Automodel: Include all candidates, even manually modelled ones.

.bazelrc

@@ -11,17 +11,7 @@ common --override_module=semmle_code=%workspace%/misc/bazel/semmle_code_stub
 build --repo_env=CC=clang --repo_env=CXX=clang++
 
 build:linux --cxxopt=-std=c++20
-# we currently cannot built the swift extractor for ARM
-build:macos --cxxopt=-std=c++20 --copt=-arch --copt=x86_64 --linkopt=-arch --linkopt=x86_64
+build:macos --cxxopt=-std=c++20 --cpu=darwin_x86_64
 build:windows --cxxopt=/std:c++20 --cxxopt=/Zc:preprocessor
 
-# this requires developer mode, but is required to have pack installer functioning
-startup --windows_enable_symlinks
-common --enable_runfiles
-
-common --registry=file:///%workspace%/misc/bazel/registry
-common --registry=https://bcr.bazel.build
-
-common --@rules_dotnet//dotnet/settings:strict_deps=false
-
 try-import %workspace%/local.bazelrc

.bazelrc.internal

@@ -1,10 +0,0 @@
-# this file should contain bazel settings required to build things from `semmle-code`
-
-common --registry=file:///%workspace%/ql/misc/bazel/registry
-common --registry=https://bcr.bazel.build
-
-# See bazelbuild/rules_dotnet#413: strict_deps in C# also appliy to 3rd-party deps, and when we pull
-# in (for example) the xunit package, there's no code in this at all, it just depends transitively on
-# its implementation packages without providing any code itself.
-# We either can depend on internal implementation details, or turn of strict deps.
-common --@rules_dotnet//dotnet/settings:strict_deps=false

.bazelversion

@@ -1 +1 @@
-7.1.2
+7.1.0

.gitattributes

@@ -73,10 +73,3 @@ python/ql/lib/semmle/python/frameworks/data/internal/subclass-capture/*.yml ling
 # auto-generated bazel lock file
 ruby/extractor/cargo-bazel-lock.json linguist-generated=true
 ruby/extractor/cargo-bazel-lock.json -merge
-
-# auto-generated files for the C# build
-csharp/paket.lock linguist-generated=true
-# needs eol=crlf, as `paket` touches this file and saves it als crlf
-csharp/.paket/Paket.Restore.targets linguist-generated=true eol=crlf
-csharp/paket.main.bzl linguist-generated=true
-csharp/paket.main_extension.bzl linguist-generated=true

.github/workflows/build-ripunzip.yml

@@ -1,74 +0,0 @@
-name: Build runzip
-
-on:
-  workflow_dispatch:
-    inputs:
-      ripunzip-version:
-        description: "what reference to checktout from google/runzip"
-        required: false
-        default: v1.2.1
-      openssl-version:
-        description: "what reference to checkout from openssl/openssl for Linux"
-        required: false
-        default: openssl-3.3.0
-
-jobs:
-  build:
-    strategy:
-      fail-fast: false
-      matrix:
-        os: [ubuntu-20.04, macos-12, windows-2019]
-    runs-on: ${{ matrix.os }}
-    steps:
-      - uses: actions/checkout@v4
-        with:
-          repository: google/ripunzip
-          ref: ${{ inputs.ripunzip-version }}
-      # we need to avoid ripunzip dynamically linking into libssl
-      # see https://github.com/sfackler/rust-openssl/issues/183
-      - if: runner.os == 'Linux'
-        name: checkout openssl
-        uses: actions/checkout@v4
-        with:
-          repository: openssl/openssl
-          path: openssl
-          ref: ${{ inputs.openssl-version }}
-      - if: runner.os == 'Linux'
-        name: build and install openssl with fPIC
-        shell: bash
-        working-directory: openssl
-        run: |
-          ./config -fPIC --prefix=$HOME/.local --openssldir=$HOME/.local/ssl
-          make -j $(nproc)
-          make install_sw -j $(nproc)
-      - if: runner.os == 'Linux'
-        name: build (linux)
-        shell: bash
-        run: |
-          env OPENSSL_LIB_DIR=$HOME/.local/lib64 OPENSSL_INCLUDE_DIR=$HOME/.local/include OPENSSL_STATIC=yes cargo build --release
-          mv target/release/ripunzip ripunzip-linux
-      - if: runner.os == 'Windows'
-        name: build (windows)
-        shell: bash
-        run: |
-          cargo build --release
-          mv target/release/ripunzip ripunzip-windows
-      - name: build (macOS)
-        if: runner.os == 'macOS'
-        shell: bash
-        run: |
-          rustup target install x86_64-apple-darwin
-          rustup target install aarch64-apple-darwin
-          cargo build --target x86_64-apple-darwin --release
-          cargo build --target aarch64-apple-darwin --release
-          lipo -create -output ripunzip-macos \
-            -arch x86_64 target/x86_64-apple-darwin/release/ripunzip \
-            -arch arm64 target/aarch64-apple-darwin/release/ripunzip
-      - uses: actions/upload-artifact@v4
-        with:
-          name: ripunzip-${{ runner.os }}
-          path: ripunzip-*
-      - name: Check built binary
-        shell: bash
-        run: |
-          ./ripunzip-* --version

.github/workflows/buildifier.yml

@@ -1,28 +0,0 @@
-name: Check bazel formatting
-
-on:
-  pull_request:
-    paths:
-      - "**.bazel"
-      - "**.bzl"
-    branches:
-      - main
-      - "rc/*"
-
-permissions:
-  contents: read
-
-jobs:
-  check:
-    runs-on: ubuntu-latest
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-      - name: Check bazel formatting
-        uses: pre-commit/action@646c83fcd040023954eafda54b4db0192ce70507
-        with:
-          extra_args: >
-            buildifier --all-files 2>&1 ||
-            (
-              echo -e "In order to format all bazel files, please run:\n  bazel run //misc/bazel:buildifier"; exit 1
-            )

.github/workflows/codeql-analysis.yml

@@ -56,9 +56,7 @@ jobs:
     #    uses a compiled language
 
     - run: |
-       cd csharp
-       dotnet tool restore
-       dotnet build .
+       dotnet build csharp
 
     - name: Perform CodeQL Analysis
       uses: github/codeql-action/analyze@main

.github/workflows/csharp-qltest.yml

@@ -81,11 +81,10 @@ jobs:
           dotnet-version: 8.0.101
       - name: Extractor unit tests
         run: |
-          dotnet tool restore
           dotnet test -p:RuntimeFrameworkVersion=8.0.1 extractor/Semmle.Util.Tests
           dotnet test -p:RuntimeFrameworkVersion=8.0.1 extractor/Semmle.Extraction.Tests
           dotnet test -p:RuntimeFrameworkVersion=8.0.1 autobuilder/Semmle.Autobuild.CSharp.Tests
-          dotnet test -p:RuntimeFrameworkVersion=8.0.1 autobuilder/Semmle.Autobuild.Cpp.Tests
+          dotnet test -p:RuntimeFrameworkVersion=8.0.1 "${{ github.workspace }}/cpp/autobuilder/Semmle.Autobuild.Cpp.Tests"
         shell: bash
   stubgentest:
     runs-on: ubuntu-latest

.github/workflows/go-tests-other-os.yml

@@ -7,6 +7,8 @@ on:
       - .github/workflows/go-tests-other-os.yml
       - .github/actions/**
       - codeql-workspace.yml
+env:
+  GO_VERSION: '~1.22.0'
 
 permissions:
   contents: read
@@ -16,17 +18,72 @@ jobs:
     name: Test MacOS
     runs-on: macos-latest
     steps:
+      - name: Set up Go ${{ env.GO_VERSION }}
+        uses: actions/setup-go@v5
+        with:
+          go-version: ${{ env.GO_VERSION }}
+          cache: false
+        id: go
+
       - name: Check out code
         uses: actions/checkout@v4
-      - name: Run tests
-        uses: ./go/actions/test
+
+      - name: Set up CodeQL CLI
+        uses: ./.github/actions/fetch-codeql
+
+      - name: Enable problem matchers in repository
+        shell: bash
+        run: 'find .github/problem-matchers -name \*.json -exec echo "::add-matcher::{}" \;'
+
+      - name: Build
+        run: |
+          cd go
+          make
+
+      - name: Cache compilation cache
+        id: query-cache
+        uses: ./.github/actions/cache-query-compilation
+        with:
+          key: go-qltest
+      - name: Test
+        run: |
+          cd go
+          make test cache="${{ steps.query-cache.outputs.cache-dir }}"
 
   test-win:
     if: github.repository_owner == 'github'
     name: Test Windows
     runs-on: windows-latest-xl
     steps:
+      - name: Set up Go ${{ env.GO_VERSION }}
+        uses: actions/setup-go@v5
+        with:
+          go-version: ${{ env.GO_VERSION }}
+          cache: false
+        id: go
+
       - name: Check out code
         uses: actions/checkout@v4
-      - name: Run tests
-        uses: ./go/actions/test
+
+      - name: Set up CodeQL CLI
+        uses: ./.github/actions/fetch-codeql
+
+      - name: Enable problem matchers in repository
+        shell: bash
+        run: 'find .github/problem-matchers -name \*.json -exec echo "::add-matcher::{}" \;'
+
+      - name: Build
+        run: |
+          cd go
+          make
+
+      - name: Cache compilation cache
+        id: query-cache
+        uses: ./.github/actions/cache-query-compilation
+        with:
+          key: go-qltest
+
+      - name: Test
+        run: |
+          cd go
+          make test cache="${{ steps.query-cache.outputs.cache-dir }}"

.github/workflows/go-tests.yml

@@ -16,6 +16,9 @@ on:
       - .github/actions/**
       - codeql-workspace.yml
 
+env:
+  GO_VERSION: '~1.22.0'
+
 permissions:
   contents: read
 
@@ -25,9 +28,51 @@ jobs:
     name: Test Linux (Ubuntu)
     runs-on: ubuntu-latest-xl
     steps:
+      - name: Set up Go ${{ env.GO_VERSION }}
+        uses: actions/setup-go@v5
+        with:
+          go-version: ${{ env.GO_VERSION }}
+          cache: false
+        id: go
+
       - name: Check out code
         uses: actions/checkout@v4
-      - name: Run tests
-        uses: ./go/actions/test
+
+      - name: Set up CodeQL CLI
+        uses: ./.github/actions/fetch-codeql
+
+      - name: Enable problem matchers in repository
+        shell: bash
+        run: 'find .github/problem-matchers -name \*.json -exec echo "::add-matcher::{}" \;'
+
+      - name: Build
+        run: |
+          cd go
+          make
+
+      - name: Check that all Go code is autoformatted
+        run: |
+          cd go
+          make check-formatting
+
+      - name: Compile qhelp files to markdown
+        run: |
+          cd go
+          env QHELP_OUT_DIR=qhelp-out make qhelp-to-markdown
+
+      - name: Upload qhelp markdown
+        uses: actions/upload-artifact@v3
         with:
-          run-code-checks: true
+          name: qhelp-markdown
+          path: go/qhelp-out/**/*.md
+
+      - name: Cache compilation cache
+        id: query-cache
+        uses: ./.github/actions/cache-query-compilation
+        with:
+          key: go-qltest
+
+      - name: Test
+        run: |
+          cd go
+          make test cache="${{ steps.query-cache.outputs.cache-dir }}"

.lfsconfig

@@ -1,5 +0,0 @@
-[lfs]
-# codeql is publicly forked by many users, and we don't want any LFS file polluting their working
-# copies. We therefore exclude everything by default.
-# For files required by bazel builds, use rules in `misc/bazel/lfs.bzl` to download them on demand.
-fetchinclude = /nothing

.pre-commit-config.yaml

@@ -20,23 +20,13 @@ repos:
       - id: autopep8
         files: ^misc/codegen/.*\.py
 
-  - repo: local
+  - repo: https://github.com/warchant/pre-commit-buildifier
+    rev: 0.0.2
     hooks:
       - id: buildifier
-        name: Format bazel files
-        files: \.(bazel|bzl)
-        language: system
-        entry: bazel run //misc/bazel:buildifier
-        pass_filenames: false
-
-#      DISABLED: can be enabled by copying this config and installing `pre-commit` with `--config` on the copy
-#      - id: go-gen
-#        name: Check checked in generated files in go
-#        files: ^go/.*
-#        language: system
-#        entry: bazel run //go:gen
-#        pass_filenames: false
 
+  - repo: local
+    hooks:
       - id: codeql-format
         name: Fix QL file formatting
         files: \.qll?$

BUILD.bazel

@@ -1 +0,0 @@
-exports_files(["LICENSE"])

CODEOWNERS

@@ -1,7 +1,6 @@
 /cpp/ @github/codeql-c-analysis
+/cpp/autobuilder/ @github/codeql-c-extractor
 /csharp/ @github/codeql-csharp
-/csharp/autobuilder/Semmle.Autobuild.Cpp @github/codeql-c-extractor
-/csharp/autobuilder/Semmle.Autobuild.Cpp.Tests @github/codeql-c-extractor
 /go/ @github/codeql-go
 /java/ @github/codeql-java
 /javascript/ @github/codeql-javascript

CONTRIBUTING.md

@@ -4,8 +4,6 @@ We welcome contributions to our CodeQL libraries and queries. Got an idea for a
 
 There is lots of useful documentation to help you write queries, ranging from information about query file structure to tutorials for specific target languages. For more information on the documentation available, see [CodeQL queries](https://codeql.github.com/docs/writing-codeql-queries/codeql-queries) on [codeql.github.com](https://codeql.github.com).
 
-Note that the CodeQL for Visual Studio Code documentation has been migrated to https://docs.github.com/en/code-security/codeql-for-vs-code/, but you can still contribute to it via a different repository. For more information, see [Contributing to GitHub Docs documentation](https://docs.github.com/en/contributing)."
-
 ## Change notes
 
 Any nontrivial user-visible change to a query pack or library pack should have a change note. For details on how to add a change note for your change, see [this guide](docs/change-notes.md).
@@ -45,7 +43,7 @@ If you have an idea for a query that you would like to share with other CodeQL u
 
 3. **Formatting**
 
-    - The queries and libraries must be autoformatted, for example using the "Format Document" command in [CodeQL for Visual Studio Code](https://docs.github.com/en/code-security/codeql-for-vs-code/).
+    - The queries and libraries must be autoformatted, for example using the "Format Document" command in [CodeQL for Visual Studio Code](https://codeql.github.com/docs/codeql-for-visual-studio-code/about-codeql-for-visual-studio-code).
 
     If you prefer, you can either:
     1. install the [pre-commit framework](https://pre-commit.com/) and install the configured hooks on this repo via `pre-commit install`, or

MODULE.bazel

@@ -13,8 +13,7 @@ local_path_override(
 
 # see https://registry.bazel.build/ for a list of available packages
 
-bazel_dep(name = "platforms", version = "0.0.9")
-bazel_dep(name = "rules_go", version = "0.47.0")
+bazel_dep(name = "platforms", version = "0.0.8")
 bazel_dep(name = "rules_pkg", version = "0.10.1")
 bazel_dep(name = "rules_nodejs", version = "6.0.3")
 bazel_dep(name = "rules_python", version = "0.31.0")
@@ -22,19 +21,6 @@ bazel_dep(name = "bazel_skylib", version = "1.5.0")
 bazel_dep(name = "abseil-cpp", version = "20240116.0", repo_name = "absl")
 bazel_dep(name = "nlohmann_json", version = "3.11.3", repo_name = "json")
 bazel_dep(name = "fmt", version = "10.0.0")
-bazel_dep(name = "gazelle", version = "0.36.0")
-bazel_dep(name = "rules_dotnet", version = "0.15.1")
-
-bazel_dep(name = "buildifier_prebuilt", version = "6.4.0", dev_dependency = True)
-
-dotnet = use_extension("@rules_dotnet//dotnet:extensions.bzl", "dotnet")
-dotnet.toolchain(dotnet_version = "8.0.101")
-use_repo(dotnet, "dotnet_toolchains")
-
-register_toolchains("@dotnet_toolchains//:all")
-
-csharp_main_extension = use_extension("//csharp:paket.main_extension.bzl", "main_extension")
-use_repo(csharp_main_extension, "paket.main")
 
 pip = use_extension("@rules_python//python/extensions:pip.bzl", "pip")
 pip.parse(
@@ -64,9 +50,6 @@ node.toolchain(
 )
 use_repo(node, "nodejs", "nodejs_toolchains")
 
-go_sdk = use_extension("@rules_go//go:extensions.bzl", "go_sdk")
-go_sdk.download(version = "1.22.2")
-
 register_toolchains(
     "@nodejs_toolchains//:all",
 )

README.md

@@ -4,7 +4,7 @@ This open source repository contains the standard CodeQL libraries and queries t
 
 ## How do I learn CodeQL and run queries?
 
-There is extensive documentation about the [CodeQL language](https://codeql.github.com/docs/), writing CodeQL using the [CodeQL extension for Visual Studio Code](https://docs.github.com/en/code-security/codeql-for-vs-code/) and using the [CodeQL CLI](https://docs.github.com/en/code-security/codeql-cli).
+There is [extensive documentation](https://codeql.github.com/docs/) on getting started with writing CodeQL using the [CodeQL extension for Visual Studio Code](https://codeql.github.com/docs/codeql-for-visual-studio-code/) and the [CodeQL CLI](https://codeql.github.com/docs/codeql-cli/).
 
 ## Contributing
 

config/dbscheme-fragments.json

@@ -28,7 +28,6 @@
     "/*- Yaml dbscheme -*/",
     "/*- Blame dbscheme -*/",
     "/*- JSON dbscheme -*/",
-    "/*- Python dbscheme -*/",
-    "/*- Empty location -*/"
+    "/*- Python dbscheme -*/"
   ]
 }

config/identical-files.json

@@ -362,11 +362,7 @@
     "java/ql/lib/semmle/code/java/security/internal/EncryptionKeySizes.qll"
   ],
   "Python model summaries test extension": [
-    "python/ql/test/library-tests/dataflow/model-summaries/InlineTaintTest.ext.yml",
-    "python/ql/test/library-tests/dataflow/model-summaries/NormalDataflowTest.ext.yml"
-  ],
-  "shared tree-sitter extractor cargo.toml": [
-    "shared/tree-sitter-extractor/Cargo.toml",
-    "ruby/extractor/codeql-extractor-fake-crate/Cargo.toml"
+    "python/ql/test/experimental/dataflow/model-summaries/InlineTaintTest.ext.yml",
+    "python/ql/test/experimental/dataflow/model-summaries/NormalDataflowTest.ext.yml"
   ]
-}
+}

cpp/autobuilder/.gitignore

@@ -0,0 +1,13 @@
+obj/
+TestResults/
+*.manifest
+*.pdb
+*.suo
+*.mdb
+*.vsmdi
+csharp.log
+**/bin/Debug
+**/bin/Release
+*.tlog
+.vs
+*.user

cpp/autobuilder/README.md

@@ -1 +0,0 @@
-The Windows autobuilder that used to live in this directory moved to `csharp/autobuilder/Semmle.Autobuild.Cpp`.

cpp/autobuilder/Semmle.Autobuild.Cpp.Tests/BuildScripts.cs

@@ -200,9 +200,9 @@ namespace Semmle.Autobuild.Cpp.Tests
 
     internal class TestDiagnosticWriter : IDiagnosticsWriter
     {
-        public IList<Semmle.Util.DiagnosticMessage> Diagnostics { get; } = new List<Semmle.Util.DiagnosticMessage>();
+        public IList<DiagnosticMessage> Diagnostics { get; } = new List<DiagnosticMessage>();
 
