Load canonical-path DLL from CODEQL_EXTRACTOR_GO_ROOT, not CODEQL_DIST

Replace std::rand() with std::mt19937 seeded by std::random_device for
less predictable cache eviction. Also use .first from emplace() instead
of structured binding with ignored second element.

Cargo.lock

@@ -2850,10 +2850,6 @@ version = "0.11.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "7da8b5736845d9f2fcb837ea5d9e2628564b3b043a70948a3f0b778838c5fb4f"
 
-[[package]]
-name = "swift-syntax-rs"
-version = "0.1.0"
-
 [[package]]
 name = "syn"
 version = "2.0.106"

Cargo.toml

@@ -9,7 +9,6 @@ members = [
     "ruby/extractor",
     "unified/extractor",
     "unified/extractor/tree-sitter-swift",
-    "unified/swift-syntax-rs",
     "rust/extractor",
     "rust/extractor/macros",
     "rust/ast-generator",

MODULE.bazel

@@ -24,15 +24,13 @@ bazel_dep(name = "rules_python", version = "1.9.0")
 bazel_dep(name = "rules_shell", version = "0.7.1")
 bazel_dep(name = "bazel_skylib", version = "1.9.0")
 bazel_dep(name = "abseil-cpp", version = "20260107.1", repo_name = "absl")
-bazel_dep(name = "nlohmann_json", version = "3.12.0.bcr.1", repo_name = "json")
+bazel_dep(name = "nlohmann_json", version = "3.11.3", repo_name = "json")
 bazel_dep(name = "fmt", version = "12.1.0-codeql.1")
 bazel_dep(name = "rules_kotlin", version = "2.2.2-codeql.1")
 bazel_dep(name = "gazelle", version = "0.50.0")
 bazel_dep(name = "rules_dotnet", version = "0.21.5-codeql.1")
 bazel_dep(name = "googletest", version = "1.17.0.bcr.2")
 bazel_dep(name = "rules_rust", version = "0.69.0")
-bazel_dep(name = "rules_swift", version = "4.0.0-rc4")
-bazel_dep(name = "swift-syntax", version = "602.0.0.bcr.2")
 bazel_dep(name = "zstd", version = "1.5.7.bcr.1")
 
 bazel_dep(name = "buildifier_prebuilt", version = "6.4.0", dev_dependency = True)
@@ -221,37 +219,6 @@ use_repo(
     "swift-resource-dir-macos",
 )
 
-# Hermetic Swift toolchain (from swift.org) for building the `swift-syntax`
-# based Rust wrapper in `unified/swift-syntax-rs`. This is the official
-# `rules_swift` standalone-toolchain extension and is independent of the
-# patched prebuilt toolchain wired up via `swift_deps` above.
-#
-# Bazel cannot auto-select between Linux distributions, so the toolchain is
-# registered explicitly for the distribution our CI runs on (ubuntu24.04,
-# x86_64). Add further platforms here if CI grows to cover them.
-#
-# We register the `exec` toolchain (normal host compilation, targeting
-# linux/x86_64) rather than the `embedded` one, which targets `os:none`
-# (Embedded Swift) and would not match a normal host build.
-#
-# The Swift version is read from `unified/swift-syntax-rs/.swift-version`, which
-# is the single source of truth shared with the local `cargo` build (via
-# swiftly / any Swift install) and `swift/Package.swift`.
-swift = use_extension("@rules_swift//swift:extensions.bzl", "swift")
-swift.toolchain(
-    name = "swift_toolchain",
-    swift_version_file = "//unified/swift-syntax-rs:.swift-version",
-)
-use_repo(
-    swift,
-    "swift_toolchain",
-    "swift_toolchain_ubuntu24.04",
-)
-
-register_toolchains(
-    "@swift_toolchain//:swift_toolchain_exec_ubuntu24.04",
-)
-
 node = use_extension("@rules_nodejs//nodejs:extensions.bzl", "node")
 node.toolchain(
     name = "nodejs",

cpp/ql/lib/semmle/code/cpp/dataflow/internal/FlowSummaryImpl.qll

@@ -6,7 +6,6 @@ private import cpp as Cpp
 private import codeql.dataflow.internal.FlowSummaryImpl
 private import codeql.dataflow.internal.AccessPathSyntax as AccessPath
 private import semmle.code.cpp.ir.dataflow.internal.DataFlowPrivate
-private import semmle.code.cpp.ir.dataflow.internal.DataFlowNodes
 private import semmle.code.cpp.ir.dataflow.internal.DataFlowUtil
 private import semmle.code.cpp.ir.dataflow.internal.DataFlowImplSpecific as DataFlowImplSpecific
 private import semmle.code.cpp.dataflow.ExternalFlow
@@ -21,22 +20,8 @@ module Input implements InputSig<Location, DataFlowImplSpecific::CppDataFlow> {
 
   class SinkBase = Void;
 
-  class FlowSummaryCallBase = CallInstruction;
-
   predicate callableFromSource(SummarizedCallableBase c) { exists(c.getBlock()) }
 
-  FlowSummaryCallBase getASourceCall(SummarizedCallableBase sc) {
-    result.getStaticCallTarget() = sc
-  }
-
-  DataFlowCallable getSummarizedCallableAsDataFlowCallable(SummarizedCallableBase c) {
-    result.asSummarizedCallable() = c
-  }
-
-  DataFlowCallable getSourceCallEnclosingCallable(FlowSummaryCallBase call) {
-    result.asSourceCallable() = call.getEnclosingFunction()
-  }
-
   ArgumentPosition callbackSelfParameterPosition() { result = TDirectPosition(-1) }
 
   ReturnKind getStandardReturnValueKind() { result = getReturnValueKind("") }
@@ -45,10 +30,6 @@ module Input implements InputSig<Location, DataFlowImplSpecific::CppDataFlow> {
     arg = repeatStars(result.(NormalReturnKind).getIndirectionIndex())
   }
 
-  ParameterPosition getFlowSummaryParameterPosition(ReturnKind rk) {
-    result = TFlowSummaryPosition(rk)
-  }
-
   string encodeParameterPosition(ParameterPosition pos) { result = pos.toString() }
 
   string encodeArgumentPosition(ArgumentPosition pos) { result = pos.toString() }
@@ -133,22 +114,10 @@ module Input implements InputSig<Location, DataFlowImplSpecific::CppDataFlow> {
 private import Make<Location, DataFlowImplSpecific::CppDataFlow, Input> as Impl
 
 private module StepsInput implements Impl::Private::StepsInputSig {
-  Impl::Private::SummaryNode getSummaryNode(Node n) {
-    result = n.(FlowSummaryNode).getSummaryNode()
-  }
-
   DataFlowCall getACall(Public::SummarizedCallable sc) {
     result.getStaticCallTarget().getUnderlyingCallable() = sc
   }
 
-  Node getSourceOutNode(Input::FlowSummaryCallBase call, ReturnKind rk) {
-    exists(IndirectReturnOutNode out | result = out |
-      out.getCallInstruction() = call and
-      pragma[only_bind_out](rk.(NormalReturnKind).getIndirectionIndex()) =
-        pragma[only_bind_out](out.getIndirectionIndex())
-    )
-  }
-
   DataFlowCallable getSourceNodeEnclosingCallable(Input::SourceBase source) { none() }
 
   Node getSourceNode(Input::SourceBase source, Impl::Private::SummaryComponentStack s) { none() }

cpp/ql/lib/semmle/code/cpp/ir/dataflow/internal/DataFlowNodes.qll

@@ -1534,8 +1534,12 @@ class FlowSummaryNode extends Node, TFlowSummaryNode {
     result = this.getSummaryNode().getSummarizedCallable()
   }
 
+  /**
+   * Gets the enclosing callable. For a `FlowSummaryNode` this is always the
+   * summarized function this node is part of.
+   */
   override DataFlowCallable getEnclosingCallable() {
-    result = FlowSummaryImpl::Private::getEnclosingCallable(this.getSummaryNode())
+    result.asSummarizedCallable() = this.getSummarizedCallable()
   }
 
   override Location getLocationImpl() { result = this.getSummarizedCallable().getLocation() }

cpp/ql/lib/semmle/code/cpp/ir/dataflow/internal/DataFlowPrivate.qll

@@ -561,21 +561,6 @@ class SummaryArgumentNode extends ArgumentNode, FlowSummaryNode {
   }
 }
 
-/** An argument node that re-enters return output as input to a flow summary. */
-private class FlowSummaryArgumentNode extends ArgumentNode, FlowSummaryNode {
-  private CallInstruction callInstruction;
-  private ReturnKind rk;
-
-  FlowSummaryArgumentNode() {
-    this.getSummaryNode() = FlowSummaryImpl::Private::summaryArgumentNode(callInstruction, rk)
-  }
-
-  override predicate argumentOf(DataFlowCall call, ArgumentPosition pos) {
-    call.asCallInstruction() = callInstruction and
-    pos = TFlowSummaryPosition(rk)
-  }
-}
-
 /** A parameter position represented by an integer. */
 class ParameterPosition = Position;
 
@@ -631,18 +616,6 @@ class IndirectionPosition extends Position, TIndirectionPosition {
   final override int getIndirectionIndex() { result = indirectionIndex }
 }
 
-class FlowSummaryPosition extends Position, TFlowSummaryPosition {
-  ReturnKind rk;
-
-  FlowSummaryPosition() { this = TFlowSummaryPosition(rk) }
-
-  override string toString() { result = "write to: " + rk.toString() }
-
-  override int getArgumentIndex() { none() }
-
-  final override int getIndirectionIndex() { result = rk.getIndirectionIndex() }
-}
-
 newtype TPosition =
   TDirectPosition(int argumentIndex) {
     exists(any(CallInstruction c).getArgument(argumentIndex))
@@ -661,8 +634,7 @@ newtype TPosition =
       p = f.getParameter(argumentIndex) and
       indirectionIndex = [1 .. Ssa::getMaxIndirectionsForType(p.getUnspecifiedType()) - 1]
     )
-  } or
-  TFlowSummaryPosition(ReturnKind rk) { FlowSummaryImpl::Private::relevantFlowSummaryPosition(rk) }
+  }
 
 private newtype TReturnKind =
   TNormalReturnKind(int indirectionIndex) {

cpp/ql/lib/semmle/code/cpp/ir/dataflow/internal/DataFlowUtil.qll

@@ -158,7 +158,7 @@ private module Cached {
     model = ""
     or
     // models-as-data summarized flow
-    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom,
+    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom.(FlowSummaryNode).getSummaryNode(),
       nodeTo.(FlowSummaryNode).getSummaryNode(), true, model)
   }
 

cpp/ql/lib/semmle/code/cpp/ir/dataflow/internal/TaintTrackingUtil.qll

@@ -67,7 +67,7 @@ private module Cached {
     model = ""
     or
     // models-as-data summarized flow
-    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom,
+    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom.(FlowSummaryNode).getSummaryNode(),
       nodeTo.(FlowSummaryNode).getSummaryNode(), false, model)
     or
     // object->field conflation for content that is a `TaintInheritingContent`.

cpp/ql/test/library-tests/dataflow/external-models/flow.expected

@@ -53,17 +53,14 @@ models
 | 52 | Summary: ; ; false; ymlStepGenerated; ; ; Argument[0]; ReturnValue; taint; df-generated |
 | 53 | Summary: ; ; false; ymlStepManual; ; ; Argument[0]; ReturnValue; taint; manual |
 | 54 | Summary: ; ; false; ymlStepManual_with_body; ; ; Argument[0]; ReturnValue; taint; manual |
-| 55 | Summary: ; MyString; true; operator[]; ; ; Argument[-1]; ReturnValue[*]; taint; manual |
-| 56 | Summary: ; MyString; true; operator[]; ; ; ReturnValue[*]; Argument[-1]; taint; manual |
-| 57 | Summary: ; ReverseFlow; true; get_ptr; ; ; ReturnValue[*]; Argument[-1].Field[ReverseFlow::value]; value; manual |
-| 58 | Summary: ; TemplateClass1; true; templateFunction2<U,V>; (U,V); ; Argument[1]; ReturnValue; value; manual |
-| 59 | Summary: ; TemplateClass1<T>; false; templateFunction<U>; (T,U); ; Argument[0]; ReturnValue; value; manual |
-| 60 | Summary: ; TemplateClass2<T,U>; true; function; (U,T); ; Argument[1]; ReturnValue; value; manual |
-| 61 | Summary: Azure::Core::IO; BodyStream; true; Read; ; ; Argument[-1]; Argument[*0]; taint; manual |
-| 62 | Summary: Azure::Core::IO; BodyStream; true; ReadToCount; ; ; Argument[-1]; Argument[*0]; taint; manual |
-| 63 | Summary: Azure::Core::IO; BodyStream; true; ReadToEnd; ; ; Argument[-1]; ReturnValue.Element; taint; manual |
-| 64 | Summary: Azure; Nullable; true; Value; ; ; Argument[-1]; ReturnValue[*]; taint; manual |
-| 65 | Summary: boost::asio; ; false; buffer; ; ; Argument[*0]; ReturnValue; taint; manual |
+| 55 | Summary: ; TemplateClass1; true; templateFunction2<U,V>; (U,V); ; Argument[1]; ReturnValue; value; manual |
+| 56 | Summary: ; TemplateClass1<T>; false; templateFunction<U>; (T,U); ; Argument[0]; ReturnValue; value; manual |
+| 57 | Summary: ; TemplateClass2<T,U>; true; function; (U,T); ; Argument[1]; ReturnValue; value; manual |
+| 58 | Summary: Azure::Core::IO; BodyStream; true; Read; ; ; Argument[-1]; Argument[*0]; taint; manual |
+| 59 | Summary: Azure::Core::IO; BodyStream; true; ReadToCount; ; ; Argument[-1]; Argument[*0]; taint; manual |
+| 60 | Summary: Azure::Core::IO; BodyStream; true; ReadToEnd; ; ; Argument[-1]; ReturnValue.Element; taint; manual |
+| 61 | Summary: Azure; Nullable; true; Value; ; ; Argument[-1]; ReturnValue[*]; taint; manual |
+| 62 | Summary: boost::asio; ; false; buffer; ; ; Argument[*0]; ReturnValue; taint; manual |
 edges
 | asio_streams.cpp:87:34:87:44 | read_until output argument | asio_streams.cpp:91:7:91:17 | recv_buffer | provenance | Src:MaD:32  |
 | asio_streams.cpp:87:34:87:44 | read_until output argument | asio_streams.cpp:93:29:93:39 | *recv_buffer | provenance | Src:MaD:32 Sink:MaD:2 |
@@ -72,16 +69,16 @@ edges
 | asio_streams.cpp:100:44:100:62 | call to buffer | asio_streams.cpp:100:44:100:62 | call to buffer | provenance |  |
 | asio_streams.cpp:100:44:100:62 | call to buffer | asio_streams.cpp:101:7:101:17 | send_buffer | provenance |  |
 | asio_streams.cpp:100:44:100:62 | call to buffer | asio_streams.cpp:103:29:103:39 | *send_buffer | provenance | Sink:MaD:2 |
-| asio_streams.cpp:100:64:100:71 | *send_str | asio_streams.cpp:100:44:100:62 | call to buffer | provenance | MaD:65 |
+| asio_streams.cpp:100:64:100:71 | *send_str | asio_streams.cpp:100:44:100:62 | call to buffer | provenance | MaD:62 |
 | azure.cpp:253:48:253:60 | *call to GetBodyStream | azure.cpp:253:48:253:60 | *call to GetBodyStream | provenance | Src:MaD:29  |
 | azure.cpp:253:48:253:60 | *call to GetBodyStream | azure.cpp:257:5:257:8 | *resp | provenance |  |
 | azure.cpp:253:48:253:60 | *call to GetBodyStream | azure.cpp:262:5:262:8 | *resp | provenance |  |
 | azure.cpp:253:48:253:60 | *call to GetBodyStream | azure.cpp:266:38:266:41 | *resp | provenance |  |
-| azure.cpp:257:5:257:8 | *resp | azure.cpp:257:16:257:21 | Read output argument | provenance | MaD:61 |
+| azure.cpp:257:5:257:8 | *resp | azure.cpp:257:16:257:21 | Read output argument | provenance | MaD:58 |
 | azure.cpp:257:16:257:21 | Read output argument | azure.cpp:258:10:258:16 | * ... | provenance |  |
-| azure.cpp:262:5:262:8 | *resp | azure.cpp:262:23:262:28 | ReadToCount output argument | provenance | MaD:62 |
+| azure.cpp:262:5:262:8 | *resp | azure.cpp:262:23:262:28 | ReadToCount output argument | provenance | MaD:59 |
 | azure.cpp:262:23:262:28 | ReadToCount output argument | azure.cpp:263:10:263:16 | * ... | provenance |  |
-| azure.cpp:266:38:266:41 | *resp | azure.cpp:266:44:266:52 | call to ReadToEnd [element] | provenance | MaD:63 |
+| azure.cpp:266:38:266:41 | *resp | azure.cpp:266:44:266:52 | call to ReadToEnd [element] | provenance | MaD:60 |
 | azure.cpp:266:44:266:52 | call to ReadToEnd [element] | azure.cpp:266:44:266:52 | call to ReadToEnd [element] | provenance |  |
 | azure.cpp:266:44:266:52 | call to ReadToEnd [element] | azure.cpp:267:10:267:12 | vec [element] | provenance |  |
 | azure.cpp:267:10:267:12 | vec [element] | azure.cpp:267:10:267:12 | vec | provenance |  |
@@ -97,10 +94,10 @@ edges
 | azure.cpp:278:10:278:13 | body | azure.cpp:278:10:278:13 | body | provenance |  |
 | azure.cpp:281:68:281:84 | *call to ExtractBodyStream | azure.cpp:281:68:281:84 | *call to ExtractBodyStream | provenance | Src:MaD:26  |
 | azure.cpp:281:68:281:84 | *call to ExtractBodyStream | azure.cpp:282:21:282:23 | *call to get | provenance |  |
-| azure.cpp:282:21:282:23 | *call to get | azure.cpp:282:28:282:36 | call to ReadToEnd [element] | provenance | MaD:63 |
+| azure.cpp:282:21:282:23 | *call to get | azure.cpp:282:28:282:36 | call to ReadToEnd [element] | provenance | MaD:60 |
 | azure.cpp:282:28:282:36 | call to ReadToEnd [element] | azure.cpp:282:10:282:38 | call to ReadToEnd | provenance |  |
 | azure.cpp:282:28:282:36 | call to ReadToEnd [element] | azure.cpp:282:28:282:36 | call to ReadToEnd [element] | provenance |  |
-| azure.cpp:289:24:289:56 | call to GetHeader | azure.cpp:289:63:289:65 | call to Value | provenance | MaD:64 |
+| azure.cpp:289:24:289:56 | call to GetHeader | azure.cpp:289:63:289:65 | call to Value | provenance | MaD:61 |
 | azure.cpp:289:32:289:40 | call to GetHeader | azure.cpp:289:24:289:56 | call to GetHeader | provenance |  |
 | azure.cpp:289:32:289:40 | call to GetHeader | azure.cpp:289:32:289:40 | call to GetHeader | provenance | Src:MaD:30  |
 | azure.cpp:289:63:289:65 | call to Value | azure.cpp:289:63:289:65 | call to Value | provenance |  |
@@ -162,27 +159,27 @@ edges
 | test.cpp:133:10:133:18 | call to ymlSource | test.cpp:134:45:134:45 | x | provenance |  |
 | test.cpp:134:13:134:43 | call to templateFunction | test.cpp:134:13:134:43 | call to templateFunction | provenance |  |
 | test.cpp:134:13:134:43 | call to templateFunction | test.cpp:135:10:135:10 | y | provenance | Sink:MaD:1 |
-| test.cpp:134:45:134:45 | x | test.cpp:134:13:134:43 | call to templateFunction | provenance | MaD:59 |
+| test.cpp:134:45:134:45 | x | test.cpp:134:13:134:43 | call to templateFunction | provenance | MaD:56 |
 | test.cpp:146:10:146:18 | call to ymlSource | test.cpp:146:10:146:18 | call to ymlSource | provenance | Src:MaD:25  |
 | test.cpp:146:10:146:18 | call to ymlSource | test.cpp:148:26:148:26 | x | provenance |  |
 | test.cpp:148:10:148:27 | call to function | test.cpp:148:10:148:27 | call to function | provenance |  |
 | test.cpp:148:10:148:27 | call to function | test.cpp:149:10:149:10 | z | provenance | Sink:MaD:1 |
-| test.cpp:148:26:148:26 | x | test.cpp:148:10:148:27 | call to function | provenance | MaD:60 |
+| test.cpp:148:26:148:26 | x | test.cpp:148:10:148:27 | call to function | provenance | MaD:57 |
 | test.cpp:155:10:155:18 | call to ymlSource | test.cpp:155:10:155:18 | call to ymlSource | provenance | Src:MaD:25  |
 | test.cpp:155:10:155:18 | call to ymlSource | test.cpp:157:26:157:26 | x | provenance |  |
 | test.cpp:157:13:157:20 | call to function | test.cpp:157:13:157:20 | call to function | provenance |  |
 | test.cpp:157:13:157:20 | call to function | test.cpp:158:10:158:10 | z | provenance | Sink:MaD:1 |
-| test.cpp:157:26:157:26 | x | test.cpp:157:13:157:20 | call to function | provenance | MaD:60 |
+| test.cpp:157:26:157:26 | x | test.cpp:157:13:157:20 | call to function | provenance | MaD:57 |
 | test.cpp:164:34:164:34 | x | test.cpp:165:69:165:69 | x | provenance |  |
 | test.cpp:165:12:165:64 | call to templateFunction2 | test.cpp:164:7:164:7 | *templateFunction3 | provenance |  |
 | test.cpp:165:12:165:64 | call to templateFunction2 | test.cpp:165:12:165:64 | call to templateFunction2 | provenance |  |
-| test.cpp:165:69:165:69 | x | test.cpp:165:12:165:64 | call to templateFunction2 | provenance | MaD:58 |
+| test.cpp:165:69:165:69 | x | test.cpp:165:12:165:64 | call to templateFunction2 | provenance | MaD:55 |
 | test.cpp:170:10:170:18 | call to ymlSource | test.cpp:170:10:170:18 | call to ymlSource | provenance | Src:MaD:25  |
 | test.cpp:170:10:170:18 | call to ymlSource | test.cpp:172:51:172:51 | x | provenance |  |
 | test.cpp:172:13:172:44 | call to templateFunction3 | test.cpp:172:13:172:44 | call to templateFunction3 | provenance |  |
 | test.cpp:172:13:172:44 | call to templateFunction3 | test.cpp:173:10:173:10 | y | provenance | Sink:MaD:1 |
 | test.cpp:172:51:172:51 | x | test.cpp:164:34:164:34 | x | provenance |  |
-| test.cpp:172:51:172:51 | x | test.cpp:172:13:172:44 | call to templateFunction3 | provenance | MaD:58 |
+| test.cpp:172:51:172:51 | x | test.cpp:172:13:172:44 | call to templateFunction3 | provenance | MaD:55 |
 | test.cpp:186:2:186:2 | *s [post update] [myField] | test.cpp:187:33:187:34 | *& ... [myField] | provenance |  |
 | test.cpp:186:2:186:24 | ... = ... | test.cpp:186:2:186:2 | *s [post update] [myField] | provenance |  |
 | test.cpp:186:14:186:22 | call to ymlSource | test.cpp:186:2:186:24 | ... = ... | provenance | Src:MaD:25  |
@@ -195,18 +192,6 @@ edges
 | test.cpp:200:10:200:33 | call to read_field_from_struct_2 | test.cpp:200:10:200:33 | call to read_field_from_struct_2 | provenance |  |
 | test.cpp:200:10:200:33 | call to read_field_from_struct_2 | test.cpp:201:10:201:10 | x | provenance | Sink:MaD:1 |
 | test.cpp:200:35:200:36 | *& ... [myField] | test.cpp:200:10:200:33 | call to read_field_from_struct_2 | provenance | MaD:51 |
-| test.cpp:216:3:216:4 | get_ptr output argument [value] | test.cpp:217:11:217:12 | *rf [value] | provenance |  |
-| test.cpp:216:3:216:28 | ... = ... | test.cpp:216:3:216:4 | get_ptr output argument [value] | provenance | MaD:57 |
-| test.cpp:216:18:216:26 | call to ymlSource | test.cpp:216:3:216:28 | ... = ... | provenance | Src:MaD:25  |
-| test.cpp:217:11:217:12 | *rf [value] | test.cpp:217:14:217:18 | value | provenance |  |
-| test.cpp:217:14:217:18 | value | test.cpp:217:14:217:18 | value | provenance |  |
-| test.cpp:217:14:217:18 | value | test.cpp:218:11:218:11 | x | provenance | Sink:MaD:1 |
-| test.cpp:222:3:222:3 | operator[] output argument | test.cpp:223:12:223:12 | *s | provenance |  |
-| test.cpp:222:3:222:20 | ... = ... | test.cpp:222:3:222:3 | operator[] output argument | provenance | MaD:56 |
-| test.cpp:222:10:222:20 | call to ymlSource | test.cpp:222:3:222:20 | ... = ... | provenance | Src:MaD:25  |
-| test.cpp:223:12:223:12 | *s | test.cpp:223:13:223:15 | call to operator[] | provenance | MaD:55 |
-| test.cpp:223:13:223:15 | call to operator[] | test.cpp:223:13:223:15 | call to operator[] | provenance |  |
-| test.cpp:223:13:223:15 | call to operator[] | test.cpp:224:11:224:11 | c | provenance | Sink:MaD:1 |
 | windows.cpp:22:15:22:29 | *call to GetCommandLineA | windows.cpp:22:15:22:29 | *call to GetCommandLineA | provenance | Src:MaD:3  |
 | windows.cpp:22:15:22:29 | *call to GetCommandLineA | windows.cpp:24:8:24:11 | * ... | provenance |  |
 | windows.cpp:22:15:22:29 | *call to GetCommandLineA | windows.cpp:27:36:27:38 | *cmd | provenance |  |
@@ -485,20 +470,6 @@ nodes
 | test.cpp:200:10:200:33 | call to read_field_from_struct_2 | semmle.label | call to read_field_from_struct_2 |
 | test.cpp:200:35:200:36 | *& ... [myField] | semmle.label | *& ... [myField] |
 | test.cpp:201:10:201:10 | x | semmle.label | x |
-| test.cpp:216:3:216:4 | get_ptr output argument [value] | semmle.label | get_ptr output argument [value] |
-| test.cpp:216:3:216:28 | ... = ... | semmle.label | ... = ... |
-| test.cpp:216:18:216:26 | call to ymlSource | semmle.label | call to ymlSource |
-| test.cpp:217:11:217:12 | *rf [value] | semmle.label | *rf [value] |
-| test.cpp:217:14:217:18 | value | semmle.label | value |
-| test.cpp:217:14:217:18 | value | semmle.label | value |
-| test.cpp:218:11:218:11 | x | semmle.label | x |
-| test.cpp:222:3:222:3 | operator[] output argument | semmle.label | operator[] output argument |
-| test.cpp:222:3:222:20 | ... = ... | semmle.label | ... = ... |
-| test.cpp:222:10:222:20 | call to ymlSource | semmle.label | call to ymlSource |
-| test.cpp:223:12:223:12 | *s | semmle.label | *s |
-| test.cpp:223:13:223:15 | call to operator[] | semmle.label | call to operator[] |
-| test.cpp:223:13:223:15 | call to operator[] | semmle.label | call to operator[] |
-| test.cpp:224:11:224:11 | c | semmle.label | c |
 | windows.cpp:22:15:22:29 | *call to GetCommandLineA | semmle.label | *call to GetCommandLineA |
 | windows.cpp:22:15:22:29 | *call to GetCommandLineA | semmle.label | *call to GetCommandLineA |
 | windows.cpp:24:8:24:11 | * ... | semmle.label | * ... |

