C++: Filter duplicate (source, sink)-pairs

cpp/ql/src/Security/CWE/CWE-022/TaintedPath.ql

@@ -48,7 +48,9 @@ class FileFunction extends FunctionWithWrappers {
 }
 
 Expr asSourceExpr(DataFlow::Node node) {
-  result in [node.asConvertedExpr(), node.asDefiningArgument()]
+  result = node.asConvertedExpr()
+  or
+  result = node.asDefiningArgument()
 }
 
 Expr asSinkExpr(DataFlow::Node node) {
@@ -79,6 +81,19 @@ class TaintedPathConfiguration extends TaintTracking::Configuration {
   override predicate isSanitizer(DataFlow::Node node) {
     node.asExpr().(Call).getTarget().getUnspecifiedType() instanceof ArithmeticType
   }
+
+  predicate hasFilteredFlowPath(DataFlow::PathNode source, DataFlow::PathNode sink) {
+    this.hasFlowPath(source, sink) and
+    not exists(DataFlow::PathNode source2, DataFlow::PathNode sink2 |
+      this.hasFlowPath(source2, sink2) and
+      asSourceExpr(source.getNode()) = asSourceExpr(source2.getNode()) and
+      asSinkExpr(sink.getNode()) = asSinkExpr(sink2.getNode())
+    |
+      not exists(source.getNode().asConvertedExpr()) and exists(source2.getNode().asConvertedExpr())
+      or
+      not exists(sink.getNode().asConvertedExpr()) and exists(sink2.getNode().asConvertedExpr())
+    )
+  }
 }
 
 from
@@ -87,7 +102,7 @@ from
 where
   taintedArg = asSinkExpr(sinkNode.getNode()) and
   fileFunction.outermostWrapperFunctionCall(taintedArg, callChain) and
-  cfg.hasFlowPath(sourceNode, sinkNode) and
+  cfg.hasFilteredFlowPath(sourceNode, sinkNode) and
   taintSource = asSourceExpr(sourceNode.getNode()) and
   isUserInput(taintSource, taintCause)
 select taintedArg, sourceNode, sinkNode,

cpp/ql/test/query-tests/Security/CWE/CWE-022/SAMATE/TaintedPath/TaintedPath.expected

@@ -8,4 +8,3 @@ nodes
 subpaths
 #select
 | CWE23_Relative_Path_Traversal__char_console_fopen_11.cpp:77:23:77:26 | data | CWE23_Relative_Path_Traversal__char_console_fopen_11.cpp:55:27:55:38 | fgets output argument | CWE23_Relative_Path_Traversal__char_console_fopen_11.cpp:77:23:77:26 | data | This argument to a file access function is derived from $@ and then passed to fopen(filename). | CWE23_Relative_Path_Traversal__char_console_fopen_11.cpp:55:27:55:38 | ... + ... | user input (fgets) |
-| CWE23_Relative_Path_Traversal__char_console_fopen_11.cpp:77:23:77:26 | data | CWE23_Relative_Path_Traversal__char_console_fopen_11.cpp:55:27:55:38 | fgets output argument | CWE23_Relative_Path_Traversal__char_console_fopen_11.cpp:77:23:77:26 | data indirection | This argument to a file access function is derived from $@ and then passed to fopen(filename). | CWE23_Relative_Path_Traversal__char_console_fopen_11.cpp:55:27:55:38 | ... + ... | user input (fgets) |

cpp/ql/test/query-tests/Security/CWE/CWE-022/semmle/tests/TaintedPath.expected

@@ -26,12 +26,6 @@ nodes
 subpaths
 #select
 | test.c:17:11:17:18 | fileName | test.c:9:23:9:26 | argv | test.c:17:11:17:18 | fileName | This argument to a file access function is derived from $@ and then passed to fopen(filename). | test.c:9:23:9:26 | argv | user input (argv) |
-| test.c:17:11:17:18 | fileName | test.c:9:23:9:26 | argv | test.c:17:11:17:18 | fileName indirection | This argument to a file access function is derived from $@ and then passed to fopen(filename). | test.c:9:23:9:26 | argv | user input (argv) |
 | test.c:32:11:32:18 | fileName | test.c:31:22:31:25 | argv | test.c:32:11:32:18 | fileName | This argument to a file access function is derived from $@ and then passed to fopen(filename). | test.c:31:22:31:25 | argv | user input (argv) |
-| test.c:32:11:32:18 | fileName | test.c:31:22:31:25 | argv | test.c:32:11:32:18 | fileName indirection | This argument to a file access function is derived from $@ and then passed to fopen(filename). | test.c:31:22:31:25 | argv | user input (argv) |
 | test.c:38:11:38:18 | fileName | test.c:37:17:37:24 | fileName | test.c:38:11:38:18 | fileName indirection | This argument to a file access function is derived from $@ and then passed to fopen(filename). | test.c:37:17:37:24 | fileName | user input (scanf) |
-| test.c:38:11:38:18 | fileName | test.c:37:17:37:24 | scanf output argument | test.c:38:11:38:18 | fileName indirection | This argument to a file access function is derived from $@ and then passed to fopen(filename). | test.c:37:17:37:24 | fileName | user input (scanf) |
 | test.c:44:11:44:18 | fileName | test.c:43:17:43:24 | fileName | test.c:44:11:44:18 | fileName | This argument to a file access function is derived from $@ and then passed to fopen(filename). | test.c:43:17:43:24 | fileName | user input (scanf) |
-| test.c:44:11:44:18 | fileName | test.c:43:17:43:24 | fileName | test.c:44:11:44:18 | fileName indirection | This argument to a file access function is derived from $@ and then passed to fopen(filename). | test.c:43:17:43:24 | fileName | user input (scanf) |
-| test.c:44:11:44:18 | fileName | test.c:43:17:43:24 | scanf output argument | test.c:44:11:44:18 | fileName | This argument to a file access function is derived from $@ and then passed to fopen(filename). | test.c:43:17:43:24 | fileName | user input (scanf) |
-| test.c:44:11:44:18 | fileName | test.c:43:17:43:24 | scanf output argument | test.c:44:11:44:18 | fileName indirection | This argument to a file access function is derived from $@ and then passed to fopen(filename). | test.c:43:17:43:24 | fileName | user input (scanf) |