-        public void AddEntry(Semmle.Util.DiagnosticMessage message) => this.Diagnostics.Add(message);
+        public void AddEntry(DiagnosticMessage message) => this.Diagnostics.Add(message);
 
         public void Dispose() { }
     }

cpp/autobuilder/Semmle.Autobuild.Cpp.Tests/Semmle.Autobuild.Cpp.Tests.csproj

@@ -0,0 +1,26 @@
+<Project Sdk="Microsoft.NET.Sdk">
+
+  <PropertyGroup>
+    <OutputType>Exe</OutputType>
+    <TargetFramework>net8.0</TargetFramework>
+    <GenerateAssemblyInfo>false</GenerateAssemblyInfo>
+    <RuntimeIdentifiers>win-x64;linux-x64;osx-x64</RuntimeIdentifiers>
+    <Nullable>enable</Nullable>
+  </PropertyGroup>
+
+  <ItemGroup>
+    <PackageReference Include="System.IO.FileSystem" Version="4.3.0" />
+    <PackageReference Include="System.IO.FileSystem.Primitives" Version="4.3.0" />
+    <PackageReference Include="xunit" Version="2.6.2" />
+    <PackageReference Include="xunit.runner.visualstudio" Version="2.5.4">
+      <PrivateAssets>all</PrivateAssets>
+      <IncludeAssets>runtime; build; native; contentfiles; analyzers</IncludeAssets>
+    </PackageReference>
+    <PackageReference Include="Microsoft.NET.Test.Sdk" Version="17.8.0" />
+  </ItemGroup>
+
+  <ItemGroup>
+    <ProjectReference Include="..\Semmle.Autobuild.Cpp\Semmle.Autobuild.Cpp.csproj" />
+    <ProjectReference Include="..\..\..\csharp\autobuilder\Semmle.Autobuild.Shared\Semmle.Autobuild.Shared.csproj" />
+  </ItemGroup>
+</Project>

cpp/autobuilder/Semmle.Autobuild.Cpp/Properties/AssemblyInfo.cs

@@ -0,0 +1,32 @@
+using System.Reflection;
+using System.Runtime.InteropServices;
+
+// General Information about an assembly is controlled through the following
+// set of attributes. Change these attribute values to modify the information
+// associated with an assembly.
+[assembly: AssemblyTitle("Semmle.Autobuild.Cpp")]
+[assembly: AssemblyDescription("")]
+[assembly: AssemblyConfiguration("")]
+[assembly: AssemblyCompany("GitHub")]
+[assembly: AssemblyProduct("CodeQL autobuilder for C++")]
+[assembly: AssemblyCopyright("Copyright © GitHub 2020")]
+[assembly: AssemblyTrademark("")]
+[assembly: AssemblyCulture("")]
+
+// Setting ComVisible to false makes the types in this assembly not visible
+// to COM components.  If you need to access a type in this assembly from
+// COM, set the ComVisible attribute to true on that type.
+[assembly: ComVisible(false)]
+
+// Version information for an assembly consists of the following four values:
+//
+//      Major Version
+//      Minor Version
+//      Build Number
+//      Revision
+//
+// You can specify all the values or you can default the Build and Revision Numbers
+// by using the '*' as shown below:
+// [assembly: AssemblyVersion("1.0.*")]
+[assembly: AssemblyVersion("1.0.0.0")]
+[assembly: AssemblyFileVersion("1.0.0.0")]

cpp/autobuilder/Semmle.Autobuild.Cpp/Semmle.Autobuild.Cpp.csproj

@@ -0,0 +1,28 @@
+<Project Sdk="Microsoft.NET.Sdk">
+
+  <PropertyGroup>
+    <TargetFramework>net8.0</TargetFramework>
+    <AssemblyName>Semmle.Autobuild.Cpp</AssemblyName>
+    <RootNamespace>Semmle.Autobuild.Cpp</RootNamespace>
+    <ApplicationIcon />
+    <OutputType>Exe</OutputType>
+    <StartupObject />
+    <GenerateAssemblyInfo>false</GenerateAssemblyInfo>
+    <RuntimeIdentifiers>win-x64;linux-x64;osx-x64</RuntimeIdentifiers>
+    <Nullable>enable</Nullable>
+  </PropertyGroup>
+
+  <ItemGroup>
+    <Folder Include="Properties\" />
+  </ItemGroup>
+
+  <ItemGroup>
+    <PackageReference Include="Microsoft.Build" Version="17.8.3" />
+  </ItemGroup>
+
+  <ItemGroup>
+    <ProjectReference Include="..\..\..\csharp\extractor\Semmle.Util\Semmle.Util.csproj" />
+    <ProjectReference Include="..\..\..\csharp\autobuilder\Semmle.Autobuild.Shared\Semmle.Autobuild.Shared.csproj" />
+  </ItemGroup>
+
+</Project>

cpp/downgrades/19887dbd33327fb07d54251786e0cb2578539775/upgrade.properties

@@ -1,4 +1,4 @@
 description: Revert support for repeated initializers, which are allowed in C with designated initializers.
 compatibility: full
-aggregate_field_init.rel: reorder aggregate_field_init.rel (@aggregateliteral aggregate, @expr initializer, @membervariable field, int position) aggregate initializer field
-aggregate_array_init.rel: reorder aggregate_array_init.rel (@aggregateliteral aggregate, @expr initializer, int element_index, int position) aggregate initializer element_index
+aggregate_field_init.rel: reorder aggregate_field_init.rel (int aggregate, int initializer, int field, int position) aggregate initializer field
+aggregate_array_init.rel: reorder aggregate_array_init.rel (int aggregate, int initializer, int element_index, int position) aggregate initializer element_index

cpp/ql/lib/CHANGELOG.md

@@ -1,30 +1,3 @@
-## 1.0.0
-
-### Breaking Changes
-
-* CodeQL package management is now generally available, and all GitHub-produced CodeQL packages have had their version numbers increased to 1.0.0.
-
-## 0.13.1
-
-No user-facing changes.
-
-## 0.13.0
-
-### Breaking Changes
-
-* Deleted the deprecated `GlobalValueNumberingImpl.qll` implementation.
-
-### New Features
-
-* Models-as-Data support has been added for C/C++. This feature allows flow sources, sinks and summaries to be expressed in compact strings as an alternative to modelling each source / sink / summary with explicit QL. See `dataflow/ExternalFlow.qll` for documentation and specification of the model format, and `models/implementations/ZMQ.qll` for a simple example of models. Importing models from `.yml` is not yet supported.
-
-### Minor Analysis Improvements
-
-* Source models have been added for the standard library function `getc` (and variations).
-* Source, sink and flow models for the ZeroMQ (ZMQ) networking library have been added.
-* Parameters of functions without definitions now have `ParameterNode`s.
-* The alias analysis used internally by various libraries has been improved to answer alias questions more conservatively. As a result, some queries may report fewer false positives.
-
 ## 0.12.11
 
 No user-facing changes.

cpp/ql/lib/change-notes/2024-04-05-sound-ir.md

@@ -0,0 +1,4 @@
+---
+category: minorAnalysis
+---
+* The alias analysis used internally by various libraries has been improved to answer alias questions more conservatively. As a result, some queries may report fewer false positives.

cpp/ql/lib/change-notes/2024-04-18-param-nodes.md

@@ -0,0 +1,4 @@
+---
+category: minorAnalysis
+---
+* Parameters of functions without definitions now have `ParameterNode`s.

cpp/ql/lib/change-notes/2024-10-04-getc.md

@@ -0,0 +1,4 @@
+---
+category: minorAnalysis
+---
+* Source models have been added for the standard library function `getc` (and variations).

cpp/ql/lib/change-notes/2024-10-04-models-as-data.md

@@ -0,0 +1,4 @@
+---
+category: feature
+---
+* Models-as-Data support has been added for C/C++. This feature allows flow sources, sinks and summaries to be expressed in compact strings as an alternative to modelling each source / sink / summary with explicit QL. See `dataflow/ExternalFlow.qll` for documentation and specification of the model format, and `models/implementations/ZMQ.qll` for a simple example of models. Importing models from `.yml` is not yet supported.

cpp/ql/lib/change-notes/2024-10-04-zmq.md

@@ -0,0 +1,4 @@
+---
+category: minorAnalysis
+---
+* Source, sink and flow models for the ZeroMQ (ZMQ) networking library have been added.

cpp/ql/lib/change-notes/released/0.13.0.md

@@ -1,16 +0,0 @@
-## 0.13.0
-
-### Breaking Changes
-
-* Deleted the deprecated `GlobalValueNumberingImpl.qll` implementation.
-
-### New Features
-
-* Models-as-Data support has been added for C/C++. This feature allows flow sources, sinks and summaries to be expressed in compact strings as an alternative to modelling each source / sink / summary with explicit QL. See `dataflow/ExternalFlow.qll` for documentation and specification of the model format, and `models/implementations/ZMQ.qll` for a simple example of models. Importing models from `.yml` is not yet supported.
-
-### Minor Analysis Improvements
-
-* Source models have been added for the standard library function `getc` (and variations).
-* Source, sink and flow models for the ZeroMQ (ZMQ) networking library have been added.
-* Parameters of functions without definitions now have `ParameterNode`s.
-* The alias analysis used internally by various libraries has been improved to answer alias questions more conservatively. As a result, some queries may report fewer false positives.

cpp/ql/lib/change-notes/released/0.13.1.md

@@ -1,3 +0,0 @@
-## 0.13.1
-
-No user-facing changes.

cpp/ql/lib/change-notes/released/1.0.0.md

@@ -1,5 +0,0 @@
-## 1.0.0
-
-### Breaking Changes
-
-* CodeQL package management is now generally available, and all GitHub-produced CodeQL packages have had their version numbers increased to 1.0.0.

cpp/ql/lib/codeql-pack.release.yml

@@ -1,2 +1,2 @@
 ---
-lastReleaseVersion: 1.0.0
+lastReleaseVersion: 0.12.11

cpp/ql/lib/qlpack.yml

@@ -1,5 +1,5 @@
 name: codeql/cpp-all
-version: 1.0.0
+version: 0.12.12-dev
 groups: cpp
 dbscheme: semmlecode.cpp.dbscheme
 extractor: cpp

cpp/ql/lib/semmle/code/cpp/PrintAST.qll

@@ -463,25 +463,6 @@ class StmtNode extends AstNode {
   }
 }
 
-/**
- * A node representing a child of a `Stmt` that is itself a `Stmt`.
- */
-class ChildStmtNode extends StmtNode {
-  Stmt childStmt;
-
-  ChildStmtNode() { exists(Stmt parent | parent.getAChild() = childStmt and childStmt = ast) }
-
-  override BaseAstNode getChildInternal(int childIndex) {
-    result = super.getChildInternal(childIndex)
-    or
-    exists(int destructorIndex |
-      result.getAst() = childStmt.getImplicitDestructorCall(destructorIndex) and
-      childIndex =
-        destructorIndex + max(int index | exists(childStmt.getChild(index)) or index = 0) + 1
-    )
-  }
-}
-
 /**
  * A node representing a `DeclStmt`.
  */
@@ -693,13 +674,6 @@ class FunctionNode extends FunctionOrGlobalOrNamespaceVariableNode {
 private string getChildAccessorWithoutConversions(Locatable parent, Element child) {
   shouldPrintDeclaration(getAnEnclosingDeclaration(parent)) and
   (
-    exists(Stmt s, int i | s.getChild(i) = parent |
-      exists(int n |
-        s.getChild(i).(Stmt).getImplicitDestructorCall(n) = child and
-        result = "getImplicitDestructorCall(" + n + ")"
-      )
-    )
-    or
     exists(Stmt s | s = parent |
       namedStmtChildPredicates(s, child, result)
       or

cpp/ql/lib/semmle/code/cpp/controlflow/IRGuards.qll

@@ -565,7 +565,7 @@ class IRGuardCondition extends Instruction {
   /** Holds if (determined by this guard) `op == k` evaluates to `areEqual` if this expression evaluates to `value`. */
   cached
   predicate comparesEq(Operand op, int k, boolean areEqual, AbstractValue value) {
-    unary_compares_eq(this, op, k, areEqual, false, value)
+    compares_eq(this, op, k, areEqual, value)
   }
 
   /**
@@ -586,7 +586,7 @@ class IRGuardCondition extends Instruction {
   cached
   predicate ensuresEq(Operand op, int k, IRBlock block, boolean areEqual) {
     exists(AbstractValue value |
-      unary_compares_eq(this, op, k, areEqual, false, value) and this.valueControls(block, value)
+      compares_eq(this, op, k, areEqual, value) and this.valueControls(block, value)
     )
   }
 
@@ -611,7 +611,7 @@ class IRGuardCondition extends Instruction {
   cached
   predicate ensuresEqEdge(Operand op, int k, IRBlock pred, IRBlock succ, boolean areEqual) {
     exists(AbstractValue value |
-      unary_compares_eq(this, op, k, areEqual, false, value) and
+      compares_eq(this, op, k, areEqual, value) and
       this.valueControlsEdge(pred, succ, value)
     )
   }
@@ -737,66 +737,26 @@ private predicate compares_eq(
   )
 }
 
-/**
- * Holds if `op == k` is `areEqual` given that `test` is equal to `value`.
- *
- * Many internal predicates in this file have a `inNonZeroCase` column.
- * Ideally, the `k` column would be a type such as `Option<int>::Option`, to
- * represent whether we have a concrete value `k` such that `op == k`, or whether
- * we only know that `op != 0`.
- * However, cannot instantiate `Option` with an infinite type. Thus the boolean
- * `inNonZeroCase` is used to distinquish the `Some` (where we have a concrete
- * value `k`) and `None` cases (where we only know that `op != 0`).
- *
- * Thus, if `inNonZeroCase = true` then `op != 0` and the value of `k` is
- * meaningless.
- *
- * To see why `inNonZeroCase` is needed consider the following C program:
- * ```c
- * char* p = ...;
- * if(p) {
- *   use(p);
- * }
- * ```
- * in C++ there would be an int-to-bool conversion on `p`. However, since C
- * does not have booleans there is no conversion. We want to be able to
- * conclude that `p` is non-zero in the true branch, so we need to give `k`
- * some value. However, simply setting `k = 1` would make the rest of the
- * analysis think that `k == 1` holds inside the branch. So we distinquish
- * between the above case and
- * ```c
- * if(p == 1) {
- *   use(p)
- * }
- * ```
- * by setting `inNonZeroCase` to `true` in the former case, but not in the
- * latter.
- */
-private predicate unary_compares_eq(
-  Instruction test, Operand op, int k, boolean areEqual, boolean inNonZeroCase, AbstractValue value
+/** Holds if `op == k` is `areEqual` given that `test` is equal to `value`. */
+private predicate compares_eq(
+  Instruction test, Operand op, int k, boolean areEqual, AbstractValue value
 ) {
   /* The simple case where the test *is* the comparison so areEqual = testIsTrue xor eq. */
-  exists(AbstractValue v | unary_simple_comparison_eq(test, op, k, inNonZeroCase, v) |
+  exists(AbstractValue v | simple_comparison_eq(test, op, k, v) |
     areEqual = true and value = v
     or
     areEqual = false and value = v.getDualValue()
   )
   or
-  unary_complex_eq(test, op, k, areEqual, inNonZeroCase, value)
+  complex_eq(test, op, k, areEqual, value)
   or
   /* (x is true => (op == k)) => (!x is false => (op == k)) */
-  exists(AbstractValue dual, boolean inNonZeroCase0 |
-    value = dual.getDualValue() and
-    unary_compares_eq(test.(LogicalNotInstruction).getUnary(), op, k, inNonZeroCase0, areEqual, dual)
-  |
-    k = 0 and inNonZeroCase = inNonZeroCase0
-    or
-    k != 0 and inNonZeroCase = true
+  exists(AbstractValue dual | value = dual.getDualValue() |
+    compares_eq(test.(LogicalNotInstruction).getUnary(), op, k, areEqual, dual)
   )
   or
   // ((test is `areEqual` => op == const + k2) and const == `k1`) =>
   // test is `areEqual` => op == k1 + k2
-  inNonZeroCase = false and
   exists(int k1, int k2, ConstantInstruction const |
     compares_eq(test, op, const.getAUse(), k2, areEqual, value) and
     int_value(const) = k1 and
@@ -821,53 +781,14 @@ private predicate simple_comparison_eq(
   value.(BooleanValue).getValue() = false
 }
 
-/**
- * Holds if `test` is an instruction that is part of test that eventually is
- * used in a conditional branch.
- */
-private predicate relevantUnaryComparison(Instruction test) {
-  not test instanceof CompareInstruction and
-  exists(IRType type, ConditionalBranchInstruction branch |
-    type instanceof IRAddressType or type instanceof IRIntegerType
-  |
-    type = test.getResultIRType() and
-    branch.getCondition() = test
-  )
-  or
-  exists(LogicalNotInstruction logicalNot |
-    relevantUnaryComparison(logicalNot) and
-    test = logicalNot.getUnary()
-  )
-}
-
-/**
- * Rearrange various simple comparisons into `op == k` form.
- */
-private predicate unary_simple_comparison_eq(
-  Instruction test, Operand op, int k, boolean inNonZeroCase, AbstractValue value
-) {
+/** Rearrange various simple comparisons into `op == k` form. */
+private predicate simple_comparison_eq(Instruction test, Operand op, int k, AbstractValue value) {
   exists(SwitchInstruction switch, CaseEdge case |
     test = switch.getExpression() and
     op.getDef() = test and
     case = value.(MatchValue).getCase() and
     exists(switch.getSuccessor(case)) and
-    case.getValue().toInt() = k and
-    inNonZeroCase = false
-  )
-  or
-  // There's no implicit CompareInstruction in files compiled as C since C
-  // doesn't have implicit boolean conversions. So instead we check whether
-  // there's a branch on a value of pointer or integer type.
-  relevantUnaryComparison(test) and
-  op.getDef() = test and
-  (
-    k = 1 and
-    value.(BooleanValue).getValue() = true and
-    inNonZeroCase = true
-    or
-    k = 0 and
-    value.(BooleanValue).getValue() = false and
-    inNonZeroCase = false
+    case.getValue().toInt() = k
   )
 }
 
@@ -879,12 +800,12 @@ private predicate complex_eq(
   add_eq(cmp, left, right, k, areEqual, value)
 }
 
-private predicate unary_complex_eq(
-  Instruction test, Operand op, int k, boolean areEqual, boolean inNonZeroCase, AbstractValue value
+private predicate complex_eq(
+  Instruction test, Operand op, int k, boolean areEqual, AbstractValue value
 ) {
-  unary_sub_eq(test, op, k, areEqual, inNonZeroCase, value)
+  sub_eq(test, op, k, areEqual, value)
   or
-  unary_add_eq(test, op, k, areEqual, inNonZeroCase, value)
+  add_eq(test, op, k, areEqual, value)
 }
 
 /*
@@ -1148,20 +1069,16 @@ private predicate sub_eq(
 }
 
 // op - x == c => op == (c+x)
-private predicate unary_sub_eq(
-  Instruction test, Operand op, int k, boolean areEqual, boolean inNonZeroCase, AbstractValue value
-) {
-  inNonZeroCase = false and
+private predicate sub_eq(Instruction test, Operand op, int k, boolean areEqual, AbstractValue value) {
   exists(SubInstruction sub, int c, int x |
-    unary_compares_eq(test, sub.getAUse(), c, areEqual, inNonZeroCase, value) and
+    compares_eq(test, sub.getAUse(), c, areEqual, value) and
     op = sub.getLeftOperand() and
     x = int_value(sub.getRight()) and
     k = c + x
   )
   or
-  inNonZeroCase = false and
   exists(PointerSubInstruction sub, int c, int x |
-    unary_compares_eq(test, sub.getAUse(), c, areEqual, inNonZeroCase, value) and
+    compares_eq(test, sub.getAUse(), c, areEqual, value) and
     op = sub.getLeftOperand() and
     x = int_value(sub.getRight()) and
     k = c + x
@@ -1215,13 +1132,11 @@ private predicate add_eq(
 }
 
 // left + x == right + c => left == right + (c-x)
-private predicate unary_add_eq(
-  Instruction test, Operand left, int k, boolean areEqual, boolean inNonZeroCase,
-  AbstractValue value
+private predicate add_eq(
+  Instruction test, Operand left, int k, boolean areEqual, AbstractValue value
 ) {
-  inNonZeroCase = false and
   exists(AddInstruction lhs, int c, int x |
-    unary_compares_eq(test, lhs.getAUse(), c, areEqual, inNonZeroCase, value) and
+    compares_eq(test, lhs.getAUse(), c, areEqual, value) and
     (
       left = lhs.getLeftOperand() and x = int_value(lhs.getRight())
       or
@@ -1230,9 +1145,8 @@ private predicate unary_add_eq(
     k = c - x
   )
   or
-  inNonZeroCase = false and
   exists(PointerAddInstruction lhs, int c, int x |
-    unary_compares_eq(test, lhs.getAUse(), c, areEqual, inNonZeroCase, value) and
+    compares_eq(test, lhs.getAUse(), c, areEqual, value) and
     (
       left = lhs.getLeftOperand() and x = int_value(lhs.getRight())
       or
@@ -1242,14 +1156,5 @@ private predicate unary_add_eq(
   )
 }
 
-private class IntegerOrPointerConstantInstruction extends ConstantInstruction {
-  IntegerOrPointerConstantInstruction() {
-    this instanceof IntegerConstantInstruction or
-    this instanceof PointerConstantInstruction
-  }
-}
-
 /** The int value of integer constant expression. */
-private int int_value(Instruction i) {
-  result = i.(IntegerOrPointerConstantInstruction).getValue().toInt()
-}
+private int int_value(Instruction i) { result = i.(IntegerConstantInstruction).getValue().toInt() }

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

@@ -9,7 +9,7 @@ private import DataFlowUtil
 /**
  * Gets a function that might be called by `call`.
  */
-DataFlowCallable viableCallable(DataFlowCall call) {
+Function viableCallable(DataFlowCall call) {
   result = call.(Call).getTarget()
   or
   // If the target of the call does not have a body in the snapshot, it might

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

@@ -242,17 +242,7 @@ class CastNode extends Node {
   CastNode() { none() } // stub implementation
 }
 
-class DataFlowCallable extends Function {
-  /** Gets a best-effort total ordering. */
-  int totalorder() {
-    this =
-      rank[result](DataFlowCallable c, string file, int startline, int startcolumn |
-        c.getLocation().hasLocationInfo(file, startline, startcolumn, _, _)
-      |
-        c order by file, startline, startcolumn
-      )
-  }
-}
+class DataFlowCallable = Function;
 
 class DataFlowExpr = Expr;
 
@@ -271,28 +261,10 @@ class DataFlowCall extends Expr instanceof Call {
   ExprNode getNode() { result.getExpr() = this }
 
   /** Gets the enclosing callable of this call. */
-  DataFlowCallable getEnclosingCallable() { result = this.getEnclosingFunction() }
-
-  /** Gets a best-effort total ordering. */
-  int totalorder() {
-    this =
-      rank[result](DataFlowCall c, int startline, int startcolumn |
-        c.getLocation().hasLocationInfo(_, startline, startcolumn, _, _)
-      |
-        c order by startline, startcolumn
-      )
-  }
+  Function getEnclosingCallable() { result = this.getEnclosingFunction() }
 }
 
-class NodeRegion instanceof Unit {
-  string toString() { result = "NodeRegion" }
-
-  predicate contains(Node n) { none() }
-
-  int totalOrder() { result = 1 }
-}
-
-predicate isUnreachableInCall(NodeRegion nr, DataFlowCall call) { none() } // stub implementation
+predicate isUnreachableInCall(Node n, DataFlowCall call) { none() } // stub implementation
 
 /**
  * Holds if access paths with `c` at their head always should be tracked at high
@@ -318,8 +290,6 @@ predicate knownSourceModel(Node source, string model) { none() }
 
 predicate knownSinkModel(Node sink, string model) { none() }
 
-class DataFlowSecondLevelScope = Unit;
-
 /**
  * Holds if flow is allowed to pass from parameter `p` and back to itself as a
  * side-effect, resulting in a summary from `p` to itself.

cpp/ql/lib/semmle/code/cpp/exprs/Expr.qll

@@ -1338,24 +1338,6 @@ class CoAwaitExpr extends UnaryOperation, @co_await {
   override string getOperator() { result = "co_await" }
 
   override int getPrecedence() { result = 16 }
-
-  /**
-   * Gets the Boolean expression that is used to decide if the enclosing
-   * coroutine should be suspended.
-   */
-  Expr getAwaitReady() { result = this.getChild(1) }
-
-  /**
-   * Gets the expression that represents the resume point if the enclosing
-   * coroutine was suspended.
-   */
-  Expr getAwaitResume() { result = this.getChild(2) }
-
-  /**
-   * Gets the expression that is evaluated when the enclosing coroutine is
-   * suspended.
-   */
-  Expr getAwaitSuspend() { result = this.getChild(3) }
 }
 
 /**
@@ -1370,24 +1352,6 @@ class CoYieldExpr extends UnaryOperation, @co_yield {
   override string getOperator() { result = "co_yield" }
 
   override int getPrecedence() { result = 2 }
-
-  /**
-   * Gets the Boolean expression that is used to decide if the enclosing
-   * coroutine should be suspended.
-   */
-  Expr getAwaitReady() { result = this.getChild(1) }
-
-  /**
-   * Gets the expression that represents the resume point if the enclosing
-   * coroutine was suspended.
-   */
-  Expr getAwaitResume() { result = this.getChild(2) }
-
-  /**
-   * Gets the expression that is evaluated when the enclosing coroutine is
-   * suspended.
-   */
-  Expr getAwaitSuspend() { result = this.getChild(3) }
 }
 
 /**

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

@@ -22,8 +22,6 @@ module CppDataFlow implements InputSig<Location> {
 
   predicate getAdditionalFlowIntoCallNodeTerm = Private::getAdditionalFlowIntoCallNodeTerm/2;
 
-  predicate getSecondLevelScope = Private::getSecondLevelScope/1;
-
   predicate validParameterAliasStep = Private::validParameterAliasStep/2;
 
   predicate mayBenefitFromCallContext = Private::mayBenefitFromCallContext/1;