cpp/ql/test/library-tests/dataflow/external-models/flow.ext.yml

@@ -23,7 +23,4 @@ extensions:
       - ["", "TemplateClass1", True, "templateFunction2<U,V>", "(U,V)", "", "Argument[1]", "ReturnValue", "value", "manual"]
       - ["", "TemplateClass2<T,U>", True, "function", "(U,T)", "", "Argument[1]", "ReturnValue", "value", "manual"]
       - ["", "", False, "read_field_from_struct", "", "", "Argument[*0].Field[MyNamespace::MyStructInNamespace::myField]", "ReturnValue", "value", "manual"]
-      - ["", "", False, "read_field_from_struct_2", "", "", "Argument[*0].Field[MyGlobalStruct::myField]", "ReturnValue", "value", "manual"]
-      - ["", "ReverseFlow", True, "get_ptr", "", "", "ReturnValue[*]", "Argument[-1].Field[ReverseFlow::value]", "value", "manual"]
-      - ["", "MyString", True, "operator[]", "", "", "ReturnValue[*]", "Argument[-1]", "taint", "manual"]
-      - ["", "MyString", True, "operator[]", "", "", "Argument[-1]", "ReturnValue[*]", "taint", "manual"]
+      - ["", "", False, "read_field_from_struct_2", "", "", "Argument[*0].Field[MyGlobalStruct::myField]", "ReturnValue", "value", "manual"]

cpp/ql/test/library-tests/dataflow/external-models/sinks.expected

@@ -21,5 +21,3 @@
 | test.cpp:173:10:173:10 | y | test-sink |
 | test.cpp:188:10:188:10 | x | test-sink |
 | test.cpp:201:10:201:10 | x | test-sink |
-| test.cpp:218:11:218:11 | x | test-sink |
-| test.cpp:224:11:224:11 | c | test-sink |

cpp/ql/test/library-tests/dataflow/external-models/sources.expected

@@ -15,8 +15,6 @@
 | test.cpp:170:10:170:18 | call to ymlSource | local |
 | test.cpp:186:14:186:22 | call to ymlSource | local |
 | test.cpp:199:14:199:22 | call to ymlSource | local |
-| test.cpp:216:18:216:26 | call to ymlSource | local |
-| test.cpp:222:10:222:20 | call to ymlSource | local |
 | windows.cpp:22:15:22:29 | *call to GetCommandLineA | local |
 | windows.cpp:34:17:34:38 | *call to GetEnvironmentStringsA | local |
 | windows.cpp:39:36:39:38 | GetEnvironmentVariableA output argument | local |

cpp/ql/test/library-tests/dataflow/external-models/test.cpp

@@ -199,28 +199,4 @@ void test_fully_qualified_field_test_2() {
 	s.myField = ymlSource();
 	int x = read_field_from_struct_2(&s);
 	ymlSink(x); // $ ir
-}
-
-struct ReverseFlow {
-	int value;
-	int& get_ptr();
-};
-
-struct MyString {
-	char& operator[](unsigned);
-};
-
-void test_reverse_flow(unsigned i, unsigned j) {
-	{
-		ReverseFlow rf;
-		rf.get_ptr() = ymlSource();
-		int x = rf.value;
-		ymlSink(x); // $ ir
-	}
-	{
-		MyString s;
-		s[i] = ymlSource();
-		char c = s[j];
-		ymlSink(c); // $ ir
-	}
 }

csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowPrivate.qll

@@ -714,7 +714,7 @@ predicate simpleLocalFlowStep(Node nodeFrom, Node nodeTo, string model) {
   ) and
   model = ""
   or
-  FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom,
+  FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom.(FlowSummaryNode).getSummaryNode(),
     nodeTo.(FlowSummaryNode).getSummaryNode(), true, model)
 }
 

csharp/ql/lib/semmle/code/csharp/dataflow/internal/FlowSummaryImpl.qll

@@ -34,8 +34,6 @@ module Input implements InputSig<Location, DataFlowImplSpecific::CsharpDataFlow>
 
   class SinkBase = Void;
 
-  class FlowSummaryCallBase = Void;
-
   predicate neutralElement(SummarizedCallableBase c, string kind, string provenance, boolean isExact) {
     interpretNeutral(c, kind, provenance, isExact)
   }
@@ -203,10 +201,6 @@ private module TypesInput implements Impl::Private::TypesInputSig {
 }
 
 private module StepsInput implements Impl::Private::StepsInputSig {
-  Impl::Private::SummaryNode getSummaryNode(Node n) {
-    result = n.(FlowSummaryNode).getSummaryNode()
-  }
-
   DataFlowCall getACall(Public::SummarizedCallable sc) {
     sc = viableCallable(result).asSummarizedCallable()
   }

csharp/ql/lib/semmle/code/csharp/dataflow/internal/TaintTrackingPrivate.qll

@@ -171,7 +171,7 @@ private module Cached {
     ) and
     model = ""
     or
-    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom,
+    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom.(FlowSummaryNode).getSummaryNode(),
       nodeTo.(FlowSummaryNode).getSummaryNode(), false, model)
   }
 }

csharp/ql/lib/semmle/code/csharp/security/dataflow/XSSSinks.qll

@@ -186,6 +186,13 @@ private Expr aspWrittenValue(AspInlineMember m) {
   m.getMember().(Callable).canReturn(result)
 }
 
+private string makeUrl(Location l) {
+  exists(string path, int sl, int sc, int el, int ec |
+    l.hasLocationInfo(path, sl, sc, el, ec) and
+    result = "file://" + path + ":" + sl + ":" + sc + ":" + el + ":" + ec
+  )
+}
+
 /**
  * A sink for writes to properties that are accessed in ASP pages.
  *
@@ -201,7 +208,10 @@ private class AspxCodeSink extends Sink {
 
   AspxCodeSink() { this.getExpr() = aspWrittenValue(inline) }
 
-  override string explanation() { result = "member is accessed inline in an ASPX page" }
+  override string explanation() {
+    result =
+      "member is [[\"accessed inline\"|\"" + makeUrl(inline.getLocation()) + "\"]] in an ASPX page"
+  }
 }
 
 /** A sink for the output stream associated with a `HttpListenerResponse`. */

docs/codeql/codeql-language-guides/customizing-library-models-for-java-and-kotlin.rst

@@ -69,26 +69,6 @@ The CodeQL library for Java and Kotlin analysis exposes the following extensible
 
 The extensible predicates are populated using the models defined in data extension files.
 
-Specifying types in Java and Kotlin models
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-
-**Nested and inner classes** are denoted by joining the enclosing type and the nested type with a dollar sign (``$``), for example ``Outer$Inner``. This applies both to the type column and to nested types in a signature. For example, the ``Level`` enum nested inside the ``Logger`` interface, nested inside the ``System`` class, is written as ``System$Logger$Level``:
-
-.. code-block:: yaml
-
-   - ["java.lang", "System$Logger", True, "log", "(System$Logger$Level,String)", "", "Argument[1]", "log-injection", "manual"]
-
-**Generics** are erased, so type parameters are removed:
-
-- In the type column, leave out any type parameters, so ``List<E>`` becomes ``List``.
-- In the signature, replace each type parameter with its upper bound, or ``Object`` if it has none. So ``T`` from ``<T>`` becomes ``Object``, and ``T`` from ``<T extends Number>`` becomes ``Number``.
-
-For example, ``forEach`` on ``Iterable<T>`` takes a ``Consumer<? super T>`` argument, so the type is ``Iterable`` and the signature is ``(Consumer)``:
-
-.. code-block:: yaml
-
-   - ["java.lang", "Iterable", True, "forEach", "(Consumer)", "", "Argument[this].Element", "Argument[0].Parameter[0]", "value", "manual"]
-
 Examples of custom model definitions
 ------------------------------------
 

go/BUILD.bazel

@@ -56,9 +56,19 @@ codeql_pkg_files(
     prefix = "tools/{CODEQL_PLATFORM}",
 )
 
+codeql_pkg_files(
+    name = "canonicalize-dll",
+    srcs = select({
+        "@platforms//os:windows": ["//shared/canonicalize:pkg"],
+        "//conditions:default": [],
+    }),
+    prefix = "tools/{CODEQL_PLATFORM}",
+)
+
 codeql_pack(
     name = "go",
     srcs = [
+        ":canonicalize-dll",
         ":extractor-pack-arch",
         ":resources",
         "//go/codeql-tools",

go/extractor/BUILD.bazel

@@ -16,6 +16,7 @@ go_library(
     importpath = "github.com/github/codeql-go/extractor",
     visibility = ["//visibility:public"],
     deps = [
+        "//go/extractor/canonicalize",
         "//go/extractor/dbscheme",
         "//go/extractor/diagnostics",
         "//go/extractor/srcarchive",

go/extractor/canonicalize/BUILD.bazel

@@ -0,0 +1,11 @@
+load("@rules_go//go:def.bzl", "go_library")
+
+go_library(
+    name = "canonicalize",
+    srcs = [
+        "canonicalize_other.go",
+        "canonicalize_windows.go",
+    ],
+    importpath = "github.com/github/codeql-go/extractor/canonicalize",
+    visibility = ["//visibility:public"],
+)

go/extractor/canonicalize/canonicalize_other.go

@@ -0,0 +1,5 @@
+//go:build !windows
+
+package canonicalize
+
+func CanonicalizePath(path string) string { return path }

go/extractor/canonicalize/canonicalize_windows.go

@@ -0,0 +1,65 @@
+//go:build windows
+
+package canonicalize
+
+import (
+	"os"
+	"path/filepath"
+	"syscall"
+	"unsafe"
+)
+
+var (
+	dll              *syscall.DLL
+	procCanonicalize *syscall.Proc
+	procFree         *syscall.Proc
+	available        bool
+)
+
+func init() {
+	root := os.Getenv("CODEQL_EXTRACTOR_GO_ROOT")
+	if root == "" {
+		return
+	}
+	dllPath := filepath.Join(root, "tools", "win64", "codeql_canonical_path.dll")
+	d, err := syscall.LoadDLL(dllPath)
+	if err != nil {
+		return
+	}
+	p, err := d.FindProc("canonicalize_path_u8")
+	if err != nil {
+		return
+	}
+	f, _ := d.FindProc("canonicalize_free_u8")
+	dll = d
+	procCanonicalize = p
+	procFree = f
+	available = true
+}
+
+func CanonicalizePath(path string) string {
+	if !available {
+		return path
+	}
+	pathBytes := append([]byte(path), 0)
+	ret, _, _ := procCanonicalize.Call(uintptr(unsafe.Pointer(&pathBytes[0])))
+	if ret == 0 {
+		return path
+	}
+	result := bytePtrToString((*byte)(unsafe.Pointer(ret)))
+	if procFree != nil {
+		procFree.Call(ret)
+	}
+	return result
+}
+
+func bytePtrToString(p *byte) string {
+	if p == nil {
+		return ""
+	}
+	var n int
+	for ptr := unsafe.Pointer(p); *(*byte)(ptr) != 0; n++ {
+		ptr = unsafe.Add(ptr, 1)
+	}
+	return string(unsafe.Slice(p, n))
+}

go/extractor/extractor.go

@@ -22,6 +22,7 @@ import (
 	"sync"
 	"time"
 
+	"github.com/github/codeql-go/extractor/canonicalize"
 	"github.com/github/codeql-go/extractor/dbscheme"
 	"github.com/github/codeql-go/extractor/diagnostics"
 	"github.com/github/codeql-go/extractor/srcarchive"
@@ -766,7 +767,7 @@ func normalizedPath(ast *ast.File, fset *token.FileSet) string {
 	if err != nil {
 		return file
 	}
-	return path
+	return canonicalize.CanonicalizePath(path)
 }
 
 // extractFile extracts AST information for the given file

go/ql/lib/semmle/go/dataflow/internal/DataFlowUtil.qll

@@ -141,7 +141,7 @@ predicate simpleLocalFlowStep(Node nodeFrom, Node nodeTo, string model) {
   any(FunctionModel m).flowStep(nodeFrom, nodeTo) and
   model = "FunctionModel"
   or
-  FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom,
+  FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom.(FlowSummaryNode).getSummaryNode(),
     nodeTo.(FlowSummaryNode).getSummaryNode(), true, model)
 }
 

go/ql/lib/semmle/go/dataflow/internal/FlowSummaryImpl.qll

@@ -31,8 +31,6 @@ module Input implements InputSig<Location, DataFlowImplSpecific::GoDataFlow> {
 
   class SinkBase = Void;
 
-  class FlowSummaryCallBase = Void;
-
   predicate callableFromSource(SummarizedCallableBase c) { exists(c.getFuncDef()) }
 
   predicate neutralElement(
@@ -115,10 +113,6 @@ module Input implements InputSig<Location, DataFlowImplSpecific::GoDataFlow> {
 private import Make<Location, DataFlowImplSpecific::GoDataFlow, Input> as Impl
 
 private module StepsInput implements Impl::Private::StepsInputSig {
-  Impl::Private::SummaryNode getSummaryNode(Node n) {
-    result = n.(FlowSummaryNode).getSummaryNode()
-  }
-
   DataFlowCall getACall(Public::SummarizedCallable sc) {
     exists(DataFlow::CallNode call |
       call.asExpr() = result and

go/ql/lib/semmle/go/dataflow/internal/TaintTrackingUtil.qll

@@ -109,8 +109,8 @@ private predicate localAdditionalForwardTaintStep(
   or
   any(AdditionalTaintStep a).step(pred, succ) and model = "AdditionalTaintStep"
   or
-  FlowSummaryImpl::Private::Steps::summaryLocalStep(pred,
-    succ.(DataFlowPrivate::FlowSummaryNode).getSummaryNode(), false, model)
+  FlowSummaryImpl::Private::Steps::summaryLocalStep(pred.(DataFlowPrivate::FlowSummaryNode)
+        .getSummaryNode(), succ.(DataFlowPrivate::FlowSummaryNode).getSummaryNode(), false, model)
 }
 
 /**

java/documentation/library-coverage/coverage.csv

@@ -188,8 +188,6 @@ org.apache.hadoop.hive.ql.metadata,1,,,,,,,,,,,,,,,,,,,,,,,,,,1,,,,,,,,,,,,,,,,,
 org.apache.hc.client5.http.async.methods,84,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,84,,,,,,,,,,,,,,,,,,
 org.apache.hc.client5.http.classic.methods,37,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,37,,,,,,,,,,,,,,,,,,
 org.apache.hc.client5.http.fluent,19,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,19,,,,,,,,,,,,,,,,,,
-org.apache.hc.client5.http.protocol,,,1,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1
-org.apache.hc.client5.http.utils,,,7,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,7,
 org.apache.hc.core5.benchmark,1,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1,,,,,,,,,,,,,,,,,,
 org.apache.hc.core5.function,,,1,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1,
 org.apache.hc.core5.http,73,2,45,,,,,,,,,,,,,1,,,,,,,,,,,,,,,,,,,,,72,,,,,,,,,,,,,,,,2,45,

java/documentation/library-coverage/coverage.rst

@@ -40,6 +40,6 @@ Java framework & library support
    `Spring <https://spring.io/>`_,``org.springframework.*``,46,494,143,26,,28,14,,35
    `Thymeleaf <https://www.thymeleaf.org/>`_,``org.thymeleaf``,,2,2,,,,,,
    `jOOQ <https://www.jooq.org/>`_,``org.jooq``,,,1,,,1,,,
-   Others,"``actions.osgi``, ``antlr``, ``ch.ethz.ssh2``, ``cn.hutool.core.codec``, ``com.alibaba.com.caucho.hessian.io``, ``com.alibaba.druid.sql``, ``com.alibaba.fastjson2``, ``com.amazonaws.auth``, ``com.auth0.jwt.algorithms``, ``com.azure.identity``, ``com.caucho.burlap.io``, ``com.caucho.hessian.io``, ``com.cedarsoftware.util.io``, ``com.esotericsoftware.kryo.io``, ``com.esotericsoftware.kryo5.io``, ``com.esotericsoftware.yamlbeans``, ``com.hubspot.jinjava``, ``com.jcraft.jsch``, ``com.microsoft.sqlserver.jdbc``, ``com.mitchellbosecke.pebble``, ``com.opensymphony.xwork2``, ``com.sshtools.j2ssh.authentication``, ``com.sun.crypto.provider``, ``com.sun.jndi.ldap``, ``com.sun.net.httpserver``, ``com.sun.net.ssl``, ``com.sun.rowset``, ``com.sun.security.auth.module``, ``com.sun.security.ntlm``, ``com.sun.security.sasl.digest``, ``com.thoughtworks.xstream``, ``com.trilead.ssh2``, ``com.unboundid.ldap.sdk``, ``com.zaxxer.hikari``, ``flexjson``, ``hudson``, ``io.jsonwebtoken``, ``io.undertow.server.handlers.resource``, ``javafx.scene.web``, ``jenkins``, ``jodd.json``, ``liquibase.database.jvm``, ``liquibase.statement.core``, ``net.lingala.zip4j``, ``net.schmizz.sshj``, ``net.sf.json``, ``net.sf.saxon.s9api``, ``ognl``, ``org.acegisecurity``, ``org.antlr.runtime``, ``org.apache.avro``, ``org.apache.commons.codec``, ``org.apache.commons.compress.archivers.tar``, ``org.apache.commons.exec``, ``org.apache.commons.fileupload``, ``org.apache.commons.httpclient.util``, ``org.apache.commons.jelly``, ``org.apache.commons.jexl2``, ``org.apache.commons.jexl3``, ``org.apache.commons.lang``, ``org.apache.commons.logging``, ``org.apache.commons.net``, ``org.apache.commons.ognl``, ``org.apache.cxf.catalog``, ``org.apache.cxf.common.classloader``, ``org.apache.cxf.common.jaxb``, ``org.apache.cxf.common.logging``, ``org.apache.cxf.configuration.jsse``, ``org.apache.cxf.helpers``, ``org.apache.cxf.resource``, ``org.apache.cxf.staxutils``, ``org.apache.cxf.tools.corba.utils``, ``org.apache.cxf.tools.util``, ``org.apache.cxf.transform``, ``org.apache.directory.ldap.client.api``, ``org.apache.hadoop.fs``, ``org.apache.hadoop.hive.metastore``, ``org.apache.hadoop.hive.ql.exec``, ``org.apache.hadoop.hive.ql.metadata``, ``org.apache.hc.client5.http.async.methods``, ``org.apache.hc.client5.http.classic.methods``, ``org.apache.hc.client5.http.fluent``, ``org.apache.hc.client5.http.protocol``, ``org.apache.hc.client5.http.utils``, ``org.apache.hive.hcatalog.templeton``, ``org.apache.ibatis.jdbc``, ``org.apache.ibatis.mapping``, ``org.apache.log4j``, ``org.apache.shiro.authc``, ``org.apache.shiro.codec``, ``org.apache.shiro.jndi``, ``org.apache.shiro.mgt``, ``org.apache.sshd.client.session``, ``org.apache.tools.ant``, ``org.apache.tools.zip``, ``org.codehaus.cargo.container.installer``, ``org.dom4j``, ``org.exolab.castor.xml``, ``org.fusesource.leveldbjni``, ``org.geogebra.web.full.main``, ``org.gradle.api.file``, ``org.ho.yaml``, ``org.influxdb``, ``org.jabsorb``, ``org.jboss.vfs``, ``org.jdbi.v3.core``, ``org.jenkins.ui.icon``, ``org.jenkins.ui.symbol``, ``org.keycloak.models.map.storage``, ``org.kohsuke.stapler``, ``org.lastaflute.web``, ``org.mvel2``, ``org.openjdk.jmh.runner.options``, ``org.owasp.esapi``, ``org.pac4j.jwt.config.encryption``, ``org.pac4j.jwt.config.signature``, ``org.scijava.log``, ``org.xml.sax``, ``org.xmlpull.v1``, ``play.libs.ws``, ``play.mvc``, ``ratpack.core.form``, ``ratpack.core.handling``, ``ratpack.core.http``, ``ratpack.exec``, ``ratpack.form``, ``ratpack.func``, ``ratpack.handling``, ``ratpack.http``, ``ratpack.util``, ``software.amazon.awssdk.transfer.s3.model``, ``sun.jvmstat.perfdata.monitor.protocol.local``, ``sun.jvmstat.perfdata.monitor.protocol.rmi``, ``sun.misc``, ``sun.net.ftp``, ``sun.net.www.protocol.http``, ``sun.security.acl``, ``sun.security.jgss.krb5``, ``sun.security.krb5``, ``sun.security.pkcs``, ``sun.security.pkcs11``, ``sun.security.provider``, ``sun.security.ssl``, ``sun.security.x509``, ``sun.tools.jconsole``",127,6042,775,148,6,14,18,,186
-   Totals,,382,26411,2707,421,16,137,33,1,415
+   Others,"``actions.osgi``, ``antlr``, ``ch.ethz.ssh2``, ``cn.hutool.core.codec``, ``com.alibaba.com.caucho.hessian.io``, ``com.alibaba.druid.sql``, ``com.alibaba.fastjson2``, ``com.amazonaws.auth``, ``com.auth0.jwt.algorithms``, ``com.azure.identity``, ``com.caucho.burlap.io``, ``com.caucho.hessian.io``, ``com.cedarsoftware.util.io``, ``com.esotericsoftware.kryo.io``, ``com.esotericsoftware.kryo5.io``, ``com.esotericsoftware.yamlbeans``, ``com.hubspot.jinjava``, ``com.jcraft.jsch``, ``com.microsoft.sqlserver.jdbc``, ``com.mitchellbosecke.pebble``, ``com.opensymphony.xwork2``, ``com.sshtools.j2ssh.authentication``, ``com.sun.crypto.provider``, ``com.sun.jndi.ldap``, ``com.sun.net.httpserver``, ``com.sun.net.ssl``, ``com.sun.rowset``, ``com.sun.security.auth.module``, ``com.sun.security.ntlm``, ``com.sun.security.sasl.digest``, ``com.thoughtworks.xstream``, ``com.trilead.ssh2``, ``com.unboundid.ldap.sdk``, ``com.zaxxer.hikari``, ``flexjson``, ``hudson``, ``io.jsonwebtoken``, ``io.undertow.server.handlers.resource``, ``javafx.scene.web``, ``jenkins``, ``jodd.json``, ``liquibase.database.jvm``, ``liquibase.statement.core``, ``net.lingala.zip4j``, ``net.schmizz.sshj``, ``net.sf.json``, ``net.sf.saxon.s9api``, ``ognl``, ``org.acegisecurity``, ``org.antlr.runtime``, ``org.apache.avro``, ``org.apache.commons.codec``, ``org.apache.commons.compress.archivers.tar``, ``org.apache.commons.exec``, ``org.apache.commons.fileupload``, ``org.apache.commons.httpclient.util``, ``org.apache.commons.jelly``, ``org.apache.commons.jexl2``, ``org.apache.commons.jexl3``, ``org.apache.commons.lang``, ``org.apache.commons.logging``, ``org.apache.commons.net``, ``org.apache.commons.ognl``, ``org.apache.cxf.catalog``, ``org.apache.cxf.common.classloader``, ``org.apache.cxf.common.jaxb``, ``org.apache.cxf.common.logging``, ``org.apache.cxf.configuration.jsse``, ``org.apache.cxf.helpers``, ``org.apache.cxf.resource``, ``org.apache.cxf.staxutils``, ``org.apache.cxf.tools.corba.utils``, ``org.apache.cxf.tools.util``, ``org.apache.cxf.transform``, ``org.apache.directory.ldap.client.api``, ``org.apache.hadoop.fs``, ``org.apache.hadoop.hive.metastore``, ``org.apache.hadoop.hive.ql.exec``, ``org.apache.hadoop.hive.ql.metadata``, ``org.apache.hc.client5.http.async.methods``, ``org.apache.hc.client5.http.classic.methods``, ``org.apache.hc.client5.http.fluent``, ``org.apache.hive.hcatalog.templeton``, ``org.apache.ibatis.jdbc``, ``org.apache.ibatis.mapping``, ``org.apache.log4j``, ``org.apache.shiro.authc``, ``org.apache.shiro.codec``, ``org.apache.shiro.jndi``, ``org.apache.shiro.mgt``, ``org.apache.sshd.client.session``, ``org.apache.tools.ant``, ``org.apache.tools.zip``, ``org.codehaus.cargo.container.installer``, ``org.dom4j``, ``org.exolab.castor.xml``, ``org.fusesource.leveldbjni``, ``org.geogebra.web.full.main``, ``org.gradle.api.file``, ``org.ho.yaml``, ``org.influxdb``, ``org.jabsorb``, ``org.jboss.vfs``, ``org.jdbi.v3.core``, ``org.jenkins.ui.icon``, ``org.jenkins.ui.symbol``, ``org.keycloak.models.map.storage``, ``org.kohsuke.stapler``, ``org.lastaflute.web``, ``org.mvel2``, ``org.openjdk.jmh.runner.options``, ``org.owasp.esapi``, ``org.pac4j.jwt.config.encryption``, ``org.pac4j.jwt.config.signature``, ``org.scijava.log``, ``org.xml.sax``, ``org.xmlpull.v1``, ``play.libs.ws``, ``play.mvc``, ``ratpack.core.form``, ``ratpack.core.handling``, ``ratpack.core.http``, ``ratpack.exec``, ``ratpack.form``, ``ratpack.func``, ``ratpack.handling``, ``ratpack.http``, ``ratpack.util``, ``software.amazon.awssdk.transfer.s3.model``, ``sun.jvmstat.perfdata.monitor.protocol.local``, ``sun.jvmstat.perfdata.monitor.protocol.rmi``, ``sun.misc``, ``sun.net.ftp``, ``sun.net.www.protocol.http``, ``sun.security.acl``, ``sun.security.jgss.krb5``, ``sun.security.krb5``, ``sun.security.pkcs``, ``sun.security.pkcs11``, ``sun.security.provider``, ``sun.security.ssl``, ``sun.security.x509``, ``sun.tools.jconsole``",127,6034,775,148,6,14,18,,186
+   Totals,,382,26403,2707,421,16,137,33,1,415
 

