nodeReaches(source, scReads, scStores, ret, reads, stores) and
DF::PathGraph::subpaths(arg, source, ret, out)
)
}
}

View File

@@ -788,24 +788,31 @@ private module Cached {
cached
predicate readSet(Node node1, ContentSet c, Node node2) { readStep(node1, c, node2) }
cached
predicate storeSet(
Node node1, ContentSet c, Node node2, DataFlowType contentType, DataFlowType containerType
) {
storeStep(node1, c, node2) and
contentType = getNodeDataFlowType(node1) and
containerType = getNodeDataFlowType(node2)
or
exists(Node n1, Node n2 |
n1 = node1.(PostUpdateNode).getPreUpdateNode() and
n2 = node2.(PostUpdateNode).getPreUpdateNode()
|
argumentValueFlowsThrough(n2, TReadStepTypesSome(containerType, c, contentType), n1)
or
readSet(n2, c, n1) and
contentType = getNodeDataFlowType(n1) and
containerType = getNodeDataFlowType(n2)
)
}
private predicate store(
Node node1, Content c, Node node2, DataFlowType contentType, DataFlowType containerType
) {
exists(ContentSet cs | c = cs.getAStoreContent() |
storeStep(node1, cs, node2) and
contentType = getNodeDataFlowType(node1) and
containerType = getNodeDataFlowType(node2)
or
exists(Node n1, Node n2 |
n1 = node1.(PostUpdateNode).getPreUpdateNode() and
n2 = node2.(PostUpdateNode).getPreUpdateNode()
|
argumentValueFlowsThrough(n2, TReadStepTypesSome(containerType, cs, contentType), n1)
or
readSet(n2, cs, n1) and
contentType = getNodeDataFlowType(n1) and
containerType = getNodeDataFlowType(n2)
)
exists(ContentSet cs |
c = cs.getAStoreContent() and storeSet(node1, cs, node2, contentType, containerType)
)
}

View File

@@ -0,0 +1,51 @@
using System;
public class ContentFlow
{
public class A
{
public A FieldA;
public B FieldB;
}
public class B
{
public A FieldA;
public B FieldB;
}
public void M(A a, B b)
{
var a1 = new A();
Sink(Through(a1.FieldA.FieldB));
a.FieldA.FieldB = new B();
Sink(Through(a));
var a2 = new A();
b.FieldB.FieldA = a2.FieldB.FieldA;
Sink(Through(b));
Sink(Through(Out()));
In(new A().FieldA.FieldB);
}
public static void Sink<T>(T t) { }
public T Through<T>(T t)
{
Sink(t);
return t;
}
public void In<T>(T t)
{
Sink(t);
}
public B Out()
{
var a = new A();
return a.FieldA.FieldB;
}
}

View File

@@ -0,0 +1,9 @@
| ContentFlow.cs:18:18:18:24 | object creation of type A | field FieldB.field FieldA. | ContentFlow.cs:19:14:19:38 | call to method Through<B> | | true |
| ContentFlow.cs:18:18:18:24 | object creation of type A | field FieldB.field FieldA. | ContentFlow.cs:37:14:37:14 | access to parameter t | | true |
| ContentFlow.cs:21:27:21:33 | object creation of type B | | ContentFlow.cs:22:14:22:23 | call to method Through<A> | field FieldA.field FieldB. | true |
| ContentFlow.cs:21:27:21:33 | object creation of type B | | ContentFlow.cs:37:14:37:14 | access to parameter t | field FieldA.field FieldB. | true |
| ContentFlow.cs:24:18:24:24 | object creation of type A | field FieldA.field FieldB. | ContentFlow.cs:26:14:26:23 | call to method Through<B> | field FieldB.field FieldA. | true |
| ContentFlow.cs:24:18:24:24 | object creation of type A | field FieldA.field FieldB. | ContentFlow.cs:37:14:37:14 | access to parameter t | field FieldB.field FieldA. | true |
| ContentFlow.cs:30:12:30:18 | object creation of type A | field FieldB.field FieldA. | ContentFlow.cs:43:14:43:14 | access to parameter t | | true |
| ContentFlow.cs:48:17:48:23 | object creation of type A | field FieldB.field FieldA. | ContentFlow.cs:28:14:28:27 | call to method Through<B> | | true |
| ContentFlow.cs:48:17:48:23 | object creation of type A | field FieldB.field FieldA. | ContentFlow.cs:37:14:37:14 | access to parameter t | | true |

View File

@@ -0,0 +1,23 @@
import csharp
import semmle.code.csharp.dataflow.internal.ContentDataFlow
class Conf extends ContentDataFlow::Configuration {
Conf() { this = "ContentFlowConf" }
override predicate isSource(DataFlow::Node src) { src.asExpr() instanceof ObjectCreation }
override predicate isSink(DataFlow::Node sink) {
exists(MethodCall mc |
mc.getTarget().hasUndecoratedName("Sink") and
mc.getAnArgument() = sink.asExpr()
)
}
override int accessPathLimit() { result = 2 }
}
from
Conf conf, ContentDataFlow::Node source, ContentDataFlow::AccessPath sourceAp,
ContentDataFlow::Node sink, ContentDataFlow::AccessPath sinkAp, boolean preservesValue
where conf.hasFlow(source, sourceAp, sink, sinkAp, preservesValue)
select source, sourceAp, sink, sinkAp, preservesValue