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

@@ -1062,16 +1062,6 @@ class DataFlowCallable extends TDataFlowCallable {
     result = this.asSummarizedCallable() or // SummarizedCallable = Function (in CPP)
     result = this.asSourceCallable()
   }
-
-  /** Gets a best-effort total ordering. */
-  int totalorder() {
-    this =
-      rank[result](DataFlowCallable c, string file, int startline, int startcolumn |
-        c.getLocation().hasLocationInfo(file, startline, startcolumn, _, _)
-      |
-        c order by file, startline, startcolumn
-      )
-  }
 }
 
 /**
@@ -1169,16 +1159,6 @@ class DataFlowCall extends TDataFlowCall {
    * Gets the location of this call.
    */
   Location getLocation() { none() }
-
-  /** Gets a best-effort total ordering. */
-  int totalorder() {
-    this =
-      rank[result](DataFlowCall c, int startline, int startcolumn |
-        c.getLocation().hasLocationInfo(_, startline, startcolumn, _, _)
-      |
-        c order by startline, startcolumn
-      )
-  }
 }
 
 /**
@@ -1267,53 +1247,43 @@ module IsUnreachableInCall {
     any(G::IRGuardCondition guard).ensuresLt(left, right, k, block, areEqual)
   }
 
-  class NodeRegion instanceof IRBlock {
-    string toString() { result = "NodeRegion" }
-
-    predicate contains(Node n) { this = n.getBasicBlock() }
-
-    int totalOrder() {
-      this =
-        rank[result](IRBlock b, int startline, int startcolumn |
-          b.getLocation().hasLocationInfo(_, startline, startcolumn, _, _)
-        |
-          b order by startline, startcolumn
-        )
-    }
-  }
-
-  predicate isUnreachableInCall(NodeRegion block, DataFlowCall call) {
+  predicate isUnreachableInCall(Node n, DataFlowCall call) {
     exists(
       InstructionDirectParameterNode paramNode, ConstantIntegralTypeArgumentNode arg,
-      IntegerConstantInstruction constant, int k, Operand left, Operand right, int argval
+      IntegerConstantInstruction constant, int k, Operand left, Operand right, IRBlock block
     |
       // arg flows into `paramNode`
-      DataFlowImplCommon::viableParamArg(call, pragma[only_bind_into](paramNode),
-        pragma[only_bind_into](arg)) and
+      DataFlowImplCommon::viableParamArg(call, paramNode, arg) and
       left = constant.getAUse() and
       right = valueNumber(paramNode.getInstruction()).getAUse() and
-      argval = arg.getValue()
+      block = n.getBasicBlock()
     |
       // and there's a guard condition which ensures that the result of `left == right + k` is `areEqual`
-      exists(boolean areEqual | ensuresEq(left, right, k, block, areEqual) |
+      exists(boolean areEqual |
+        ensuresEq(pragma[only_bind_into](left), pragma[only_bind_into](right),
+          pragma[only_bind_into](k), pragma[only_bind_into](block), areEqual)
+      |
         // this block ensures that left = right + k, but it holds that `left != right + k`
         areEqual = true and
-        constant.getValue().toInt() != argval + k
+        constant.getValue().toInt() != arg.getValue() + k
         or
         // this block ensures that or `left != right + k`, but it holds that `left = right + k`
         areEqual = false and
-        constant.getValue().toInt() = argval + k
+        constant.getValue().toInt() = arg.getValue() + k
       )
       or
       // or there's a guard condition which ensures that the result of `left < right + k` is `isLessThan`
-      exists(boolean isLessThan | ensuresLt(left, right, k, block, isLessThan) |
+      exists(boolean isLessThan |
+        ensuresLt(pragma[only_bind_into](left), pragma[only_bind_into](right),
+          pragma[only_bind_into](k), pragma[only_bind_into](block), isLessThan)
+      |
         isLessThan = true and
         // this block ensures that `left < right + k`, but it holds that `left >= right + k`
-        constant.getValue().toInt() >= argval + k
+        constant.getValue().toInt() >= arg.getValue() + k
         or
         // this block ensures that `left >= right + k`, but it holds that `left < right + k`
         isLessThan = false and
-        constant.getValue().toInt() < argval + k
+        constant.getValue().toInt() < arg.getValue() + k
       )
     )
   }
@@ -1613,393 +1583,3 @@ predicate validParameterAliasStep(Node node1, Node node2) {
     )
   )
 }
-
-private predicate isTopLevel(Cpp::Stmt s) { any(Function f).getBlock().getAStmt() = s }
-
-private Cpp::Stmt getAChainedBranch(Cpp::IfStmt s) {
-  result = s.getThen()
-  or
-  exists(Cpp::Stmt elseBranch | s.getElse() = elseBranch |
-    result = getAChainedBranch(elseBranch)
-    or
-    result = elseBranch and not elseBranch instanceof Cpp::IfStmt
-  )
-}
-
-private Instruction getAnInstruction(Node n) {
-  result = n.asInstruction()
-  or
-  not n instanceof InstructionNode and
-  result = n.asOperand().getUse()
-  or
-  result = n.(SsaPhiNode).getPhiNode().getBasicBlock().getFirstInstruction()
-  or
-  n.(IndirectInstruction).hasInstructionAndIndirectionIndex(result, _)
-  or
-  not n instanceof IndirectInstruction and
-  exists(Operand operand |
-    n.(IndirectOperand).hasOperandAndIndirectionIndex(operand, _) and
-    result = operand.getUse()
-  )
-  or
-  result = getAnInstruction(n.(PostUpdateNode).getPreUpdateNode())
-}
-
-private newtype TDataFlowSecondLevelScope =
-  TTopLevelIfBranch(Cpp::Stmt s) {
-    exists(Cpp::IfStmt ifstmt | s = getAChainedBranch(ifstmt) and isTopLevel(ifstmt))
-  } or
-  TTopLevelSwitchCase(Cpp::SwitchCase s) {
-    exists(Cpp::SwitchStmt switchstmt | s = switchstmt.getASwitchCase() and isTopLevel(switchstmt))
-  }
-
-/**
- * A second-level control-flow scope in a `switch` or a chained `if` statement.
- *
- * This is a `switch` case or a branch of a chained `if` statement, given that
- * the `switch` or `if` statement is top level, that is, it is not nested inside
- * other CFG constructs.
- */
-class DataFlowSecondLevelScope extends TDataFlowSecondLevelScope {
-  /** Gets a textual representation of this element. */
-  string toString() {
-    exists(Cpp::Stmt s | this = TTopLevelIfBranch(s) | result = s.toString())
-    or
-    exists(Cpp::SwitchCase s | this = TTopLevelSwitchCase(s) | result = s.toString())
-  }
-
-  /** Gets the primary location of this element. */
-  Cpp::Location getLocation() {
-    exists(Cpp::Stmt s | this = TTopLevelIfBranch(s) | result = s.getLocation())
-    or
-    exists(Cpp::SwitchCase s | this = TTopLevelSwitchCase(s) | result = s.getLocation())
-  }
-
-  /**
-   * Gets a statement directly contained in this scope. For an `if` branch, this
-   * is the branch itself, and for a `switch case`, this is one the statements
-   * of that case branch.
-   */
-  private Cpp::Stmt getAStmt() {
-    exists(Cpp::Stmt s | this = TTopLevelIfBranch(s) | result = s)
-    or
-    exists(Cpp::SwitchCase s | this = TTopLevelSwitchCase(s) | result = s.getAStmt())
-  }
-
-  /** Gets a data-flow node nested within this scope. */
-  Node getANode() {
-    getAnInstruction(result).getAst().(Cpp::ControlFlowNode).getEnclosingStmt().getParentStmt*() =
-      this.getAStmt()
-  }
-}
-
-/** Gets the second-level scope containing the node `n`, if any. */
-DataFlowSecondLevelScope getSecondLevelScope(Node n) { result.getANode() = n }
-
-/**
- * Module that defines flow through iterators.
- * For example,
- * ```cpp
- * auto it = v.begin();
- * *it = source();
- * ...
- * sink(v[0]);
- * ```
- */
-module IteratorFlow {
-  private import codeql.ssa.Ssa as SsaImpl
-  private import semmle.code.cpp.models.interfaces.Iterator as Interface
-  private import semmle.code.cpp.models.implementations.Iterator as Impl
-
-  /**
-   * A variable of some type that can produce an iterator.
-   */
-  class SourceVariable extends Ssa::SourceVariable {
-    SourceVariable() {
-      exists(Interface::GetIteratorFunction gets, Cpp::FunctionInput input, int i |
-        input.isParameterDerefOrQualifierObject(i) and
-        gets.getsIterator(input, _)
-      |
-        this.getType().stripType() = gets.getParameter(i).getType().stripType()
-        or
-        i = -1 and
-        this.getType().stripType() = gets.getDeclaringType()
-      )
-    }
-  }
-
-  private module SsaInput implements SsaImpl::InputSig<Location> {
-    import Ssa::InputSigCommon
-
-    class SourceVariable = IteratorFlow::SourceVariable;
-
-    /** A call to function that dereferences an iterator. */
-    private class IteratorPointerDereferenceCall extends CallInstruction {
-      IteratorPointerDereferenceCall() {
-        this.getStaticCallTarget() instanceof Impl::IteratorPointerDereferenceOperator
-      }
-    }
-
-    /** A call to a function that obtains an iterator. */
-    private class GetsIteratorCall extends CallInstruction {
-      GetsIteratorCall() { this.getStaticCallTarget() instanceof Impl::GetIteratorFunction }
-    }
-
-    /** A call to `operator++` or `operator--` on an iterator. */
-    private class IteratorCrementCall extends CallInstruction {
-      IteratorCrementCall() { this.getStaticCallTarget() instanceof Impl::IteratorCrementOperator }
-    }
-
-    /**
-     * Gets an ultimate definition of `def`.
-     *
-     * Note: Unlike `def.getAnUltimateDefinition()` this predicate also
-     * traverses back through iterator increment and decrement operations.
-     */
-    private Ssa::Def getAnUltimateDefinition(Ssa::Def def) {
-      result = def.getAnUltimateDefinition()
-      or
-      exists(IRBlock bb, int i, IteratorCrementCall crementCall, Ssa::SourceVariable sv |
-        crementCall = def.getValue().asInstruction().(StoreInstruction).getSourceValue() and
-        sv = def.getSourceVariable() and
-        bb.getInstruction(i) = crementCall and
-        Ssa::ssaDefReachesRead(sv, result.asDef(), bb, i)
-      )
-    }
-
-    /**
-     * Holds if `write` is an instruction that writes to address `address`
-     */
-    private predicate isIteratorWrite(Instruction write, Operand address) {
-      exists(Ssa::DefImpl writeDef, IRBlock bb, int i |
-        writeDef.hasIndexInBlock(bb, i, _) and
-        bb.getInstruction(i) = write and
-        address = writeDef.getAddressOperand()
-      )
-    }
-
-    /**
-     * Holds if `writeToDeref` is a write to an iterator that was obtained
-     * by `beginCall`. That is, the following instruction sequence holds:
-     * ```cpp
-     * it = container.begin(); // or a similar iterator-obtaining function call
-     * ...
-     * *it = value;
-     * ```
-     */
-    private predicate isIteratorStoreInstruction(
-      GetsIteratorCall beginCall, Instruction writeToDeref
-    ) {
-      exists(
-        StoreInstruction beginStore, IRBlock bbStar, int iStar, Ssa::Def def,
-        IteratorPointerDereferenceCall starCall, Ssa::Def ultimate, Operand address
-      |
-        isIteratorWrite(writeToDeref, address) and
-        operandForFullyConvertedCall(address, starCall) and
-        bbStar.getInstruction(iStar) = starCall and
-        Ssa::ssaDefReachesRead(_, def.asDef(), bbStar, iStar) and
-        ultimate = getAnUltimateDefinition*(def) and
-        beginStore = ultimate.getValue().asInstruction() and
-        operandForFullyConvertedCall(beginStore.getSourceValueOperand(), beginCall)
-      )
-    }
-
-    /**
-     * Holds if `(bb, i)` contains a write to an iterator that may have been obtained
-     * by calling `begin` (or related functions) on the variable `v`.
-     */
-    predicate variableWrite(IRBlock bb, int i, SourceVariable v, boolean certain) {
-      certain = false and
-      exists(GetsIteratorCall beginCall, Instruction writeToDeref, IRBlock bbQual, int iQual |
-        isIteratorStoreInstruction(beginCall, writeToDeref) and
-        bb.getInstruction(i) = writeToDeref and
-        bbQual.getInstruction(iQual) = beginCall and
-        Ssa::variableRead(bbQual, iQual, v, _)
-      )
-    }
-
-    /** Holds if `(bb, i)` reads the container variable `v`. */
-    predicate variableRead(IRBlock bb, int i, SourceVariable v, boolean certain) {
-      Ssa::variableRead(bb, i, v, certain)
-    }
-  }
-
-  private module IteratorSsa = SsaImpl::Make<Location, SsaInput>;
-
-  cached
-  private newtype TSsaDef =
-    TDef(IteratorSsa::DefinitionExt def) or
-    TPhi(PhiNode phi)
-
-  abstract private class SsaDef extends TSsaDef {
-    /** Gets a textual representation of this element. */
-    string toString() { none() }
-
-    /** Gets the underlying non-phi definition or use. */
-    IteratorSsa::DefinitionExt asDef() { none() }
-
-    /** Gets the underlying phi node. */
-    PhiNode asPhi() { none() }
-
-    /** Gets the location of this element. */
-    abstract Location getLocation();
-  }
-
-  private class Def extends TDef, SsaDef {
-    IteratorSsa::DefinitionExt def;
-
-    Def() { this = TDef(def) }
-
-    final override IteratorSsa::DefinitionExt asDef() { result = def }
-
-    final override Location getLocation() { result = this.getImpl().getLocation() }
-
-    /** Gets the variable written to by this definition. */
-    final SourceVariable getSourceVariable() { result = def.getSourceVariable() }
-
-    override string toString() { result = def.toString() }
-
-    /**
-     * Holds if this definition (or use) has index `index` in block `block`,
-     * and is a definition (or use) of the variable `sv`.
-     */
-    predicate hasIndexInBlock(IRBlock block, int index, SourceVariable sv) {
-      def.definesAt(sv, block, index, _)
-    }
-
-    private Ssa::DefImpl getImpl() {
-      exists(IRBlock bb, int i |
-        this.hasIndexInBlock(bb, i, _) and
-        result.hasIndexInBlock(bb, i)
-      )
-    }
-
-    /** Gets the value written by this definition (i.e., the "right-hand side"). */
-    Node0Impl getValue() { result = this.getImpl().getValue() }
-
-    /** Gets the indirection index of this definition. */
-    int getIndirectionIndex() { result = this.getImpl().getIndirectionIndex() }
-  }
-
-  private class Phi extends TPhi, SsaDef {
-    PhiNode phi;
-
-    Phi() { this = TPhi(phi) }
-
-    final override PhiNode asPhi() { result = phi }
-
-    final override Location getLocation() { result = phi.getBasicBlock().getLocation() }
-
-    override string toString() { result = phi.toString() }
-
-    SsaIteratorNode getNode() { result.getIteratorFlowNode() = phi }
-  }
-
-  private class PhiNode extends IteratorSsa::DefinitionExt {
-    PhiNode() {
-      this instanceof IteratorSsa::PhiNode or
-      this instanceof IteratorSsa::PhiReadNode
-    }
-
-    SsaIteratorNode getNode() { result.getIteratorFlowNode() = this }
-  }
-
-  cached
-  private module IteratorSsaCached {
-    cached
-    predicate adjacentDefRead(IRBlock bb1, int i1, SourceVariable sv, IRBlock bb2, int i2) {
-      IteratorSsa::adjacentDefReadExt(_, sv, bb1, i1, bb2, i2)
-      or
-      exists(PhiNode phi |
-        IteratorSsa::lastRefRedefExt(_, sv, bb1, i1, phi) and
-        phi.definesAt(sv, bb2, i2, _)
-      )
-    }
-
-    cached
-    Node getAPriorDefinition(IteratorSsa::DefinitionExt next) {
-      exists(IRBlock bb, int i, SourceVariable sv, IteratorSsa::DefinitionExt def |
-        IteratorSsa::lastRefRedefExt(pragma[only_bind_into](def), pragma[only_bind_into](sv),
-          pragma[only_bind_into](bb), pragma[only_bind_into](i), next) and
-        nodeToDefOrUse(result, sv, bb, i, _)
-      )
-    }
-  }
-
-  /** The set of nodes necessary for iterator flow. */
-  class IteratorFlowNode instanceof PhiNode {
-    /** Gets a textual representation of this node. */
-    string toString() { result = super.toString() }
-
-    /** Gets the type of this node. */
-    DataFlowType getType() {
-      exists(Ssa::SourceVariable sv |
-        super.definesAt(sv, _, _, _) and
-        result = sv.getType()
-      )
-    }
-
-    /** Gets the `Declaration` that contains this block. */
-    Declaration getFunction() { result = super.getBasicBlock().getEnclosingFunction() }
-
-    /** Gets the locatino of this node. */
-    Location getLocation() { result = super.getBasicBlock().getLocation() }
-  }
-
-  private import IteratorSsaCached
-
-  private predicate defToNode(Node node, Def def, boolean uncertain) {
-    (
-      nodeHasOperand(node, def.getValue().asOperand(), def.getIndirectionIndex())
-      or
-      nodeHasInstruction(node, def.getValue().asInstruction(), def.getIndirectionIndex())
-    ) and
-    uncertain = false
-  }
-
-  private predicate nodeToDefOrUse(
-    Node node, SourceVariable sv, IRBlock bb, int i, boolean uncertain
-  ) {
-    exists(Def def |
-      def.hasIndexInBlock(bb, i, sv) and
-      defToNode(node, def, uncertain)
-    )
-    or
-    useToNode(bb, i, sv, node) and
-    uncertain = false
-  }
-
-  private predicate useToNode(IRBlock bb, int i, SourceVariable sv, Node nodeTo) {
-    exists(PhiNode phi |
-      phi.definesAt(sv, bb, i, _) and
-      nodeTo = phi.getNode()
-    )
-    or
-    exists(Ssa::UseImpl use |
-      use.hasIndexInBlock(bb, i, sv) and
-      nodeTo = use.getNode()
-    )
-  }
-
-  /**
-   * Holds if `nodeFrom` flows to `nodeTo` in a single step.
-   */
-  predicate localFlowStep(Node nodeFrom, Node nodeTo) {
-    exists(
-      Node nFrom, SourceVariable sv, IRBlock bb1, int i1, IRBlock bb2, int i2, boolean uncertain
-    |
-      adjacentDefRead(bb1, i1, sv, bb2, i2) and
-      nodeToDefOrUse(nFrom, sv, bb1, i1, uncertain) and
-      useToNode(bb2, i2, sv, nodeTo)
-    |
-      if uncertain = true
-      then
-        nodeFrom =
-          [
-            nFrom,
-            getAPriorDefinition(any(IteratorSsa::DefinitionExt next | next.definesAt(sv, bb1, i1, _)))
-          ]
-      else nFrom = nodeFrom
-    )
-  }
-}

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

@@ -46,7 +46,6 @@ private newtype TIRDataFlowNode =
     Ssa::isModifiableByCall(operand, indirectionIndex)
   } or
   TSsaPhiNode(Ssa::PhiNode phi) or
-  TSsaIteratorNode(IteratorFlow::IteratorFlowNode n) or
   TRawIndirectOperand0(Node0Impl node, int indirectionIndex) {
     Ssa::hasRawIndirectOperand(node.asOperand(), indirectionIndex)
   } or
@@ -654,30 +653,6 @@ class SsaPhiNode extends Node, TSsaPhiNode {
   predicate isPhiRead() { phi.isPhiRead() }
 }
 
-/**
- * INTERNAL: do not use.
- *
- * Dataflow nodes necessary for iterator flow
- */
-class SsaIteratorNode extends Node, TSsaIteratorNode {
-  IteratorFlow::IteratorFlowNode node;
-
-  SsaIteratorNode() { this = TSsaIteratorNode(node) }
-
-  /** Gets the phi node associated with this node. */
-  IteratorFlow::IteratorFlowNode getIteratorFlowNode() { result = node }
-
-  override Declaration getEnclosingCallable() { result = this.getFunction() }
-
-  override Declaration getFunction() { result = node.getFunction() }
-
-  override DataFlowType getType() { result = node.getType() }
-
-  final override Location getLocationImpl() { result = node.getLocation() }
-
-  override string toStringImpl() { result = node.toString() }
-}
-
 /**
  * INTERNAL: do not use.
  *
@@ -1215,11 +1190,11 @@ class UninitializedNode extends Node {
   LocalVariable v;
 
   UninitializedNode() {
-    exists(Ssa::Def def, Ssa::SourceVariable sv |
+    exists(Ssa::Def def |
       def.getIndirectionIndex() = 0 and
       def.getValue().asInstruction() instanceof UninitializedInstruction and
-      Ssa::defToNode(this, def, sv, _, _, _) and
-      v = sv.getBaseVariable().(Ssa::BaseIRVariable).getIRVariable().getAst()
+      Ssa::nodeToDefOrUse(this, def, _) and
+      v = def.getSourceVariable().getBaseVariable().(Ssa::BaseIRVariable).getIRVariable().getAst()
     )
   }
 
@@ -2176,8 +2151,6 @@ private module Cached {
       // Def-use/Use-use flow
       Ssa::ssaFlow(nodeFrom, nodeTo)
       or
-      IteratorFlow::localFlowStep(nodeFrom, nodeTo)
-      or
       // Operand -> Instruction flow
       simpleInstructionLocalFlowStep(nodeFrom.asOperand(), nodeTo.asInstruction())
       or

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

@@ -546,7 +546,7 @@ module ProductFlow {
       Flow1::PathGraph::edges(pred1, succ1, _, _) and
       exists(ReturnKindExt returnKind |
         succ1.getNode() = returnKind.getAnOutNode(call) and
-        paramReturnNode(_, pred1.asParameterReturnNode(), _, returnKind)
+        pred1.getNode().(ReturnNodeExt).getKind() = returnKind
       )
     }
 
@@ -574,7 +574,7 @@ module ProductFlow {
       Flow2::PathGraph::edges(pred2, succ2, _, _) and
       exists(ReturnKindExt returnKind |
         succ2.getNode() = returnKind.getAnOutNode(call) and
-        paramReturnNode(_, pred2.asParameterReturnNode(), _, returnKind)
+        pred2.getNode().(ReturnNodeExt).getKind() = returnKind
       )
     }
 

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

@@ -246,6 +246,14 @@ private module IteratorIndirections {
       baseType = super.getValueType()
     }
 