java/kotlin-extractor/src/main/java/com/semmle/util/files/FileUtil.java

@@ -1242,11 +1242,11 @@ public class FileUtil
 	public static File tryMakeCanonical (File f)
 	{
 		try {
-			return f.getCanonicalFile();
+			return NativeCanonicalizer.resolve(f.getCanonicalFile());
 		}
 		catch (IOException ignored) {
 			Exceptions.ignore(ignored, "Can't log error: Could be too verbose.");
-			return new File(simplifyPath(f));
+			return NativeCanonicalizer.resolve(new File(simplifyPath(f)));
 		}
 	}
 

java/kotlin-extractor/src/main/java/com/semmle/util/files/NativeCanonicalizer.java

@@ -0,0 +1,41 @@
+package com.semmle.util.files;
+
+import java.io.File;
+import java.nio.file.Path;
+import java.nio.file.Paths;
+
+public class NativeCanonicalizer {
+  private static final boolean available;
+
+  static {
+    boolean loaded = false;
+    if (File.separatorChar == '\\') {
+      String dist = System.getenv("CODEQL_DIST");
+      if (dist != null && !dist.isEmpty()) {
+        try {
+          Path library = Paths.get(dist).resolve("tools").resolve("win64")
+              .resolve("codeql_canonical_path.dll").toAbsolutePath();
+          System.load(library.toString());
+          loaded = true;
+        } catch (RuntimeException | UnsatisfiedLinkError ignored) {
+        }
+      }
+    }
+    available = loaded;
+  }
+
+  private NativeCanonicalizer() {}
+
+  // UTF-16 JNI interface - no encoding conversion
+  private static native String nativeCanonicalizePath(String path);
+
+  public static File resolve(File path) {
+    if (!available) return path;
+    String result = nativeCanonicalizePath(path.getAbsolutePath());
+    return result != null ? new File(result) : path;
+  }
+
+  public static boolean isAvailable() {
+    return available;
+  }
+}

java/ql/lib/ext/org.apache.hc.client5.http.protocol.model.yml

@@ -1,10 +1,4 @@
 extensions:
-  - addsTo:
-      pack: codeql/java-all
-      extensible: summaryModel
-    data:
-      - ["org.apache.hc.client5.http.protocol", "RedirectLocations", True, "add", "(URI)", "", "Argument[0]", "Argument[this].Element", "value", "hq-manual"]
-
   - addsTo:
       pack: codeql/java-all
       extensible: neutralModel

java/ql/lib/ext/org.apache.hc.client5.http.protocol.yml

@@ -0,0 +1,6 @@
+extensions:
+  - addsTo:
+      pack: codeql/java-all
+      extensible: summaryModel
+    data:
+      - ["org.apache.hc.client5.http.protocol", "RedirectLocations", True, "add", "(URI)", "", "Argument[0]", "Argument[this].Element", "value", "hq-manual"]

java/ql/lib/semmle/code/java/dataflow/internal/DataFlowUtil.qll

@@ -247,8 +247,8 @@ private predicate simpleLocalFlowStep0(Node node1, Node node2, string model) {
   or
   cloneStep(node1, node2) and model = "CloneStep"
   or
-  FlowSummaryImpl::Private::Steps::summaryLocalStep(node1, node2.(FlowSummaryNode).getSummaryNode(),
-    true, model)
+  FlowSummaryImpl::Private::Steps::summaryLocalStep(node1.(FlowSummaryNode).getSummaryNode(),
+    node2.(FlowSummaryNode).getSummaryNode(), true, model)
 }
 
 /**

java/ql/lib/semmle/code/java/dataflow/internal/FlowSummaryImpl.qll

@@ -41,8 +41,6 @@ module Input implements InputSig<Location, DataFlowImplSpecific::JavaDataFlow> {
 
   class SinkBase = Void;
 
-  class FlowSummaryCallBase = Void;
-
   predicate neutralElement(
     Input::SummarizedCallableBase c, string kind, string provenance, boolean isExact
   ) {
@@ -146,10 +144,6 @@ private module TypesInput implements Impl::Private::TypesInputSig {
 }
 
 private module StepsInput implements Impl::Private::StepsInputSig {
-  Impl::Private::SummaryNode getSummaryNode(Node n) {
-    result = n.(FlowSummaryNode).getSummaryNode()
-  }
-
   DataFlowCall getACall(Public::SummarizedCallable sc) {
     sc = viableCallable(result).asSummarizedCallable()
   }

java/ql/lib/semmle/code/java/dataflow/internal/TaintTrackingUtil.qll

@@ -145,8 +145,8 @@ private module Cached {
       )
     )
     or
-    FlowSummaryImpl::Private::Steps::summaryLocalStep(src,
-      sink.(DataFlowPrivate::FlowSummaryNode).getSummaryNode(), false, model)
+    FlowSummaryImpl::Private::Steps::summaryLocalStep(src.(DataFlowPrivate::FlowSummaryNode)
+          .getSummaryNode(), sink.(DataFlowPrivate::FlowSummaryNode).getSummaryNode(), false, model)
   }
 
   /**

java/ql/test/library-tests/dataflow/local-additional-taint/localAdditionalTaintStep.ql

@@ -13,7 +13,8 @@ predicate taintFlowUpdate(DataFlow::ParameterNode p1, DataFlow::ParameterNode p2
 }
 
 predicate summaryStep(FlowSummaryNode src, FlowSummaryNode sink) {
-  FlowSummaryImpl::Private::Steps::summaryLocalStep(src, sink.getSummaryNode(), false, _) or
+  FlowSummaryImpl::Private::Steps::summaryLocalStep(src.getSummaryNode(), sink.getSummaryNode(),
+    false, _) or
   FlowSummaryImpl::Private::Steps::summaryReadStep(src.getSummaryNode(), _, sink.getSummaryNode()) or
   FlowSummaryImpl::Private::Steps::summaryStoreStep(src.getSummaryNode(), _, sink.getSummaryNode())
 }

javascript/ql/lib/semmle/javascript/dataflow/internal/DataFlowPrivate.qll

@@ -1212,8 +1212,8 @@ private predicate valuePreservingStep(Node node1, Node node2) {
   or
   node2 = FlowSteps::getThrowTarget(node1)
   or
-  FlowSummaryPrivate::Steps::summaryLocalStep(node1, node2.(FlowSummaryNode).getSummaryNode(), true,
-    _) // TODO: preserve 'model'
+  FlowSummaryPrivate::Steps::summaryLocalStep(node1.(FlowSummaryNode).getSummaryNode(),
+    node2.(FlowSummaryNode).getSummaryNode(), true, _) // TODO: preserve 'model'
 }
 
 predicate knownSourceModel(Node sink, string model) { none() }

javascript/ql/lib/semmle/javascript/dataflow/internal/FlowSummaryPrivate.qll

@@ -142,10 +142,6 @@ string encodeArgumentPosition(ArgumentPosition pos) {
 ReturnKind getStandardReturnValueKind() { result = MkNormalReturnKind() and Stage::ref() }
 
 private module FlowSummaryStepInput implements Private::StepsInputSig {
-  Private::SummaryNode getSummaryNode(DataFlow::Node n) {
-    result = n.(FlowSummaryNode).getSummaryNode()
-  }
-
   overlay[global]
   DataFlowCall getACall(SummarizedCallable sc) {
     exists(LibraryCallable callable | callable = sc |

javascript/ql/lib/semmle/javascript/dataflow/internal/TaintTrackingPrivate.qll

@@ -12,8 +12,8 @@ cached
 predicate defaultAdditionalTaintStep(DataFlow::Node node1, DataFlow::Node node2) {
   TaintTracking::AdditionalTaintStep::step(node1, node2)
   or
-  FlowSummaryPrivate::Steps::summaryLocalStep(node1, node2.(FlowSummaryNode).getSummaryNode(),
-    false, _) // TODO: preserve 'model' parameter
+  FlowSummaryPrivate::Steps::summaryLocalStep(node1.(FlowSummaryNode).getSummaryNode(),
+    node2.(FlowSummaryNode).getSummaryNode(), false, _) // TODO: preserve 'model' parameter
   or
   // Convert steps out of array elements to plain taint steps
   FlowSummaryPrivate::Steps::summaryReadStep(node1.(FlowSummaryNode).getSummaryNode(),

javascript/ql/lib/semmle/javascript/dataflow/internal/sharedlib/DataFlowArg.qll

@@ -3,7 +3,6 @@ private import DataFlowImplSpecific
 private import codeql.dataflow.DataFlow as SharedDataFlow
 private import codeql.dataflow.TaintTracking as SharedTaintTracking
 private import codeql.dataflow.internal.FlowSummaryImpl as FlowSummaryImpl
-private import codeql.util.Void
 
 module JSDataFlow implements SharedDataFlow::InputSig<Location> {
   import Private
@@ -29,8 +28,6 @@ module JSFlowSummary implements FlowSummaryImpl::InputSig<Location, JSDataFlow>
   private import semmle.javascript.dataflow.internal.FlowSummaryPrivate as FlowSummaryPrivate
   import FlowSummaryPrivate
 
-  class FlowSummaryCallBase = Void;
-
   overlay[local]
   predicate callableFromSource(SummarizedCallableBase c) { none() }
 

python/ql/lib/CHANGELOG.md

@@ -70,7 +70,7 @@ No user-facing changes.
 
 ### Minor Analysis Improvements
 
-* Added new full SSRF sanitization barrier from the new AntiSSRF library.
+* Added new full SSRF sanitization barrier from the new AntiSSRF library. 
 * When a guard such as `isSafe(x)` is defined, we now also automatically handle `isSafe(x) == true` and `isSafe(x) != false`.
 
 ## 6.1.1
@@ -169,7 +169,7 @@ No user-facing changes.
 ### Minor Analysis Improvements
 
 - The modelling of Psycopg2 now supports the use of `psycopg2.pool` connection pools for handling database connections.
-* Removed `lxml` as an XML bomb sink. The underlying libxml2 library now includes [entity reference loop detection](https://github.com/lxml/lxml/blob/f33ac2c2f5f9c4c4c1fc47f363be96db308f2fa6/doc/FAQ.txt#L1077) that prevents XML bomb attacks.
+* Removed `lxml` as an XML bomb sink. The underlying libxml2 library now includes [entity reference loop detection](https://github.com/lxml/lxml/blob/f33ac2c2f5f9c4c4c1fc47f363be96db308f2fa6/doc/FAQ.txt#L1077) that prevents XML bomb attacks. 
 
 ## 4.0.13
 
@@ -262,7 +262,7 @@ No user-facing changes.
 ### Minor Analysis Improvements
 
 * The sensitive data library has been improved so that `snake_case` style variable names are recognized more reliably. This may result in more sensitive data being identified, and more results from queries that use the sensitive data library.
-- Additional taint steps through methods of `lxml.etree.Element` and `lxml.etree.ElementTree` objects from the `lxml` PyPI package have been modeled.
+- Additional taint steps through methods of `lxml.etree.Element` and `lxml.etree.ElementTree` objects from the `lxml` PyPI package have been modeled. 
 
 ## 3.1.0
 
@@ -316,7 +316,7 @@ No user-facing changes.
 
 ### 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.
+* 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
 
@@ -545,7 +545,7 @@ No user-facing changes.
 
 ### New Features
 
-* The `DataFlow::StateConfigSig` signature module has gained default implementations for `isBarrier/2` and `isAdditionalFlowStep/4`.
+* The `DataFlow::StateConfigSig` signature module has gained default implementations for `isBarrier/2` and `isAdditionalFlowStep/4`. 
   Hence it is no longer needed to provide `none()` implementations of these predicates if they are not needed.
 
 ### Minor Analysis Improvements
@@ -572,7 +572,7 @@ No user-facing changes.
 * Deleted many deprecated predicates and classes with uppercase `API`, `HTTP`, `XSS`, `SQL`, etc. in their names. Use the PascalCased versions instead.
 * Deleted the deprecated `getName()` predicate from the `Container` class, use `getAbsolutePath()` instead.
 * Deleted many deprecated module names that started with a lowercase letter, use the versions that start with an uppercase letter instead.
-* Deleted many deprecated predicates in `PointsTo.qll`.
+* Deleted many deprecated predicates in `PointsTo.qll`. 
 * Deleted many deprecated files from the `semmle.python.security` package.
 * Deleted the deprecated `BottleRoutePointToExtension` class from `Extensions.qll`.
 * Type tracking is now aware of flow summaries. This leads to a richer API graph, and may lead to more results in some queries.
@@ -729,7 +729,7 @@ No user-facing changes.
 ### Deprecated APIs
 
 * Some unused predicates in `SsaDefinitions.qll`, `TObject.qll`, `protocols.qll`, and the `pointsto/` folder have been deprecated.
-* Some classes/modules with upper-case acronyms in their name have been renamed to follow our style-guide.
+* Some classes/modules with upper-case acronyms in their name have been renamed to follow our style-guide. 
   The old name still exists as a deprecated alias.
 
 ### Minor Analysis Improvements
@@ -748,9 +748,9 @@ No user-facing changes.
 
 ### Deprecated APIs
 
-* Many classes/predicates/modules with upper-case acronyms in their name have been renamed to follow our style-guide.
+* Many classes/predicates/modules with upper-case acronyms in their name have been renamed to follow our style-guide. 
   The old name still exists as a deprecated alias.
-* The utility files previously in the `semmle.python.security.performance` package have been moved to the `semmle.python.security.regexp` package.
+* The utility files previously in the `semmle.python.security.performance` package have been moved to the `semmle.python.security.regexp` package.  
   The previous files still exist as deprecated aliases.
 
 ### Minor Analysis Improvements
@@ -843,9 +843,9 @@ No user-facing changes.
 
 ### Deprecated APIs
 
-* Many classes/predicates/modules that had upper-case acronyms have been renamed to follow our style-guide.
+* Many classes/predicates/modules that had upper-case acronyms have been renamed to follow our style-guide. 
   The old name still exists as a deprecated alias.
-* Some modules that started with a lowercase letter have been renamed to follow our style-guide.
+* Some modules that started with a lowercase letter have been renamed to follow our style-guide. 
   The old name still exists as a deprecated alias.
 
 ### New Features

python/ql/lib/semmle/python/dataflow/new/internal/DataFlowPrivate.qll

@@ -529,7 +529,7 @@ predicate simpleLocalFlowStepForTypetracking(Node nodeFrom, Node nodeTo) {
 }
 
 private predicate summaryLocalStep(Node nodeFrom, Node nodeTo, string model) {
-  FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom,
+  FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom.(FlowSummaryNode).getSummaryNode(),
     nodeTo.(FlowSummaryNode).getSummaryNode(), true, model)
 }
 
@@ -1138,9 +1138,7 @@ predicate clearsContent(Node n, ContentSet cs) {
  * Holds if the value that is being tracked is expected to be stored inside content `c`
  * at node `n`.
  */
-predicate expectsContent(Node n, ContentSet c) {
-  FlowSummaryImpl::Private::Steps::summaryExpectsContent(n.(FlowSummaryNode).getSummaryNode(), c)
-}
+predicate expectsContent(Node n, ContentSet c) { none() }
 
 /**
  * Holds if values stored inside attribute `c` are cleared at node `n`.

python/ql/lib/semmle/python/dataflow/new/internal/FlowSummaryImpl.qll

@@ -20,8 +20,6 @@ module Input implements InputSig<Location, DataFlowImplSpecific::PythonDataFlow>
 
   class SinkBase = Void;
 
-  class FlowSummaryCallBase = Void;
-
   predicate callableFromSource(SummarizedCallableBase c) { none() }
 
   ArgumentPosition callbackSelfParameterPosition() { result.isLambdaSelf() }
@@ -93,8 +91,6 @@ module Input implements InputSig<Location, DataFlowImplSpecific::PythonDataFlow>
     cs.isAnyTupleOrDictionaryElement() and result = "AnyTupleOrDictionaryElement" and arg = ""
   }
 
-  string encodeWithContent(ContentSet c, string arg) { result = "With" + encodeContent(c, arg) }
-
   bindingset[token]
   ParameterPosition decodeUnknownParameterPosition(AccessPath::AccessPathTokenBase token) {
     // needed to support `Argument[x..y]` ranges
@@ -113,10 +109,6 @@ module Input implements InputSig<Location, DataFlowImplSpecific::PythonDataFlow>
 private import Make<Location, DataFlowImplSpecific::PythonDataFlow, Input> as Impl
 
 private module StepsInput implements Impl::Private::StepsInputSig {
-  Impl::Private::SummaryNode getSummaryNode(Node n) {
-    result = n.(FlowSummaryNode).getSummaryNode()
-  }
-
   overlay[global]
   DataFlowCall getACall(Public::SummarizedCallable sc) {
     result =

python/ql/lib/semmle/python/dataflow/new/internal/TaintTrackingPrivate.qll

@@ -80,8 +80,10 @@ private module Cached {
     ) and
     model = ""
     or
-    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom,
-      nodeTo.(DataFlowPrivate::FlowSummaryNode).getSummaryNode(), false, model)
+    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom
+          .(DataFlowPrivate::FlowSummaryNode)
+          .getSummaryNode(), nodeTo.(DataFlowPrivate::FlowSummaryNode).getSummaryNode(), false,
+      model)
   }
 }
 

python/ql/lib/semmle/python/dataflow/new/internal/TypeTrackingImpl.qll

@@ -170,13 +170,7 @@ module TypeTrackingInput implements Shared::TypeTrackingInput<Location> {
 
