Files
codeql/python/ql/test/experimental/dataflow/coverage/argumentRoutingTest.ql
Rasmus Lerchedahl Petersen 11c71fdd18 Python: remove EssaNodes
This commit removes SSA nodes from the data flow graph. Specifically, for a definition and use such as
```python
  x = expr
  y = x + 2
```
we used to have flow from `expr` to an SSA variable representing x and from that SSA variable to the use of `x` in the definition of `y`. Now we instead have flow from `expr` to the control flow node for `x` at line 1 and from there to the control flow node for `x` at line 2.

Specific changes:
- `EssaNode` from the data flow layer no longer exists.
- Several glue steps between `EssaNode`s and `CfgNode`s have been deleted.
- Entry nodes are now admitted as `CfgNodes` in the data flow layer (they were filtered out before).
- Entry nodes now have a new `toString` taking into account that the module name may be ambigous.
- Some tests have been rewritten to accomodate the changes, but only `python/ql/test/experimental/dataflow/basic/maximalFlowsConfig.qll` should have semantic changes.
- Comments have been updated
- Test output has been updated, but apart from `python/ql/test/experimental/dataflow/basic/maximalFlows.expected` only `python/ql/test/experimental/dataflow/typetracking-summaries/summaries.py` should have a semantic change. This is a bonus fix, probably meaning that something was never connected up correctly.
2023-11-20 21:35:32 +01:00

146 lines
5.0 KiB
Plaintext

import python
import semmle.python.dataflow.new.DataFlow
private import semmle.python.dataflow.new.internal.DataFlowPrivate as DataFlowPrivate
import experimental.dataflow.TestUtil.RoutingTest
module Argument1RoutingTest implements RoutingTestSig {
class Argument = Unit;
string flowTag(Argument arg) { result = "arg1" and exists(arg) }
predicate relevantFlow(DataFlow::Node source, DataFlow::Node sink, Argument arg) {
(
exists(Argument1ExtraRoutingConfig cfg | cfg.hasFlow(source, sink))
or
exists(ArgumentRoutingConfig cfg |
cfg.hasFlow(source, sink) and
cfg.isArgSource(source, 1) and
cfg.isGoodSink(sink, 1)
)
) and
exists(arg)
}
}
class ArgNumber extends int {
ArgNumber() { this in [1 .. 7] }
}
class ArgumentRoutingConfig extends DataFlow::Configuration {
ArgumentRoutingConfig() { this = "ArgumentRoutingConfig" }
predicate isArgSource(DataFlow::Node node, ArgNumber argNumber) {
node.(DataFlow::CfgNode).getNode().(NameNode).getId() = "arg" + argNumber
}
override predicate isSource(DataFlow::Node node) { this.isArgSource(node, _) }
predicate isGoodSink(DataFlow::Node node, ArgNumber argNumber) {
exists(CallNode call |
call.getFunction().(NameNode).getId() = "SINK" + argNumber and
node.(DataFlow::CfgNode).getNode() = call.getAnArg()
)
}
predicate isBadSink(DataFlow::Node node, ArgNumber argNumber) {
exists(CallNode call |
call.getFunction().(NameNode).getId() = "SINK" + argNumber + "_F" and
node.(DataFlow::CfgNode).getNode() = call.getAnArg()
)
}
override predicate isSink(DataFlow::Node node) {
this.isGoodSink(node, _) or this.isBadSink(node, _)
}
/**
* We want to be able to use `arg` in a sequence of calls such as `func(kw=arg); ... ; func(arg)`.
* Use-use flow lets the argument to the first call reach the sink inside the second call,
* making it seem like we handle all cases even if we only handle the last one.
* We make the test honest by preventing flow into source nodes.
*/
override predicate isBarrierIn(DataFlow::Node node) { this.isSource(node) }
}
class Argument1ExtraRoutingConfig extends DataFlow::Configuration {
Argument1ExtraRoutingConfig() { this = "Argument1ExtraRoutingConfig" }
override predicate isSource(DataFlow::Node node) {
exists(AssignmentDefinition def, DataFlow::CallCfgNode call |
// def.getVariable() = node.(DataFlow::EssaNode).getVar() and
def.getDefiningNode() = node.(DataFlow::CfgNode).getNode() and
def.getValue() = call.getNode() and
call.getFunction().asCfgNode().(NameNode).getId().matches("With\\_%")
) and
// node.(DataFlow::EssaNode).getVar().getName().matches("with\\_%")
node.(DataFlow::CfgNode).getNode().(NameNode).getId().matches("with\\_%")
}
override predicate isSink(DataFlow::Node node) {
exists(CallNode call |
call.getFunction().(NameNode).getId() = "SINK1" and
node.(DataFlow::CfgNode).getNode() = call.getAnArg()
)
}
/**
* We want to be able to use `arg` in a sequence of calls such as `func(kw=arg); ... ; func(arg)`.
* Use-use flow lets the argument to the first call reach the sink inside the second call,
* making it seem like we handle all cases even if we only handle the last one.
* We make the test honest by preventing flow into source nodes.
*/
override predicate isBarrierIn(DataFlow::Node node) { this.isSource(node) }
}
module RestArgumentRoutingTest implements RoutingTestSig {
class Argument = ArgNumber;
string flowTag(Argument arg) { result = "arg" + arg }
predicate relevantFlow(DataFlow::Node source, DataFlow::Node sink, Argument arg) {
exists(ArgumentRoutingConfig cfg |
cfg.hasFlow(source, sink) and
cfg.isArgSource(source, arg) and
cfg.isGoodSink(sink, arg)
) and
arg > 1
}
}
/** Bad flow from `arg<n>` to `SINK<N>_F` */
module BadArgumentRoutingTestSinkF implements RoutingTestSig {
class Argument = ArgNumber;
string flowTag(Argument arg) { result = "bad" + arg }
predicate relevantFlow(DataFlow::Node source, DataFlow::Node sink, Argument arg) {
exists(ArgumentRoutingConfig cfg |
cfg.hasFlow(source, sink) and
cfg.isArgSource(source, arg) and
cfg.isBadSink(sink, arg)
)
}
}
/** Bad flow from `arg<n>` to `SINK<M>` or `SINK<M>_F`, where `n != m`. */
module BadArgumentRoutingTestWrongSink implements RoutingTestSig {
class Argument = ArgNumber;
string flowTag(Argument arg) { result = "bad" + arg }
predicate relevantFlow(DataFlow::Node source, DataFlow::Node sink, Argument arg) {
exists(ArgumentRoutingConfig cfg |
cfg.hasFlow(source, sink) and
cfg.isArgSource(source, any(ArgNumber i | not i = arg)) and
(
cfg.isGoodSink(sink, arg)
or
cfg.isBadSink(sink, arg)
)
)
}
}
import MakeTest<MergeTests4<MakeTestSig<Argument1RoutingTest>, MakeTestSig<RestArgumentRoutingTest>,
MakeTestSig<BadArgumentRoutingTestSinkF>, MakeTestSig<BadArgumentRoutingTestWrongSink>>>