+    override predicate isAdditionalDereference(Instruction deref, Operand address) {
+      exists(CallInstruction call |
+        operandForFullyConvertedCall(getAUse(deref), call) and
+        this = call.getStaticCallTarget().getClassAndName("operator*") and
+        address = call.getThisArgumentOperand()
+      )
+    }
+
     override predicate isAdditionalWrite(Node0Impl value, Operand address, boolean certain) {
       exists(CallInstruction call | call.getArgumentOperand(0) = value.asOperand() |
         this = call.getStaticCallTarget().getClassAndName("operator=") and
@@ -254,6 +262,16 @@ private module IteratorIndirections {
       )
     }
 
+    override predicate isAdditionalTaintStep(Node node1, Node node2) {
+      exists(CallInstruction call |
+        // Taint through `operator+=` and `operator-=` on iterators.
+        call.getStaticCallTarget() instanceof Iterator::IteratorAssignArithmeticOperator and
+        node2.(IndirectArgumentOutNode).getPreUpdateNode() = node1 and
+        node1.(IndirectOperand).hasOperandAndIndirectionIndex(call.getArgumentOperand(0), _) and
+        node1.getType().getUnspecifiedType() = this
+      )
+    }
+
     override predicate isAdditionalConversionFlow(Operand opFrom, Instruction instrTo) {
       // This is a bit annoying: Consider the following snippet:
       // ```
@@ -571,6 +589,230 @@ private class BaseCallInstruction extends BaseSourceVariableInstruction, CallIns
 
 cached
 private module Cached {
+  private import semmle.code.cpp.models.interfaces.Iterator as Interfaces
+  private import semmle.code.cpp.models.implementations.Iterator as Iterator
+  private import semmle.code.cpp.models.interfaces.FunctionInputsAndOutputs as IO
+
+  /**
+   * Holds if `next` is a instruction with a memory result that potentially
+   * updates the memory produced by `prev`.
+   */
+  private predicate memorySucc(Instruction prev, Instruction next) {
+    prev = next.(ChiInstruction).getTotal()
+    or
+    // Phi inputs can be inexact.
+    prev = next.(PhiInstruction).getAnInputOperand().getAnyDef()
+    or
+    prev = next.(CopyInstruction).getSourceValue()
+    or
+    exists(ReadSideEffectInstruction read |
+      next = read.getPrimaryInstruction() and
+      isAdditionalConversionFlow(_, next) and
+      prev = read.getSideEffectOperand().getAnyDef()
+    )
+  }
+
+  /**
+   * Holds if `iteratorDerefAddress` is an address of an iterator dereference (i.e., `*it`)
+   * that is used for a write operation that writes the value `value`. The `memory` instruction
+   * represents the memory that the IR's SSA analysis determined was read by the call to `operator*`.
+   *
+   * The `numberOfLoads` integer represents the number of dereferences this write corresponds to
+   * on the underlying container that produced the iterator.
+   */
+  private predicate isChiAfterIteratorDef(
+    Instruction memory, Operand iteratorDerefAddress, Node0Impl value, int numberOfLoads
+  ) {
+    exists(
+      BaseSourceVariableInstruction iteratorBase, ReadSideEffectInstruction read,
+      Operand iteratorAddress
+    |
+      numberOfLoads >= 0 and
+      isDef(_, value, iteratorDerefAddress, iteratorBase, numberOfLoads + 2, 0) and
+      isUse(_, iteratorAddress, iteratorBase, numberOfLoads + 1, 0) and
+      iteratorBase.getResultType() instanceof Interfaces::Iterator and
+      isDereference(iteratorAddress.getDef(), read.getArgumentDef().getAUse(), _) and
+      memory = read.getSideEffectOperand().getAnyDef()
+    )
+  }
+
+  private predicate isSource(Instruction instr, Operand iteratorAddress, int numberOfLoads) {
+    getAUse(instr) = iteratorAddress and
+    exists(BaseSourceVariableInstruction iteratorBase |
+      iteratorBase.getResultType() instanceof Interfaces::Iterator and
+      not iteratorBase.getResultType() instanceof Cpp::PointerType and
+      isUse(_, iteratorAddress, iteratorBase, numberOfLoads - 1, 0)
+    )
+  }
+
+  private predicate isSink(Instruction instr, CallInstruction call) {
+    getAUse(instr).(ArgumentOperand).getCall() = call and
+    // Only include operations that may modify the object that the iterator points to.
+    // The following is a non-exhaustive list of things that may modify the value of the
+    // iterator, but never the value of what the iterator points to.
+    // The more things we can exclude here, the faster the small dataflow-like analysis
+    // done by `convertsIntoArgument` will converge.
+    not exists(Function f | f = call.getStaticCallTarget() |
+      f instanceof Iterator::IteratorCrementOperator or
+      f instanceof Iterator::IteratorBinaryArithmeticOperator or
+      f instanceof Iterator::IteratorAssignArithmeticOperator or
+      f instanceof Iterator::IteratorCrementMemberOperator or
+      f instanceof Iterator::IteratorBinaryArithmeticMemberOperator or
+      f instanceof Iterator::IteratorAssignArithmeticMemberOperator or
+      f instanceof Iterator::IteratorAssignmentMemberOperator
+    )
+  }
+
+  private predicate convertsIntoArgumentFwd(Instruction instr) {
+    isSource(instr, _, _)
+    or
+    exists(Instruction prev | convertsIntoArgumentFwd(prev) |
+      conversionFlow(unique( | | getAUse(prev)), instr, false, _)
+    )
+  }
+
+  private predicate convertsIntoArgumentRev(Instruction instr) {
+    convertsIntoArgumentFwd(instr) and
+    (
+      isSink(instr, _)
+      or
+      exists(Instruction next | convertsIntoArgumentRev(next) |
+        conversionFlow(unique( | | getAUse(instr)), next, false, _)
+      )
+    )
+  }
+
+  private predicate convertsIntoArgument(
+    Operand iteratorAddress, CallInstruction call, int numberOfLoads
+  ) {
+    exists(Instruction iteratorAddressDef |
+      isSource(iteratorAddressDef, iteratorAddress, numberOfLoads) and
+      isSink(iteratorAddressDef, call) and
+      convertsIntoArgumentRev(pragma[only_bind_into](iteratorAddressDef))
+    )
+  }
+
+  private predicate isChiAfterIteratorArgument(
+    Instruction memory, Operand iteratorAddress, int numberOfLoads
+  ) {
+    // Ideally, `iteratorAddress` would be an `ArgumentOperand`, but there might be
+    // various conversions applied to it before it becomes an argument.
+    // So we do a small amount of flow to find the call that the iterator is passed to.
+    exists(CallInstruction call | convertsIntoArgument(iteratorAddress, call, numberOfLoads) |
+      exists(ReadSideEffectInstruction read |
+        read.getPrimaryInstruction() = call and
+        read.getSideEffectOperand().getAnyDef() = memory
+      )
+      or
+      exists(LoadInstruction load |
+        iteratorAddress.getDef() = load and
+        memory = load.getSourceValueOperand().getAnyDef()
+      )
+    )
+  }
+
+  /**
+   * Holds if `iterator` is a `StoreInstruction` that stores the result of some function
+   * returning an iterator into an address computed started at `containerBase`.
+   *
+   * For example, given a declaration like `std::vector<int>::iterator it = v.begin()`,
+   * the `iterator` will be the `StoreInstruction` generated by the write to `it`, and
+   * `containerBase` will be the address of `v`.
+   */
+  private predicate isChiAfterBegin(
+    BaseSourceVariableInstruction containerBase, StoreInstruction iterator
+  ) {
+    exists(
+      CallInstruction getIterator, Iterator::GetIteratorFunction getIteratorFunction,
+      IO::FunctionInput input, int i
+    |
+      getIterator = iterator.getSourceValue() and
+      getIteratorFunction = getIterator.getStaticCallTarget() and
+      getIteratorFunction.getsIterator(input, _) and
+      isDef(_, any(Node0Impl n | n.asInstruction() = iterator), _, _, 1, 0) and
+      input.isParameterDerefOrQualifierObject(i) and
+      isUse(_, getIterator.getArgumentOperand(i), containerBase, 0, 0)
+    )
+  }
+
+  /**
+   * Holds if `iteratorAddress` is an address of an iterator that is used for
+   * a read operation. The `memory` instruction represents the memory that
+   * the IR's SSA analysis determined was read by the call to `operator*`.
+   *
+   * Finally, the `numberOfLoads` integer represents the number of dereferences
+   * this read corresponds to on the underlying container that produced the iterator.
+   */
+  private predicate isChiBeforeIteratorUse(
+    Operand iteratorAddress, Instruction memory, int numberOfLoads
+  ) {
+    exists(
+      BaseSourceVariableInstruction iteratorBase, LoadInstruction load,
+      ReadSideEffectInstruction read, Operand iteratorDerefAddress
+    |
+      numberOfLoads >= 0 and
+      isUse(_, iteratorAddress, iteratorBase, numberOfLoads + 1, 0) and
+      isUse(_, iteratorDerefAddress, iteratorBase, numberOfLoads + 2, 0) and
+      iteratorBase.getResultType() instanceof Interfaces::Iterator and
+      load.getSourceAddressOperand() = iteratorDerefAddress and
+      read.getPrimaryInstruction() = load.getSourceAddress() and
+      memory = read.getSideEffectOperand().getAnyDef()
+    )
+  }
+
+  /**
+   * Holds if `iteratorDerefAddress` is an address of an iterator dereference (i.e., `*it`)
+   * that is used for a write operation that writes the value `value` to a container that
+   * created the iterator. `container` represents the base of the address of the container
+   * that was used to create the iterator.
+   */
+  cached
+  predicate isIteratorDef(
+    BaseSourceVariableInstruction container, Operand iteratorDerefAddress, Node0Impl value,
+    int numberOfLoads, int indirectionIndex
+  ) {
+    exists(Instruction memory, Instruction begin, int upper, int ind |
+      isChiAfterIteratorDef(memory, iteratorDerefAddress, value, numberOfLoads) and
+      memorySucc*(begin, memory) and
+      isChiAfterBegin(container, begin) and
+      upper = countIndirectionsForCppType(getResultLanguageType(container)) and
+      ind = numberOfLoads + [1 .. upper] and
+      indirectionIndex = ind - (numberOfLoads + 1)
+    )
+  }
+
+  /**
+   * Holds if `iteratorAddress` is an address of an iterator that is used for a
+   * read operation to read a value from a container that created the iterator.
+   * `container` represents the base of the address of the container that was used
+   * to create the iterator.
+   */
+  cached
+  predicate isIteratorUse(
+    BaseSourceVariableInstruction container, Operand iteratorAddress, int numberOfLoads,
+    int indirectionIndex
+  ) {
+    // Direct use
+    exists(Instruction begin, Instruction memory, int upper, int ind |
+      isChiBeforeIteratorUse(iteratorAddress, memory, numberOfLoads) and
+      memorySucc*(begin, memory) and
+      isChiAfterBegin(container, begin) and
+      upper = countIndirectionsForCppType(getResultLanguageType(container)) and
+      ind = numberOfLoads + [1 .. upper] and
+      indirectionIndex = ind - (numberOfLoads + 1)
+    )
+    or
+    // Use through function output
+    exists(Instruction memory, Instruction begin, int upper, int ind |
+      isChiAfterIteratorArgument(memory, iteratorAddress, numberOfLoads) and
+      memorySucc*(begin, memory) and
+      isChiAfterBegin(container, begin) and
+      upper = countIndirectionsForCppType(getResultLanguageType(container)) and
+      ind = numberOfLoads + [1 .. upper] and
+      indirectionIndex = ind - (numberOfLoads - 1)
+    )
+  }
+
   /** Holds if `op` is the only use of its defining instruction, and that op is used in a conversation */
   private predicate isConversion(Operand op) {
     exists(Instruction def, Operand use |

cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/IRVariable.qll

@@ -17,11 +17,18 @@ private import Imports::IRType
  * The variable may be a user-declared variable (`IRUserVariable`) or a temporary variable generated
  * by the AST-to-IR translation (`IRTempVariable`).
  */
-abstract private class AbstractIRVariable extends TIRVariable {
+class IRVariable extends TIRVariable {
   Language::Declaration func;
 
+  IRVariable() {
+    this = TIRUserVariable(_, _, func) or
+    this = TIRTempVariable(func, _, _, _) or
+    this = TIRStringLiteral(func, _, _, _) or
+    this = TIRDynamicInitializationFlag(func, _, _)
+  }
+
   /** Gets a textual representation of this element. */
-  abstract string toString();
+  string toString() { none() }
 
   /**
    * Holds if this variable's value cannot be changed within a function. Currently used for string
@@ -42,13 +49,13 @@ abstract private class AbstractIRVariable extends TIRVariable {
   /**
    * Gets the type of the variable.
    */
-  abstract Language::LanguageType getLanguageType();
+  Language::LanguageType getLanguageType() { none() }
 
   /**
    * Gets the AST node that declared this variable, or that introduced this
    * variable as part of the AST-to-IR translation.
    */
-  abstract Language::AST getAst();
+  Language::AST getAst() { none() }
 
   /** DEPRECATED: Alias for getAst */
   deprecated Language::AST getAST() { result = this.getAst() }
@@ -57,7 +64,7 @@ abstract private class AbstractIRVariable extends TIRVariable {
    * Gets an identifier string for the variable. This identifier is unique
    * within the function.
    */
-  abstract string getUniqueId();
+  string getUniqueId() { none() }
 
   /**
    * Gets the source location of this variable.
@@ -67,7 +74,7 @@ abstract private class AbstractIRVariable extends TIRVariable {
   /**
    * Gets the IR for the function that references this variable.
    */
-  final IRFunction getEnclosingIRFunction() { result.getFunction() = this.getEnclosingFunction() }
+  final IRFunction getEnclosingIRFunction() { result.getFunction() = func }
 
   /**
    * Gets the function that references this variable.
@@ -75,18 +82,10 @@ abstract private class AbstractIRVariable extends TIRVariable {
   final Language::Declaration getEnclosingFunction() { result = func }
 }
 
-/**
- * A variable referenced by the IR for a function.
- *
- * The variable may be a user-declared variable (`IRUserVariable`) or a temporary variable generated
- * by the AST-to-IR translation (`IRTempVariable`).
- */
-final class IRVariable = AbstractIRVariable;
-
 /**
  * A user-declared variable referenced by the IR for a function.
  */
-class IRUserVariable extends AbstractIRVariable, TIRUserVariable {
+class IRUserVariable extends IRVariable, TIRUserVariable {
   Language::Variable var;
   Language::LanguageType type;
 
@@ -115,29 +114,26 @@ class IRUserVariable extends AbstractIRVariable, TIRUserVariable {
  * A variable (user-declared or temporary) that is allocated on the stack. This includes all
  * parameters, non-static local variables, and temporary variables.
  */
-abstract private class AbstractIRAutomaticVariable extends AbstractIRVariable { }
-
-/**
- * A variable (user-declared or temporary) that is allocated on the stack. This includes all
- * parameters, non-static local variables, and temporary variables.
- */
-final class IRAutomaticVariable = AbstractIRAutomaticVariable;
-
-/**
- * A user-declared variable that is allocated on the stack. This includes all parameters and
- * non-static local variables.
- */
-private class AbstractIRAutomaticUserVariable extends IRUserVariable, AbstractIRAutomaticVariable {
-  override Language::AutomaticVariable var;
-
-  final override Language::AutomaticVariable getVariable() { result = var }
+class IRAutomaticVariable extends IRVariable {
+  IRAutomaticVariable() {
+    exists(Language::Variable var |
+      this = TIRUserVariable(var, _, func) and
+      Language::isVariableAutomatic(var)
+    )
+    or
+    this = TIRTempVariable(func, _, _, _)
+  }
 }
 
 /**
  * A user-declared variable that is allocated on the stack. This includes all parameters and
  * non-static local variables.
  */
-final class IRAutomaticUserVariable = AbstractIRAutomaticUserVariable;
+class IRAutomaticUserVariable extends IRUserVariable, IRAutomaticVariable {
+  override Language::AutomaticVariable var;
+
+  final override Language::AutomaticVariable getVariable() { result = var }
+}
 
 /**
  * A user-declared variable that is not allocated on the stack. This includes all global variables,
@@ -155,10 +151,16 @@ class IRStaticUserVariable extends IRUserVariable {
  * A variable that is not user-declared. This includes temporary variables generated as part of IR
  * construction, as well as string literals.
  */
-abstract private class AbstractIRGeneratedVariable extends AbstractIRVariable {
+class IRGeneratedVariable extends IRVariable {
   Language::AST ast;
   Language::LanguageType type;
 
+  IRGeneratedVariable() {
+    this = TIRTempVariable(func, ast, _, type) or
+    this = TIRStringLiteral(func, ast, type, _) or
+    this = TIRDynamicInitializationFlag(func, ast, type)
+  }
+
   final override Language::LanguageType getLanguageType() { result = type }
 
   final override Language::AST getAst() { result = ast }
@@ -194,20 +196,12 @@ abstract private class AbstractIRGeneratedVariable extends AbstractIRVariable {
   string getBaseString() { none() }
 }
 
-/**
- * A variable that is not user-declared. This includes temporary variables generated as part of IR
- * construction, as well as string literals.
- */
-final class IRGeneratedVariable = AbstractIRGeneratedVariable;
-
 /**
  * A temporary variable introduced by IR construction. The most common examples are the variable
  * generated to hold the return value of a function, or the variable generated to hold the result of
  * a condition operator (`a ? b : c`).
  */
-class IRTempVariable extends AbstractIRGeneratedVariable, AbstractIRAutomaticVariable,
-  TIRTempVariable
-{
+class IRTempVariable extends IRGeneratedVariable, IRAutomaticVariable, TIRTempVariable {
   TempVariableTag tag;
 
   IRTempVariable() { this = TIRTempVariable(func, ast, tag, type) }
@@ -247,7 +241,7 @@ class IRThrowVariable extends IRTempVariable {
  * A temporary variable generated to hold the contents of all arguments passed to the `...` of a
  * function that accepts a variable number of arguments.
  */
-class IREllipsisVariable extends IRTempVariable, AbstractIRParameter {
+class IREllipsisVariable extends IRTempVariable, IRParameter {
   IREllipsisVariable() { tag = EllipsisTempVar() }
 
   final override string toString() { result = "#ellipsis" }
@@ -258,7 +252,7 @@ class IREllipsisVariable extends IRTempVariable, AbstractIRParameter {
 /**
  * A temporary variable generated to hold the `this` pointer.
  */
-class IRThisVariable extends IRTempVariable, AbstractIRParameter {
+class IRThisVariable extends IRTempVariable, IRParameter {
   IRThisVariable() { tag = ThisTempVar() }
 
   final override string toString() { result = "#this" }
@@ -270,7 +264,7 @@ class IRThisVariable extends IRTempVariable, AbstractIRParameter {
  * A variable generated to represent the contents of a string literal. This variable acts much like
  * a read-only global variable.
  */
-class IRStringLiteral extends AbstractIRGeneratedVariable, TIRStringLiteral {
+class IRStringLiteral extends IRGeneratedVariable, TIRStringLiteral {
   Language::StringLiteral literal;
 
   IRStringLiteral() { this = TIRStringLiteral(func, ast, type, literal) }
@@ -294,7 +288,7 @@ class IRStringLiteral extends AbstractIRGeneratedVariable, TIRStringLiteral {
  * used to model the runtime initialization of static local variables in C++, as well as static
  * fields in C#.
  */
-class IRDynamicInitializationFlag extends AbstractIRGeneratedVariable, TIRDynamicInitializationFlag {
+class IRDynamicInitializationFlag extends IRGeneratedVariable, TIRDynamicInitializationFlag {
   Language::Variable var;
 
   IRDynamicInitializationFlag() {
@@ -320,24 +314,24 @@ class IRDynamicInitializationFlag extends AbstractIRGeneratedVariable, TIRDynami
  * An IR variable which acts like a function parameter, including positional parameters and the
  * temporary variables generated for `this` and ellipsis parameters.
  */
-abstract private class AbstractIRParameter extends AbstractIRAutomaticVariable {
+class IRParameter extends IRAutomaticVariable {
+  IRParameter() {
+    this.(IRAutomaticUserVariable).getVariable() instanceof Language::Parameter
+    or
+    this = TIRTempVariable(_, _, ThisTempVar(), _)
+    or
+    this = TIRTempVariable(_, _, EllipsisTempVar(), _)
+  }
+
   /**
    * Gets the zero-based index of this parameter. The `this` parameter has index -1.
    */
   int getIndex() { none() }
 }
 
-/**
- * An IR variable which acts like a function parameter, including positional parameters and the
- * temporary variables generated for `this` and ellipsis parameters.
- */
-final class IRParameter = AbstractIRParameter;
-
 /**
  * An IR variable representing a positional parameter.
  */
-class IRPositionalParameter extends AbstractIRParameter, AbstractIRAutomaticUserVariable {
-  IRPositionalParameter() { this.getVariable() instanceof Language::Parameter }
-
+class IRPositionalParameter extends IRParameter, IRAutomaticUserVariable {
   final override int getIndex() { result = this.getVariable().(Language::Parameter).getIndex() }
 }

cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/Instruction.qll

@@ -247,7 +247,8 @@ class Instruction extends Construction::TStageInstruction {
    * Gets the type of the result produced by this instruction. If the instruction does not produce
    * a result, its result type will be `IRVoidType`.
    */
-  final IRType getResultIRType() { result = Construction::getInstructionResultIRType(this) }
+  cached
+  final IRType getResultIRType() { result = this.getResultLanguageType().getIRType() }
 
   /**
    * Gets the type of the result produced by this instruction. If the
@@ -994,8 +995,9 @@ class ConstantInstruction extends ConstantValueInstruction {
  */
 class IntegerConstantInstruction extends ConstantInstruction {
   IntegerConstantInstruction() {
-    exists(IRType resultType | resultType = this.getResultIRType() |
-      resultType instanceof IRIntegerType or resultType instanceof IRBooleanType
+    exists(IRType resultType |
+      resultType = this.getResultIRType() and
+      (resultType instanceof IRIntegerType or resultType instanceof IRBooleanType)
     )
   }
 }
@@ -1007,17 +1009,6 @@ class FloatConstantInstruction extends ConstantInstruction {
   FloatConstantInstruction() { this.getResultIRType() instanceof IRFloatingPointType }
 }
 
-/**
- * An instruction whose result is a constant value of a pointer type.
- */
-class PointerConstantInstruction extends ConstantInstruction {
-  PointerConstantInstruction() {
-    exists(IRType resultType | resultType = this.getResultIRType() |
-      resultType instanceof IRAddressType or resultType instanceof IRFunctionAddressType
-    )
-  }
-}
-
 /**
  * An instruction whose result is the address of a string literal.
  */

cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/SSAConstruction.qll

@@ -429,11 +429,6 @@ private module Cached {
     instr = unreachedInstruction(_) and result = Language::getVoidType()
   }
 