   /** Holds if there is a level step from `nodeFrom` to `nodeTo`, which may depend on the call graph. */
   predicate levelStepCall(Node nodeFrom, LocalSourceNode nodeTo) {
-    // HOTFIX: `instanceFieldStep` is temporarily disabled (via `and none()`).
-    // It uses `classInstanceTracker(cls)` -- itself a type-tracker run --
-    // from inside `levelStepCall`, creating a structural mutual recursion
-    // that causes catastrophic query slowdowns on some OOP-heavy Python
-    // codebases (e.g. mypy and dask). The `and none()` should be removed
-    // once that recursion is redesigned.
-    instanceFieldStep(nodeFrom, nodeTo) and none()
+    instanceFieldStep(nodeFrom, nodeTo)
     or
     inheritedFieldStep(nodeFrom, nodeTo)
   }

python/ql/lib/semmle/python/frameworks/Stdlib.qll

@@ -4199,9 +4199,11 @@ module StdlibPrivate {
       // The positional argument contains a mapping.
       // TODO: these values can be overwritten by keyword arguments
       //  - dict mapping
-      input = "Argument[0].WithAnyDictionaryElement" and
-      output = "ReturnValue" and
-      preservesValue = true
+      exists(DataFlow::DictionaryElementContent dc, string key | key = dc.getKey() |
+        input = "Argument[0].DictionaryElement[" + key + "]" and
+        output = "ReturnValue.DictionaryElement[" + key + "]" and
+        preservesValue = true
+      )
       or
       //  - list-of-pairs mapping
       input = "Argument[0].ListElement.TupleElement[1]" and
@@ -4238,7 +4240,9 @@ module StdlibPrivate {
         or
         input = "Argument[0].SetElement"
         or
-        input = "Argument[0].AnyTupleElement"
+        exists(DataFlow::TupleElementContent tc, int i | i = tc.getIndex() |
+          input = "Argument[0].TupleElement[" + i.toString() + "]"
+        )
         // TODO: Once we have DictKeyContent, we need to transform that into ListElementContent
       ) and
       // Element content is mutated into list element content
@@ -4262,9 +4266,11 @@ module StdlibPrivate {
     }
 
     override predicate propagatesFlow(string input, string output, boolean preservesValue) {
-      input = "Argument[0].WithAnyTupleElement" and
-      output = "ReturnValue" and
-      preservesValue = true
+      exists(DataFlow::TupleElementContent tc, int i | i = tc.getIndex() |
+        input = "Argument[0].TupleElement[" + i.toString() + "]" and
+        output = "ReturnValue.TupleElement[" + i.toString() + "]" and
+        preservesValue = true
+      )
       or
       input = "Argument[0].ListElement" and
       output = "ReturnValue" and
@@ -4288,7 +4294,9 @@ module StdlibPrivate {
         or
         input = "Argument[0].SetElement"
         or
-        input = "Argument[0].AnyTupleElement"
+        exists(DataFlow::TupleElementContent tc, int i | i = tc.getIndex() |
+          input = "Argument[0].TupleElement[" + i.toString() + "]"
+        )
         // TODO: Once we have DictKeyContent, we need to transform that into ListElementContent
       ) and
       output = "ReturnValue.SetElement" and
@@ -4334,7 +4342,9 @@ module StdlibPrivate {
         or
         input = "Argument[0].SetElement"
         or
-        input = "Argument[0].AnyTupleElement"
+        exists(DataFlow::TupleElementContent tc, int i | i = tc.getIndex() |
+          input = "Argument[0].TupleElement[" + i.toString() + "]"
+        )
         // TODO: Once we have DictKeyContent, we need to transform that into ListElementContent
       ) and
       output = "ReturnValue.ListElement" and
@@ -4362,7 +4372,9 @@ module StdlibPrivate {
         or
         content = "SetElement"
         or
-        content = "AnyTupleElement"
+        exists(DataFlow::TupleElementContent tc, int i | i = tc.getIndex() |
+          content = "TupleElement[" + i.toString() + "]"
+        )
       |
         // TODO: Once we have DictKeyContent, we need to transform that into ListElementContent
         input = "Argument[0]." + content and
@@ -4392,7 +4404,9 @@ module StdlibPrivate {
         or
         input = "Argument[0].SetElement"
         or
-        input = "Argument[0].AnyTupleElement"
+        exists(DataFlow::TupleElementContent tc, int i | i = tc.getIndex() |
+          input = "Argument[0].TupleElement[" + i.toString() + "]"
+        )
         // TODO: Once we have DictKeyContent, we need to transform that into ListElementContent
       ) and
       output = "ReturnValue.ListElement" and
@@ -4420,7 +4434,9 @@ module StdlibPrivate {
         or
         input = "Argument[0].SetElement"
         or
-        input = "Argument[0].AnyTupleElement"
+        exists(DataFlow::TupleElementContent tc, int i | i = tc.getIndex() |
+          input = "Argument[0].TupleElement[" + i.toString() + "]"
+        )
         // TODO: Once we have DictKeyContent, we need to transform that into ListElementContent
       ) and
       output = "ReturnValue" and
@@ -4452,7 +4468,9 @@ module StdlibPrivate {
           // We reduce generality slightly by not tracking tuple contents on list arguments beyond the first, for performance.
           // TODO: Once we have TupleElementAny, this generality can be increased.
           i = 0 and
-          input = "Argument[1].AnyTupleElement"
+          exists(DataFlow::TupleElementContent tc, int j | j = tc.getIndex() |
+            input = "Argument[1].TupleElement[" + j.toString() + "]"
+          )
           // TODO: Once we have DictKeyContent, we need to transform that into ListElementContent
         ) and
         output = "Argument[0].Parameter[" + i.toString() + "]" and
@@ -4481,7 +4499,9 @@ module StdlibPrivate {
         or
         input = "Argument[1].SetElement"
         or
-        input = "Argument[1].AnyTupleElement"
+        exists(DataFlow::TupleElementContent tc, int i | i = tc.getIndex() |
+          input = "Argument[1].TupleElement[" + i.toString() + "]"
+        )
         // TODO: Once we have DictKeyContent, we need to transform that into ListElementContent
       ) and
       (output = "Argument[0].Parameter[0]" or output = "ReturnValue.ListElement") and
@@ -4505,7 +4525,9 @@ module StdlibPrivate {
         or
         input = "Argument[0].SetElement"
         or
-        input = "Argument[0].AnyTupleElement"
+        exists(DataFlow::TupleElementContent tc, int i | i = tc.getIndex() |
+          input = "Argument[0].TupleElement[" + i.toString() + "]"
+        )
         // TODO: Once we have DictKeyContent, we need to transform that into ListElementContent
       ) and
       output = "ReturnValue.ListElement.TupleElement[1]" and
@@ -4530,7 +4552,12 @@ module StdlibPrivate {
           or
           input = "Argument[" + i.toString() + "].SetElement"
           or
-          input = "Argument[" + i.toString() + "].AnyTupleElement"
+          // We reduce generality slightly by not tracking tuple contents on arguments beyond the first two, for performance.
+          // TODO: Once we have TupleElementAny, this generality can be increased.
+          i in [0 .. 1] and
+          exists(DataFlow::TupleElementContent tc, int j | j = tc.getIndex() |
+            input = "Argument[" + i.toString() + "].TupleElement[" + j.toString() + "]"
+          )
           // TODO: Once we have DictKeyContent, we need to transform that into ListElementContent
         ) and
         output = "ReturnValue.ListElement.TupleElement[" + i.toString() + "]" and
@@ -4553,6 +4580,12 @@ module StdlibPrivate {
     override DataFlow::ArgumentNode getACallback() { none() }
 
     override predicate propagatesFlow(string input, string output, boolean preservesValue) {
+      exists(DataFlow::Content c |
+        input = "Argument[self]." + c.getMaDRepresentation() and
+        output = "ReturnValue." + c.getMaDRepresentation() and
+        preservesValue = true
+      )
+      or
       input = "Argument[self]" and
       output = "ReturnValue" and
       preservesValue = true
@@ -4708,10 +4741,12 @@ module StdlibPrivate {
     override DataFlow::ArgumentNode getACallback() { none() }
 
     override predicate propagatesFlow(string input, string output, boolean preservesValue) {
-      input = "Argument[self].AnyDictionaryElement" and
-      output = "ReturnValue.TupleElement[1]" and
-      preservesValue = true
-      // TODO: put `key` into "ReturnValue.TupleElement[0]"
+      exists(DataFlow::DictionaryElementContent dc, string key | key = dc.getKey() |
+        input = "Argument[self].DictionaryElement[" + key + "]" and
+        output = "ReturnValue.TupleElement[1]" and
+        preservesValue = true
+        // TODO: put `key` into "ReturnValue.TupleElement[0]"
+      )
     }
   }
 
@@ -4790,9 +4825,11 @@ module StdlibPrivate {
     }
 
     override predicate propagatesFlow(string input, string output, boolean preservesValue) {
-      input = "Argument[self].AnyDictionaryElement" and
-      output = "ReturnValue.ListElement" and
-      preservesValue = true
+      exists(DataFlow::DictionaryElementContent dc, string key | key = dc.getKey() |
+        input = "Argument[self].DictionaryElement[" + key + "]" and
+        output = "ReturnValue.ListElement" and
+        preservesValue = true
+      )
       or
       input = "Argument[self]" and
       output = "ReturnValue" and
@@ -4839,9 +4876,11 @@ module StdlibPrivate {
     }
 
     override predicate propagatesFlow(string input, string output, boolean preservesValue) {
-      input = "Argument[self].AnyDictionaryElement" and
-      output = "ReturnValue.ListElement.TupleElement[1]" and
-      preservesValue = true
+      exists(DataFlow::DictionaryElementContent dc, string key | key = dc.getKey() |
+        input = "Argument[self].DictionaryElement[" + key + "]" and
+        output = "ReturnValue.ListElement.TupleElement[1]" and
+        preservesValue = true
+      )
       or
       // TODO: Add the keys to output list
       input = "Argument[self]" and

python/ql/test/library-tests/dataflow/coverage/test_builtins.py

@@ -589,11 +589,11 @@ def test_zip_tuple():
 
     SINK(z[0][0])  # $ flow="SOURCE, l:-7 -> z[0][0]"
     SINK(z[0][1])  # $ flow="SOURCE, l:-7 -> z[0][1]"
-    SINK_F(z[0][2]) # $ SPURIOUS: flow="SOURCE, l:-7 -> z[0][2]"
+    SINK_F(z[0][2])
     SINK_F(z[0][3])
     SINK(z[1][0])  # $ flow="SOURCE, l:-11 -> z[1][0]"
     SINK_F(z[1][1]) # $ SPURIOUS: flow="SOURCE, l:-11 -> z[1][1]"
-    SINK(z[1][2]) # $ flow="SOURCE, l:-11 -> z[1][2]"
+    SINK(z[1][2]) # $ MISSING: flow="SOURCE, l:-11 -> z[1][2]" # Tuple contents are not tracked beyond the first two arguments for performance.
     SINK_F(z[1][3])
 
 @expects(4)

python/ql/test/library-tests/dataflow/typetracking/attribute_tests.py

@@ -157,7 +157,7 @@ class MyClass2(object):
         print(self.foo) # $ tracked MISSING: tracked=foo
 
 instance = MyClass2()
-print(instance.foo) # $ MISSING: tracked=foo tracked
+print(instance.foo) # $ tracked MISSING: tracked=foo
 instance.print_foo() # $ MISSING: tracked=foo
 
 
@@ -195,7 +195,7 @@ class Sub1(Base1):
 
 sub1 = Sub1()
 sub1.read_foo()
-print(sub1.foo) # $ MISSING: tracked=foo tracked
+print(sub1.foo) # $ tracked MISSING: tracked=foo
 
 
 # attribute written in a subclass method, read in an inherited base class method
@@ -210,7 +210,7 @@ class Sub2(Base2):
 
 sub2 = Sub2()
 sub2.read_bar()
-print(sub2.bar) # $ MISSING: tracked=bar tracked
+print(sub2.bar) # $ tracked MISSING: tracked=bar
 
 
 # attribute written in a base class method, read on an instance of the subclass
@@ -223,4 +223,4 @@ class Sub3(Base3):
     pass
 
 sub3 = Sub3()
-print(sub3.baz) # $ MISSING: tracked=baz tracked
+print(sub3.baz) # $ tracked MISSING: tracked=baz

python/ql/test/library-tests/frameworks/django-orm/testapp/orm_tests.py

@@ -362,7 +362,7 @@ def test_load_in_bulk():
     # see https://docs.djangoproject.com/en/4.0/ref/models/querysets/#in-bulk
     d = TestLoad.objects.in_bulk([1])
     for val in d.values():
-        SINK(val.text) # $ flow="SOURCE, l:-65 -> val.text"
+        SINK(val.text) # $ MISSING: flow
     SINK(d[1].text) # $ flow="SOURCE, l:-66 -> d[1].text"
 
 

python/ql/test/query-tests/Security/CWE-089-SqlInjection/SqlInjection.expected

@@ -1,6 +1,7 @@
 #select
 | app.py:23:20:23:24 | ControlFlowNode for query | app.py:20:18:20:21 | ControlFlowNode for name | app.py:23:20:23:24 | ControlFlowNode for query | This SQL query depends on a $@. | app.py:20:18:20:21 | ControlFlowNode for name | user-provided value |
 | app.py:30:20:30:24 | ControlFlowNode for query | app.py:27:19:27:22 | ControlFlowNode for name | app.py:30:20:30:24 | ControlFlowNode for query | This SQL query depends on a $@. | app.py:27:19:27:22 | ControlFlowNode for name | user-provided value |
+| app.py:37:20:37:24 | ControlFlowNode for query | app.py:34:19:34:22 | ControlFlowNode for name | app.py:37:20:37:24 | ControlFlowNode for query | This SQL query depends on a $@. | app.py:34:19:34:22 | ControlFlowNode for name | user-provided value |
 | app.py:44:20:44:24 | ControlFlowNode for query | app.py:41:19:41:22 | ControlFlowNode for name | app.py:44:20:44:24 | ControlFlowNode for query | This SQL query depends on a $@. | app.py:41:19:41:22 | ControlFlowNode for name | user-provided value |
 | app.py:51:20:51:24 | ControlFlowNode for query | app.py:48:19:48:22 | ControlFlowNode for name | app.py:51:20:51:24 | ControlFlowNode for query | This SQL query depends on a $@. | app.py:48:19:48:22 | ControlFlowNode for name | user-provided value |
 | sql_injection.py:21:24:21:77 | ControlFlowNode for BinaryExpr | sql_injection.py:14:15:14:22 | ControlFlowNode for username | sql_injection.py:21:24:21:77 | ControlFlowNode for BinaryExpr | This SQL query depends on a $@. | sql_injection.py:14:15:14:22 | ControlFlowNode for username | user-provided value |
@@ -24,6 +25,8 @@ edges
 | app.py:21:5:21:9 | ControlFlowNode for query | app.py:23:20:23:24 | ControlFlowNode for query | provenance |  |
 | app.py:27:19:27:22 | ControlFlowNode for name | app.py:28:5:28:9 | ControlFlowNode for query | provenance |  |
 | app.py:28:5:28:9 | ControlFlowNode for query | app.py:30:20:30:24 | ControlFlowNode for query | provenance |  |
+| app.py:34:19:34:22 | ControlFlowNode for name | app.py:35:5:35:9 | ControlFlowNode for query | provenance |  |
+| app.py:35:5:35:9 | ControlFlowNode for query | app.py:37:20:37:24 | ControlFlowNode for query | provenance |  |
 | app.py:41:19:41:22 | ControlFlowNode for name | app.py:42:5:42:9 | ControlFlowNode for query | provenance |  |
 | app.py:42:5:42:9 | ControlFlowNode for query | app.py:44:20:44:24 | ControlFlowNode for query | provenance |  |
 | app.py:48:19:48:22 | ControlFlowNode for name | app.py:49:5:49:9 | ControlFlowNode for query | provenance |  |
@@ -51,6 +54,9 @@ nodes
 | app.py:27:19:27:22 | ControlFlowNode for name | semmle.label | ControlFlowNode for name |
 | app.py:28:5:28:9 | ControlFlowNode for query | semmle.label | ControlFlowNode for query |
 | app.py:30:20:30:24 | ControlFlowNode for query | semmle.label | ControlFlowNode for query |
+| app.py:34:19:34:22 | ControlFlowNode for name | semmle.label | ControlFlowNode for name |
+| app.py:35:5:35:9 | ControlFlowNode for query | semmle.label | ControlFlowNode for query |
+| app.py:37:20:37:24 | ControlFlowNode for query | semmle.label | ControlFlowNode for query |
 | app.py:41:19:41:22 | ControlFlowNode for name | semmle.label | ControlFlowNode for name |
 | app.py:42:5:42:9 | ControlFlowNode for query | semmle.label | ControlFlowNode for query |
 | app.py:44:20:44:24 | ControlFlowNode for query | semmle.label | ControlFlowNode for query |

python/ql/test/query-tests/Security/CWE-089-SqlInjection/app.py

@@ -31,10 +31,10 @@ async def unsafe2(name: str): # $ Source
     cursor.close()
 
 @app.get("/unsafe3/")
-async def unsafe3(name: str): # $ MISSING: Source
+async def unsafe3(name: str): # $ Source
     query = "select * from users where name=" + name
     cursor = hdb_con3.cursor()
-    cursor.execute(query) # $ MISSING: Alert
+    cursor.execute(query) # $ Alert
     cursor.close()
 
 @app.get("/unsafe4/")

ruby/ql/lib/codeql/ruby/dataflow/internal/DataFlowPrivate.qll

@@ -198,7 +198,8 @@ module LocalFlow {
     FlowSummaryNode nodeFrom, FlowSummaryNode nodeTo, FlowSummaryImpl::Public::SummarizedCallable c,
     string model
   ) {
-    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom, nodeTo.getSummaryNode(), true, model) and
+    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom.getSummaryNode(),
+      nodeTo.getSummaryNode(), true, model) and
     c = nodeFrom.getSummarizedCallable()
   }
 

ruby/ql/lib/codeql/ruby/dataflow/internal/FlowSummaryImpl.qll

@@ -18,8 +18,6 @@ module Input implements InputSig<Location, DataFlowImplSpecific::RubyDataFlow> {
 
   class SinkBase = Void;
 
-  class FlowSummaryCallBase = Void;
-
   predicate callableFromSource(SummarizedCallableBase c) { none() }
 
   ArgumentPosition callbackSelfParameterPosition() { result.isLambdaSelf() }
@@ -159,10 +157,6 @@ module Input implements InputSig<Location, DataFlowImplSpecific::RubyDataFlow> {
 private import Make<Location, DataFlowImplSpecific::RubyDataFlow, Input> as Impl
 
 private module StepsInput implements Impl::Private::StepsInputSig {
-  Impl::Private::SummaryNode getSummaryNode(Node n) {
-    result = n.(FlowSummaryNode).getSummaryNode()
-  }
-
   DataFlowCall getACall(Public::SummarizedCallable sc) {
     result.asCall().getAstNode() = sc.(LibraryCallable).getACall()
     or

ruby/ql/lib/codeql/ruby/dataflow/internal/TaintTrackingPrivate.qll

@@ -109,7 +109,7 @@ private module Cached {
     ) and
     model = ""
     or
-    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom,
+    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom.(FlowSummaryNode).getSummaryNode(),
       nodeTo.(FlowSummaryNode).getSummaryNode(), false, model)
     or
     any(FlowSteps::AdditionalTaintStep s).step(nodeFrom, nodeTo) and model = "AdditionalTaintStep"

ruby/ql/test/library-tests/frameworks/active_support/ActiveSupportDataFlow.expected

@@ -231,11 +231,11 @@ edges
 | hash_extensions.rb:122:5:122:10 | single : Array [element 0] | hash_extensions.rb:125:10:125:15 | single : Array [element 0] | provenance |  |
 | hash_extensions.rb:122:14:122:26 | call to [] : Array [element 0] | hash_extensions.rb:122:5:122:10 | single : Array [element 0] | provenance |  |
 | hash_extensions.rb:122:15:122:25 | call to source | hash_extensions.rb:122:14:122:26 | call to [] : Array [element 0] | provenance |  |
-| hash_extensions.rb:123:5:123:9 | multi : Array [element 0] | hash_extensions.rb:127:10:127:14 | multi : Array [element 0] | provenance |  |
+| hash_extensions.rb:123:5:123:9 | multi : Array [element 0] | hash_extensions.rb:126:10:126:14 | multi : Array [element 0] | provenance |  |
 | hash_extensions.rb:123:13:123:38 | call to [] : Array [element 0] | hash_extensions.rb:123:5:123:9 | multi : Array [element 0] | provenance |  |
 | hash_extensions.rb:123:14:123:24 | call to source | hash_extensions.rb:123:13:123:38 | call to [] : Array [element 0] | provenance |  |
 | hash_extensions.rb:125:10:125:15 | single : Array [element 0] | hash_extensions.rb:125:10:125:20 | call to sole | provenance |  |
-| hash_extensions.rb:127:10:127:14 | multi : Array [element 0] | hash_extensions.rb:127:10:127:19 | call to sole | provenance |  |
+| hash_extensions.rb:126:10:126:14 | multi : Array [element 0] | hash_extensions.rb:126:10:126:19 | call to sole | provenance |  |
 nodes
 | active_support.rb:180:5:180:5 | x : Array [element 0] | semmle.label | x : Array [element 0] |
 | active_support.rb:180:9:180:18 | call to [] : Array [element 0] | semmle.label | call to [] : Array [element 0] |
@@ -493,10 +493,11 @@ nodes
 | hash_extensions.rb:123:14:123:24 | call to source | semmle.label | call to source |
 | hash_extensions.rb:125:10:125:15 | single : Array [element 0] | semmle.label | single : Array [element 0] |
 | hash_extensions.rb:125:10:125:20 | call to sole | semmle.label | call to sole |
-| hash_extensions.rb:127:10:127:14 | multi : Array [element 0] | semmle.label | multi : Array [element 0] |
-| hash_extensions.rb:127:10:127:19 | call to sole | semmle.label | call to sole |
+| hash_extensions.rb:126:10:126:14 | multi : Array [element 0] | semmle.label | multi : Array [element 0] |
+| hash_extensions.rb:126:10:126:19 | call to sole | semmle.label | call to sole |
 subpaths
 testFailures
+| hash_extensions.rb:126:10:126:19 | call to sole | Unexpected result: hasValueFlow=b |
 #select
 | active_support.rb:182:10:182:13 | ...[...] | active_support.rb:180:10:180:17 | call to source | active_support.rb:182:10:182:13 | ...[...] | $@ | active_support.rb:180:10:180:17 | call to source | call to source |
 | active_support.rb:188:10:188:13 | ...[...] | active_support.rb:186:10:186:18 | call to source | active_support.rb:188:10:188:13 | ...[...] | $@ | active_support.rb:186:10:186:18 | call to source | call to source |
@@ -557,4 +558,4 @@ testFailures
 | hash_extensions.rb:115:10:115:39 | ...[...] | hash_extensions.rb:110:21:110:31 | call to source | hash_extensions.rb:115:10:115:39 | ...[...] | $@ | hash_extensions.rb:110:21:110:31 | call to source | call to source |
 | hash_extensions.rb:115:10:115:39 | ...[...] | hash_extensions.rb:110:65:110:75 | call to source | hash_extensions.rb:115:10:115:39 | ...[...] | $@ | hash_extensions.rb:110:65:110:75 | call to source | call to source |
 | hash_extensions.rb:125:10:125:20 | call to sole | hash_extensions.rb:122:15:122:25 | call to source | hash_extensions.rb:125:10:125:20 | call to sole | $@ | hash_extensions.rb:122:15:122:25 | call to source | call to source |
-| hash_extensions.rb:127:10:127:19 | call to sole | hash_extensions.rb:123:14:123:24 | call to source | hash_extensions.rb:127:10:127:19 | call to sole | $@ | hash_extensions.rb:123:14:123:24 | call to source | call to source |
+| hash_extensions.rb:126:10:126:19 | call to sole | hash_extensions.rb:123:14:123:24 | call to source | hash_extensions.rb:126:10:126:19 | call to sole | $@ | hash_extensions.rb:123:14:123:24 | call to source | call to source |

ruby/ql/test/library-tests/frameworks/active_support/hash_extensions.rb

@@ -123,8 +123,7 @@ def m_sole
     multi = [source("b"), source("c")]
     sink(empty.sole)
     sink(single.sole) # $ hasValueFlow=a
-    # TODO: model that 'sole' does not return if the receiver has multiple elements
-    sink(multi.sole) # $ SPURIOUS: hasValueFlow=b
+    sink(multi.sole) # TODO: model that 'sole' does not return if the receiver has multiple elements
 end
 
 m_sole()

rust/ql/lib/codeql/rust/dataflow/internal/DataFlowImpl.qll

@@ -180,7 +180,7 @@ Expr getPostUpdateReverseStep(Expr e, boolean preservesValue) {
 module LocalFlow {
   predicate flowSummaryLocalStep(Node nodeFrom, Node nodeTo, string model) {
     exists(FlowSummaryImpl::Public::SummarizedCallable c |
-      FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom,
+      FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom.(FlowSummaryNode).getSummaryNode(),
         nodeTo.(FlowSummaryNode).getSummaryNode(), true, model) and
       c = nodeFrom.(FlowSummaryNode).getSummarizedCallable()
     )

rust/ql/lib/codeql/rust/dataflow/internal/FlowSummaryImpl.qll

@@ -11,7 +11,6 @@ private import codeql.rust.dataflow.FlowSummary
 private import codeql.rust.dataflow.Ssa
 private import codeql.rust.dataflow.internal.ModelsAsData
 private import Content
-private import Node
 
 predicate encodeContentTupleField(TupleFieldContent c, string arg) {
   exists(Addressable a, int pos, string prefix |
@@ -29,12 +28,9 @@ predicate encodeContentStructField(StructFieldContent c, string arg) {
 
 module Input implements InputSig<Location, RustDataFlow> {
   private import codeql.rust.frameworks.stdlib.Stdlib
-  private import codeql.util.Void
 
   class SummarizedCallableBase = Function;
 
-  class FlowSummaryCallBase = Void;
-
   predicate callableFromSource(SummarizedCallableBase c) { c.fromSource() }
 
   abstract private class SourceSinkBase extends AstNode {
@@ -148,10 +144,6 @@ module Input implements InputSig<Location, RustDataFlow> {
 private import Make<Location, RustDataFlow, Input> as Impl
 
 module StepsInput implements Impl::Private::StepsInputSig {
-  Impl::Private::SummaryNode getSummaryNode(RustDataFlow::Node n) {
-    result = n.(FlowSummaryNode).getSummaryNode()
-  }
-
   DataFlowCall getACall(Public::SummarizedCallable sc) { result.asCall().getStaticTarget() = sc }
 
   /** Gets the argument of `source` described by `sc`, if any. */

rust/ql/lib/codeql/rust/dataflow/internal/TaintTrackingImpl.qll

@@ -83,7 +83,7 @@ module RustTaintTrackingGen<DataFlowImpl::RustDataFlowInputSig I> implements
         pred.(Node::PostUpdateNode).getPreUpdateNode().asExpr(), _, succ, _)
     )
     or
-    FlowSummaryImpl::Private::Steps::summaryLocalStep(pred,
+    FlowSummaryImpl::Private::Steps::summaryLocalStep(pred.(Node::FlowSummaryNode).getSummaryNode(),
       succ.(Node::FlowSummaryNode).getSummaryNode(), false, model)
   }
 

rust/ql/lib/codeql/rust/security/HardcodedCryptographicValueExtensions.qll

@@ -62,34 +62,24 @@ module HardcodedCryptographicValue {
   abstract class Barrier extends DataFlow::Node { }
 