View File

@@ -788,24 +788,31 @@ private module Cached {
cached
predicate readSet(Node node1, ContentSet c, Node node2) { readStep(node1, c, node2) }
cached
predicate storeSet(
Node node1, ContentSet c, Node node2, DataFlowType contentType, DataFlowType containerType
) {
storeStep(node1, c, node2) and
contentType = getNodeDataFlowType(node1) and
containerType = getNodeDataFlowType(node2)
or
exists(Node n1, Node n2 |
n1 = node1.(PostUpdateNode).getPreUpdateNode() and
n2 = node2.(PostUpdateNode).getPreUpdateNode()
|
argumentValueFlowsThrough(n2, TReadStepTypesSome(containerType, c, contentType), n1)
or
readSet(n2, c, n1) and
contentType = getNodeDataFlowType(n1) and
containerType = getNodeDataFlowType(n2)
)
}
private predicate store(
Node node1, Content c, Node node2, DataFlowType contentType, DataFlowType containerType
) {
exists(ContentSet cs | c = cs.getAStoreContent() |
storeStep(node1, cs, node2) and
contentType = getNodeDataFlowType(node1) and
containerType = getNodeDataFlowType(node2)
or
exists(Node n1, Node n2 |
n1 = node1.(PostUpdateNode).getPreUpdateNode() and
n2 = node2.(PostUpdateNode).getPreUpdateNode()
|
argumentValueFlowsThrough(n2, TReadStepTypesSome(containerType, cs, contentType), n1)
or
readSet(n2, cs, n1) and
contentType = getNodeDataFlowType(n1) and
containerType = getNodeDataFlowType(n2)
)
exists(ContentSet cs |
c = cs.getAStoreContent() and storeSet(node1, cs, node2, contentType, containerType)
)
}

View File

@@ -788,24 +788,31 @@ private module Cached {
cached
predicate readSet(Node node1, ContentSet c, Node node2) { readStep(node1, c, node2) }
cached
predicate storeSet(
Node node1, ContentSet c, Node node2, DataFlowType contentType, DataFlowType containerType
) {
storeStep(node1, c, node2) and
contentType = getNodeDataFlowType(node1) and
containerType = getNodeDataFlowType(node2)
or
exists(Node n1, Node n2 |
n1 = node1.(PostUpdateNode).getPreUpdateNode() and
n2 = node2.(PostUpdateNode).getPreUpdateNode()
|
argumentValueFlowsThrough(n2, TReadStepTypesSome(containerType, c, contentType), n1)
or
readSet(n2, c, n1) and
contentType = getNodeDataFlowType(n1) and
containerType = getNodeDataFlowType(n2)
)
}
private predicate store(
Node node1, Content c, Node node2, DataFlowType contentType, DataFlowType containerType
) {
exists(ContentSet cs | c = cs.getAStoreContent() |
storeStep(node1, cs, node2) and
contentType = getNodeDataFlowType(node1) and
containerType = getNodeDataFlowType(node2)
or
exists(Node n1, Node n2 |
n1 = node1.(PostUpdateNode).getPreUpdateNode() and
n2 = node2.(PostUpdateNode).getPreUpdateNode()
|
argumentValueFlowsThrough(n2, TReadStepTypesSome(containerType, cs, contentType), n1)
or
readSet(n2, cs, n1) and
contentType = getNodeDataFlowType(n1) and
containerType = getNodeDataFlowType(n2)
)
exists(ContentSet cs |
c = cs.getAStoreContent() and storeSet(node1, cs, node2, contentType, containerType)
)
}

View File

@@ -788,24 +788,31 @@ private module Cached {
cached
predicate readSet(Node node1, ContentSet c, Node node2) { readStep(node1, c, node2) }
cached
predicate storeSet(
Node node1, ContentSet c, Node node2, DataFlowType contentType, DataFlowType containerType
) {
storeStep(node1, c, node2) and
contentType = getNodeDataFlowType(node1) and
containerType = getNodeDataFlowType(node2)
or
exists(Node n1, Node n2 |
n1 = node1.(PostUpdateNode).getPreUpdateNode() and
n2 = node2.(PostUpdateNode).getPreUpdateNode()
|
argumentValueFlowsThrough(n2, TReadStepTypesSome(containerType, c, contentType), n1)
or
readSet(n2, c, n1) and
contentType = getNodeDataFlowType(n1) and
containerType = getNodeDataFlowType(n2)
)
}
private predicate store(
Node node1, Content c, Node node2, DataFlowType contentType, DataFlowType containerType
) {
exists(ContentSet cs | c = cs.getAStoreContent() |
storeStep(node1, cs, node2) and
contentType = getNodeDataFlowType(node1) and
containerType = getNodeDataFlowType(node2)
or
exists(Node n1, Node n2 |
n1 = node1.(PostUpdateNode).getPreUpdateNode() and
n2 = node2.(PostUpdateNode).getPreUpdateNode()
|
argumentValueFlowsThrough(n2, TReadStepTypesSome(containerType, cs, contentType), n1)
or
readSet(n2, cs, n1) and
contentType = getNodeDataFlowType(n1) and
containerType = getNodeDataFlowType(n2)
)
exists(ContentSet cs |
c = cs.getAStoreContent() and storeSet(node1, cs, node2, contentType, containerType)
)
}