-  cached
-  IRType getInstructionResultIRType(Instruction instr) {
-    result = instr.getResultLanguageType().getIRType()
-  }
-
   /**
    * Holds if `opcode` is the opcode that specifies the operation performed by `instr`.
    *

cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/IRVariable.qll

@@ -17,11 +17,18 @@ private import Imports::IRType
  * The variable may be a user-declared variable (`IRUserVariable`) or a temporary variable generated
  * by the AST-to-IR translation (`IRTempVariable`).
  */
-abstract private class AbstractIRVariable extends TIRVariable {
+class IRVariable extends TIRVariable {
   Language::Declaration func;
 
+  IRVariable() {
+    this = TIRUserVariable(_, _, func) or
+    this = TIRTempVariable(func, _, _, _) or
+    this = TIRStringLiteral(func, _, _, _) or
+    this = TIRDynamicInitializationFlag(func, _, _)
+  }
+
   /** Gets a textual representation of this element. */
-  abstract string toString();
+  string toString() { none() }
 
   /**
    * Holds if this variable's value cannot be changed within a function. Currently used for string
@@ -42,13 +49,13 @@ abstract private class AbstractIRVariable extends TIRVariable {
   /**
    * Gets the type of the variable.
    */
-  abstract Language::LanguageType getLanguageType();
+  Language::LanguageType getLanguageType() { none() }
 
   /**
    * Gets the AST node that declared this variable, or that introduced this
    * variable as part of the AST-to-IR translation.
    */
-  abstract Language::AST getAst();
+  Language::AST getAst() { none() }
 
   /** DEPRECATED: Alias for getAst */
   deprecated Language::AST getAST() { result = this.getAst() }
@@ -57,7 +64,7 @@ abstract private class AbstractIRVariable extends TIRVariable {
    * Gets an identifier string for the variable. This identifier is unique
    * within the function.
    */
-  abstract string getUniqueId();
+  string getUniqueId() { none() }
 
   /**
    * Gets the source location of this variable.
@@ -67,7 +74,7 @@ abstract private class AbstractIRVariable extends TIRVariable {
   /**
    * Gets the IR for the function that references this variable.
    */
-  final IRFunction getEnclosingIRFunction() { result.getFunction() = this.getEnclosingFunction() }
+  final IRFunction getEnclosingIRFunction() { result.getFunction() = func }
 
   /**
    * Gets the function that references this variable.
@@ -75,18 +82,10 @@ abstract private class AbstractIRVariable extends TIRVariable {
   final Language::Declaration getEnclosingFunction() { result = func }
 }
 
-/**
- * A variable referenced by the IR for a function.
- *
- * The variable may be a user-declared variable (`IRUserVariable`) or a temporary variable generated
- * by the AST-to-IR translation (`IRTempVariable`).
- */
-final class IRVariable = AbstractIRVariable;
-
 /**
  * A user-declared variable referenced by the IR for a function.
  */
-class IRUserVariable extends AbstractIRVariable, TIRUserVariable {
+class IRUserVariable extends IRVariable, TIRUserVariable {
   Language::Variable var;
   Language::LanguageType type;
 
@@ -115,29 +114,26 @@ class IRUserVariable extends AbstractIRVariable, TIRUserVariable {
  * A variable (user-declared or temporary) that is allocated on the stack. This includes all
  * parameters, non-static local variables, and temporary variables.
  */
-abstract private class AbstractIRAutomaticVariable extends AbstractIRVariable { }
-
-/**
- * A variable (user-declared or temporary) that is allocated on the stack. This includes all
- * parameters, non-static local variables, and temporary variables.
- */
-final class IRAutomaticVariable = AbstractIRAutomaticVariable;
-
-/**
- * A user-declared variable that is allocated on the stack. This includes all parameters and
- * non-static local variables.
- */
-private class AbstractIRAutomaticUserVariable extends IRUserVariable, AbstractIRAutomaticVariable {
-  override Language::AutomaticVariable var;
-
-  final override Language::AutomaticVariable getVariable() { result = var }
+class IRAutomaticVariable extends IRVariable {
+  IRAutomaticVariable() {
+    exists(Language::Variable var |
+      this = TIRUserVariable(var, _, func) and
+      Language::isVariableAutomatic(var)
+    )
+    or
+    this = TIRTempVariable(func, _, _, _)
+  }
 }
 
 /**
  * A user-declared variable that is allocated on the stack. This includes all parameters and
  * non-static local variables.
  */
-final class IRAutomaticUserVariable = AbstractIRAutomaticUserVariable;
+class IRAutomaticUserVariable extends IRUserVariable, IRAutomaticVariable {
+  override Language::AutomaticVariable var;
+
+  final override Language::AutomaticVariable getVariable() { result = var }
+}
 
 /**
  * A user-declared variable that is not allocated on the stack. This includes all global variables,
@@ -155,10 +151,16 @@ class IRStaticUserVariable extends IRUserVariable {
  * A variable that is not user-declared. This includes temporary variables generated as part of IR
  * construction, as well as string literals.
  */
-abstract private class AbstractIRGeneratedVariable extends AbstractIRVariable {
+class IRGeneratedVariable extends IRVariable {
   Language::AST ast;
   Language::LanguageType type;
 
+  IRGeneratedVariable() {
+    this = TIRTempVariable(func, ast, _, type) or
+    this = TIRStringLiteral(func, ast, type, _) or
+    this = TIRDynamicInitializationFlag(func, ast, type)
+  }
+
   final override Language::LanguageType getLanguageType() { result = type }
 
   final override Language::AST getAst() { result = ast }
@@ -194,20 +196,12 @@ abstract private class AbstractIRGeneratedVariable extends AbstractIRVariable {
   string getBaseString() { none() }
 }
 
-/**
- * A variable that is not user-declared. This includes temporary variables generated as part of IR
- * construction, as well as string literals.
- */
-final class IRGeneratedVariable = AbstractIRGeneratedVariable;
-
 /**
  * A temporary variable introduced by IR construction. The most common examples are the variable
  * generated to hold the return value of a function, or the variable generated to hold the result of
  * a condition operator (`a ? b : c`).
  */
-class IRTempVariable extends AbstractIRGeneratedVariable, AbstractIRAutomaticVariable,
-  TIRTempVariable
-{
+class IRTempVariable extends IRGeneratedVariable, IRAutomaticVariable, TIRTempVariable {
   TempVariableTag tag;
 
   IRTempVariable() { this = TIRTempVariable(func, ast, tag, type) }
@@ -247,7 +241,7 @@ class IRThrowVariable extends IRTempVariable {
  * A temporary variable generated to hold the contents of all arguments passed to the `...` of a
  * function that accepts a variable number of arguments.
  */
-class IREllipsisVariable extends IRTempVariable, AbstractIRParameter {
+class IREllipsisVariable extends IRTempVariable, IRParameter {
   IREllipsisVariable() { tag = EllipsisTempVar() }
 
   final override string toString() { result = "#ellipsis" }
@@ -258,7 +252,7 @@ class IREllipsisVariable extends IRTempVariable, AbstractIRParameter {
 /**
  * A temporary variable generated to hold the `this` pointer.
  */
-class IRThisVariable extends IRTempVariable, AbstractIRParameter {
+class IRThisVariable extends IRTempVariable, IRParameter {
   IRThisVariable() { tag = ThisTempVar() }
 
   final override string toString() { result = "#this" }
@@ -270,7 +264,7 @@ class IRThisVariable extends IRTempVariable, AbstractIRParameter {
  * A variable generated to represent the contents of a string literal. This variable acts much like
  * a read-only global variable.
  */
-class IRStringLiteral extends AbstractIRGeneratedVariable, TIRStringLiteral {
+class IRStringLiteral extends IRGeneratedVariable, TIRStringLiteral {
   Language::StringLiteral literal;
 
   IRStringLiteral() { this = TIRStringLiteral(func, ast, type, literal) }
@@ -294,7 +288,7 @@ class IRStringLiteral extends AbstractIRGeneratedVariable, TIRStringLiteral {
  * used to model the runtime initialization of static local variables in C++, as well as static
  * fields in C#.
  */
-class IRDynamicInitializationFlag extends AbstractIRGeneratedVariable, TIRDynamicInitializationFlag {
+class IRDynamicInitializationFlag extends IRGeneratedVariable, TIRDynamicInitializationFlag {
   Language::Variable var;
 
   IRDynamicInitializationFlag() {
@@ -320,24 +314,24 @@ class IRDynamicInitializationFlag extends AbstractIRGeneratedVariable, TIRDynami
  * An IR variable which acts like a function parameter, including positional parameters and the
  * temporary variables generated for `this` and ellipsis parameters.
  */
-abstract private class AbstractIRParameter extends AbstractIRAutomaticVariable {
+class IRParameter extends IRAutomaticVariable {
+  IRParameter() {
+    this.(IRAutomaticUserVariable).getVariable() instanceof Language::Parameter
+    or
+    this = TIRTempVariable(_, _, ThisTempVar(), _)
+    or
+    this = TIRTempVariable(_, _, EllipsisTempVar(), _)
+  }
+
   /**
    * Gets the zero-based index of this parameter. The `this` parameter has index -1.
    */
   int getIndex() { none() }
 }
 
-/**
- * An IR variable which acts like a function parameter, including positional parameters and the
- * temporary variables generated for `this` and ellipsis parameters.
- */
-final class IRParameter = AbstractIRParameter;
-
 /**
  * An IR variable representing a positional parameter.
  */
-class IRPositionalParameter extends AbstractIRParameter, AbstractIRAutomaticUserVariable {
-  IRPositionalParameter() { this.getVariable() instanceof Language::Parameter }
-
+class IRPositionalParameter extends IRParameter, IRAutomaticUserVariable {
   final override int getIndex() { result = this.getVariable().(Language::Parameter).getIndex() }
 }

cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/Instruction.qll

@@ -247,7 +247,8 @@ class Instruction extends Construction::TStageInstruction {
    * Gets the type of the result produced by this instruction. If the instruction does not produce
    * a result, its result type will be `IRVoidType`.
    */
-  final IRType getResultIRType() { result = Construction::getInstructionResultIRType(this) }
+  cached
+  final IRType getResultIRType() { result = this.getResultLanguageType().getIRType() }
 
   /**
    * Gets the type of the result produced by this instruction. If the
@@ -994,8 +995,9 @@ class ConstantInstruction extends ConstantValueInstruction {
  */
 class IntegerConstantInstruction extends ConstantInstruction {
   IntegerConstantInstruction() {
-    exists(IRType resultType | resultType = this.getResultIRType() |
-      resultType instanceof IRIntegerType or resultType instanceof IRBooleanType
+    exists(IRType resultType |
+      resultType = this.getResultIRType() and
+      (resultType instanceof IRIntegerType or resultType instanceof IRBooleanType)
     )
   }
 }
@@ -1007,17 +1009,6 @@ class FloatConstantInstruction extends ConstantInstruction {
   FloatConstantInstruction() { this.getResultIRType() instanceof IRFloatingPointType }
 }
 
-/**
- * An instruction whose result is a constant value of a pointer type.
- */
-class PointerConstantInstruction extends ConstantInstruction {
-  PointerConstantInstruction() {
-    exists(IRType resultType | resultType = this.getResultIRType() |
-      resultType instanceof IRAddressType or resultType instanceof IRFunctionAddressType
-    )
-  }
-}
-
 /**
  * An instruction whose result is the address of a string literal.
  */

cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/IRConstruction.qll

@@ -377,10 +377,6 @@ CppType getInstructionResultType(TStageInstruction instr) {
   result = getVoidType()
 }
 
-IRType getInstructionResultIRType(Instruction instr) {
-  result = instr.getResultLanguageType().getIRType()
-}
-
 predicate getInstructionOpcode(Opcode opcode, TStageInstruction instr) {
   getInstructionTranslatedElement(instr).hasInstruction(opcode, getInstructionTag(instr), _)
   or

cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/InstructionTag.qll

@@ -89,8 +89,7 @@ newtype TInstructionTag =
   ImplicitDestructorTag(int index) {
     exists(Expr e | exists(e.getImplicitDestructorCall(index))) or
     exists(Stmt s | exists(s.getImplicitDestructorCall(index)))
-  } or
-  CoAwaitBranchTag()
+  }
 
 class InstructionTag extends TInstructionTag {
   final string toString() { result = getInstructionTagId(this) }
@@ -187,8 +186,6 @@ string getInstructionTagId(TInstructionTag tag) {
   or
   tag = BoolConversionCompareTag() and result = "BoolConvComp"
   or
-  tag = ResultCopyTag() and result = "ResultCopy"
-  or
   tag = LoadTag() and result = "Load" // Implicit load due to lvalue-to-rvalue conversion
   or
   tag = CatchTag() and result = "Catch"
@@ -266,6 +263,4 @@ string getInstructionTagId(TInstructionTag tag) {
   exists(int index |
     tag = ImplicitDestructorTag(index) and result = "ImplicitDestructor(" + index + ")"
   )
-  or
-  tag = CoAwaitBranchTag() and result = "CoAwaitBranch"
 }

cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/TranslatedElement.qll

@@ -830,12 +830,6 @@ newtype TTranslatedElement =
       not ignoreExpr(dc)
     )
   } or
-  // The set of destructors to invoke after a handler for a `try` statement. These
-  // need to be special cased because the destructors need to run following an
-  // `ExceptionEdge`, but not following a `GotoEdge` edge.
-  TTranslatedDestructorsAfterHandler(Handler handler) {
-    exists(handler.getAnImplicitDestructorCall())
-  } or
   // A precise side effect of an argument to a `Call`
   TTranslatedArgumentExprSideEffect(Call call, Expr expr, int n, SideEffectOpcode opcode) {
     not ignoreExpr(expr) and

cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/TranslatedExpr.qll

@@ -538,11 +538,6 @@ class TranslatedResultCopy extends TranslatedExpr, TTranslatedResultCopy {
   final override predicate producesExprResult() { any() }
 
   private TranslatedCoreExpr getOperand() { result.getExpr() = expr }
-
-  override predicate handlesDestructorsExplicitly() {
-    // The destructor calls will already have been generated by the translation of `expr`.
-    any()
-  }
 }
 
 class TranslatedCommaExpr extends TranslatedNonConstantExpr {
@@ -1264,7 +1259,9 @@ class TranslatedUnaryExpr extends TranslatedSingleInstructionExpr {
     expr instanceof NotExpr or
     expr instanceof ComplementExpr or
     expr instanceof UnaryPlusExpr or
-    expr instanceof UnaryMinusExpr
+    expr instanceof UnaryMinusExpr or
+    expr instanceof CoAwaitExpr or
+    expr instanceof CoYieldExpr
   }
 
   final override Instruction getFirstInstruction(EdgeKind kind) {
@@ -1304,6 +1301,12 @@ class TranslatedUnaryExpr extends TranslatedSingleInstructionExpr {
     expr instanceof UnaryPlusExpr and result instanceof Opcode::CopyValue
     or
     expr instanceof UnaryMinusExpr and result instanceof Opcode::Negate
+    or
+    // TODO: Use a new opcode to represent "awaiting the value"
+    expr instanceof CoAwaitExpr and result instanceof Opcode::CopyValue
+    or
+    // TODO: Use a new opcode to represent "awaiting the value"
+    expr instanceof CoYieldExpr and result instanceof Opcode::CopyValue
   }
 
   private TranslatedExpr getOperand() {
@@ -1311,146 +1314,6 @@ class TranslatedUnaryExpr extends TranslatedSingleInstructionExpr {
   }
 }
 
-/**
- * IR translation of a `co_await` or `co_yield` expression.
- *
- * The translation of `x = co_await ...` is essentially:
- * ```cpp
- * if (!awaiter.await_ready()) {
- *   awaiter.await_suspend();
- * }
- * x = awaiter.await_resume();
- * ```
- * where `awaiter` is an object constructed from programmer-supplied
- * input, and for IR construction purposes these are resolved by the C/C++
- * front-end.
- *
- * See https://en.cppreference.com/w/cpp/language/coroutines#co_await for the
- * specification on how `awaiter` is obtained.
- */
-abstract private class TranslatedCoExpr extends TranslatedNonConstantExpr {
-  /** Gets the operand of this operation. */
-  abstract Expr getOperand();
-
-  /**
-   * Gets the expression that decides if the enclosing coroutine should be
-   * suspended.
-   */
-  abstract Expr getAwaitReady();
-
-  /**
-   * Gets the expression that is evaluated when the enclosing coroutine is
-   * suspended.
-   */
-  abstract Expr getAwaitSuspend();
-
-  /**
-   * Gets the expression that represents the resume point if the enclosing
-   * coroutine was suspended.
-   */
-  abstract Expr getAwaitResume();
-
-  final override Instruction getFirstInstruction(EdgeKind kind) {
-    result = this.getTranslatedOperand().getFirstInstruction(kind)
-  }
-
-  override Instruction getALastInstructionInternal() {
-    result = this.getTranslatedAwaitResume().getALastInstruction()
-  }
-
-  final override TranslatedElement getChildInternal(int id) {
-    id = 0 and result = this.getTranslatedOperand()
-    or
-    id = 1 and result = this.getTranslatedAwaitReady()
-    or
-    id = 2 and result = this.getTranslatedAwaitResume()
-    or
-    id = 3 and result = this.getTranslatedAwaitSuspend()
-  }
-
-  final override Instruction getInstructionSuccessorInternal(InstructionTag tag, EdgeKind kind) {
-    tag = CoAwaitBranchTag() and
-    (
-      kind instanceof TrueEdge and
-      result = this.getTranslatedAwaitResume().getFirstInstruction(any(GotoEdge goto))
-      or
-      kind instanceof FalseEdge and
-      result = this.getTranslatedAwaitSuspend().getFirstInstruction(any(GotoEdge goto))
-    )
-  }
-
-  override Instruction getResult() { result = this.getTranslatedAwaitResume().getResult() }
-
-  final override Instruction getChildSuccessorInternal(TranslatedElement child, EdgeKind kind) {
-    child = this.getTranslatedOperand() and
-    result = this.getTranslatedAwaitReady().getFirstInstruction(kind)
-    or
-    child = this.getTranslatedAwaitReady() and
-    kind instanceof GotoEdge and
-    result = this.getInstruction(CoAwaitBranchTag())
-    or
-    child = this.getTranslatedAwaitSuspend() and
-    result = this.getTranslatedAwaitResume().getFirstInstruction(kind)
-    or
-    child = this.getTranslatedAwaitResume() and
-    result = this.getParent().getChildSuccessor(this, kind)
-  }
-
-  override predicate hasInstruction(Opcode opcode, InstructionTag tag, CppType resultType) {
-    tag = CoAwaitBranchTag() and
-    opcode instanceof Opcode::ConditionalBranch and
-    resultType = getVoidType()
-  }
-
-  override Instruction getInstructionRegisterOperand(InstructionTag tag, OperandTag operandTag) {
-    tag = CoAwaitBranchTag() and
-    operandTag instanceof ConditionOperandTag and
-    result = this.getTranslatedAwaitReady().getResult()
-  }
-
-  private TranslatedExpr getTranslatedOperand() {
-    result = getTranslatedExpr(this.getOperand().getFullyConverted())
-  }
-
-  private TranslatedExpr getTranslatedAwaitReady() {
-    result = getTranslatedExpr(this.getAwaitReady().getFullyConverted())
-  }
-
-  private TranslatedExpr getTranslatedAwaitResume() {
-    result = getTranslatedExpr(this.getAwaitResume().getFullyConverted())
-  }
-
-  private TranslatedExpr getTranslatedAwaitSuspend() {
-    result = getTranslatedExpr(this.getAwaitSuspend().getFullyConverted())
-  }
-}
-
-/** IR translation of `co_await`. */
-class TranslatedCoAwaitExpr extends TranslatedCoExpr {
-  override CoAwaitExpr expr;
-
-  final override Expr getOperand() { result = expr.getOperand() }
-
-  final override Expr getAwaitReady() { result = expr.getAwaitReady() }
-
-  final override Expr getAwaitSuspend() { result = expr.getAwaitSuspend() }
-
-  final override Expr getAwaitResume() { result = expr.getAwaitResume() }
-}
-
-/** IR translation of `co_yield`. */
-class TranslatedCoYieldxpr extends TranslatedCoExpr {
-  override CoYieldExpr expr;
-
-  final override Expr getOperand() { result = expr.getOperand() }
-
-  final override Expr getAwaitReady() { result = expr.getAwaitReady() }
-
-  final override Expr getAwaitSuspend() { result = expr.getAwaitSuspend() }
-
-  final override Expr getAwaitResume() { result = expr.getAwaitResume() }
-}
-
 abstract class TranslatedConversion extends TranslatedNonConstantExpr {
   override Conversion expr;
 
@@ -1844,6 +1707,9 @@ class TranslatedAssignExpr extends TranslatedNonConstantExpr {
     child = this.getRightOperand() and
     result = this.getLeftOperand().getFirstInstruction(kind)
     or
+    child = this.getRightOperand() and
+    result = this.getLeftOperand().getFirstInstruction(kind)
+    or
     kind instanceof GotoEdge and
     child = this.getLeftOperand() and
     result = this.getInstruction(AssignmentStoreTag())

cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/TranslatedStmt.qll

@@ -777,72 +777,6 @@ abstract class TranslatedHandler extends TranslatedStmt {
   TranslatedStmt getBlock() { result = getTranslatedStmt(stmt.getBlock()) }
 }
 
-/**
- * The IR translation of the destructor calls of the parent `TranslatedCatchByTypeHandler`.
- *
- * This object does not itself generate the destructor calls. Instead, its
- * children provide the actual calls.
- */
-class TranslatedDestructorsAfterHandler extends TranslatedElement,
-  TTranslatedDestructorsAfterHandler
-{
-  Handler handler;
-
-  TranslatedDestructorsAfterHandler() { this = TTranslatedDestructorsAfterHandler(handler) }
-
-  override string toString() { result = "Destructor calls after handler: " + handler }
-
-  private TranslatedCall getTranslatedImplicitDestructorCall(int id) {
-    result.getExpr() = handler.getImplicitDestructorCall(id)
-  }
-
-  override Instruction getFirstInstruction(EdgeKind kind) {
-    result = this.getChild(0).getFirstInstruction(kind)
-  }
-
-  override Handler getAst() { result = handler }
-
-  override Instruction getInstructionSuccessorInternal(InstructionTag tag, EdgeKind kind) { none() }
-
-  override TranslatedElement getChild(int id) {
-    result = this.getTranslatedImplicitDestructorCall(id)
-  }
-
-  override predicate handlesDestructorsExplicitly() { any() }
-
-  override Declaration getFunction() { result = handler.getEnclosingFunction() }
-
-  override Instruction getChildSuccessorInternal(TranslatedElement child, EdgeKind kind) {
-    exists(int id | child = this.getChild(id) |
-      // Transition to the next child, if any.
-      result = this.getChild(id + 1).getFirstInstruction(kind)
-      or
-      // And otherwise go to the next handler, if any.
-      not exists(this.getChild(id + 1)) and
-      result =
-        getTranslatedStmt(handler)
-            .getParent()
-            .(TranslatedTryStmt)
-            .getNextHandler(getTranslatedStmt(handler), kind)
-    )
-  }
-
-  override TranslatedElement getLastChild() {
-    result =
-      this.getTranslatedImplicitDestructorCall(max(int id |
-          exists(handler.getImplicitDestructorCall(id))
-        ))
-  }
-
-  override Instruction getALastInstructionInternal() {
-    result = this.getLastChild().getALastInstruction()
-  }
-
-  override predicate hasInstruction(Opcode opcode, InstructionTag tag, CppType resultType) {
-    none()
-  }
-}
-
 /**
  * The IR translation of a C++ `catch` block that catches an exception with a
  * specific type (e.g. `catch (const std::exception&)`).
@@ -856,14 +790,10 @@ class TranslatedCatchByTypeHandler extends TranslatedHandler {
     resultType = getVoidType()
   }
 