   /**
-   * Holds if `e` is a literal or a combination of literals that is constant.
+   * A literal, considered as a flow source.
    */
-  private predicate isConstant(Expr e) {
-    e instanceof LiteralExpr // e.g. `0`
-    or
-    forex(Expr elem | elem = e.(ArrayListExpr).getExpr(_) | isConstant(elem)) // e.g. `[0, 0, 0, 0]`
-    or
-    isConstant(e.(ArrayRepeatExpr).getRepeatOperand()) // e.g. `[0; 10]`
-    or
-    // e.g. `const MY_CONST: u64 = ...`
-    // the constant initializer / body is the preferred source location for flow paths, when available.
-    e = any(Const c).getBody()
-    or
-    // e.g. `u64::MAX`
-    // when the constant initializer is not available as a source location (case above), use the access instead.
-    e instanceof ConstAccess and
-    not exists(e.(ConstAccess).getConst().getBody())
-    or
-    // e.g. `1 << 4`
-    isConstant(e.(BinaryExpr).getLhs()) and
-    isConstant(e.(BinaryExpr).getRhs())
+  private class LiteralSource extends Source {
+    LiteralSource() { this.asExpr() instanceof LiteralExpr }
   }
 
   /**
-   * A constant, considered as a flow source.
+   * An array initialized from a list of literals, considered as a single flow source. For example:
+   * ```
+   * [0, 0, 0, 0]
+   * [0; 10]
+   * ```
    */
-  private class ConstantSource extends Source {
-    ConstantSource() { isConstant(this.asExpr()) }
+  private class ArrayListSource extends Source {
+    ArrayListSource() {
+      this.asExpr().(ArrayListExpr).getExpr(_) instanceof LiteralExpr or
+      this.asExpr().(ArrayRepeatExpr).getRepeatOperand() instanceof LiteralExpr
+    }
   }
 
   /**
@@ -165,24 +155,4 @@ module HardcodedCryptographicValue {
       )
     }
   }
-
-  /**
-   * An arithmetic or bitwise operation that acts as a barrier.
-   *
-   * This prevents false positives where a hard-coded value is combined with
-   * non-constant data through operations like `+`, `^`, or `+=` (including string concatenation).
-   */
-  private class ArithmeticOperationBarrier extends Barrier {
-    ArithmeticOperationBarrier() {
-      // binary operations (e.g. `a + b`, `a ^ b`)
-      this.asExpr() = any(BinaryArithmeticOperation a).getAnOperand()
-      or
-      this.asExpr() = any(BinaryBitwiseOperation a).getAnOperand()
-      or
-      // compound assignments (e.g. `a += b`, `a ^= b`)
-      this.asExpr() = any(AssignArithmeticOperation a).getAnOperand()
-      or
-      this.asExpr() = any(AssignBitwiseOperation a).getAnOperand()
-    }
-  }
 }

rust/ql/src/change-notes/2026-06-25-hard-coded-cryptographic-value-arithmetic-barrier.md

@@ -1,4 +0,0 @@
----
-category: minorAnalysis
----
-* The `rust/hard-coded-cryptographic-value` query now treats arithmetic and bitwise operations, including string append operations, as barriers. This addresses false positive results where hard-coded constants are combined with non-constant data, such as incrementing a nonce or appending variable data to a constant prefix.

rust/ql/test/library-tests/type-inference/CONSISTENCY/PathResolutionConsistency.expected

@@ -1,5 +1,5 @@
 multipleResolvedTargets
-| main.rs:2240:9:2240:31 | ... .my_add(...) |
-| main.rs:2242:9:2242:29 | ... .my_add(...) |
-| main.rs:2757:13:2757:17 | x.f() |
+| main.rs:2223:9:2223:31 | ... .my_add(...) |
+| main.rs:2225:9:2225:29 | ... .my_add(...) |
+| main.rs:2740:13:2740:17 | x.f() |
 | regressions.rs:179:17:179:27 | ... + ... |

rust/ql/test/library-tests/type-inference/main.rs

@@ -264,10 +264,6 @@ mod method_non_parametric_trait_impl {
         }
     }
 
-    trait MyTrait2<B> {
-        fn m3(self) -> B;
-    }
-
     trait MyProduct<A, B> {
         // MyProduct::fst
         fn fst(self) -> A;
@@ -279,10 +275,6 @@ mod method_non_parametric_trait_impl {
         x.m1() // $ target=m1
     }
 
-    fn call_trait_m1_trait2_m3<T1, T2: MyTrait<T1>, T3: MyTrait2<T2>>(x: T3) -> T1 {
-        x.m3().m1() // $ target=m1 target=m3
-    }
-
     impl MyTrait<S1> for MyThing<S1> {
         // MyThing<S1>::m1
         fn m1(self) -> S1 {
@@ -297,13 +289,6 @@ mod method_non_parametric_trait_impl {
         }
     }
 
-    impl MyTrait2<MyThing<S1>> for MyThing<S2> {
-        // MyThing<S2>::m3
-        fn m3(self) -> MyThing<S1> {
-            MyThing { a: S1 }
-        }
-    }
-
     // Implementation where the type parameter `TD` only occurs in the
     // implemented trait and not the implementing type.
     impl<TD> MyTrait<TD> for MyThing<S3>
@@ -468,8 +453,6 @@ mod method_non_parametric_trait_impl {
         let thing = MyThing { a: S1 };
         let i = thing.convert_to(); // $ type=i:S1 target=T::convert_to
         let j = convert_to(thing); // $ target=convert_to $ MISSING: type=j:S1 -- the blanket implementation `impl<T: MyTrait<S1>> ConvertTo<S1> for T` is currently not included in the constraint analysis
-
-        let x = call_trait_m1_trait2_m3(MyThing { a: S2 }); // $ target=call_trait_m1_trait2_m3 type=x:S1
     }
 }
 

rust/ql/test/query-tests/security/CWE-798/HardcodedCryptographicValue.expected

@@ -10,50 +10,45 @@
 | test_cookie.rs:21:28:21:34 | [0; 64] | test_cookie.rs:21:28:21:34 | [0; 64] | test_cookie.rs:22:16:22:24 | ...::from | This hard-coded value is used as $@. | test_cookie.rs:22:16:22:24 | ...::from | a key |
 | test_cookie.rs:38:28:38:36 | [0u8; 64] | test_cookie.rs:38:28:38:36 | [0u8; 64] | test_cookie.rs:42:14:42:32 | ...::from | This hard-coded value is used as $@. | test_cookie.rs:42:14:42:32 | ...::from | a key |
 | test_cookie.rs:49:23:49:25 | 0u8 | test_cookie.rs:49:23:49:25 | 0u8 | test_cookie.rs:53:14:53:32 | ...::from | This hard-coded value is used as $@. | test_cookie.rs:53:14:53:32 | ...::from | a key |
-| test_heuristic.rs:38:25:38:30 | 0xFFFF | test_heuristic.rs:38:25:38:30 | 0xFFFF | test_heuristic.rs:81:22:81:31 | MY_CONST_1 | This hard-coded value is used as $@. | test_heuristic.rs:81:22:81:31 | MY_CONST_1 | a salt |
-| test_heuristic.rs:39:25:39:59 | ... as u64 | test_heuristic.rs:39:25:39:59 | ... as u64 | test_heuristic.rs:82:22:82:31 | MY_CONST_2 | This hard-coded value is used as $@. | test_heuristic.rs:82:22:82:31 | MY_CONST_2 | a salt |
-| test_heuristic.rs:40:27:40:32 | 0xFFFF | test_heuristic.rs:40:27:40:32 | 0xFFFF | test_heuristic.rs:83:22:83:32 | MY_STATIC_3 | This hard-coded value is used as $@. | test_heuristic.rs:83:22:83:32 | MY_STATIC_3 | a salt |
-| test_heuristic.rs:49:31:49:38 | [0u8; 16] | test_heuristic.rs:49:31:49:38 | [0u8; 16] | test_heuristic.rs:50:41:50:48 | const_iv | This hard-coded value is used as $@. | test_heuristic.rs:50:41:50:48 | const_iv | an initialization vector |
-| test_heuristic.rs:68:30:68:37 | "secret" | test_heuristic.rs:68:30:68:37 | "secret" | test_heuristic.rs:68:30:68:37 | "secret" | This hard-coded value is used as $@. | test_heuristic.rs:68:30:68:37 | "secret" | a password |
-| test_heuristic.rs:69:20:69:27 | [0u8; 16] | test_heuristic.rs:69:20:69:27 | [0u8; 16] | test_heuristic.rs:69:19:69:27 | &... | This hard-coded value is used as $@. | test_heuristic.rs:69:19:69:27 | &... | a nonce |
-| test_heuristic.rs:70:31:70:38 | [0u8; 16] | test_heuristic.rs:70:31:70:38 | [0u8; 16] | test_heuristic.rs:70:30:70:38 | &... | This hard-coded value is used as $@. | test_heuristic.rs:70:30:70:38 | &... | a salt |
-| test_heuristic.rs:72:22:72:22 | 0 | test_heuristic.rs:72:22:72:22 | 0 | test_heuristic.rs:72:22:72:22 | 0 | This hard-coded value is used as $@. | test_heuristic.rs:72:22:72:22 | 0 | a salt |
-| test_heuristic.rs:76:22:76:27 | ... << ... | test_heuristic.rs:76:22:76:27 | ... << ... | test_heuristic.rs:76:22:76:27 | ... << ... | This hard-coded value is used as $@. | test_heuristic.rs:76:22:76:27 | ... << ... | a salt |
-| test_heuristic.rs:78:22:78:29 | ...::MAX | test_heuristic.rs:78:22:78:29 | ...::MAX | test_heuristic.rs:78:22:78:29 | ...::MAX | This hard-coded value is used as $@. | test_heuristic.rs:78:22:78:29 | ...::MAX | a salt |
-| test_heuristic.rs:79:22:79:33 | ... / ... | test_heuristic.rs:79:22:79:33 | ... / ... | test_heuristic.rs:79:22:79:33 | ... / ... | This hard-coded value is used as $@. | test_heuristic.rs:79:22:79:33 | ... / ... | a salt |
-| test_heuristic.rs:86:29:86:32 | 1u64 | test_heuristic.rs:86:29:86:32 | 1u64 | test_heuristic.rs:87:22:87:31 | MY_CONST_5 | This hard-coded value is used as $@. | test_heuristic.rs:87:22:87:31 | MY_CONST_5 | a salt |
-| test_heuristic.rs:88:29:88:33 | ... + ... | test_heuristic.rs:88:29:88:33 | ... + ... | test_heuristic.rs:89:22:89:31 | MY_CONST_6 | This hard-coded value is used as $@. | test_heuristic.rs:89:22:89:31 | MY_CONST_6 | a salt |
+| test_heuristic.rs:44:31:44:38 | [0u8; 16] | test_heuristic.rs:44:31:44:38 | [0u8; 16] | test_heuristic.rs:45:41:45:48 | const_iv | This hard-coded value is used as $@. | test_heuristic.rs:45:41:45:48 | const_iv | an initialization vector |
+| test_heuristic.rs:63:30:63:37 | "secret" | test_heuristic.rs:63:30:63:37 | "secret" | test_heuristic.rs:63:30:63:37 | "secret" | This hard-coded value is used as $@. | test_heuristic.rs:63:30:63:37 | "secret" | a password |
+| test_heuristic.rs:64:20:64:27 | [0u8; 16] | test_heuristic.rs:64:20:64:27 | [0u8; 16] | test_heuristic.rs:64:19:64:27 | &... | This hard-coded value is used as $@. | test_heuristic.rs:64:19:64:27 | &... | a nonce |
+| test_heuristic.rs:65:31:65:38 | [0u8; 16] | test_heuristic.rs:65:31:65:38 | [0u8; 16] | test_heuristic.rs:65:30:65:38 | &... | This hard-coded value is used as $@. | test_heuristic.rs:65:30:65:38 | &... | a salt |
+| test_heuristic.rs:67:22:67:22 | 0 | test_heuristic.rs:67:22:67:22 | 0 | test_heuristic.rs:67:22:67:22 | 0 | This hard-coded value is used as $@. | test_heuristic.rs:67:22:67:22 | 0 | a salt |
+| test_heuristic.rs:69:32:69:32 | 1 | test_heuristic.rs:69:32:69:32 | 1 | test_heuristic.rs:69:22:69:32 | ... + ... | This hard-coded value is used as $@. | test_heuristic.rs:69:22:69:32 | ... + ... | a salt |
+| test_heuristic.rs:70:34:70:35 | 32 | test_heuristic.rs:70:34:70:35 | 32 | test_heuristic.rs:70:22:70:62 | ... ^ ... | This hard-coded value is used as $@. | test_heuristic.rs:70:22:70:62 | ... ^ ... | a salt |
+| test_heuristic.rs:70:52:70:61 | 0xFFFFFFFF | test_heuristic.rs:70:52:70:61 | 0xFFFFFFFF | test_heuristic.rs:70:22:70:62 | ... ^ ... | This hard-coded value is used as $@. | test_heuristic.rs:70:22:70:62 | ... ^ ... | a salt |
 edges
 | test_cipher.rs:18:9:18:14 | const1 [&ref] | test_cipher.rs:19:73:19:78 | const1 [&ref] | provenance |  |
 | test_cipher.rs:18:28:18:36 | &... [&ref] | test_cipher.rs:18:9:18:14 | const1 [&ref] | provenance |  |
 | test_cipher.rs:18:29:18:36 | [0u8; 16] | test_cipher.rs:18:28:18:36 | &... [&ref] | provenance |  |
 | test_cipher.rs:19:49:19:79 | ...::from_slice(...) [&ref] | test_cipher.rs:19:30:19:47 | ...::new | provenance | MaD:3 Sink:MaD:3 |
-| test_cipher.rs:19:73:19:78 | const1 [&ref] | test_cipher.rs:19:49:19:79 | ...::from_slice(...) [&ref] | provenance | MaD:13 |
+| test_cipher.rs:19:73:19:78 | const1 [&ref] | test_cipher.rs:19:49:19:79 | ...::from_slice(...) [&ref] | provenance | MaD:17 |
 | test_cipher.rs:25:9:25:14 | const4 [&ref] | test_cipher.rs:26:66:26:71 | const4 [&ref] | provenance |  |
 | test_cipher.rs:25:28:25:36 | &... [&ref] | test_cipher.rs:25:9:25:14 | const4 [&ref] | provenance |  |
 | test_cipher.rs:25:29:25:36 | [0u8; 16] | test_cipher.rs:25:28:25:36 | &... [&ref] | provenance |  |
 | test_cipher.rs:26:42:26:72 | ...::from_slice(...) [&ref] | test_cipher.rs:26:30:26:40 | ...::new | provenance | MaD:4 Sink:MaD:4 |
-| test_cipher.rs:26:66:26:71 | const4 [&ref] | test_cipher.rs:26:42:26:72 | ...::from_slice(...) [&ref] | provenance | MaD:13 |
+| test_cipher.rs:26:66:26:71 | const4 [&ref] | test_cipher.rs:26:42:26:72 | ...::from_slice(...) [&ref] | provenance | MaD:17 |
 | test_cipher.rs:29:9:29:14 | const5 [&ref] | test_cipher.rs:30:95:30:100 | const5 [&ref] | provenance |  |
 | test_cipher.rs:29:28:29:36 | &... [&ref] | test_cipher.rs:29:9:29:14 | const5 [&ref] | provenance |  |
 | test_cipher.rs:29:29:29:36 | [0u8; 16] | test_cipher.rs:29:28:29:36 | &... [&ref] | provenance |  |
 | test_cipher.rs:30:72:30:101 | ...::from_slice(...) [&ref] | test_cipher.rs:30:30:30:40 | ...::new | provenance | MaD:5 Sink:MaD:5 |
-| test_cipher.rs:30:95:30:100 | const5 [&ref] | test_cipher.rs:30:72:30:101 | ...::from_slice(...) [&ref] | provenance | MaD:13 |
+| test_cipher.rs:30:95:30:100 | const5 [&ref] | test_cipher.rs:30:72:30:101 | ...::from_slice(...) [&ref] | provenance | MaD:17 |
 | test_cipher.rs:37:9:37:14 | const7 | test_cipher.rs:38:74:38:79 | const7 | provenance |  |
 | test_cipher.rs:37:27:37:74 | [...] | test_cipher.rs:37:9:37:14 | const7 | provenance |  |
 | test_cipher.rs:38:49:38:80 | ...::from_slice(...) [&ref] | test_cipher.rs:38:30:38:47 | ...::new | provenance | MaD:3 Sink:MaD:3 |
-| test_cipher.rs:38:73:38:79 | &const7 [&ref] | test_cipher.rs:38:49:38:80 | ...::from_slice(...) [&ref] | provenance | MaD:13 |
+| test_cipher.rs:38:73:38:79 | &const7 [&ref] | test_cipher.rs:38:49:38:80 | ...::from_slice(...) [&ref] | provenance | MaD:17 |
 | test_cipher.rs:38:74:38:79 | const7 | test_cipher.rs:38:73:38:79 | &const7 [&ref] | provenance |  |
 | test_cipher.rs:41:9:41:14 | const8 [&ref] | test_cipher.rs:42:73:42:78 | const8 [&ref] | provenance |  |
 | test_cipher.rs:41:28:41:76 | &... [&ref] | test_cipher.rs:41:9:41:14 | const8 [&ref] | provenance |  |
 | test_cipher.rs:41:29:41:76 | [...] | test_cipher.rs:41:28:41:76 | &... [&ref] | provenance |  |
 | test_cipher.rs:42:49:42:79 | ...::from_slice(...) [&ref] | test_cipher.rs:42:30:42:47 | ...::new | provenance | MaD:3 Sink:MaD:3 |
-| test_cipher.rs:42:73:42:78 | const8 [&ref] | test_cipher.rs:42:49:42:79 | ...::from_slice(...) [&ref] | provenance | MaD:13 |
+| test_cipher.rs:42:73:42:78 | const8 [&ref] | test_cipher.rs:42:49:42:79 | ...::from_slice(...) [&ref] | provenance | MaD:17 |
 | test_cipher.rs:50:9:50:15 | const10 [element] | test_cipher.rs:51:75:51:81 | const10 [element] | provenance |  |
 | test_cipher.rs:50:37:50:52 | ...::zeroed | test_cipher.rs:50:37:50:54 | ...::zeroed(...) [element] | provenance | Src:MaD:7  |
 | test_cipher.rs:50:37:50:54 | ...::zeroed(...) [element] | test_cipher.rs:50:9:50:15 | const10 [element] | provenance |  |
 | test_cipher.rs:51:50:51:82 | ...::from_slice(...) [&ref, element] | test_cipher.rs:51:31:51:48 | ...::new | provenance | MaD:3 Sink:MaD:3 Sink:MaD:3 |
-| test_cipher.rs:51:74:51:81 | &const10 [&ref, element] | test_cipher.rs:51:50:51:82 | ...::from_slice(...) [&ref, element] | provenance | MaD:13 |
+| test_cipher.rs:51:74:51:81 | &const10 [&ref, element] | test_cipher.rs:51:50:51:82 | ...::from_slice(...) [&ref, element] | provenance | MaD:17 |
 | test_cipher.rs:51:75:51:81 | const10 [element] | test_cipher.rs:51:74:51:81 | &const10 [&ref, element] | provenance |  |
 | test_cipher.rs:73:9:73:14 | const2 [&ref] | test_cipher.rs:74:46:74:51 | const2 [&ref] | provenance |  |
 | test_cipher.rs:73:18:73:26 | &... [&ref] | test_cipher.rs:73:9:73:14 | const2 [&ref] | provenance |  |
@@ -69,27 +64,26 @@ edges
 | test_cookie.rs:22:27:22:32 | array2 | test_cookie.rs:22:26:22:32 | &array2 [&ref] | provenance |  |
 | test_cookie.rs:38:9:38:14 | array2 | test_cookie.rs:42:34:42:39 | array2 | provenance |  |
 | test_cookie.rs:38:18:38:37 | ...::from(...) | test_cookie.rs:38:9:38:14 | array2 | provenance |  |
-| test_cookie.rs:38:28:38:36 | [0u8; 64] | test_cookie.rs:38:18:38:37 | ...::from(...) | provenance | MaD:8 |
-| test_cookie.rs:38:28:38:36 | [0u8; 64] | test_cookie.rs:38:18:38:37 | ...::from(...) | provenance | MaD:9 |
-| test_cookie.rs:38:28:38:36 | [0u8; 64] | test_cookie.rs:38:18:38:37 | ...::from(...) | provenance | MaD:10 |
-| test_cookie.rs:38:28:38:36 | [0u8; 64] | test_cookie.rs:38:18:38:37 | ...::from(...) | provenance | MaD:11 |
 | test_cookie.rs:38:28:38:36 | [0u8; 64] | test_cookie.rs:38:18:38:37 | ...::from(...) | provenance | MaD:12 |
+| test_cookie.rs:38:28:38:36 | [0u8; 64] | test_cookie.rs:38:18:38:37 | ...::from(...) | provenance | MaD:13 |
+| test_cookie.rs:38:28:38:36 | [0u8; 64] | test_cookie.rs:38:18:38:37 | ...::from(...) | provenance | MaD:14 |
+| test_cookie.rs:38:28:38:36 | [0u8; 64] | test_cookie.rs:38:18:38:37 | ...::from(...) | provenance | MaD:15 |
+| test_cookie.rs:38:28:38:36 | [0u8; 64] | test_cookie.rs:38:18:38:37 | ...::from(...) | provenance | MaD:16 |
 | test_cookie.rs:42:34:42:39 | array2 | test_cookie.rs:42:14:42:32 | ...::from | provenance | MaD:2 Sink:MaD:2 |
 | test_cookie.rs:49:9:49:14 | array3 [element] | test_cookie.rs:53:34:53:39 | array3 [element] | provenance |  |
-| test_cookie.rs:49:23:49:25 | 0u8 | test_cookie.rs:49:23:49:29 | ...::from_elem(...) [element] | provenance | MaD:14 |
+| test_cookie.rs:49:23:49:25 | 0u8 | test_cookie.rs:49:23:49:29 | ...::from_elem(...) [element] | provenance | MaD:18 |
 | test_cookie.rs:49:23:49:29 | ...::from_elem(...) [element] | test_cookie.rs:49:9:49:14 | array3 [element] | provenance |  |
 | test_cookie.rs:53:34:53:39 | array3 [element] | test_cookie.rs:53:14:53:32 | ...::from | provenance | MaD:2 Sink:MaD:2 |
-| test_heuristic.rs:38:25:38:30 | 0xFFFF | test_heuristic.rs:81:22:81:31 | MY_CONST_1 | provenance |  |
-| test_heuristic.rs:39:25:39:59 | ... as u64 | test_heuristic.rs:82:22:82:31 | MY_CONST_2 | provenance |  |
-| test_heuristic.rs:39:62:40:33 | static MY_STATIC_3 | test_heuristic.rs:83:22:83:32 | MY_STATIC_3 | provenance |  |
-| test_heuristic.rs:40:27:40:32 | 0xFFFF | test_heuristic.rs:39:62:40:33 | static MY_STATIC_3 | provenance |  |
-| test_heuristic.rs:49:9:49:16 | const_iv [&ref] | test_heuristic.rs:50:41:50:48 | const_iv | provenance |  |
-| test_heuristic.rs:49:30:49:38 | &... [&ref] | test_heuristic.rs:49:9:49:16 | const_iv [&ref] | provenance |  |
-| test_heuristic.rs:49:31:49:38 | [0u8; 16] | test_heuristic.rs:49:30:49:38 | &... [&ref] | provenance |  |
-| test_heuristic.rs:69:20:69:27 | [0u8; 16] | test_heuristic.rs:69:19:69:27 | &... | provenance |  |
-| test_heuristic.rs:70:31:70:38 | [0u8; 16] | test_heuristic.rs:70:30:70:38 | &... | provenance |  |
-| test_heuristic.rs:86:29:86:32 | 1u64 | test_heuristic.rs:87:22:87:31 | MY_CONST_5 | provenance |  |
-| test_heuristic.rs:88:29:88:33 | ... + ... | test_heuristic.rs:89:22:89:31 | MY_CONST_6 | provenance |  |
+| test_heuristic.rs:44:9:44:16 | const_iv [&ref] | test_heuristic.rs:45:41:45:48 | const_iv | provenance |  |
+| test_heuristic.rs:44:30:44:38 | &... [&ref] | test_heuristic.rs:44:9:44:16 | const_iv [&ref] | provenance |  |
+| test_heuristic.rs:44:31:44:38 | [0u8; 16] | test_heuristic.rs:44:30:44:38 | &... [&ref] | provenance |  |
+| test_heuristic.rs:64:20:64:27 | [0u8; 16] | test_heuristic.rs:64:19:64:27 | &... | provenance |  |
+| test_heuristic.rs:65:31:65:38 | [0u8; 16] | test_heuristic.rs:65:30:65:38 | &... | provenance |  |
+| test_heuristic.rs:69:32:69:32 | 1 | test_heuristic.rs:69:22:69:32 | ... + ... | provenance | MaD:8 |
+| test_heuristic.rs:70:23:70:35 | ... << ... | test_heuristic.rs:70:22:70:62 | ... ^ ... | provenance | MaD:10 |
+| test_heuristic.rs:70:34:70:35 | 32 | test_heuristic.rs:70:23:70:35 | ... << ... | provenance | MaD:11 |
+| test_heuristic.rs:70:41:70:61 | ... & ... | test_heuristic.rs:70:22:70:62 | ... ^ ... | provenance | MaD:10 |
+| test_heuristic.rs:70:52:70:61 | 0xFFFFFFFF | test_heuristic.rs:70:41:70:61 | ... & ... | provenance | MaD:9 |
 models
 | 1 | Sink: <_ as crypto_common::KeyInit>::new_from_slice; Argument[0]; credentials-key |
 | 2 | Sink: <biscotti::crypto::master::Key>::from; Argument[0]; credentials-key |
@@ -98,13 +92,17 @@ models
 | 5 | Sink: <cipher::stream_wrapper::StreamCipherCoreWrapper as crypto_common::KeyIvInit>::new; Argument[1]; credentials-iv |
 | 6 | Sink: <cookie::secure::key::Key>::from; Argument[0].Reference; credentials-key |
 | 7 | Source: core::mem::zeroed; ReturnValue.Element; constant-source |
-| 8 | Summary: <alloc::vec::Vec as core::convert::From>::from; Argument[0].Field[alloc::borrow::Cow::Owned(0)]; ReturnValue; value |
-| 9 | Summary: <alloc::vec::Vec as core::convert::From>::from; Argument[0].Field[alloc::bstr::ByteString(0)]; ReturnValue; value |
-| 10 | Summary: <alloc::vec::Vec as core::convert::From>::from; Argument[0].Field[alloc::collections::binary_heap::BinaryHeap::data]; ReturnValue; value |
-| 11 | Summary: <alloc::vec::Vec as core::convert::From>::from; Argument[0].Field[alloc::string::String::vec]; ReturnValue; value |
-| 12 | Summary: <alloc::vec::Vec as core::convert::From>::from; Argument[0]; ReturnValue; taint |
-| 13 | Summary: <generic_array::GenericArray>::from_slice; Argument[0].Reference; ReturnValue.Reference; value |
-| 14 | Summary: alloc::vec::from_elem; Argument[0]; ReturnValue.Element; value |
+| 8 | Summary: <_ as core::ops::arith::Add>::add; Argument[self,0]; ReturnValue; taint |
+| 9 | Summary: <_ as core::ops::bit::BitAnd>::bitand; Argument[self,0]; ReturnValue; taint |
+| 10 | Summary: <_ as core::ops::bit::BitXor>::bitxor; Argument[self,0]; ReturnValue; taint |
+| 11 | Summary: <_ as core::ops::bit::Shl>::shl; Argument[self,0]; ReturnValue; taint |
+| 12 | Summary: <alloc::vec::Vec as core::convert::From>::from; Argument[0].Field[alloc::borrow::Cow::Owned(0)]; ReturnValue; value |
+| 13 | Summary: <alloc::vec::Vec as core::convert::From>::from; Argument[0].Field[alloc::bstr::ByteString(0)]; ReturnValue; value |
+| 14 | Summary: <alloc::vec::Vec as core::convert::From>::from; Argument[0].Field[alloc::collections::binary_heap::BinaryHeap::data]; ReturnValue; value |
+| 15 | Summary: <alloc::vec::Vec as core::convert::From>::from; Argument[0].Field[alloc::string::String::vec]; ReturnValue; value |
+| 16 | Summary: <alloc::vec::Vec as core::convert::From>::from; Argument[0]; ReturnValue; taint |
+| 17 | Summary: <generic_array::GenericArray>::from_slice; Argument[0].Reference; ReturnValue.Reference; value |
+| 18 | Summary: alloc::vec::from_elem; Argument[0]; ReturnValue.Element; value |
 nodes
 | test_cipher.rs:18:9:18:14 | const1 [&ref] | semmle.label | const1 [&ref] |
 | test_cipher.rs:18:28:18:36 | &... [&ref] | semmle.label | &... [&ref] |
@@ -168,28 +166,21 @@ nodes
 | test_cookie.rs:49:23:49:29 | ...::from_elem(...) [element] | semmle.label | ...::from_elem(...) [element] |
 | test_cookie.rs:53:14:53:32 | ...::from | semmle.label | ...::from |
 | test_cookie.rs:53:34:53:39 | array3 [element] | semmle.label | array3 [element] |
-| test_heuristic.rs:38:25:38:30 | 0xFFFF | semmle.label | 0xFFFF |
-| test_heuristic.rs:39:25:39:59 | ... as u64 | semmle.label | ... as u64 |
-| test_heuristic.rs:39:62:40:33 | static MY_STATIC_3 | semmle.label | static MY_STATIC_3 |
-| test_heuristic.rs:40:27:40:32 | 0xFFFF | semmle.label | 0xFFFF |
-| test_heuristic.rs:49:9:49:16 | const_iv [&ref] | semmle.label | const_iv [&ref] |
-| test_heuristic.rs:49:30:49:38 | &... [&ref] | semmle.label | &... [&ref] |
-| test_heuristic.rs:49:31:49:38 | [0u8; 16] | semmle.label | [0u8; 16] |
-| test_heuristic.rs:50:41:50:48 | const_iv | semmle.label | const_iv |
-| test_heuristic.rs:68:30:68:37 | "secret" | semmle.label | "secret" |
-| test_heuristic.rs:69:19:69:27 | &... | semmle.label | &... |
-| test_heuristic.rs:69:20:69:27 | [0u8; 16] | semmle.label | [0u8; 16] |
-| test_heuristic.rs:70:30:70:38 | &... | semmle.label | &... |
-| test_heuristic.rs:70:31:70:38 | [0u8; 16] | semmle.label | [0u8; 16] |
-| test_heuristic.rs:72:22:72:22 | 0 | semmle.label | 0 |
-| test_heuristic.rs:76:22:76:27 | ... << ... | semmle.label | ... << ... |
-| test_heuristic.rs:78:22:78:29 | ...::MAX | semmle.label | ...::MAX |
-| test_heuristic.rs:79:22:79:33 | ... / ... | semmle.label | ... / ... |
-| test_heuristic.rs:81:22:81:31 | MY_CONST_1 | semmle.label | MY_CONST_1 |
-| test_heuristic.rs:82:22:82:31 | MY_CONST_2 | semmle.label | MY_CONST_2 |
-| test_heuristic.rs:83:22:83:32 | MY_STATIC_3 | semmle.label | MY_STATIC_3 |
-| test_heuristic.rs:86:29:86:32 | 1u64 | semmle.label | 1u64 |
-| test_heuristic.rs:87:22:87:31 | MY_CONST_5 | semmle.label | MY_CONST_5 |
-| test_heuristic.rs:88:29:88:33 | ... + ... | semmle.label | ... + ... |
-| test_heuristic.rs:89:22:89:31 | MY_CONST_6 | semmle.label | MY_CONST_6 |
+| test_heuristic.rs:44:9:44:16 | const_iv [&ref] | semmle.label | const_iv [&ref] |
+| test_heuristic.rs:44:30:44:38 | &... [&ref] | semmle.label | &... [&ref] |
+| test_heuristic.rs:44:31:44:38 | [0u8; 16] | semmle.label | [0u8; 16] |
+| test_heuristic.rs:45:41:45:48 | const_iv | semmle.label | const_iv |
+| test_heuristic.rs:63:30:63:37 | "secret" | semmle.label | "secret" |
+| test_heuristic.rs:64:19:64:27 | &... | semmle.label | &... |
+| test_heuristic.rs:64:20:64:27 | [0u8; 16] | semmle.label | [0u8; 16] |
+| test_heuristic.rs:65:30:65:38 | &... | semmle.label | &... |
+| test_heuristic.rs:65:31:65:38 | [0u8; 16] | semmle.label | [0u8; 16] |
+| test_heuristic.rs:67:22:67:22 | 0 | semmle.label | 0 |
+| test_heuristic.rs:69:22:69:32 | ... + ... | semmle.label | ... + ... |
+| test_heuristic.rs:69:32:69:32 | 1 | semmle.label | 1 |
+| test_heuristic.rs:70:22:70:62 | ... ^ ... | semmle.label | ... ^ ... |
+| test_heuristic.rs:70:23:70:35 | ... << ... | semmle.label | ... << ... |
+| test_heuristic.rs:70:34:70:35 | 32 | semmle.label | 32 |
+| test_heuristic.rs:70:41:70:61 | ... & ... | semmle.label | ... & ... |
+| test_heuristic.rs:70:52:70:61 | 0xFFFFFFFF | semmle.label | 0xFFFFFFFF |
 subpaths