-  override predicate handlesDestructorsExplicitly() { any() }
-
   override TranslatedElement getChildInternal(int id) {
     result = super.getChildInternal(id)
     or
     id = 0 and result = this.getParameter()
-    or
-    id = 1 and result = this.getDestructors()
   }
 
   override Instruction getChildSuccessorInternal(TranslatedElement child, EdgeKind kind) {
@@ -880,9 +810,7 @@ class TranslatedCatchByTypeHandler extends TranslatedHandler {
       result = this.getParameter().getFirstInstruction(kind)
       or
       kind instanceof ExceptionEdge and
-      if exists(this.getDestructors())
-      then result = this.getDestructors().getFirstInstruction(any(GotoEdge edge))
-      else result = this.getParent().(TranslatedTryStmt).getNextHandler(this, any(GotoEdge edge))
+      result = this.getParent().(TranslatedTryStmt).getNextHandler(this, any(GotoEdge edge))
     )
   }
 
@@ -894,8 +822,6 @@ class TranslatedCatchByTypeHandler extends TranslatedHandler {
   private TranslatedParameter getParameter() {
     result = getTranslatedParameter(stmt.getParameter())
   }
-
-  private TranslatedDestructorsAfterHandler getDestructors() { result.getAst() = stmt }
 }
 
 /**
@@ -916,7 +842,9 @@ class TranslatedCatchAnyHandler extends TranslatedHandler {
   }
 }
 
-abstract class TranslatedIfLikeStmt extends TranslatedStmt, ConditionContext {
+class TranslatedIfStmt extends TranslatedStmt, ConditionContext {
+  override IfStmt stmt;
+
   override Instruction getFirstInstruction(EdgeKind kind) {
     if this.hasInitialization()
     then result = this.getInitialization().getFirstInstruction(kind)
@@ -929,8 +857,6 @@ abstract class TranslatedIfLikeStmt extends TranslatedStmt, ConditionContext {
 
   override TranslatedElement getLastChild() { result = this.getElse() or result = this.getThen() }
 
-  override predicate handlesDestructorsExplicitly() { any() }
-
   override TranslatedElement getChildInternal(int id) {
     id = 0 and result = this.getInitialization()
     or
@@ -941,21 +867,25 @@ abstract class TranslatedIfLikeStmt extends TranslatedStmt, ConditionContext {
     id = 3 and result = this.getElse()
   }
 
-  abstract predicate hasInitialization();
+  private predicate hasInitialization() { exists(stmt.getInitialization()) }
 
-  abstract TranslatedStmt getInitialization();
+  private TranslatedStmt getInitialization() {
+    result = getTranslatedStmt(stmt.getInitialization())
+  }
 
-  abstract TranslatedCondition getCondition();
+  private TranslatedCondition getCondition() {
+    result = getTranslatedCondition(stmt.getCondition().getFullyConverted())
+  }
 
   private Instruction getFirstConditionInstruction(EdgeKind kind) {
     result = this.getCondition().getFirstInstruction(kind)
   }
 
-  abstract TranslatedStmt getThen();
+  private TranslatedStmt getThen() { result = getTranslatedStmt(stmt.getThen()) }
 
-  abstract TranslatedStmt getElse();
+  private TranslatedStmt getElse() { result = getTranslatedStmt(stmt.getElse()) }
 
-  abstract predicate hasElse();
+  private predicate hasElse() { exists(stmt.getElse()) }
 
   override Instruction getInstructionSuccessorInternal(InstructionTag tag, EdgeKind kind) { none() }
 
@@ -968,11 +898,7 @@ abstract class TranslatedIfLikeStmt extends TranslatedStmt, ConditionContext {
     child = this.getCondition() and
     if this.hasElse()
     then result = this.getElse().getFirstInstruction(kind)
-    else (
-      if this.hasAnImplicitDestructorCall()
-      then result = this.getChild(this.getFirstDestructorCallIndex()).getFirstInstruction(kind)
-      else result = this.getParent().getChildSuccessor(this, kind)
-    )
+    else result = this.getParent().getChildSuccessor(this, kind)
   }
 
   override Instruction getChildSuccessorInternal(TranslatedElement child, EdgeKind kind) {
@@ -980,24 +906,7 @@ abstract class TranslatedIfLikeStmt extends TranslatedStmt, ConditionContext {
     result = this.getFirstConditionInstruction(kind)
     or
     (child = this.getThen() or child = this.getElse()) and
-    (
-      if this.hasAnImplicitDestructorCall()
-      then result = this.getChild(this.getFirstDestructorCallIndex()).getFirstInstruction(kind)
-      else result = this.getParent().getChildSuccessor(this, kind)
-    )
-    or
-    exists(int destructorId |
-      destructorId >= this.getFirstDestructorCallIndex() and
-      child = this.getChild(destructorId) and
-      result = this.getChild(destructorId + 1).getFirstInstruction(kind)
-    )
-    or
-    exists(int lastDestructorIndex |
-      lastDestructorIndex =
-        max(int n | exists(this.getChild(n)) and n >= this.getFirstDestructorCallIndex()) and
-      child = this.getChild(lastDestructorIndex) and
-      result = this.getParent().getChildSuccessor(this, kind)
-    )
+    result = this.getParent().getChildSuccessor(this, kind)
   }
 
   override predicate hasInstruction(Opcode opcode, InstructionTag tag, CppType resultType) {
@@ -1005,44 +914,76 @@ abstract class TranslatedIfLikeStmt extends TranslatedStmt, ConditionContext {
   }
 }
 
-class TranslatedIfStmt extends TranslatedIfLikeStmt {
-  override IfStmt stmt;
-
-  override predicate hasInitialization() { exists(stmt.getInitialization()) }
-
-  override TranslatedStmt getInitialization() {
-    result = getTranslatedStmt(stmt.getInitialization())
-  }
-
-  override TranslatedCondition getCondition() {
-    result = getTranslatedCondition(stmt.getCondition().getFullyConverted())
-  }
-
-  override TranslatedStmt getThen() { result = getTranslatedStmt(stmt.getThen()) }
-
-  override TranslatedStmt getElse() { result = getTranslatedStmt(stmt.getElse()) }
-
-  override predicate hasElse() { exists(stmt.getElse()) }
-}
-
-class TranslatedConstExprIfStmt extends TranslatedIfLikeStmt {
+class TranslatedConstExprIfStmt extends TranslatedStmt, ConditionContext {
   override ConstexprIfStmt stmt;
 
-  override predicate hasInitialization() { exists(stmt.getInitialization()) }
+  override Instruction getFirstInstruction(EdgeKind kind) {
+    if this.hasInitialization()
+    then result = this.getInitialization().getFirstInstruction(kind)
+    else result = this.getFirstConditionInstruction(kind)
+  }
 
-  override TranslatedStmt getInitialization() {
+  override TranslatedElement getChildInternal(int id) {
+    id = 0 and result = this.getInitialization()
+    or
+    id = 1 and result = this.getCondition()
+    or
+    id = 2 and result = this.getThen()
+    or
+    id = 3 and result = this.getElse()
+  }
+
+  private predicate hasInitialization() { exists(stmt.getInitialization()) }
+
+  private TranslatedStmt getInitialization() {
     result = getTranslatedStmt(stmt.getInitialization())
   }
 
-  override TranslatedCondition getCondition() {
+  private TranslatedCondition getCondition() {
     result = getTranslatedCondition(stmt.getCondition().getFullyConverted())
   }
 
-  override TranslatedStmt getThen() { result = getTranslatedStmt(stmt.getThen()) }
+  private Instruction getFirstConditionInstruction(EdgeKind kind) {
+    result = this.getCondition().getFirstInstruction(kind)
+  }
 
-  override TranslatedStmt getElse() { result = getTranslatedStmt(stmt.getElse()) }
+  private TranslatedStmt getThen() { result = getTranslatedStmt(stmt.getThen()) }
 
-  override predicate hasElse() { exists(stmt.getElse()) }
+  private TranslatedStmt getElse() { result = getTranslatedStmt(stmt.getElse()) }
+
+  private predicate hasElse() { exists(stmt.getElse()) }
+
+  override Instruction getInstructionSuccessorInternal(InstructionTag tag, EdgeKind kind) { none() }
+
+  override Instruction getChildTrueSuccessor(TranslatedCondition child, EdgeKind kind) {
+    child = this.getCondition() and
+    result = this.getThen().getFirstInstruction(kind)
+  }
+
+  override Instruction getChildFalseSuccessor(TranslatedCondition child, EdgeKind kind) {
+    child = this.getCondition() and
+    if this.hasElse()
+    then result = this.getElse().getFirstInstruction(kind)
+    else result = this.getParent().getChildSuccessor(this, kind)
+  }
+
+  override Instruction getChildSuccessorInternal(TranslatedElement child, EdgeKind kind) {
+    child = this.getInitialization() and
+    result = this.getFirstConditionInstruction(kind)
+    or
+    (child = this.getThen() or child = this.getElse()) and
+    result = this.getParent().getChildSuccessor(this, kind)
+  }
+
+  override predicate hasInstruction(Opcode opcode, InstructionTag tag, CppType resultType) {
+    none()
+  }
+
+  override Instruction getALastInstructionInternal() {
+    result = this.getThen().getALastInstruction()
+    or
+    result = this.getElse().getALastInstruction()
+  }
 }
 
 abstract class TranslatedLoop extends TranslatedStmt, ConditionContext {
@@ -1222,8 +1163,6 @@ class TranslatedForStmt extends TranslatedLoop {
 class TranslatedRangeBasedForStmt extends TranslatedStmt, ConditionContext {
   override RangeBasedForStmt stmt;
 
-  override predicate handlesDestructorsExplicitly() { any() }
-
   override TranslatedElement getChildInternal(int id) {
     id = 0 and result = this.getInitialization()
     or
@@ -1277,19 +1216,6 @@ class TranslatedRangeBasedForStmt extends TranslatedStmt, ConditionContext {
     or
     child = this.getUpdate() and
     result = this.getCondition().getFirstInstruction(kind)
-    or
-    exists(int destructorId |
-      destructorId >= this.getFirstDestructorCallIndex() and
-      child = this.getChild(destructorId) and
-      result = this.getChild(destructorId + 1).getFirstInstruction(kind)
-    )
-    or
-    exists(int lastDestructorIndex |
-      lastDestructorIndex =
-        max(int n | exists(this.getChild(n)) and n >= this.getFirstDestructorCallIndex()) and
-      child = this.getChild(lastDestructorIndex) and
-      result = this.getParent().getChildSuccessor(this, kind)
-    )
   }
 
   override predicate hasInstruction(Opcode opcode, InstructionTag tag, CppType resultType) {
@@ -1305,9 +1231,7 @@ class TranslatedRangeBasedForStmt extends TranslatedStmt, ConditionContext {
 
   override Instruction getChildFalseSuccessor(TranslatedCondition child, EdgeKind kind) {
     child = this.getCondition() and
-    if this.hasAnImplicitDestructorCall()
-    then result = this.getChild(this.getFirstDestructorCallIndex()).getFirstInstruction(kind)
-    else result = this.getParent().getChildSuccessor(this, kind)
+    result = this.getParent().getChildSuccessor(this, kind)
   }
 
   private TranslatedDeclStmt getRangeVariableDeclStmt() {
@@ -1352,11 +1276,6 @@ class TranslatedJumpStmt extends TranslatedStmt {
   override JumpStmt stmt;
 
   override Instruction getFirstInstruction(EdgeKind kind) {
-    // The first instruction is a destructor call, if any.
-    result = this.getChildInternal(0).getFirstInstruction(kind)
-    or
-    // Otherwise, the first (and only) instruction is a `NoOp`
-    not exists(this.getChildInternal(0)) and
     result = this.getInstruction(OnlyInstructionTag()) and
     kind instanceof GotoEdge
   }
@@ -1365,20 +1284,7 @@ class TranslatedJumpStmt extends TranslatedStmt {
     result = this.getInstruction(OnlyInstructionTag())
   }
 
-  private TranslatedCall getTranslatedImplicitDestructorCall(int id) {
-    result.getExpr() = stmt.getImplicitDestructorCall(id)
-  }
-
-  override TranslatedElement getLastChild() {
-    result =
-      this.getTranslatedImplicitDestructorCall(max(int id |
-          exists(stmt.getImplicitDestructorCall(id))
-        ))
-  }
-
-  override TranslatedElement getChildInternal(int id) {
-    result = this.getTranslatedImplicitDestructorCall(id)
-  }
+  override TranslatedElement getChildInternal(int id) { none() }
 
   override predicate hasInstruction(Opcode opcode, InstructionTag tag, CppType resultType) {
     tag = OnlyInstructionTag() and
@@ -1391,19 +1297,7 @@ class TranslatedJumpStmt extends TranslatedStmt {
     result = getTranslatedStmt(stmt.getTarget()).getFirstInstruction(kind)
   }
 
-  final override predicate handlesDestructorsExplicitly() { any() }
-
-  override Instruction getChildSuccessorInternal(TranslatedElement child, EdgeKind kind) {
-    exists(int id | child = this.getChildInternal(id) |
-      // Transition to the next destructor call, if any.
-      result = this.getChildInternal(id + 1).getFirstInstruction(kind)
-      or
-      // And otherwise, exit this element by flowing to the target of the jump.
-      not exists(this.getChildInternal(id + 1)) and
-      kind instanceof GotoEdge and
-      result = this.getInstruction(OnlyInstructionTag())
-    )
-  }
+  override Instruction getChildSuccessorInternal(TranslatedElement child, EdgeKind kind) { none() }
 }
 
 private EdgeKind getCaseEdge(SwitchCase switchCase) {

cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/IRVariable.qll

@@ -17,11 +17,18 @@ private import Imports::IRType
  * The variable may be a user-declared variable (`IRUserVariable`) or a temporary variable generated
  * by the AST-to-IR translation (`IRTempVariable`).
  */
-abstract private class AbstractIRVariable extends TIRVariable {
+class IRVariable extends TIRVariable {
   Language::Declaration func;
 
+  IRVariable() {
+    this = TIRUserVariable(_, _, func) or
+    this = TIRTempVariable(func, _, _, _) or
+    this = TIRStringLiteral(func, _, _, _) or
+    this = TIRDynamicInitializationFlag(func, _, _)
+  }
+
   /** Gets a textual representation of this element. */
-  abstract string toString();
+  string toString() { none() }
 
   /**
    * Holds if this variable's value cannot be changed within a function. Currently used for string
@@ -42,13 +49,13 @@ abstract private class AbstractIRVariable extends TIRVariable {
   /**
    * Gets the type of the variable.
    */
-  abstract Language::LanguageType getLanguageType();
+  Language::LanguageType getLanguageType() { none() }
 
   /**
    * Gets the AST node that declared this variable, or that introduced this
    * variable as part of the AST-to-IR translation.
    */
-  abstract Language::AST getAst();
+  Language::AST getAst() { none() }
 
   /** DEPRECATED: Alias for getAst */
   deprecated Language::AST getAST() { result = this.getAst() }
@@ -57,7 +64,7 @@ abstract private class AbstractIRVariable extends TIRVariable {
    * Gets an identifier string for the variable. This identifier is unique
    * within the function.
    */
-  abstract string getUniqueId();
+  string getUniqueId() { none() }
 
   /**
    * Gets the source location of this variable.
@@ -67,7 +74,7 @@ abstract private class AbstractIRVariable extends TIRVariable {
   /**
    * Gets the IR for the function that references this variable.
    */
-  final IRFunction getEnclosingIRFunction() { result.getFunction() = this.getEnclosingFunction() }
+  final IRFunction getEnclosingIRFunction() { result.getFunction() = func }
 
   /**
    * Gets the function that references this variable.
@@ -75,18 +82,10 @@ abstract private class AbstractIRVariable extends TIRVariable {
   final Language::Declaration getEnclosingFunction() { result = func }
 }
 
-/**
- * A variable referenced by the IR for a function.
- *
- * The variable may be a user-declared variable (`IRUserVariable`) or a temporary variable generated
- * by the AST-to-IR translation (`IRTempVariable`).
- */
-final class IRVariable = AbstractIRVariable;
-
 /**
  * A user-declared variable referenced by the IR for a function.
  */
-class IRUserVariable extends AbstractIRVariable, TIRUserVariable {
+class IRUserVariable extends IRVariable, TIRUserVariable {
   Language::Variable var;
   Language::LanguageType type;
 
@@ -115,29 +114,26 @@ class IRUserVariable extends AbstractIRVariable, TIRUserVariable {
  * A variable (user-declared or temporary) that is allocated on the stack. This includes all
  * parameters, non-static local variables, and temporary variables.
  */
-abstract private class AbstractIRAutomaticVariable extends AbstractIRVariable { }
-
-/**
- * A variable (user-declared or temporary) that is allocated on the stack. This includes all
- * parameters, non-static local variables, and temporary variables.
- */
-final class IRAutomaticVariable = AbstractIRAutomaticVariable;
-
-/**
- * A user-declared variable that is allocated on the stack. This includes all parameters and
- * non-static local variables.
- */
-private class AbstractIRAutomaticUserVariable extends IRUserVariable, AbstractIRAutomaticVariable {
-  override Language::AutomaticVariable var;
-
-  final override Language::AutomaticVariable getVariable() { result = var }
+class IRAutomaticVariable extends IRVariable {
+  IRAutomaticVariable() {
+    exists(Language::Variable var |
+      this = TIRUserVariable(var, _, func) and
+      Language::isVariableAutomatic(var)
+    )
+    or
+    this = TIRTempVariable(func, _, _, _)
+  }
 }
 
 /**
  * A user-declared variable that is allocated on the stack. This includes all parameters and
  * non-static local variables.
  */
-final class IRAutomaticUserVariable = AbstractIRAutomaticUserVariable;
+class IRAutomaticUserVariable extends IRUserVariable, IRAutomaticVariable {
+  override Language::AutomaticVariable var;
+
+  final override Language::AutomaticVariable getVariable() { result = var }
+}
 
 /**
  * A user-declared variable that is not allocated on the stack. This includes all global variables,
@@ -155,10 +151,16 @@ class IRStaticUserVariable extends IRUserVariable {
  * A variable that is not user-declared. This includes temporary variables generated as part of IR
  * construction, as well as string literals.
  */
-abstract private class AbstractIRGeneratedVariable extends AbstractIRVariable {
+class IRGeneratedVariable extends IRVariable {
   Language::AST ast;
   Language::LanguageType type;
 
+  IRGeneratedVariable() {
+    this = TIRTempVariable(func, ast, _, type) or
+    this = TIRStringLiteral(func, ast, type, _) or
+    this = TIRDynamicInitializationFlag(func, ast, type)
+  }
+
   final override Language::LanguageType getLanguageType() { result = type }
 
   final override Language::AST getAst() { result = ast }
@@ -194,20 +196,12 @@ abstract private class AbstractIRGeneratedVariable extends AbstractIRVariable {
   string getBaseString() { none() }
 }
 
-/**
- * A variable that is not user-declared. This includes temporary variables generated as part of IR
- * construction, as well as string literals.
- */
-final class IRGeneratedVariable = AbstractIRGeneratedVariable;
-
 /**
  * A temporary variable introduced by IR construction. The most common examples are the variable
  * generated to hold the return value of a function, or the variable generated to hold the result of
  * a condition operator (`a ? b : c`).
  */
-class IRTempVariable extends AbstractIRGeneratedVariable, AbstractIRAutomaticVariable,
-  TIRTempVariable
-{
+class IRTempVariable extends IRGeneratedVariable, IRAutomaticVariable, TIRTempVariable {
   TempVariableTag tag;
 
   IRTempVariable() { this = TIRTempVariable(func, ast, tag, type) }
@@ -247,7 +241,7 @@ class IRThrowVariable extends IRTempVariable {
  * A temporary variable generated to hold the contents of all arguments passed to the `...` of a
  * function that accepts a variable number of arguments.
  */
-class IREllipsisVariable extends IRTempVariable, AbstractIRParameter {
+class IREllipsisVariable extends IRTempVariable, IRParameter {
   IREllipsisVariable() { tag = EllipsisTempVar() }
 
   final override string toString() { result = "#ellipsis" }
@@ -258,7 +252,7 @@ class IREllipsisVariable extends IRTempVariable, AbstractIRParameter {
 /**
  * A temporary variable generated to hold the `this` pointer.
  */
-class IRThisVariable extends IRTempVariable, AbstractIRParameter {
+class IRThisVariable extends IRTempVariable, IRParameter {
   IRThisVariable() { tag = ThisTempVar() }
 
   final override string toString() { result = "#this" }
@@ -270,7 +264,7 @@ class IRThisVariable extends IRTempVariable, AbstractIRParameter {
  * A variable generated to represent the contents of a string literal. This variable acts much like
  * a read-only global variable.
  */
-class IRStringLiteral extends AbstractIRGeneratedVariable, TIRStringLiteral {
+class IRStringLiteral extends IRGeneratedVariable, TIRStringLiteral {
   Language::StringLiteral literal;
 
   IRStringLiteral() { this = TIRStringLiteral(func, ast, type, literal) }
@@ -294,7 +288,7 @@ class IRStringLiteral extends AbstractIRGeneratedVariable, TIRStringLiteral {
  * used to model the runtime initialization of static local variables in C++, as well as static
  * fields in C#.
  */
-class IRDynamicInitializationFlag extends AbstractIRGeneratedVariable, TIRDynamicInitializationFlag {
+class IRDynamicInitializationFlag extends IRGeneratedVariable, TIRDynamicInitializationFlag {
   Language::Variable var;
 