rust/ql/test/query-tests/security/CWE-798/test_heuristic.rs

@@ -35,11 +35,6 @@ impl MyCryptor {
     }
 }
 
-const MY_CONST_1: u64 = 0xFFFF; // $ Alert[rust/hard-coded-cryptographic-value]
-const MY_CONST_2: u64 = std::env::consts::ARCH.len() as u64; // $ Alert[rust/hard-coded-cryptographic-value]
-static MY_STATIC_3: u64 = 0xFFFF; // $ Alert[rust/hard-coded-cryptographic-value]
-static MY_STATIC_4: u64 = std::env::consts::ARCH.len() as u64;
-
 fn test(var_string: &str, var_data: &[u8;16], var_u64: u64) {
     encrypt_with("plaintext", var_data, var_data);
 
@@ -71,32 +66,6 @@ fn test(var_string: &str, var_data: &[u8;16], var_u64: u64) {
 
     mc2.set_salt_u64(0); // $ Alert[rust/hard-coded-cryptographic-value]
     mc2.set_salt_u64(var_u64);
-    mc2.set_salt_u64(var_u64 + 1);
-    mc2.set_salt_u64((var_u64 << 32) ^ (var_u64  & 0xFFFFFFFF));
-    mc2.set_salt_u64(1 << 4); // $ Alert[rust/hard-coded-cryptographic-value]
-
-    mc2.set_salt_u64(u64::MAX); // $ Alert[rust/hard-coded-cryptographic-value]
-    mc2.set_salt_u64(u64::MAX / 4); // $ Alert[rust/hard-coded-cryptographic-value]
-
-    mc2.set_salt_u64(MY_CONST_1); // $ Sink[rust/hard-coded-cryptographic-value]
-    mc2.set_salt_u64(MY_CONST_2); // $ Sink[rust/hard-coded-cryptographic-value]
-    mc2.set_salt_u64(MY_STATIC_3); // $ Sink[rust/hard-coded-cryptographic-value]
-    mc2.set_salt_u64(MY_STATIC_4);
-
-    const MY_CONST_5: u64 = 1u64; // $ Alert[rust/hard-coded-cryptographic-value]
-    mc2.set_salt_u64(MY_CONST_5); // $ Sink[rust/hard-coded-cryptographic-value]
-    const MY_CONST_6: u64 = 2 + 3; // $ Alert[rust/hard-coded-cryptographic-value]
-    mc2.set_salt_u64(MY_CONST_6); // $ Sink[rust/hard-coded-cryptographic-value]
-
-    let mut key1 = "foo".to_string(); // $ MISSING: Alert[rust/hard-coded-cryptographic-value]
-    key1 += "bar"; // $ MISSING: Alert[rust/hard-coded-cryptographic-value]
-    let _ = MyCryptor::new(&key1);
-
-    let mut key2 = "foo".to_string();
-    key2 += var_string;
-    let _ = MyCryptor::new(&key2);
-
-    let mut key3 = var_string.to_string();
-    key3 += "bar";
-    let _ = MyCryptor::new(&key3);
+    mc2.set_salt_u64(var_u64 + 1); // $ SPURIOUS: Alert[rust/hard-coded-cryptographic-value]
+    mc2.set_salt_u64((var_u64 << 32) ^ (var_u64  & 0xFFFFFFFF)); // $ SPURIOUS: Alert[rust/hard-coded-cryptographic-value]
 }

shared/canonicalize/BUILD.bazel

@@ -0,0 +1,35 @@
+load("@rules_cc//cc:defs.bzl", "cc_binary", "cc_library")
+load("@rules_pkg//pkg:mappings.bzl", "pkg_attributes", "pkg_files")
+
+cc_binary(
+    name = "codeql_canonical_path.dll",
+    srcs = [
+        "canonicalize.cpp",
+        "canonicalize.h",
+        "canonicalize_jni.cpp",
+    ],
+    defines = ["CODEQL_CANONICALIZE_EXPORTS"],
+    linkopts = ["-lkernel32"],
+    linkshared = True,
+    target_compatible_with = ["@platforms//os:windows"],
+    visibility = ["//visibility:public"],
+    deps = ["@rules_java//toolchains:jni"],
+)
+
+cc_library(
+    name = "canonicalize",
+    srcs = ["canonicalize.cpp"],
+    hdrs = ["canonicalize.h"],
+    defines = ["CODEQL_CANONICALIZE_EXPORTS"],
+    linkopts = ["-lkernel32"],
+    target_compatible_with = ["@platforms//os:windows"],
+    visibility = ["//visibility:public"],
+)
+
+pkg_files(
+    name = "pkg",
+    srcs = [":codeql_canonical_path.dll"],
+    attributes = pkg_attributes(mode = "0755"),
+    target_compatible_with = ["@platforms//os:windows"],
+    visibility = ["//visibility:public"],
+)

shared/canonicalize/canonicalize.cpp

@@ -0,0 +1,165 @@
+#ifdef _WIN32
+#include "canonicalize.h"
+#include <windows.h>
+#include <string>
+#include <unordered_map>
+#include <shared_mutex>
+#include <random>
+
+namespace {
+
+class PathCache {
+public:
+    static PathCache& instance() {
+        static PathCache cache;
+        return cache;
+    }
+
+    const wchar_t* canonicalize(const wchar_t* path) {
+        std::wstring key(path);
+
+        // Fast path: shared (read) lock for cache hit
+        {
+            std::shared_lock lock(mutex_);
+            auto it = cache_.find(key);
+            if (it != cache_.end()) {
+                return _wcsdup(it->second.c_str());
+            }
+        }
+
+        // Slow path: resolve and insert under exclusive lock
+        std::wstring resolved = resolve(path);
+        if (resolved.empty()) return nullptr;
+
+        std::unique_lock lock(mutex_);
+        // Check again under exclusive lock (another thread may have inserted)
+        auto it = cache_.find(key);
+        if (it != cache_.end()) {
+            return _wcsdup(it->second.c_str());
+        }
+
+        // Evict a random entry if at capacity (matches C# strategy)
+        if (cache_.size() >= max_capacity_) {
+            std::uniform_int_distribution<size_t> dist(0, cache_.size() - 1);
+            auto evict = cache_.begin();
+            std::advance(evict, dist(rng_));
+            cache_.erase(evict);
+        }
+
+        auto inserted = cache_.emplace(std::move(key), std::move(resolved)).first;
+        return _wcsdup(inserted->second.c_str());
+    }
+
+private:
+    PathCache() = default;
+
+    static constexpr size_t max_capacity_ = 4096;
+    std::unordered_map<std::wstring, std::wstring> cache_;
+    std::shared_mutex mutex_;
+    std::mt19937 rng_{std::random_device{}()};
+
+    static std::wstring resolve(const wchar_t* path) {
+        HANDLE h = CreateFileW(
+            path,
+            0,
+            FILE_SHARE_READ | FILE_SHARE_WRITE | FILE_SHARE_DELETE,
+            nullptr,
+            OPEN_EXISTING,
+            FILE_FLAG_BACKUP_SEMANTICS,
+            nullptr);
+
+        if (h == INVALID_HANDLE_VALUE) {
+            return resolve_nonexistent(path);
+        }
+
+        std::wstring result = get_final_path(h);
+        CloseHandle(h);
+
+        if (result.empty()) return {};
+        return strip_prefix(result);
+    }
+
+    static std::wstring get_final_path(HANDLE h) {
+        wchar_t buf[MAX_PATH];
+        DWORD len = GetFinalPathNameByHandleW(h, buf, MAX_PATH, FILE_NAME_NORMALIZED);
+
+        if (len > 0 && len < MAX_PATH) {
+            return std::wstring(buf, len);
+        }
+        if (len >= MAX_PATH) {
+            std::wstring big(len + 1, L'\0');
+            len = GetFinalPathNameByHandleW(h, big.data(), len + 1, FILE_NAME_NORMALIZED);
+            if (len > 0) return std::wstring(big.data(), len);
+        }
+        return {};
+    }
+
+    static std::wstring strip_prefix(const std::wstring& path) {
+        constexpr std::wstring_view unc_prefix = L"\\\\?\\UNC\\";
+        constexpr std::wstring_view lp_prefix = L"\\\\?\\";
+
+        if (path.starts_with(unc_prefix)) {
+            return L"\\" + path.substr(unc_prefix.size() - 1);
+        }
+        if (path.starts_with(lp_prefix)) {
+            return std::wstring(path.substr(lp_prefix.size()));
+        }
+        return path;
+    }
+
+    // For non-existent files: canonicalize parent, append filename
+    // (matches C#'s ConstructCanonicalPath)
+    static std::wstring resolve_nonexistent(const wchar_t* path) {
+        std::wstring spath(path);
+        auto sep = spath.find_last_of(L"\\/");
+        if (sep == std::wstring::npos) return {};
+
+        std::wstring parent = spath.substr(0, sep);
+        std::wstring name = spath.substr(sep + 1);
+
+        std::wstring canonical_parent = resolve(parent.c_str());
+        if (canonical_parent.empty()) return {};
+
+        return canonical_parent + L"\\" + name;
+    }
+};
+
+} // namespace
+
+extern "C" {
+
+CODEQL_API const wchar_t* canonicalize_path_w(const wchar_t* path) {
+    if (!path || !*path) return nullptr;
+    return PathCache::instance().canonicalize(path);
+}
+
+CODEQL_API void canonicalize_free_w(const wchar_t* path) {
+    free(const_cast<wchar_t*>(path));
+}
+
+CODEQL_API const char* canonicalize_path_u8(const char* path) {
+    if (!path || !*path) return nullptr;
+
+    int wlen = MultiByteToWideChar(CP_UTF8, 0, path, -1, nullptr, 0);
+    if (wlen <= 0) return nullptr;
+    std::wstring wpath(wlen - 1, L'\0');
+    MultiByteToWideChar(CP_UTF8, 0, path, -1, wpath.data(), wlen);
+
+    const wchar_t* wresult = PathCache::instance().canonicalize(wpath.c_str());
+    if (!wresult) return nullptr;
+
+    int ulen = WideCharToMultiByte(CP_UTF8, 0, wresult, -1, nullptr, 0, nullptr, nullptr);
+    if (ulen <= 0) { free(const_cast<wchar_t*>(wresult)); return nullptr; }
+    char* result = static_cast<char*>(malloc(ulen));
+    WideCharToMultiByte(CP_UTF8, 0, wresult, -1, result, ulen, nullptr, nullptr);
+    free(const_cast<wchar_t*>(wresult));
+
+    return result;
+}
+
+CODEQL_API void canonicalize_free_u8(const char* path) {
+    free(const_cast<char*>(path));
+}
+
+} // extern "C"
+#endif

shared/canonicalize/canonicalize.h

@@ -0,0 +1,31 @@
+#ifndef CODEQL_CANONICALIZE_H
+#define CODEQL_CANONICALIZE_H
+
+#ifdef _WIN32
+
+#ifdef CODEQL_CANONICALIZE_EXPORTS
+#define CODEQL_API __declspec(dllexport)
+#else
+#define CODEQL_API __declspec(dllimport)
+#endif
+
+#include <wchar.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+// UTF-16 interface (for JNI / Java / Kotlin)
+CODEQL_API const wchar_t* canonicalize_path_w(const wchar_t* path);
+CODEQL_API void canonicalize_free_w(const wchar_t* path);
+
+// UTF-8 interface (for Go)
+CODEQL_API const char* canonicalize_path_u8(const char* path);
+CODEQL_API void canonicalize_free_u8(const char* path);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif // _WIN32
+#endif // CODEQL_CANONICALIZE_H

shared/canonicalize/canonicalize_jni.cpp

@@ -0,0 +1,25 @@
+#ifdef _WIN32
+#include <jni.h>
+#include "canonicalize.h"
+
+extern "C" {
+
+JNIEXPORT jstring JNICALL
+Java_com_semmle_util_files_NativeCanonicalizer_nativeCanonicalizePath(
+    JNIEnv *env, jclass cls, jstring jpath) {
+
+    const jchar* path = env->GetStringChars(jpath, nullptr);
+    const wchar_t* result = canonicalize_path_w(reinterpret_cast<const wchar_t*>(path));
+    env->ReleaseStringChars(jpath, path);
+
+    if (result == nullptr) return nullptr;
+
+    jstring jresult = env->NewString(
+        reinterpret_cast<const jchar*>(result),
+        static_cast<jsize>(wcslen(result)));
+    canonicalize_free_w(result);
+    return jresult;
+}
+
+} // extern "C"
+#endif

shared/dataflow/codeql/dataflow/internal/FlowSummaryImpl.qll

@@ -54,24 +54,6 @@ signature module InputSig<LocationSig Location, DF::InputSig<Location> Lang> {
     none()
   }
 
-  /**
-   * A base class of calls that are candidates for flow summary modeling.
-   */
-  class FlowSummaryCallBase {
-    string toString();
-  }
-
-  /** Gets a call that targets summarized callable `sc`. */
-  default FlowSummaryCallBase getASourceCall(SummarizedCallableBase sc) { none() }
-
-  /** Gets the callable corresponding to summarized callable `c`. */
-  default Lang::DataFlowCallable getSummarizedCallableAsDataFlowCallable(SummarizedCallableBase c) {
-    none()
-  }
-
-  /** Gets the enclosing callable of `call`. */
-  default Lang::DataFlowCallable getSourceCallEnclosingCallable(FlowSummaryCallBase call) { none() }
-
   /** Gets the parameter position representing a callback itself, if any. */
   default Lang::ArgumentPosition callbackSelfParameterPosition() { none() }
 
@@ -92,9 +74,6 @@ signature module InputSig<LocationSig Location, DF::InputSig<Location> Lang> {
     result = getStandardReturnValueKind()
   }
 
-  /** Gets the parameter position corresponding to a flow-summary return kind `rk`, if any. */
-  default Lang::ParameterPosition getFlowSummaryParameterPosition(Lang::ReturnKind rk) { none() }
-
   /** Gets the textual representation of parameter position `pos` used in MaD. */
   string encodeParameterPosition(Lang::ParameterPosition pos);
 
@@ -681,10 +660,6 @@ module Make<
       s.length() = 1 and
       s.head() instanceof TArgumentSummaryComponent
       or
-      // ReturnValue.*
-      s.length() = 1 and
-      s.head() instanceof TReturnSummaryComponent
-      or
       // Argument[n].ReturnValue.*
       s.length() = 2 and
       s.head() instanceof TReturnSummaryComponent and
@@ -1162,13 +1137,6 @@ module Make<
       outputState(c, s) and s = SummaryComponentStack::argument(_)
     }
 
-    private predicate relevantFlowSummaryPosition(SummarizedCallable c, ReturnKind rk) {
-      exists(SummaryComponentStack input |
-        summary(c, input, _, _, _) and
-        input = TSingletonSummaryComponentStack(TReturnSummaryComponent(rk))
-      )
-    }
-
     pragma[nomagic]
     private predicate sourceOutputStateEntry(
       SourceElement source, SummaryComponentStack s, string kind, string model
@@ -1304,12 +1272,6 @@ module Make<
       TSummaryParameterNode(SummarizedCallable c, ParameterPosition pos) {
         summaryParameterNodeRange(c, pos)
       } or
-      TSummaryReturnArgumentNode(FlowSummaryCallBase call, ReturnKind rk) {
-        exists(SummarizedCallable sc |
-          call = getASourceCall(sc) and
-          relevantFlowSummaryPosition(sc, rk)
-        )
-      } or
       TSourceOutputNode(SourceElement source, SummaryNodeState state, string kind, string model) {
         state.isSourceOutputState(source, _, kind, model)
       } or
@@ -1359,40 +1321,6 @@ module Make<
       override SinkElement getSinkElement() { none() }
     }
 