   IRDynamicInitializationFlag() {
@@ -320,24 +314,24 @@ class IRDynamicInitializationFlag extends AbstractIRGeneratedVariable, TIRDynami
  * An IR variable which acts like a function parameter, including positional parameters and the
  * temporary variables generated for `this` and ellipsis parameters.
  */
-abstract private class AbstractIRParameter extends AbstractIRAutomaticVariable {
+class IRParameter extends IRAutomaticVariable {
+  IRParameter() {
+    this.(IRAutomaticUserVariable).getVariable() instanceof Language::Parameter
+    or
+    this = TIRTempVariable(_, _, ThisTempVar(), _)
+    or
+    this = TIRTempVariable(_, _, EllipsisTempVar(), _)
+  }
+
   /**
    * Gets the zero-based index of this parameter. The `this` parameter has index -1.
    */
   int getIndex() { none() }
 }
 
-/**
- * An IR variable which acts like a function parameter, including positional parameters and the
- * temporary variables generated for `this` and ellipsis parameters.
- */
-final class IRParameter = AbstractIRParameter;
-
 /**
  * An IR variable representing a positional parameter.
  */
-class IRPositionalParameter extends AbstractIRParameter, AbstractIRAutomaticUserVariable {
-  IRPositionalParameter() { this.getVariable() instanceof Language::Parameter }
-
+class IRPositionalParameter extends IRParameter, IRAutomaticUserVariable {
   final override int getIndex() { result = this.getVariable().(Language::Parameter).getIndex() }
 }

cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/Instruction.qll

@@ -247,7 +247,8 @@ class Instruction extends Construction::TStageInstruction {
    * Gets the type of the result produced by this instruction. If the instruction does not produce
    * a result, its result type will be `IRVoidType`.
    */
-  final IRType getResultIRType() { result = Construction::getInstructionResultIRType(this) }
+  cached
+  final IRType getResultIRType() { result = this.getResultLanguageType().getIRType() }
 
   /**
    * Gets the type of the result produced by this instruction. If the
@@ -994,8 +995,9 @@ class ConstantInstruction extends ConstantValueInstruction {
  */
 class IntegerConstantInstruction extends ConstantInstruction {
   IntegerConstantInstruction() {
-    exists(IRType resultType | resultType = this.getResultIRType() |
-      resultType instanceof IRIntegerType or resultType instanceof IRBooleanType
+    exists(IRType resultType |
+      resultType = this.getResultIRType() and
+      (resultType instanceof IRIntegerType or resultType instanceof IRBooleanType)
     )
   }
 }
@@ -1007,17 +1009,6 @@ class FloatConstantInstruction extends ConstantInstruction {
   FloatConstantInstruction() { this.getResultIRType() instanceof IRFloatingPointType }
 }
 
-/**
- * An instruction whose result is a constant value of a pointer type.
- */
-class PointerConstantInstruction extends ConstantInstruction {
-  PointerConstantInstruction() {
-    exists(IRType resultType | resultType = this.getResultIRType() |
-      resultType instanceof IRAddressType or resultType instanceof IRFunctionAddressType
-    )
-  }
-}
-
 /**
  * An instruction whose result is the address of a string literal.
  */

cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/SSAConstruction.qll

@@ -429,11 +429,6 @@ private module Cached {
     instr = unreachedInstruction(_) and result = Language::getVoidType()
   }
 
-  cached
-  IRType getInstructionResultIRType(Instruction instr) {
-    result = instr.getResultLanguageType().getIRType()
-  }
-
   /**
    * Holds if `opcode` is the opcode that specifies the operation performed by `instr`.
    *

cpp/ql/lib/semmle/code/cpp/models/implementations/Iterator.qll

@@ -560,7 +560,7 @@ private class IteratorAssignmentMemberOperatorModel extends IteratorAssignmentMe
   TaintFunction, SideEffectFunction, AliasFunction
 {
   override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
-    (input.isParameterDeref(0) or input.isParameter(0)) and
+    input.isParameterDeref(0) and
     output.isQualifierObject()
   }
 
@@ -579,34 +579,17 @@ private class IteratorAssignmentMemberOperatorModel extends IteratorAssignmentMe
   override predicate parameterEscapesOnlyViaReturn(int index) { index = -1 }
 }
 
-/**
- * A `begin` member function, or a related function, that returns an iterator.
- */
-class BeginFunction extends MemberFunction {
-  BeginFunction() {
-    this.hasName(["begin", "cbegin", "rbegin", "crbegin", "before_begin", "cbefore_begin"]) and
-    this.getType().getUnspecifiedType() instanceof Iterator
-  }
-}
-
-/**
- * An `end` member function, or a related function, that returns an iterator.
- */
-class EndFunction extends MemberFunction {
-  EndFunction() {
-    this.hasName(["end", "cend", "rend", "crend"]) and
-    this.getType().getUnspecifiedType() instanceof Iterator
-  }
-}
-
 /**
  * A `begin` or `end` member function, or a related member function, that
  * returns an iterator.
  */
 class BeginOrEndFunction extends MemberFunction {
   BeginOrEndFunction() {
-    this instanceof BeginFunction or
-    this instanceof EndFunction
+    this.hasName([
+        "begin", "cbegin", "rbegin", "crbegin", "end", "cend", "rend", "crend", "before_begin",
+        "cbefore_begin"
+      ]) and
+    this.getType().getUnspecifiedType() instanceof Iterator
   }
 }
 

cpp/ql/lib/semmle/code/cpp/valuenumbering/GlobalValueNumberingImpl.qll

@@ -0,0 +1,616 @@
+/**
+ * DEPRECATED: This library has been replaced with a newer version which
+ * provides better performance and precision. Use
+ * `semmle.code.cpp.valuenumbering.GlobalValueNumbering` instead.
+ *
+ * Provides an implementation of Global Value Numbering.
+ * See https://en.wikipedia.org/wiki/Global_value_numbering
+ *
+ * The predicate `globalValueNumber` converts an expression into a `GVN`,
+ * which is an abstract type representing the value of the expression. If
+ * two expressions have the same `GVN` then they compute the same value.
+ * For example:
+ *
+ * ```
+ * void f(int x, int y) {
+ *   g(x+y, x+y);
+ * }
+ * ```
+ *
+ * In this example, both arguments in the call to `g` compute the same value,
+ * so both arguments have the same `GVN`. In other words, we can find
+ * this call with the following query:
+ *
+ * ```
+ * from FunctionCall call, GVN v
+ * where v = globalValueNumber(call.getArgument(0))
+ *   and v = globalValueNumber(call.getArgument(1))
+ * select call
+ * ```
+ *
+ * The analysis is conservative, so two expressions might have different
+ * `GVN`s even though the actually always compute the same value. The most
+ * common reason for this is that the analysis cannot prove that there
+ * are no side-effects that might cause the computed value to change.
+ */
+
+/*
+ * Note to developers: the correctness of this module depends on the
+ * definitions of GVN, globalValueNumber, and analyzableExpr being kept in
+ * sync with each other. If you change this module then make sure that the
+ * change is symmetric across all three.
+ */
+
+import cpp
+private import semmle.code.cpp.controlflow.SSA
+
+/**
+ * Holds if the result is a control flow node that might change the
+ * value of any global variable. This is used in the implementation
+ * of `GVN_OtherVariable`, because we need to be quite conservative when
+ * we assign a value number to a global variable. For example:
+ *
+ * ```
+ * x = g+1;
+ * dosomething();
+ * y = g+1;
+ * ```
+ *
+ * It is not safe to assign the same value number to both instances
+ * of `g+1` in this example, because the call to `dosomething` might
+ * change the value of `g`.
+ */
+private ControlFlowNode nodeWithPossibleSideEffect() {
+  result instanceof Call
+  or
+  // If the lhs of an assignment is not analyzable by SSA, then
+  // we need to treat the assignment as having a possible side-effect.
+  result instanceof Assignment and not result instanceof SsaDefinition
+  or
+  result instanceof CrementOperation and not result instanceof SsaDefinition
+  or
+  exists(LocalVariable v |
+    result = v.getInitializer().getExpr() and not result instanceof SsaDefinition
+  )
+  or
+  result instanceof AsmStmt
+}
+
+/**
+ * Gets the entry node of the control flow graph of which `node` is a
+ * member.
+ */
+cached
+private ControlFlowNode getControlFlowEntry(ControlFlowNode node) {
+  result = node.getControlFlowScope().getEntryPoint() and
+  result.getASuccessor*() = node
+}
+
+/**
+ * Holds if there is a control flow edge from `src` to `dst` or
+ * if `dst` is an expression with a possible side-effect. The idea
+ * is to treat side effects as entry points in the control flow
+ * graph so that we can use the dominator tree to find the most recent
+ * side-effect.
+ */
+private predicate sideEffectCfg(ControlFlowNode src, ControlFlowNode dst) {
+  src.getASuccessor() = dst
+  or
+  // Add an edge from the entry point to any node that might have a side
+  // effect.
+  dst = nodeWithPossibleSideEffect() and
+  src = getControlFlowEntry(dst)
+}
+
+/**
+ * Holds if `dominator` is the immediate dominator of `node` in
+ * the side-effect CFG.
+ */
+private predicate iDomEffect(ControlFlowNode dominator, ControlFlowNode node) =
+  idominance(functionEntry/1, sideEffectCfg/2)(_, dominator, node)
+
+/**
+ * Gets the most recent side effect. To be more precise, `result` is a
+ * dominator of `node` and no side-effects can occur between `result` and
+ * `node`.
+ *
+ * `sideEffectCFG` has an edge from the function entry to every node with a
+ * side-effect. This means that every node with a side-effect has the
+ * function entry as its immediate dominator. So if node `x` dominates node
+ * `y` then there can be no side effects between `x` and `y` unless `x` is
+ * the function entry. So the optimal choice for `result` has the function
+ * entry as its immediate dominator.
+ *
+ * Example:
+ *
+ * ```
+ * 000:  int f(int a, int b, int *p) {
+ * 001:    int r = 0;
+ * 002:    if (a) {
+ * 003:      if (b) {
+ * 004:        sideEffect1();
+ * 005:      }
+ * 006:    } else {
+ * 007:      sideEffect2();
+ * 008:    }
+ * 009:    if (a) {
+ * 010:      r++; // Not a side-effect, because r is an SSA variable.
+ * 011:    }
+ * 012:    if (b) {
+ * 013:      r++; // Not a side-effect, because r is an SSA variable.
+ * 014:    }
+ * 015:    return *p;
+ * 016:  }
+ * ```
+ *
+ * Suppose we want to find the most recent side-effect for the dereference
+ * of `p` on line 015. The `sideEffectCFG` has an edge from the function
+ * entry (line 000) to the side effects at lines 004 and 007. Therefore,
+ * the immediate dominator tree looks like this:
+ *
+ * 000 - 001 - 002 - 003
+ *     - 004
+ *     - 007
+ *     - 009 - 010
+ *           - 012 - 013
+ *                 - 015
+ *
+ * The immediate dominator path to line 015 is 000 - 009 - 012 - 015.
+ * Therefore, the most recent side effect for line 015 is line 009.
+ */
+cached
+private ControlFlowNode mostRecentSideEffect(ControlFlowNode node) {
+  exists(ControlFlowNode entry |
+    functionEntry(entry) and
+    iDomEffect(entry, result) and
+    iDomEffect*(result, node)
+  )
+}
+
+/** Used to represent the "global value number" of an expression. */
+cached
+private newtype GvnBase =
+  GVN_IntConst(int val, Type t) { mk_IntConst(val, t, _) } or
+  GVN_FloatConst(float val, Type t) { mk_FloatConst(val, t, _) } or
+  // If the local variable does not have a defining value, then
+  // we use the SsaDefinition as its global value number.
+  GVN_UndefinedStackVariable(StackVariable x, SsaDefinition def) {
+    mk_UndefinedStackVariable(x, def, _)
+  } or
+  // Variables with no SSA information. As a crude (but safe)
+  // approximation, we use `mostRecentSideEffect` to compute a definition
+  // location for the variable. This ensures that two instances of the same
+  // global variable will only get the same value number if they are
+  // guaranteed to have the same value.
+  GVN_OtherVariable(Variable x, ControlFlowNode dominator) { mk_OtherVariable(x, dominator, _) } or
+  deprecated GVN_FieldAccess(GVN s, Field f) {
+    mk_DotFieldAccess(s, f, _) or
+    mk_PointerFieldAccess_with_deref(s, f, _) or
+    mk_ImplicitThisFieldAccess_with_deref(s, f, _)
+  } or
+  // Dereference a pointer. The value might have changed since the last
+  // time the pointer was dereferenced, so we need to include a definition
+  // location. As a crude (but safe) approximation, we use
+  // `mostRecentSideEffect` to compute a definition location.
+  deprecated GVN_Deref(GVN p, ControlFlowNode dominator) {
+    mk_Deref(p, dominator, _) or
+    mk_PointerFieldAccess(p, _, dominator, _) or
+    mk_ImplicitThisFieldAccess_with_qualifier(p, _, dominator, _)
+  } or
+  GVN_ThisExpr(Function fcn) {
+    mk_ThisExpr(fcn, _) or
+    mk_ImplicitThisFieldAccess(fcn, _, _, _)
+  } or
+  deprecated GVN_Conversion(Type t, GVN child) { mk_Conversion(t, child, _) } or
+  deprecated GVN_BinaryOp(GVN lhs, GVN rhs, string opname) { mk_BinaryOp(lhs, rhs, opname, _) } or
+  deprecated GVN_UnaryOp(GVN child, string opname) { mk_UnaryOp(child, opname, _) } or
+  deprecated GVN_ArrayAccess(GVN x, GVN i, ControlFlowNode dominator) {
+    mk_ArrayAccess(x, i, dominator, _)
+  } or
+  // Any expression that is not handled by the cases above is
+  // given a unique number based on the expression itself.
+  GVN_Unanalyzable(Expr e) { not analyzableExpr(e) }
+
+/**
+ * A Global Value Number. A GVN is an abstract representation of the value
+ * computed by an expression. The relationship between `Expr` and `GVN` is
+ * many-to-one: every `Expr` has exactly one `GVN`, but multiple
+ * expressions can have the same `GVN`. If two expressions have the same
+ * `GVN`, it means that they compute the same value at run time. The `GVN`
+ * is an opaque value, so you cannot deduce what the run-time value of an
+ * expression will be from its `GVN`. The only use for the `GVN` of an
+ * expression is to find other expressions that compute the same value.
+ * Use the predicate `globalValueNumber` to get the `GVN` for an `Expr`.
+ *
+ * Note: `GVN` has `toString` and `getLocation` methods, so that it can be
+ * displayed in a results list. These work by picking an arbitrary
+ * expression with this `GVN` and using its `toString` and `getLocation`
+ * methods.
+ */
+deprecated class GVN extends GvnBase {
+  GVN() { this instanceof GvnBase }
+
+  /** Gets an expression that has this GVN. */
+  Expr getAnExpr() { this = globalValueNumber(result) }
+
+  /** Gets the kind of the GVN. This can be useful for debugging. */
+  string getKind() {
+    if this instanceof GVN_IntConst
+    then result = "IntConst"
+    else
+      if this instanceof GVN_FloatConst
+      then result = "FloatConst"
+      else
+        if this instanceof GVN_UndefinedStackVariable
+        then result = "UndefinedStackVariable"
+        else
+          if this instanceof GVN_OtherVariable
+          then result = "OtherVariable"
+          else
+            if this instanceof GVN_FieldAccess
+            then result = "FieldAccess"
+            else
+              if this instanceof GVN_Deref
+              then result = "Deref"
+              else
+                if this instanceof GVN_ThisExpr
+                then result = "ThisExpr"
+                else
+                  if this instanceof GVN_Conversion
+                  then result = "Conversion"
+                  else
+                    if this instanceof GVN_BinaryOp
+                    then result = "BinaryOp"
+                    else
+                      if this instanceof GVN_UnaryOp
+                      then result = "UnaryOp"
+                      else
+                        if this instanceof GVN_ArrayAccess
+                        then result = "ArrayAccess"
+                        else
+                          if this instanceof GVN_Unanalyzable
+                          then result = "Unanalyzable"
+                          else result = "error"
+  }
+
+  /**
+   * Gets an example of an expression with this GVN.
+   * This is useful for things like implementing toString().
+   */
+  private Expr exampleExpr() {
+    // Pick the expression with the minimum source location string. This is
+    // just an arbitrary way to pick an expression with this `GVN`.
+    result = min(Expr e | this = globalValueNumber(e) | e order by e.getLocation().toString())
+  }
+
+  /** Gets a textual representation of this element. */
+  string toString() { result = this.exampleExpr().toString() }
+
+  /** Gets the primary location of this element. */
+  Location getLocation() { result = this.exampleExpr().getLocation() }
+}
+
+private predicate analyzableIntConst(Expr e) {
+  strictcount(e.getValue().toInt()) = 1 and
+  strictcount(e.getUnspecifiedType()) = 1
+}
+
+private predicate mk_IntConst(int val, Type t, Expr e) {
+  analyzableIntConst(e) and
+  val = e.getValue().toInt() and
+  t = e.getUnspecifiedType()
+}
+
+private predicate analyzableFloatConst(Expr e) {
+  strictcount(e.getValue().toFloat()) = 1 and
+  strictcount(e.getUnspecifiedType()) = 1 and
+  not analyzableIntConst(e)
+}
+
+private predicate mk_FloatConst(float val, Type t, Expr e) {
+  analyzableFloatConst(e) and
+  val = e.getValue().toFloat() and
+  t = e.getUnspecifiedType()
+}
+
+private predicate analyzableStackVariable(VariableAccess access) {
+  strictcount(SsaDefinition def | def.getAUse(_) = access | def) = 1 and
+  strictcount(SsaDefinition def, Variable v | def.getAUse(v) = access | v) = 1 and
+  count(SsaDefinition def, Variable v |
+    def.getAUse(v) = access
+  |
+    def.getDefiningValue(v).getFullyConverted()
+  ) <= 1 and
+  not analyzableConst(access)
+}
+
+// Note: this predicate only has a result if the access has no
+// defining value. If there is a defining value, then there is no
+// need to generate a fresh `GVN` for the access because `globalValueNumber`
+// will follow the chain and use the GVN of the defining value.
+private predicate mk_UndefinedStackVariable(
+  StackVariable x, SsaDefinition def, VariableAccess access
+) {
+  analyzableStackVariable(access) and
+  access = def.getAUse(x) and
+  not exists(def.getDefiningValue(x))
+}
+
+private predicate analyzableDotFieldAccess(DotFieldAccess access) {
+  strictcount(access.getTarget()) = 1 and
+  strictcount(access.getQualifier().getFullyConverted()) = 1 and
+  not analyzableConst(access)
+}
+
+deprecated private predicate mk_DotFieldAccess(GVN qualifier, Field target, DotFieldAccess access) {
+  analyzableDotFieldAccess(access) and
+  target = access.getTarget() and
+  qualifier = globalValueNumber(access.getQualifier().getFullyConverted())
+}
+
+private predicate analyzablePointerFieldAccess(PointerFieldAccess access) {
+  strictcount(mostRecentSideEffect(access)) = 1 and
+  strictcount(access.getTarget()) = 1 and
+  strictcount(access.getQualifier().getFullyConverted()) = 1 and
+  not analyzableConst(access)
+}
+
+deprecated private predicate mk_PointerFieldAccess(
+  GVN qualifier, Field target, ControlFlowNode dominator, PointerFieldAccess access
+) {
+  analyzablePointerFieldAccess(access) and
+  dominator = mostRecentSideEffect(access) and
+  target = access.getTarget() and
+  qualifier = globalValueNumber(access.getQualifier().getFullyConverted())
+}
+
+/**
+ * `obj->field` is equivalent to `(*obj).field`, so we need to wrap an
+ * extra `GVN_Deref` around the qualifier.
+ */
+deprecated private predicate mk_PointerFieldAccess_with_deref(
+  GVN new_qualifier, Field target, PointerFieldAccess access
+) {
+  exists(GVN qualifier, ControlFlowNode dominator |
+    mk_PointerFieldAccess(qualifier, target, dominator, access) and
+    new_qualifier = GVN_Deref(qualifier, dominator)
+  )
+}
+
+private predicate analyzableImplicitThisFieldAccess(ImplicitThisFieldAccess access) {
+  strictcount(mostRecentSideEffect(access)) = 1 and
+  strictcount(access.getTarget()) = 1 and
+  strictcount(access.getEnclosingFunction()) = 1 and
+  not analyzableConst(access)
+}
+
+private predicate mk_ImplicitThisFieldAccess(
+  Function fcn, Field target, ControlFlowNode dominator, ImplicitThisFieldAccess access
+) {
+  analyzableImplicitThisFieldAccess(access) and
+  dominator = mostRecentSideEffect(access) and
+  target = access.getTarget() and
+  fcn = access.getEnclosingFunction()
+}
+
+deprecated private predicate mk_ImplicitThisFieldAccess_with_qualifier(
+  GVN qualifier, Field target, ControlFlowNode dominator, ImplicitThisFieldAccess access
+) {
+  exists(Function fcn |
+    mk_ImplicitThisFieldAccess(fcn, target, dominator, access) and
+    qualifier = GVN_ThisExpr(fcn)
+  )
+}
+
+deprecated private predicate mk_ImplicitThisFieldAccess_with_deref(
+  GVN new_qualifier, Field target, ImplicitThisFieldAccess access
+) {
+  exists(GVN qualifier, ControlFlowNode dominator |
+    mk_ImplicitThisFieldAccess_with_qualifier(qualifier, target, dominator, access) and
+    new_qualifier = GVN_Deref(qualifier, dominator)
+  )
+}
+
+/**
+ * Holds if `access` is an access of a variable that does
+ * not have SSA information. (For example, because the variable
+ * is global.)
+ */
+private predicate analyzableOtherVariable(VariableAccess access) {
+  not access instanceof FieldAccess and
+  not exists(SsaDefinition def | access = def.getAUse(_)) and
+  strictcount(access.getTarget()) = 1 and
+  strictcount(mostRecentSideEffect(access)) = 1 and
+  not analyzableConst(access)
+}
+
+private predicate mk_OtherVariable(Variable x, ControlFlowNode dominator, VariableAccess access) {
+  analyzableOtherVariable(access) and
+  x = access.getTarget() and
+  dominator = mostRecentSideEffect(access)
+}
+
+private predicate analyzableConversion(Conversion conv) {
+  strictcount(conv.getUnspecifiedType()) = 1 and
+  strictcount(conv.getExpr()) = 1 and
+  not analyzableConst(conv)
+}
+
+deprecated private predicate mk_Conversion(Type t, GVN child, Conversion conv) {
+  analyzableConversion(conv) and
+  t = conv.getUnspecifiedType() and
+  child = globalValueNumber(conv.getExpr())
+}
+
+private predicate analyzableBinaryOp(BinaryOperation op) {
+  op.isPure() and
+  strictcount(op.getLeftOperand().getFullyConverted()) = 1 and
+  strictcount(op.getRightOperand().getFullyConverted()) = 1 and
+  strictcount(op.getOperator()) = 1 and
+  not analyzableConst(op)
+}
+
+deprecated private predicate mk_BinaryOp(GVN lhs, GVN rhs, string opname, BinaryOperation op) {
+  analyzableBinaryOp(op) and
+  lhs = globalValueNumber(op.getLeftOperand().getFullyConverted()) and
+  rhs = globalValueNumber(op.getRightOperand().getFullyConverted()) and
+  opname = op.getOperator()
+}
+
+private predicate analyzableUnaryOp(UnaryOperation op) {
+  not op instanceof PointerDereferenceExpr and
+  op.isPure() and
+  strictcount(op.getOperand().getFullyConverted()) = 1 and
+  strictcount(op.getOperator()) = 1 and
+  not analyzableConst(op)
+}
+
+deprecated private predicate mk_UnaryOp(GVN child, string opname, UnaryOperation op) {
+  analyzableUnaryOp(op) and
+  child = globalValueNumber(op.getOperand().getFullyConverted()) and
+  opname = op.getOperator()
+}
+
+private predicate analyzableThisExpr(ThisExpr thisExpr) {
+  strictcount(thisExpr.getEnclosingFunction()) = 1 and
+  not analyzableConst(thisExpr)
+}
+
+private predicate mk_ThisExpr(Function fcn, ThisExpr thisExpr) {
+  analyzableThisExpr(thisExpr) and
+  fcn = thisExpr.getEnclosingFunction()
+}
+
+private predicate analyzableArrayAccess(ArrayExpr ae) {
+  strictcount(ae.getArrayBase().getFullyConverted()) = 1 and
+  strictcount(ae.getArrayOffset().getFullyConverted()) = 1 and
+  strictcount(mostRecentSideEffect(ae)) = 1 and
+  not analyzableConst(ae)
+}
+
+deprecated private predicate mk_ArrayAccess(
+  GVN base, GVN offset, ControlFlowNode dominator, ArrayExpr ae
+) {
+  analyzableArrayAccess(ae) and
+  base = globalValueNumber(ae.getArrayBase().getFullyConverted()) and
+  offset = globalValueNumber(ae.getArrayOffset().getFullyConverted()) and
+  dominator = mostRecentSideEffect(ae)
+}
+
+private predicate analyzablePointerDereferenceExpr(PointerDereferenceExpr deref) {
+  strictcount(deref.getOperand().getFullyConverted()) = 1 and
+  strictcount(mostRecentSideEffect(deref)) = 1 and
+  not analyzableConst(deref)
+}
+
+deprecated private predicate mk_Deref(GVN p, ControlFlowNode dominator, PointerDereferenceExpr deref) {
+  analyzablePointerDereferenceExpr(deref) and
+  p = globalValueNumber(deref.getOperand().getFullyConverted()) and
+  dominator = mostRecentSideEffect(deref)
+}
+
+/** Gets the global value number of expression `e`. */
+cached
+deprecated GVN globalValueNumber(Expr e) {
+  exists(int val, Type t |
+    mk_IntConst(val, t, e) and
+    result = GVN_IntConst(val, t)
+  )
+  or
+  exists(float val, Type t |
+    mk_FloatConst(val, t, e) and
+    result = GVN_FloatConst(val, t)
+  )
+  or
+  // Local variable with a defining value.
+  exists(StackVariable x, SsaDefinition def |
+    analyzableStackVariable(e) and
+    e = def.getAUse(x) and
+    result = globalValueNumber(def.getDefiningValue(x).getFullyConverted())
+  )
+  or
+  // Local variable without a defining value.
+  exists(StackVariable x, SsaDefinition def |
+    mk_UndefinedStackVariable(x, def, e) and
+    result = GVN_UndefinedStackVariable(x, def)
+  )
+  or
+  // Variable with no SSA information.
+  exists(Variable x, ControlFlowNode dominator |
+    mk_OtherVariable(x, dominator, e) and
+    result = GVN_OtherVariable(x, dominator)
+  )
+  or
+  exists(GVN qualifier, Field target |
+    mk_DotFieldAccess(qualifier, target, e) and
+    result = GVN_FieldAccess(qualifier, target)
+  )
+  or
+  exists(GVN qualifier, Field target |
+    mk_PointerFieldAccess_with_deref(qualifier, target, e) and
+    result = GVN_FieldAccess(qualifier, target)
+  )
+  or
+  exists(GVN qualifier, Field target |
+    mk_ImplicitThisFieldAccess_with_deref(qualifier, target, e) and
+    result = GVN_FieldAccess(qualifier, target)
+  )
+  or
+  exists(Function fcn |
+    mk_ThisExpr(fcn, e) and
+    result = GVN_ThisExpr(fcn)
+  )
+  or
+  exists(Type t, GVN child |
+    mk_Conversion(t, child, e) and
+    result = GVN_Conversion(t, child)
+  )
+  or
+  exists(GVN lhs, GVN rhs, string opname |
+    mk_BinaryOp(lhs, rhs, opname, e) and
+    result = GVN_BinaryOp(lhs, rhs, opname)
+  )
+  or
+  exists(GVN child, string opname |
+    mk_UnaryOp(child, opname, e) and
+    result = GVN_UnaryOp(child, opname)
+  )
+  or
+  exists(GVN x, GVN i, ControlFlowNode dominator |
+    mk_ArrayAccess(x, i, dominator, e) and
+    result = GVN_ArrayAccess(x, i, dominator)
+  )
+  or
+  exists(GVN p, ControlFlowNode dominator |
+    mk_Deref(p, dominator, e) and
+    result = GVN_Deref(p, dominator)
+  )
+  or
+  not analyzableExpr(e) and result = GVN_Unanalyzable(e)
+}
+
+private predicate analyzableConst(Expr e) {
+  analyzableIntConst(e) or
+  analyzableFloatConst(e)
+}
+
+/**
+ * Holds if the expression is explicitly handled by `globalValueNumber`.
+ * Unanalyzable expressions still need to be given a global value number,
+ * but it will be a unique number that is not shared with any other
+ * expression.
+ */
+private predicate analyzableExpr(Expr e) {
+  analyzableConst(e) or
+  analyzableStackVariable(e) or
+  analyzableDotFieldAccess(e) or
+  analyzablePointerFieldAccess(e) or
+  analyzableImplicitThisFieldAccess(e) or
+  analyzableOtherVariable(e) or
+  analyzableConversion(e) or
+  analyzableBinaryOp(e) or
+  analyzableUnaryOp(e) or
+  analyzableThisExpr(e) or
+  analyzableArrayAccess(e) or
+  analyzablePointerDereferenceExpr(e)
+}