-    private class SummaryReturnArgumentNode extends SummaryNode, TSummaryReturnArgumentNode {
-      private FlowSummaryCallBase call;
-      private ReturnKind rk;
-
-      SummaryReturnArgumentNode() { this = TSummaryReturnArgumentNode(call, rk) }
-
-      override string toString() { result = "[summary] value written to " + rk + " at " + call }
-
-      override SummarizedCallable getSummarizedCallable() { none() }
-
-      override SourceElement getSourceElement() { none() }
-
-      override SinkElement getSinkElement() { none() }
-    }
-
-    /**
-     * Gets the summary node that represents the argument node used to transfer
-     * flow into the caller when a value is written to the value returned by
-     * `call` with kind `rk`.
-     */
-    SummaryNode summaryArgumentNode(FlowSummaryCallBase call, ReturnKind rk) {
-      result = TSummaryReturnArgumentNode(call, rk)
-    }
-
-    /** Gets the enclosing callable for summary node `sn`. */
-    DataFlowCallable getEnclosingCallable(SummaryNode sn) {
-      result = getSummarizedCallableAsDataFlowCallable(sn.getSummarizedCallable())
-      or
-      exists(FlowSummaryCallBase call |
-        sn = TSummaryReturnArgumentNode(call, _) and
-        result = getSourceCallEnclosingCallable(call)
-      )
-    }
-
     class SourceOutputNode extends SummaryNode, TSourceOutputNode {
       private SourceElement source_;
       private SummaryNodeState state_;
@@ -1499,12 +1427,6 @@ module Make<
       SummarizedCallable c, SummaryNodeState state, ParameterPosition pos
     ) {
       state.isInputState(c, SummaryComponentStack::argument(pos))
-      or
-      exists(ReturnKind rk |
-        relevantFlowSummaryPosition(c, rk) and
-        state.isInputState(c, SummaryComponentStack::return(rk)) and
-        pos = getFlowSummaryParameterPosition(rk)
-      )
     }
 
     /**
@@ -1638,9 +1560,6 @@ module Make<
       )
     }
 
-    /** Holds if return kind `rk` is a relevant return kind for flow summary modeling. */
-    predicate relevantFlowSummaryPosition(ReturnKind rk) { relevantFlowSummaryPosition(_, rk) }
-
     /**
      * Holds if flow is allowed to pass from the parameter at position `pos` of `c`,
      * to a return node, and back out to the parameter.
@@ -1817,15 +1736,9 @@ module Make<
     }
 
     signature module StepsInputSig {
-      /** Gets the summary node represented by data-flow node `n`, if any. */
-      SummaryNode getSummaryNode(Node n);
-
       /** Gets a call that targets summarized callable `sc`. */
       DataFlowCall getACall(SummarizedCallable sc);
 
-      /** Gets the out node of kind `rk` for `call`, if any. */
-      default Node getSourceOutNode(FlowSummaryCallBase call, ReturnKind rk) { none() }
-
       /** Gets the enclosing callable of `source`. */
       DataFlowCallable getSourceNodeEnclosingCallable(SourceBase source);
 
@@ -1852,7 +1765,7 @@ module Make<
        * Holds if there is a local step from `pred` to `succ`, which is synthesized
        * from a flow summary.
        */
-      private predicate summaryLocalStepImpl(
+      predicate summaryLocalStep(
         SummaryNode pred, SummaryNode succ, boolean preservesValue, string model
       ) {
         exists(
@@ -1898,24 +1811,9 @@ module Make<
         )
       }
 
-      /** Holds if there is a local step between data-flow nodes synthesized from a flow summary. */
-      predicate summaryLocalStep(Node pred, SummaryNode succ, boolean preservesValue, string model) {
-        exists(SummaryNode predSummary |
-          predSummary = StepsInput::getSummaryNode(pred) and
-          summaryLocalStepImpl(predSummary, succ, preservesValue, model)
-        )
-        or
-        exists(FlowSummaryCallBase summaryCall, ReturnKind rk, SummarizedCallable sc |
-          pred = StepsInput::getSourceOutNode(summaryCall, rk) and
-          summaryCall = getASourceCall(sc) and
-          summary(sc, SummaryComponentStack::return(rk), _, preservesValue, model) and
-          succ = TSummaryReturnArgumentNode(summaryCall, rk)
-        )
-      }
-
       /** Holds if the value of `succ` is uniquely determined by the value of `pred`. */
       predicate summaryLocalMustFlowStep(SummaryNode pred, SummaryNode succ) {
-        pred = unique(SummaryNode n1 | summaryLocalStepImpl(n1, succ, true, _))
+        pred = unique(SummaryNode n1 | summaryLocalStep(n1, succ, true, _))
       }
 
       /**
@@ -2037,7 +1935,7 @@ module Make<
         or
         exists(SummaryNode mid, boolean clearsOrExpectsMid |
           paramReachesLocal(p, mid, clearsOrExpectsMid) and
-          summaryLocalStepImpl(mid, n, true, _) and
+          summaryLocalStep(mid, n, true, _) and
           if
             summaryClearsContent(n, _) or
             summaryExpectsContent(n, _)
@@ -2095,7 +1993,7 @@ module Make<
        */
       predicate summaryThroughStepValue(ArgNode arg, Node out, SummarizedCallable sc) {
         exists(SummaryNode ret |
-          summaryLocalStepImpl(summaryArgParamRetOut(arg, ret, out, sc), ret, true, _)
+          summaryLocalStep(summaryArgParamRetOut(arg, ret, out, sc), ret, true, _)
         )
       }
 
@@ -2108,7 +2006,7 @@ module Make<
        */
       predicate summaryThroughStepTaint(ArgNode arg, Node out, SummarizedCallable sc) {
         exists(SummaryNode ret |
-          summaryLocalStepImpl(summaryArgParamRetOut(arg, ret, out, sc), ret, false, _)
+          summaryLocalStep(summaryArgParamRetOut(arg, ret, out, sc), ret, false, _)
         )
       }
 
@@ -2122,7 +2020,7 @@ module Make<
       predicate summaryGetterStep(ArgNode arg, ContentSet c, Node out, SummarizedCallable sc) {
         exists(SummaryNode mid, SummaryNode ret |
           summaryReadStep(summaryArgParamRetOut(arg, ret, out, sc), c, mid) and
-          summaryLocalStepImpl(mid, ret, _, _)
+          summaryLocalStep(mid, ret, _, _)
         )
       }
 
@@ -2135,7 +2033,7 @@ module Make<
        */
       predicate summarySetterStep(ArgNode arg, ContentSet c, Node out, SummarizedCallable sc) {
         exists(SummaryNode mid, SummaryNode ret |
-          summaryLocalStepImpl(summaryArgParamRetOut(arg, ret, out, sc), mid, _, _) and
+          summaryLocalStep(summaryArgParamRetOut(arg, ret, out, sc), mid, _, _) and
           summaryStoreStep(mid, c, ret)
         )
       }
@@ -2851,11 +2749,9 @@ module Make<
         key = "semmle.label" and val = n.toString()
       }
 
-      private Node getNode(SummaryNode sn) { sn = StepsInput::getSummaryNode(result) }
-
       private predicate edgesComponent(NodeOrCall a, NodeOrCall b, string value) {
         exists(boolean preservesValue |
-          PrivateSteps::summaryLocalStep(getNode(a.asNode()), b.asNode(), preservesValue, _) and
+          PrivateSteps::summaryLocalStep(a.asNode(), b.asNode(), preservesValue, _) and
           if preservesValue = true then value = "value" else value = "taint"
         )
         or

shared/typeinference/codeql/typeinference/internal/TypeInference.qll

@@ -1346,14 +1346,6 @@ module Make1<LocationSig Location, InputSig1<Location> Input1> {
     module MatchingWithEnvironment<MatchingWithEnvironmentInputSig Input> {
       private import Input
 
-      pragma[nomagic]
-      private TypeParameter getDeclTypeParameter(Declaration decl, TypeArgumentPosition tapos) {
-        exists(TypeParameterPosition tppos |
-          result = decl.getTypeParameter(tppos) and
-          typeArgumentParameterPositionMatch(tapos, tppos)
-        )
-      }
-
       /**
        * Gets the type of the type argument at `path` in `a` that corresponds to
        * the type parameter `tp` in `target`, if any.
@@ -1364,11 +1356,11 @@ module Make1<LocationSig Location, InputSig1<Location> Input1> {
        */
       bindingset[a, target]
       pragma[inline_late]
-      Type getTypeArgument(Access a, Declaration target, TypeParameter tp, TypePath path) {
-        exists(TypeArgumentPosition tapos |
+      private Type getTypeArgument(Access a, Declaration target, TypeParameter tp, TypePath path) {
+        exists(TypeArgumentPosition tapos, TypeParameterPosition tppos |
           result = a.getTypeArgument(tapos, path) and
-          tp = getDeclTypeParameter(target, tapos) and
-          not isPseudoType(result)
+          tp = target.getTypeParameter(tppos) and
+          typeArgumentParameterPositionMatch(tapos, tppos)
         )
       }
 
@@ -1534,35 +1526,42 @@ module Make1<LocationSig Location, InputSig1<Location> Input1> {
 
       private module AccessConstraint {
         private predicate relevantAccessConstraint(
-          Access a, AccessEnvironment e, Declaration target, TypeParameter constrainedTp,
+          Access a, AccessEnvironment e, Declaration target, AccessPosition apos, TypePath path,
           TypeMention constraint
         ) {
           target = a.getTarget(e) and
-          typeParameterHasConstraint(target, constrainedTp, constraint)
+          typeParameterHasConstraint(target, apos, _, path, constraint)
         }
 
         private newtype TRelevantAccess =
-          MkRelevantAccess(Access a, AccessEnvironment e, TypeParameter constrainedTp) {
-            relevantAccessConstraint(a, e, _, constrainedTp, _)
+          MkRelevantAccess(Access a, AccessPosition apos, AccessEnvironment e, TypePath path) {
+            relevantAccessConstraint(a, e, _, apos, path, _)
           }
 
+        /**
+         * If the access `a` for `apos`, environment `e`, and `path` has an inferred type
+         * which type inference requires to satisfy some constraint.
+         */
         private class RelevantAccess extends MkRelevantAccess {
           Access a;
+          AccessPosition apos;
           AccessEnvironment e;
-          TypeParameter constrainedTp;
+          TypePath path;
 
-          RelevantAccess() { this = MkRelevantAccess(a, e, constrainedTp) }
+          RelevantAccess() { this = MkRelevantAccess(a, apos, e, path) }
 
           pragma[nomagic]
-          Type getTypeAt(TypePath path) { typeMatch(a, e, _, path, result, constrainedTp) }
+          Type getTypeAt(TypePath suffix) {
+            result = a.getInferredType(e, apos, path.appendInverse(suffix))
+          }
 
           /** Gets the constraint that this relevant access should satisfy. */
           TypeMention getConstraint(Declaration target) {
-            relevantAccessConstraint(a, e, target, constrainedTp, result)
+            relevantAccessConstraint(a, e, target, apos, path, result)
           }
 
           string toString() {
-            result = a.toString() + ", " + e.toString() + ", " + constrainedTp.toString()
+            result = a.toString() + ", " + apos.toString() + ", " + path.toString()
           }
 
           Location getLocation() { result = a.getLocation() }
@@ -1578,7 +1577,7 @@ module Make1<LocationSig Location, InputSig1<Location> Input1> {
           class TypeMatchingContext = Access;
 
           TypeMatchingContext getTypeMatchingContext(RelevantAccess at) {
-            at = MkRelevantAccess(result, _, _)
+            at = MkRelevantAccess(result, _, _, _)
           }
 
           pragma[nomagic]
@@ -1592,32 +1591,41 @@ module Make1<LocationSig Location, InputSig1<Location> Input1> {
             SatisfiesTypeParameterConstraintInput>;
 
         pragma[nomagic]
-        predicate argSatisfiesConstraintAtTypeParameter(
+        predicate satisfiesConstraintAtTypeParameter(
           Access a, AccessEnvironment e, Declaration target, AccessPosition apos, TypePath prefix,
           TypeMention constraint, TypePath pathToTypeParamInConstraint,
           TypePath pathToTypeParamInSub
         ) {
-          exists(RelevantAccess ra, TypeParameter constrainedTp |
-            ra = MkRelevantAccess(a, e, constrainedTp) and
-            relevantAccessConstraint(a, e, target, constrainedTp, constraint) and
+          exists(RelevantAccess ra |
+            ra = MkRelevantAccess(a, apos, e, prefix) and
             SatisfiesTypeParameterConstraint::satisfiesConstraintAtTypeParameter(ra, constraint,
               pathToTypeParamInConstraint, pathToTypeParamInSub) and
-            exists(DeclarationPosition dpos |
-              accessDeclarationPositionMatch(apos, dpos) and
-              constrainedTp = target.getDeclaredType(dpos, prefix)
-            )
+            constraint = ra.getConstraint(target)
           )
         }
 
         pragma[nomagic]
         predicate satisfiesConstraint(
-          Access a, AccessEnvironment e, Declaration target, TypeParameter constrainedTp,
+          Access a, AccessEnvironment e, Declaration target, AccessPosition apos, TypePath prefix,
           TypeMention constraint, TypePath path, Type t
         ) {
           exists(RelevantAccess ra |
-            ra = MkRelevantAccess(a, e, constrainedTp) and
-            relevantAccessConstraint(a, e, target, constrainedTp, constraint) and
-            SatisfiesTypeParameterConstraint::satisfiesConstraint(ra, constraint, path, t)
+            ra = MkRelevantAccess(a, apos, e, prefix) and
+            SatisfiesTypeParameterConstraint::satisfiesConstraint(ra, constraint, path, t) and
+            constraint = ra.getConstraint(target)
+          )
+        }
+
+        pragma[nomagic]
+        predicate satisfiesConstraintThrough(
+          Access a, AccessEnvironment e, Declaration target, AccessPosition apos, TypePath prefix,
+          TypeAbstraction abs, TypeMention constraint, TypePath path, Type t
+        ) {
+          exists(RelevantAccess ra |
+            ra = MkRelevantAccess(a, apos, e, prefix) and
+            SatisfiesTypeParameterConstraint::satisfiesConstraintThrough(ra, abs, constraint, path,
+              t) and
+            constraint = ra.getConstraint(target)
           )
         }
       }
@@ -1636,44 +1644,51 @@ module Make1<LocationSig Location, InputSig1<Location> Input1> {
       }
 
       /**
-       * Holds if the type parameter `constrainedTp` applies to `target` and the
-       * constraint `constraint` applies to `constrainedTp`.
+       * Holds if the type parameter `constrainedTp` occurs in the declared type of
+       * `target` at `apos` and `pathToConstrained`, and there is a constraint
+       * `constraint` on `constrainedTp`.
        */
       pragma[nomagic]
       private predicate typeParameterHasConstraint(
-        Declaration target, TypeParameter constrainedTp, TypeMention constraint
+        Declaration target, AccessPosition apos, TypeParameter constrainedTp,
+        TypePath pathToConstrained, TypeMention constraint
       ) {
-        constrainedTp = target.getTypeParameter(_) and
-        constraint = getATypeParameterConstraint(constrainedTp, target)
+        exists(DeclarationPosition dpos |
+          accessDeclarationPositionMatch(apos, dpos) and
+          constrainedTp = target.getTypeParameter(_) and
+          constrainedTp = target.getDeclaredType(dpos, pathToConstrained) and
+          constraint = getATypeParameterConstraint(constrainedTp, target)
+        )
       }
 
       /**
-       * Holds if the type parameter `constrainedTp` applies to `target`, the
-       * constraint `constraint` applies to `constrainedTp`, and type parameter
-       * `tp` occurs at `pathToTp` in `constraint`.
+       * Holds if the declared type of `target` contains a type parameter at
+       * `apos` and `pathToConstrained` that must satisfy `constraint` and `tp`
+       * occurs at `pathToTp` in `constraint`.
        *
        * For example, in
-       *
        * ```csharp
        * interface IFoo<A> { }
        * T1 M<T1, T2>(T2 item) where T2 : IFoo<T1> { }
        * ```
-       *
-       * with the method declaration being the target, we have the following
-       * - `constrainedTp = T2`,
+       * with the method declaration being the target and with `apos`
+       * corresponding to `item`, we have the following
+       * - `pathToConstrained = ""`,
+       * - `tp = T1`,
        * - `constraint = IFoo`,
-       * - `tp = T1`, and
        * - `pathToTp = "A"`.
        */
       pragma[nomagic]
       private predicate typeParameterConstraintHasTypeParameter(
-        Declaration target, TypeParameter constrainedTp, TypeMention constraint, TypePath pathToTp,
-        TypeParameter tp
+        Declaration target, AccessPosition apos, TypePath pathToConstrained, TypeMention constraint,
+        TypePath pathToTp, TypeParameter tp
       ) {
-        typeParameterHasConstraint(target, constrainedTp, constraint) and
-        tp = target.getTypeParameter(_) and
-        tp = constraint.getTypeAt(pathToTp) and
-        constrainedTp != tp
+        exists(TypeParameter constrainedTp |
+          typeParameterHasConstraint(target, apos, constrainedTp, pathToConstrained, constraint) and
+          tp = target.getTypeParameter(_) and
+          tp = constraint.getTypeAt(pathToTp) and
+          constrainedTp != tp
+        )
       }
 
       pragma[nomagic]
@@ -1681,9 +1696,9 @@ module Make1<LocationSig Location, InputSig1<Location> Input1> {
         Access a, AccessEnvironment e, Declaration target, TypePath path, Type t, TypeParameter tp
       ) {
         not exists(getTypeArgument(a, target, tp, _)) and
-        exists(TypeMention constraint, TypeParameter constrainedTp, TypePath pathToTp |
-          typeParameterConstraintHasTypeParameter(target, constrainedTp, constraint, pathToTp, tp) and
-          AccessConstraint::satisfiesConstraint(a, e, target, constrainedTp, constraint,
+        exists(TypeMention constraint, AccessPosition apos, TypePath pathToTp, TypePath pathToTp2 |
+          typeParameterConstraintHasTypeParameter(target, apos, pathToTp2, constraint, pathToTp, tp) and
+          AccessConstraint::satisfiesConstraint(a, e, target, apos, pathToTp2, constraint,
             pathToTp.appendInverse(path), t)
         )
       }
@@ -1766,7 +1781,7 @@ module Make1<LocationSig Location, InputSig1<Location> Input1> {
           Declaration target, TypePath prefix, TypeMention constraint,
           TypePath pathToTypeParamInConstraint, TypePath pathToTypeParamInSub
         |
-          AccessConstraint::argSatisfiesConstraintAtTypeParameter(a, e, target, apos, prefix,
+          AccessConstraint::satisfiesConstraintAtTypeParameter(a, e, target, apos, prefix,
             constraint, pathToTypeParamInConstraint, pathToTypeParamInSub)
         |
           exists(TypePath suffix |
@@ -1832,12 +1847,7 @@ module Make1<LocationSig Location, InputSig1<Location> Input1> {
              */
 
             typeMatch(a, e, target, suffix, result, tp) and
-            exists(TypeParameter constrainedTp, DeclarationPosition dpos |
-              typeParameterConstraintHasTypeParameter(target, constrainedTp, constraint, pathToTp,
-                tp) and
-              accessDeclarationPositionMatch(apos, dpos) and
-              constrainedTp = target.getDeclaredType(dpos, _)
-            ) and
+            typeParameterConstraintHasTypeParameter(target, apos, _, constraint, pathToTp, tp) and
             pathToTp = pathToTypeParamInConstraint.appendInverse(mid) and
             path = prefix.append(pathToTypeParamInSub.append(mid).append(suffix))
           )

swift/ql/lib/codeql/swift/dataflow/internal/DataFlowPrivate.qll

@@ -305,7 +305,7 @@ private module Cached {
     model = ""
     or
     // flow through a flow summary (extension of `SummaryModelCsv`)
-    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom,
+    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom.(FlowSummaryNode).getSummaryNode(),
       nodeTo.(FlowSummaryNode).getSummaryNode(), true, model)
   }
 

swift/ql/lib/codeql/swift/dataflow/internal/FlowSummaryImpl.qll

@@ -20,8 +20,6 @@ module Input implements InputSig<Location, DataFlowImplSpecific::SwiftDataFlow>
 
   class SinkBase = Void;
 
-  class FlowSummaryCallBase = Void;
-
   predicate callableFromSource(SummarizedCallableBase c) { c.hasBody() }
 
   ArgumentPosition callbackSelfParameterPosition() { result instanceof ThisArgumentPosition }
@@ -115,10 +113,6 @@ module Input implements InputSig<Location, DataFlowImplSpecific::SwiftDataFlow>
 private import Make<Location, DataFlowImplSpecific::SwiftDataFlow, Input> as Impl
 
 private module StepsInput implements Impl::Private::StepsInputSig {
-  Impl::Private::SummaryNode getSummaryNode(Node n) {
-    result = n.(FlowSummaryNode).getSummaryNode()
-  }
-
   DataFlowCall getACall(Public::SummarizedCallable sc) { result.asCall().getStaticTarget() = sc }
 
   DataFlowCallable getSourceNodeEnclosingCallable(Input::SourceBase source) { none() }

swift/ql/lib/codeql/swift/dataflow/internal/TaintTrackingPrivate.qll

@@ -76,7 +76,7 @@ private module Cached {
     model = ""
     or
     // flow through a flow summary (extension of `SummaryModelCsv`)
-    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom,
+    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom.(FlowSummaryNode).getSummaryNode(),
       nodeTo.(FlowSummaryNode).getSummaryNode(), false, model)
     or
     any(AdditionalTaintStep a).step(nodeFrom, nodeTo) and model = "AdditionalTaintStep"

unified/swift-syntax-rs/.gitignore

@@ -1,2 +0,0 @@
-/target
-/swift/.build

unified/swift-syntax-rs/.swift-version

@@ -1 +0,0 @@
-6.2.4

unified/swift-syntax-rs/BUILD.bazel

@@ -1,55 +0,0 @@
-load("@rules_rust//rust:defs.bzl", "rust_binary", "rust_library", "rust_test")
-load("@rules_swift//swift:swift.bzl", "swift_library")
-
-package(default_visibility = ["//visibility:public"])
-
-# The pinned Swift version, shared with the local `cargo` build and
-# `swift/Package.swift`. Referenced by the `swift.toolchain` extension in
-# //:MODULE.bazel via `swift_version_file`.
-exports_files([".swift-version"])
-
-# The Swift FFI shim: wraps swift-syntax and exposes a small C ABI
-# (`ssr_parse_json` / `ssr_string_free`). Built with the hermetic Swift
-# toolchain registered in //:MODULE.bazel; provides a `CcInfo` that the Rust
-# targets below link against.
-swift_library(
-    name = "swift_syntax_ffi",
-    srcs = ["swift/Sources/SwiftSyntaxFFI/SwiftSyntaxFFI.swift"],
-    module_name = "SwiftSyntaxFFI",
-    deps = [
-        "@swift-syntax//:SwiftParser",
-        "@swift-syntax//:SwiftSyntax",
-    ],
-)
-
-# Safe Rust bindings on top of the C ABI. `build.rs` is intentionally *not*
-# wired up here (no `cargo_build_script`): under Bazel the Swift side is
-# provided by `:swift_syntax_ffi`, whereas `build.rs` only exists to build the
-# Swift shim in the local `cargo` workflow.
-rust_library(
-    name = "swift_syntax_rs",
-    srcs = ["src/lib.rs"],
-    edition = "2024",
-    deps = [":swift_syntax_ffi"],
-)
-
-rust_binary(
-    name = "swift-syntax-parse",
-    srcs = ["src/main.rs"],
-    # The Swift toolchain propagates a runfiles-relative RPATH to its runtime
-    # `.so`s via `swift_syntax_ffi`'s `CcInfo`, but (unlike `swift_binary`)
-    # `rust_binary` does not copy those libraries into runfiles. Add the
-    # toolchain files as `data` so the runtime is found at execution time.
-    # (rules_swift does not yet support statically linking the runtime.)
-    data = ["@swift_toolchain_ubuntu24.04//:files"],
-    edition = "2024",
-    deps = [":swift_syntax_rs"],
-)
-
-rust_test(
-    name = "swift_syntax_rs_test",
-    size = "small",
-    crate = ":swift_syntax_rs",
-    data = ["@swift_toolchain_ubuntu24.04//:files"],
-    edition = "2024",
-)

unified/swift-syntax-rs/Cargo.toml

@@ -1,14 +0,0 @@
-[package]
-name = "swift-syntax-rs"
-description = "Rust wrapper around the swift-syntax package for parsing Swift source"
-version = "0.1.0"
-authors = ["GitHub"]
-edition = "2024"
-
-[lib]
-name = "swift_syntax_rs"
-path = "src/lib.rs"
-
-[[bin]]
-name = "swift-syntax-parse"
-path = "src/main.rs"

unified/swift-syntax-rs/README.md

@@ -1,180 +0,0 @@
-# swift-syntax-rs
-
-A Rust wrapper around the [swift-syntax](https://github.com/swiftlang/swift-syntax)
-package, allowing Swift source code to be parsed from Rust.
-
-Parsing is delegated to a small Swift shim (in [`swift/`](swift/)) that links
-against `SwiftSyntax`/`SwiftParser` and exposes a tiny C ABI. The Rust crate
-builds that shim (via `build.rs`) and provides safe bindings on top of it.
-
-## Output format
-
-The emitted JSON tree preserves the AST's named structure. Every node has a
-`kind` and a `range` with `start`/`end` positions (UTF-8 `offset` plus 1-based
-`line`/`column`). Beyond that:
-
-- **Tokens** carry `text`, `tokenKind`, and `leadingTrivia`/`trailingTrivia`
-  arrays of `{ kind, text }` pieces.
-- **Layout nodes** (e.g. `functionDecl`) embed their children directly as
-  members keyed by the child's name in the parent (`name`, `signature`,
-  `body`, …), alongside `kind`/`range`. Absent optional children are omitted.
-- **Collection nodes** (e.g. `codeBlockItemList`) are elided: a list-valued
-  field is simply a JSON array of its elements (e.g. `parameters`, `statements`).
-  This drops the collection node's own `kind`/`range`.
-
-Only meaningful trivia is kept — the four comment kinds (`lineComment`,
-`blockComment`, `docLineComment`, `docBlockComment`) and `unexpectedText`
-(source the parser skipped). Whitespace trivia is dropped, since node ranges
-already encode positions.
-
-### Example
-
-Parsing `let x = 1 // c` produces the following (each `range` object is
-abbreviated here as `…`):
-
-```jsonc
-{
-  "kind": "sourceFile",
-  "range": …,
-  "statements": [                       // collection node elided to an array
-    {
-      "kind": "codeBlockItem",
-      "range": …,
-      "item": {
-        "kind": "variableDecl",
-        "range": …,
-        "attributes": [],               // empty collection → empty array
-        "modifiers": [],
-        "bindingSpecifier": {           // a token
-          "kind": "token",
-          "text": "let",
-          "tokenKind": "keyword(SwiftSyntax.Keyword.let)",
-          "range": …,
-          "leadingTrivia": [],
-          "trailingTrivia": []
-        },
-        "bindings": [
-          {
-            "kind": "patternBinding",
-            "range": …,
-            "pattern": {
-              "kind": "identifierPattern",
-              "range": …,
-              "identifier": { "kind": "token", "text": "x", "tokenKind": "identifier(\"x\")", "range": …, "leadingTrivia": [], "trailingTrivia": [] }
-            },
-            "initializer": {
-              "kind": "initializerClause",
-              "range": …,
-              "equal": { "kind": "token", "text": "=", "tokenKind": "equal", "range": …, "leadingTrivia": [], "trailingTrivia": [] },
-              "value": {
-                "kind": "integerLiteralExpr",
-                "range": …,
-                "literal": {
-                  "kind": "token",
-                  "text": "1",
-                  "tokenKind": "integerLiteral(\"1\")",
-                  "range": …,
-                  "leadingTrivia": [],
-                  "trailingTrivia": [ { "kind": "lineComment", "text": "// c" } ]
-                }
-              }
-            }
-          }
-        ]
-      }
-    }
-  ],
-  "endOfFileToken": { "kind": "token", "text": "", "tokenKind": "endOfFile", "range": …, "leadingTrivia": [], "trailingTrivia": [] }
-}
-```
-
-Note how `statements`, `bindings`, `attributes`, and `modifiers` are plain
-arrays (their collection nodes are elided), layout children such as
-`bindingSpecifier` and `initializer` are embedded by name, and the `// c`
-comment rides along as `trailingTrivia` on the token it follows.
-
-## Prerequisites
-
-The build does not depend on any particular version manager. You need:
-
-- **Rust** — pinned to `1.88` by the repo-root [`rust-toolchain.toml`](../../rust-toolchain.toml),
-  which `rustup` picks up automatically.
-- **Swift** — pinned to the version in [`.swift-version`](.swift-version)
-  (currently `6.2.4`), used to build `swift-syntax` `602.0.0`. Install it any way
-  you like — [swift.org](https://www.swift.org/install/) or
-  [swiftly](https://www.swift.org/swiftly/) (which reads `.swift-version`), or a
-  system package. Just make sure `swift` is on your `PATH` (or point `build.rs`
-  at it with the `SWIFT` environment variable).
-
-On Debian/Ubuntu the Swift runtime also needs `libncurses6` (and related libs)
-available on the system.
-
-## Building & testing
-
-With `cargo` and `swift` on `PATH`:
-
-```sh
-cargo build
-cargo test
-```
-
-If your `swift`/`swiftc` are not on `PATH`, point the build at them explicitly:
-
-```sh
-SWIFT=/path/to/swift SWIFTC=/path/to/swiftc cargo build
-```
-
-The first build compiles `swift-syntax` and can take several minutes.
-
-## Building with Bazel (CI)
-
-CI builds this crate hermetically with Bazel. A Swift toolchain is downloaded
-from swift.org by the official `rules_swift` standalone toolchain extension
-(wired up in the repo-root `MODULE.bazel`), `swift-syntax` is pulled from the
-Bazel Central Registry, and the FFI shim is compiled as a `swift_library` that
-the Rust targets link against. `build.rs` is not used under Bazel; it only
-builds the Swift shim for the local `cargo` workflow.
-
-```sh
-bazel build //unified/swift-syntax-rs:swift-syntax-parse
-bazel test  //unified/swift-syntax-rs:swift_syntax_rs_test
-bazel run   //unified/swift-syntax-rs:swift-syntax-parse < some.swift
-```
-
-Requirements:
-
-- **`clang`** must be installed on the runner. `rules_swift` requires the Bazel
-  CC toolchain to use clang; the repo's `.bazelrc` already sets
-  `--repo_env=CC=clang`, so no extra flags are needed.
-- The registered Swift toolchain currently targets **ubuntu24.04 / x86_64**
-  only (Bazel cannot auto-select between Linux distributions). Add more
-  platforms in `MODULE.bazel` (`swift.toolchain` + `register_toolchains`) if CI
-  grows to cover them.
-
-The Swift compiler version is read from [`.swift-version`](.swift-version) by
-both the Bazel toolchain (`swift.toolchain(swift_version_file = …)`) and the
-local build, and is kept in sync with the `swift-syntax` release pinned in
-`swift/Package.swift`, so local and CI builds behave identically.
-
-## Usage
-
-Library:
-
-```rust
-let json = swift_syntax_rs::parse_to_json("let x = 1")?;
-println!("{json}");
-```
-
-CLI (reads a file argument or stdin, prints the syntax tree as JSON):
-
-```sh
-echo 'let x = 1' | cargo run --bin swift-syntax-parse
-```
-
-## Layout
-
-- `swift/` — Swift package exposing the `ssr_parse_json` / `ssr_string_free` C ABI.
-- `build.rs` — builds the Swift package and emits link/rpath flags (local `cargo` only).
-- `BUILD.bazel` — Bazel targets for the hermetic CI build (swift_library + rust targets).
-- `src/lib.rs` — safe Rust bindings (`parse_to_json`).
-- `src/main.rs` — demo CLI.

unified/swift-syntax-rs/build.rs

@@ -1,102 +0,0 @@
-use std::env;
-use std::path::PathBuf;
-use std::process::Command;
-
-fn main() {
-    let manifest_dir = PathBuf::from(env::var("CARGO_MANIFEST_DIR").unwrap());
-    let swift_dir = manifest_dir.join("swift");
-
-    println!(
-        "cargo:rerun-if-changed={}",
-        swift_dir.join("Package.swift").display()
-    );
-    println!(
-        "cargo:rerun-if-changed={}",
-        swift_dir
-            .join("Sources/SwiftSyntaxFFI/SwiftSyntaxFFI.swift")
-            .display()
-    );
-    println!("cargo:rerun-if-env-changed=SWIFT");
-    println!("cargo:rerun-if-env-changed=SWIFTC");
-
-    // Build the Swift FFI package as a release dynamic library.
-    let mut command = Command::new(swift_bin());
-    command
-        .args(["build", "-c", "release"])
-        .current_dir(&swift_dir);
-    apply_bare_repository_workaround(&mut command);
-    let status = command.status().unwrap_or_else(|e| {
-        panic!(
-            "failed to run `{swift} build`: {e}\n\
-             Install a Swift toolchain (see https://www.swift.org/install/, e.g. via \
-             swiftly) and ensure `swift` is on PATH, or set the `SWIFT` environment \
-             variable to the `swift` executable. The pinned version is in `.swift-version`.",
-            swift = swift_bin(),
-        )
-    });
-    assert!(status.success(), "`swift build` failed");
-
-    // Link against the freshly built dynamic library.
-    let build_dir = swift_dir.join(".build/release");
-    println!("cargo:rustc-link-search=native={}", build_dir.display());
-    println!("cargo:rustc-link-lib=dylib=SwiftSyntaxFFI");
-    println!("cargo:rustc-link-arg=-Wl,-rpath,{}", build_dir.display());
-
-    // The executable also needs to find the Swift runtime libraries at run time.
-    if let Some(runtime) = swift_runtime_dir() {
-        println!("cargo:rustc-link-search=native={}", runtime.display());
-        println!("cargo:rustc-link-arg=-Wl,-rpath,{}", runtime.display());
-    }
-}
-
-/// Query the active Swift toolchain for the directory containing its runtime
-/// shared libraries (e.g. `libswiftCore.so`).
-fn swift_runtime_dir() -> Option<PathBuf> {
-    let output = Command::new(swiftc_bin())
-        .arg("-print-target-info")
-        .output()
-        .ok()?;
-    if !output.status.success() {
-        return None;
-    }
-    let info = String::from_utf8_lossy(&output.stdout);
-
-    // Extract the value of `"runtimeResourcePath": "..."` without pulling in a
-    // JSON dependency.
-    let key = "\"runtimeResourcePath\"";
-    let start = info.find(key)?;
-    let rest = &info[start + key.len()..];
-    let colon = rest.find(':')?;
-    let after = &rest[colon + 1..];
-    let open = after.find('"')?;
-    let value_start = &after[open + 1..];
-    let close = value_start.find('"')?;
-    let resource_path = &value_start[..close];
-
-    Some(PathBuf::from(resource_path).join("linux"))
-}
-
-/// The `swift` driver to invoke: `$SWIFT` if set, otherwise `swift` from `PATH`.
-/// This keeps the build tool-agnostic — any Swift install works; no particular
-/// version manager is required.
-fn swift_bin() -> String {
-    env::var("SWIFT").unwrap_or_else(|_| "swift".to_string())
-}
-
-/// The `swiftc` compiler to invoke: `$SWIFTC` if set, otherwise `swiftc` from
-/// `PATH`.
-fn swiftc_bin() -> String {
-    env::var("SWIFTC").unwrap_or_else(|_| "swiftc".to_string())
-}
-
-/// Some environments (notably GitHub Codespaces) inject
-/// `GIT_CONFIG_KEY_0=safe.bareRepository` / `GIT_CONFIG_VALUE_0=explicit`, which
-/// breaks the cached bare git repositories `swift build` uses. When exactly that
-/// key is present, relax it to `all` for the `swift build` subprocess only
-/// (rather than unconditionally, which could clobber an unrelated
-/// `GIT_CONFIG_VALUE_0`).
-fn apply_bare_repository_workaround(command: &mut Command) {
-    if env::var("GIT_CONFIG_KEY_0").as_deref() == Ok("safe.bareRepository") {
-        command.env("GIT_CONFIG_VALUE_0", "all");
-    }
-}

unified/swift-syntax-rs/src/lib.rs

@@ -1,125 +0,0 @@
-//! Rust wrapper around the [swift-syntax](https://github.com/swiftlang/swift-syntax)
-//! package, allowing Swift source code to be parsed from Rust.
-//!
-//! The heavy lifting is done by a small Swift shim (see `swift/`) that links
-//! against `SwiftSyntax`/`SwiftParser` and exposes a tiny C ABI. This module
-//! provides safe Rust bindings on top of that ABI.
-
-use std::ffi::{CStr, CString};
-use std::os::raw::c_char;
-
-// C ABI exported by the `SwiftSyntaxFFI` dynamic library.
-unsafe extern "C" {
-    /// Parse a NUL-terminated Swift source string, returning a heap-allocated
-    /// NUL-terminated JSON string (or null on failure). The caller owns the
-    /// returned pointer and must release it with `ssr_string_free`.
-    fn ssr_parse_json(source: *const c_char) -> *mut c_char;
-
-    /// Free a string previously returned by `ssr_parse_json`.
-    fn ssr_string_free(ptr: *mut c_char);
-}
-
-/// Errors that can occur while parsing Swift source.
-#[derive(Debug)]
-pub enum ParseError {
-    /// The provided source contained an interior NUL byte.
-    NulByte,
-    /// The Swift side failed to produce a result.
-    SwiftFailure,
-}
-
-impl std::fmt::Display for ParseError {
-    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
-        match self {
-            ParseError::NulByte => write!(f, "source contained an interior NUL byte"),
-            ParseError::SwiftFailure => write!(f, "swift-syntax failed to parse the source"),
-        }
-    }
-}
-
-impl std::error::Error for ParseError {}
-
-/// Parse the given Swift `source` and return a JSON representation of its
-/// syntax tree.
-///
-/// # Example
-/// ```no_run
-/// let json = swift_syntax_rs::parse_to_json("let x = 1").unwrap();
-/// println!("{json}");
-/// ```
-pub fn parse_to_json(source: &str) -> Result<String, ParseError> {
-    let c_source = CString::new(source).map_err(|_| ParseError::NulByte)?;
-
-    // SAFETY: `c_source` is a valid NUL-terminated string for the duration of
-    // the call. The returned pointer, if non-null, is owned by us and freed via
-    // `ssr_string_free` before returning.
-    unsafe {
-        let ptr = ssr_parse_json(c_source.as_ptr());
-        if ptr.is_null() {
-            return Err(ParseError::SwiftFailure);
-        }
-        let json = CStr::from_ptr(ptr).to_string_lossy().into_owned();
-        ssr_string_free(ptr);
-        Ok(json)
-    }
-}
-
-#[cfg(test)]
-mod tests {
-    use super::*;
-
-    #[test]
-    fn parses_simple_declaration() {
-        let json = parse_to_json("let x = 1").expect("parsing should succeed");
-        assert!(
-            json.contains("\"kind\":\"sourceFile\""),
-            "unexpected tree: {json}"
-        );
-        assert!(json.contains("\"text\":\"x\""), "unexpected tree: {json}");
-        // Source ranges are emitted for every node.
-        assert!(json.contains("\"range\""), "missing ranges: {json}");
-        assert!(
-            json.contains("\"line\"") && json.contains("\"column\"") && json.contains("\"offset\""),
-            "missing location fields: {json}"
-        );
-    }
-
-    #[test]
-    fn preserves_named_structure() {
-        let json = parse_to_json("let x = 1").expect("parsing should succeed");
-        // Layout children are embedded directly under their field name.
-        assert!(
-            json.contains("\"initializer\""),
-            "missing initializer field: {json}"
-        );
-        // Collection-valued fields are elided to plain JSON arrays.
-        assert!(
-            json.contains("\"statements\":["),
-            "statements should be an array: {json}"
-        );
-        assert!(
-            json.contains("\"bindings\":["),
-            "bindings should be an array: {json}"
-        );
-    }
-
-    #[test]
-    fn parses_empty_source() {
-        let json = parse_to_json("").expect("parsing empty source should succeed");
-        assert!(
-            json.contains("\"kind\":\"sourceFile\""),
-            "unexpected tree: {json}"
-        );
-    }
-
-    #[test]
-    fn captures_trivia() {
-        // A leading comment is kept as trivia on the token it precedes.
-        let json = parse_to_json("// hi\nlet x = 1").expect("parsing should succeed");
-        assert!(json.contains("\"leadingTrivia\""), "missing trivia: {json}");
-        assert!(
-            json.contains("lineComment"),
-            "comment trivia not captured: {json}"
-        );
-    }
-}

unified/swift-syntax-rs/src/main.rs

@@ -1,42 +0,0 @@
-//! Small demo CLI: parse Swift source and print the syntax tree as JSON.
-//!
-//! Usage:
-//!   swift-syntax-parse [FILE]
-//!
-//! Reads from FILE if given, otherwise from standard input.
-
-use std::io::Read;
-use std::process::ExitCode;
-
-fn main() -> ExitCode {
-    let source = match read_source() {
-        Ok(s) => s,
-        Err(e) => {
-            eprintln!("error reading source: {e}");
-            return ExitCode::FAILURE;
-        }
-    };
-
-    match swift_syntax_rs::parse_to_json(&source) {
-        Ok(json) => {
-            println!("{json}");
-            ExitCode::SUCCESS
-        }
-        Err(e) => {
-            eprintln!("error: {e}");
-            ExitCode::FAILURE
-        }
-    }
-}
-
-fn read_source() -> std::io::Result<String> {
-    let mut args = std::env::args().skip(1);
-    match args.next() {
-        Some(path) => std::fs::read_to_string(path),
-        None => {
-            let mut buf = String::new();
-            std::io::stdin().read_to_string(&mut buf)?;
-            Ok(buf)
-        }
-    }
-}

unified/swift-syntax-rs/swift/Package.resolved

@@ -1,15 +0,0 @@
-{
-  "originHash" : "ad84ad340ee01aa6e38b877dd29e38fa9cb40603dfdefa4e6a69e27f2db208d1",
-  "pins" : [
-    {
-      "identity" : "swift-syntax",
-      "kind" : "remoteSourceControl",
-      "location" : "https://github.com/swiftlang/swift-syntax.git",
-      "state" : {
-        "revision" : "4799286537280063c85a32f09884cfbca301b1a1",
-        "version" : "602.0.0"
-      }
-    }
-  ],
-  "version" : 3
-}

unified/swift-syntax-rs/swift/Package.swift

@@ -1,30 +0,0 @@
-// swift-tools-version:6.0
-import PackageDescription
-
-let package = Package(
-    name: "SwiftSyntaxFFI",
-    products: [
-        // Dynamic library so the produced .so records its dependency on the
-        // Swift runtime libraries, keeping the Rust link step simple.
-        .library(
-            name: "SwiftSyntaxFFI",
-            type: .dynamic,
-            targets: ["SwiftSyntaxFFI"]
-        )
-    ],
-    dependencies: [
-        .package(
-            url: "https://github.com/swiftlang/swift-syntax.git",
-            exact: "602.0.0"
-        )
-    ],
-    targets: [
-        .target(
-            name: "SwiftSyntaxFFI",
-            dependencies: [
-                .product(name: "SwiftSyntax", package: "swift-syntax"),
-                .product(name: "SwiftParser", package: "swift-syntax"),
-            ]
-        )
-    ]
-)

unified/swift-syntax-rs/swift/Sources/SwiftSyntaxFFI/SwiftSyntaxFFI.swift

@@ -1,145 +0,0 @@
-import Foundation
-import SwiftParser
-
-// `@_spi(RawSyntax)` exposes the `childName(_:)` helper that maps a child's
-// key path to its field name in the parent layout; used to emit named fields.
-@_spi(RawSyntax) import SwiftSyntax
-
-#if canImport(Glibc)
-    import Glibc
-#elseif canImport(Darwin)
-    import Darwin
-#endif
-
-/// Convert an absolute position into an `{ offset, line, column }` dictionary.
-///
-/// `offset` is a UTF-8 byte offset; `line`/`column` are 1-based.
-private func location(
-    _ position: AbsolutePosition,
-    _ converter: SourceLocationConverter
-) -> [String: Any] {
-    let loc = converter.location(for: position)
-    return [
-        "offset": position.utf8Offset,
-        "line": loc.line,
-        "column": loc.column,
-    ]
-}
-
-/// Trivia kinds worth preserving in the serialized tree. Comments carry
-/// developer intent (including doc comments), and `unexpectedText` flags source
-/// the parser had to skip. Whitespace and multi-line-string escape markers are
-/// dropped: node ranges already encode positions, so they would only bloat the
-/// output.
-private let keptTriviaKinds: Set<String> = [
-    "lineComment",
-    "blockComment",
-    "docLineComment",
-    "docBlockComment",
-    "unexpectedText",
-]
-
-/// Serialize a trivia collection into an array of `{ kind, text }` pieces,
-/// keeping only the kinds in `keptTriviaKinds`.
-private func serializeTrivia(_ trivia: Trivia) -> [Any] {
-    trivia.pieces.compactMap { piece -> [String: Any]? in
-        // The label of an enum case mirror is the case name (e.g. "spaces",
-        // "lineComment"), which gives us a stable kind without an exhaustive
-        // switch over every `TriviaPiece` case.
-        let kind = Mirror(reflecting: piece).children.first?.label ?? "\(piece)"
-        guard keptTriviaKinds.contains(kind) else { return nil }
-        return [
-            "kind": kind,
-            "text": Trivia(pieces: [piece]).description,
-        ]
-    }
-}
-
-/// Recursively convert a SwiftSyntax node into a JSON-serializable value.
-///
-///   * Tokens and layout nodes (e.g. `functionDecl`) become objects carrying
-///     `kind` and source `range`. Layout nodes additionally embed their children
-///     directly as members keyed by the child's name in the parent (e.g. `name`,
-///     `signature`, `body`); absent optional children are omitted. Field names
-///     never collide with `kind`/`range`.
-///   * Collection nodes (e.g. `codeBlockItemList`) are *elided*: they become a
-///     plain array of their serialized elements, taking the place of the
-///     collection node itself. A list-valued layout field (e.g. `parameters`) is
-///     therefore simply a JSON array. This drops the collection node's own
-///     `kind`/`range`, which are unnamed and largely recoverable from the
-///     elements.
-private func serialize(
-    _ node: Syntax,
-    _ converter: SourceLocationConverter
-) -> Any {
-    if node.kind.isSyntaxCollection {
-        return node.children(viewMode: .sourceAccurate).map {
-            serialize($0, converter)
-        }
-    }
-
-    // Source range covering the node's content, excluding surrounding trivia.
-    let range: [String: Any] = [
-        "start": location(node.positionAfterSkippingLeadingTrivia, converter),
-        "end": location(node.endPositionBeforeTrailingTrivia, converter),
-    ]
-
-    if let token = node.as(TokenSyntax.self) {
-        return [
-            "kind": "token",
-            "tokenKind": "\(token.tokenKind)",
-            "text": token.text,
-            "range": range,
-            "leadingTrivia": serializeTrivia(token.leadingTrivia),
-            "trailingTrivia": serializeTrivia(token.trailingTrivia),
-        ]
-    }
-
-    var result: [String: Any] = [
-        "kind": "\(node.kind)",
-        "range": range,
-    ]
-    var unnamed = 0
-    for child in node.children(viewMode: .sourceAccurate) {
-        // Layout children are named; embed each under its field name in the
-        // parent (the same mechanism SwiftSyntax uses for its debug dump). A
-        // child that is a collection serializes to an array (see above).
-        if let keyPath = child.keyPathInParent, let name = childName(keyPath) {
-            result[name] = serialize(child, converter)
-        } else {
-            // Defensive fallback for any unnamed layout child.
-            result["child\(unnamed)"] = serialize(child, converter)
-            unnamed += 1
-        }
-    }
-    return result
-}
-
-/// Parse the given NUL-terminated Swift source string and return a
-/// heap-allocated, NUL-terminated JSON representation of the syntax tree.
-///
-/// The returned pointer is owned by the caller and MUST be released with
-/// `ssr_string_free`. Returns `nil` on failure.
-@_cdecl("ssr_parse_json")
-public func ssr_parse_json(_ source: UnsafePointer<CChar>?) -> UnsafeMutablePointer<CChar>? {
-    guard let source = source else { return nil }
-    let code = String(cString: source)
-    let tree = Parser.parse(source: code)
-    let converter = SourceLocationConverter(fileName: "<input>", tree: tree)
-    let json = serialize(Syntax(tree), converter)
-
-    guard
-        let data = try? JSONSerialization.data(
-            withJSONObject: json, options: [.sortedKeys]),
-        let string = String(data: data, encoding: .utf8)
-    else {
-        return nil
-    }
-    return strdup(string)
-}
-
-/// Free a string previously returned by this library.
-@_cdecl("ssr_string_free")
-public func ssr_string_free(_ ptr: UnsafeMutablePointer<CChar>?) {
-    free(ptr)
-}