cpp/ql/lib/upgrades/ddd31fd02e51ad270bc9e6712708e5a5b6881518/upgrade.properties

@@ -1,4 +1,4 @@
 description: Removed unused column from the `folders` and `files` relations
 compatibility: full
-files.rel: reorder files.rel (@file id, string name, string simple, string ext, int fromSource) id name
-folders.rel: reorder folders.rel (@folder id, string name, string simple) id name
+files.rel: reorder files.rel (int id, string name, string simple, string ext, int fromSource) id name
+folders.rel: reorder folders.rel (int id, string name, string simple) id name

cpp/ql/src/CHANGELOG.md

@@ -1,28 +1,3 @@
-## 1.0.0
-
-### Breaking Changes
-
-* CodeQL package management is now generally available, and all GitHub-produced CodeQL packages have had their version numbers increased to 1.0.0.
-
-### Minor Analysis Improvements
-
-* The "Use of unique pointer after lifetime ends" query (`cpp/use-of-unique-pointer-after-lifetime-ends`) no longer reports an alert when the pointer is converted to a boolean
-* The "Variable not initialized before use" query (`cpp/not-initialised`) no longer reports an alert on static variables.
-
-## 0.9.12
-
-### New Queries
-
-* Added a new query, `cpp/iterator-to-expired-container`, to detect the creation of iterators owned by a temporary objects that are about to be destroyed.
-
-## 0.9.11
-
-### Minor Analysis Improvements
-
-* The "Uncontrolled data used in path expression" query (`cpp/path-injection`) query produces fewer near-duplicate results.
-* The "Global variable may be used before initialization" query (`cpp/global-use-before-init`) no longer raises an alert on global variables that are initialized when they are declared.
-* The "Inconsistent null check of pointer" query (`cpp/inconsistent-nullness-testing`) query no longer raises an alert when the guarded check is in a macro expansion.
-
 ## 0.9.10
 
 No user-facing changes.

cpp/ql/src/Critical/DoubleFree.qhelp

@@ -14,32 +14,13 @@ the program, or security vulnerabilities, by allowing an attacker to overwrite a
 </overview>
 <recommendation>
 <p>
-Ensure that all execution paths deallocate the allocated memory at most once. In complex cases it may
-help to reassign a pointer to a null value after deallocating it. This will prevent double-free vulnerabilities
-since most deallocation functions will perform a null-pointer check before attempting to deallocate memory.
+Ensure that all execution paths deallocate the allocated memory at most once. If possible, reassign
+the pointer to a null value after deallocating it. This will prevent double-free vulnerabilities since
+most deallocation functions will perform a null-pointer check before attempting to deallocate the memory.
 </p>
 
 </recommendation>
-<example>
-<p>
-In the following example, <code>buff</code> is allocated and then freed twice:
-</p>
-<sample src="DoubleFreeBad.cpp" />
-<p>
-Reviewing the code above, the issue can be fixed by simply deleting the additional call to
-<code>free(buff)</code>.
-</p>
-<sample src="DoubleFreeGood.cpp" />
-<p>
-In the next example, <code>task</code> may be deleted twice, if an exception occurs inside the <code>try</code>
-block after the first <code>delete</code>:
-</p>
-<sample src="DoubleFreeBad2.cpp" />
-<p>
-The problem can be solved by assigning a null value to the pointer after the first <code>delete</code>, as
-calling <code>delete</code> a second time on the null pointer is harmless.
-</p>
-<sample src="DoubleFreeGood2.cpp" />
+<example><sample src="DoubleFree.cpp" />
 </example>
 <references>
 

cpp/ql/src/Critical/DoubleFreeBad2.cpp

@@ -1,16 +0,0 @@
-void g() {
-	MyTask *task = nullptr;
-
-	try
-	{
-		task = new MyTask;
-
-		...
-
-		delete task;
-
-		...
-	} catch (...) {
-		delete task; // BAD: potential double-free
-	}
-}

cpp/ql/src/Critical/DoubleFreeGood.cpp

@@ -1,7 +0,0 @@
-int* f() {
-	int *buff = malloc(SIZE*sizeof(int));
-	do_stuff(buff);
-	free(buff); // GOOD: buff is only freed once.
-	int *new_buffer = malloc(SIZE*sizeof(int));
-	return new_buffer;
-}

cpp/ql/src/Critical/DoubleFreeGood2.cpp

@@ -1,17 +0,0 @@
-void g() {
-	MyTask *task = nullptr;
-
-	try
-	{
-		task = new MyTask;
-
-		...
-
-		delete task;
-		task = nullptr;
-
-		...
-	} catch (...) {
-		delete task; // GOOD: harmless if task is NULL
-	}
-}

cpp/ql/src/Critical/NotInitialised.ql

@@ -54,7 +54,6 @@ predicate undefinedLocalUse(VariableAccess va) {
     // it is hard to tell when a struct or array has been initialized, so we
     // ignore them
     not isAggregateType(lv.getUnderlyingType()) and
-    not lv.isStatic() and // static variables are initialized to zero or null by default
     not lv.getType().hasName("va_list") and
     va = lv.getAnAccess() and
     noDefPath(lv, va) and
@@ -71,8 +70,7 @@ predicate uninitialisedGlobal(GlobalVariable gv) {
     va = gv.getAnAccess() and
     va.isRValue() and
     not gv.hasInitializer() and
-    not gv.hasSpecifier("extern") and
-    not gv.isStatic() // static variables are initialized to zero or null by default
+    not gv.hasSpecifier("extern")
   )
 }
 

cpp/ql/src/Likely Bugs/Format/NonConstantFormat.qhelp

@@ -42,7 +42,7 @@ in the previous example, one solution is to make the log message a trailing argu
 <p>An alternative solution is to allow <code>log_with_timestamp</code> to accept format arguments:</p>
 <sample src="NonConstantFormat-2-good.c" />
 <p>In this formulation, the non-constant format string to <code>printf</code> has been replaced with
-a non-constant format string to <code>vprintf</code>. The analysis will no longer consider the body of
+a non-constant format string to <code>vprintf</code>. Semmle will no longer consider the body of
 <code>log_with_timestamp</code> to be a problem, and will instead check that every call to
 <code>log_with_timestamp</code> passes a constant format string.</p>
 

cpp/ql/src/Likely Bugs/Format/TooManyFormatArguments.qhelp

@@ -22,8 +22,10 @@ function.
 </example>
 <references>
 
-<li>CERT C Coding Standard: <a href="https://wiki.sei.cmu.edu/confluence/display/c/FIO47-C.+Use+valid+format+strings">FIO47-C. Use valid format strings</a>.</li>
+<li>cplusplus.com: <a href="http://www.tutorialspoint.com/cplusplus/cpp_functions.htm">C++ Functions</a>.</li>
 <li>Microsoft C Runtime Library Reference: <a href="https://docs.microsoft.com/en-us/cpp/c-runtime-library/reference/printf-printf-l-wprintf-wprintf-l">printf, wprintf</a>.</li>
 
+
+
 </references>
 </qhelp>

cpp/ql/src/Likely Bugs/Format/WrongNumberOfFormatArguments.qhelp

@@ -19,8 +19,8 @@ contents.
 
 </overview>
 <recommendation>
-<p>Review the format and arguments expected by the highlighted function calls. Update either
-the format or the arguments so that the expected number of arguments are passed to the
+<p>Review the format and arguments expected by the highlighted function calls. Update either 
+the format or the arguments so that the expected number of arguments are passed to the 
 function.
 </p>
 
@@ -30,8 +30,11 @@ function.
 </example>
 <references>
 
-<li>CERT C Coding Standard: <a href="https://wiki.sei.cmu.edu/confluence/display/c/FIO47-C.+Use+valid+format+strings">FIO47-C. Use valid format strings</a>.</li>
+<li>CERT C Coding
+Standard: <a href="https://www.securecoding.cert.org/confluence/display/c/FIO30-C.+Exclude+user+input+from+format+strings">FIO30-C. Exclude user input from format strings</a>.</li>
+<li>cplusplus.com: <a href="http://www.tutorialspoint.com/cplusplus/cpp_functions.htm">C++ Functions</a>.</li>
 <li>Microsoft C Runtime Library Reference: <a href="https://docs.microsoft.com/en-us/cpp/c-runtime-library/reference/printf-printf-l-wprintf-wprintf-l">printf, wprintf</a>.</li>
 
+
 </references>
 </qhelp>

cpp/ql/src/Likely Bugs/Format/WrongTypeFormatArguments.cpp

@@ -0,0 +1,4 @@
+int main() {
+  printf("%s\n", 42); //printf will treat 42 as a char*, will most likely segfault
+  return 0;
+}

cpp/ql/src/Likely Bugs/Format/WrongTypeFormatArguments.qhelp

@@ -4,33 +4,29 @@
 <qhelp>
 <overview>
 <p>Each call to the <code>printf</code> function or a related function should include
-the type and sequence of arguments defined by the format. If the function is passed arguments
+the type and sequence of arguments defined by the format. If the function is passed arguments 
 of a different type or in a different sequence then the arguments are reinterpreted to fit the type and sequence expected, resulting in unpredictable behavior.</p>
 
 </overview>
 <recommendation>
-<p>Review the format and arguments expected by the highlighted function calls. Update either
-the format or the arguments so that the expected type and sequence of arguments are passed to
+<p>Review the format and arguments expected by the highlighted function calls. Update either 
+the format or the arguments so that the expected type and sequence of arguments are passed to 
 the function.
 </p>
 
 </recommendation>
-<example>
-
-<p>In the following example, the wrong format specifier is given for an integer format argument:</p>
-
-<sample src="WrongTypeFormatArgumentsBad.cpp" />
-
-<p>The corrected version uses <code>%i</code> as the format specifier for the integer format argument:</p>
-
-<sample src="WrongTypeFormatArgumentsGood.cpp" />
+<example><sample src="WrongTypeFormatArguments.cpp" />
 
 </example>
 <references>
 
-<li>Microsoft Learn: <a href="https://learn.microsoft.com/en-us/cpp/c-runtime-library/format-specification-syntax-printf-and-wprintf-functions?view=msvc-170">Format specification syntax: printf and wprintf functions</a>.</li>
-<li>cplusplus.com:<a href="https://cplusplus.com/reference/cstdio/printf/"></a>printf</li>
-<li>CERT C Coding Standard: <a href="https://wiki.sei.cmu.edu/confluence/display/c/FIO47-C.+Use+valid+format+strings">FIO47-C. Use valid format strings</a>.</li>
+<li>CERT C Coding
+Standard: <a href="https://www.securecoding.cert.org/confluence/display/c/FIO30-C.+Exclude+user+input+from+format+strings">FIO30-C. Exclude user input from format strings</a>.</li>
+<li>cplusplus.com: <a href="http://www.tutorialspoint.com/cplusplus/cpp_functions.htm">C++ Functions</a>.</li>
+<li>CRT Alphabetical Function Reference: <a href="https://docs.microsoft.com/en-us/cpp/c-runtime-library/reference/printf-printf-l-wprintf-wprintf-l">printf, _printf_l, wprintf, _wprintf_l</a>.</li>
+
+
+
 
 </references>
 </qhelp>

cpp/ql/src/Likely Bugs/Format/WrongTypeFormatArgumentsBad.cpp

@@ -1,4 +0,0 @@
-int main() {
-  printf("%s\n", 42); // BAD: printf will treat 42 as a char*, will most likely segfault
-  return 0;
-}

cpp/ql/src/Likely Bugs/Format/WrongTypeFormatArgumentsGood.cpp

@@ -1,4 +0,0 @@
-int main() {
-  printf("%i\n", 42); // GOOD: printf will treat 42 as an int
-  return 0;
-}

cpp/ql/src/Likely Bugs/Likely Typos/IncorrectNotOperatorUsage.cpp

@@ -2,18 +2,19 @@
 
 void f_warning(int i)
 {
-    // BAD: the usage of the logical not operator in this case is unlikely to be correct
+    // The usage of the logical not operator in this case is unlikely to be correct
     // as the output is being used as an operator for a bit-wise and operation
-    if (i & !FLAGS)
+    if (i & !FLAGS) 
     {
         // code
     }
 }
 
+
 void f_fixed(int i)
 {
-    if (i & ~FLAGS) // GOOD: Changing the logical not operator for the bit-wise not operator would fix this logic
+    if (i & ~FLAGS) // Changing the logical not operator for the bit-wise not operator would fix this logic
     {
         // code
     }
-}
+}

cpp/ql/src/Likely Bugs/Likely Typos/IncorrectNotOperatorUsage.qhelp

@@ -16,13 +16,7 @@
 <p>Carefully inspect the flagged expressions. Consider the intent in the code logic, and decide whether it is necessary to change the not operator.</p>
 </recommendation>
 
-<example>
-<p>Here is an example of this issue and how it can be fixed:</p>
-
-<sample src="IncorrectNotOperatorUsage.cpp" />
-
-<p>In other cases, particularly when the expressions have <code>bool</code> type, the fix may instead be of the form <code>a &amp;&amp; !b</code>.</p>
-</example>
+<example><sample src="IncorrectNotOperatorUsage.cpp" /></example>
 
 <references>
   <li>

cpp/ql/src/Likely Bugs/Memory Management/AllocaInLoop.ql

@@ -37,19 +37,6 @@ class AllocaCall extends FunctionCall {
   }
 }
 
-/**
- * Gets an expression associated with a dataflow node.
- */
-private Expr getExpr(DataFlow::Node node) {
-  result = node.asInstruction().getAst()
-  or
-  result = node.asOperand().getUse().getAst()
-  or
-  result = node.(DataFlow::RawIndirectInstruction).getInstruction().getAst()
-  or
-  result = node.(DataFlow::RawIndirectOperand).getOperand().getUse().getAst()
-}
-
 /**
  * A loop that contains an `alloca` call.
  */
@@ -198,6 +185,19 @@ class LoopWithAlloca extends Stmt {
     not this.conditionReachesWithoutUpdate(var, this.(Loop).getCondition())
   }
 
+  /**
+   * Gets an expression associated with a dataflow node.
+   */
+  private Expr getExpr(DataFlow::Node node) {
+    result = node.asInstruction().getAst()
+    or
+    result = node.asOperand().getUse().getAst()
+    or
+    result = node.(DataFlow::RawIndirectInstruction).getInstruction().getAst()
+    or
+    result = node.(DataFlow::RawIndirectOperand).getOperand().getUse().getAst()
+  }
+
   /**
    * Gets a definition that may be the most recent definition of the
    * controlling variable `var` before this loop.
@@ -210,7 +210,7 @@ class LoopWithAlloca extends Stmt {
       // Phi nodes will be preceded by nodes that represent actual definitions
       not result instanceof DataFlow::SsaPhiNode and
       // A source is outside the loop if it's not inside the loop
-      not exists(Expr e | e = getExpr(result) | this = getAnEnclosingLoopOfExpr(e))
+      not exists(Expr e | e = this.getExpr(result) | this = getAnEnclosingLoopOfExpr(e))
     )
   }
 
@@ -221,9 +221,9 @@ class LoopWithAlloca extends Stmt {
   private int getAControllingVarInitialValue(Variable var, DataFlow::Node source) {
     source = this.getAPrecedingDef(var) and
     (
-      result = getExpr(source).getValue().toInt()
+      result = this.getExpr(source).getValue().toInt()
       or
-      result = getExpr(source).(Assignment).getRValue().getValue().toInt()
+      result = this.getExpr(source).(Assignment).getRValue().getValue().toInt()
     )
   }
 

cpp/ql/src/Likely Bugs/Memory Management/NtohlArrayNoBound.qll

@@ -107,7 +107,7 @@ class SnprintfSizeExpr extends BufferAccess, FunctionCall {
 }
 
 class MemcmpSizeExpr extends BufferAccess, FunctionCall {
-  MemcmpSizeExpr() { this.getTarget().hasName("memcmp") }
+  MemcmpSizeExpr() { this.getTarget().hasName("Memcmp") }
 
   override Expr getPointer() {
     result = this.getArgument(0) or

cpp/ql/src/Likely Bugs/Memory Management/StrncpyFlippedArgs.cpp

@@ -0,0 +1,2 @@
+strncpy(dest, src, sizeof(src)); //wrong: size of dest should be used
+strncpy(dest, src, strlen(src)); //wrong: size of dest should be used

cpp/ql/src/Likely Bugs/Memory Management/StrncpyFlippedArgs.qhelp

@@ -3,7 +3,7 @@
   "qhelp.dtd">
 <qhelp>
 <overview>
-<p>The standard library function <code>strncpy</code> copies a source string to a destination buffer. The third argument defines the maximum number of characters to copy and should be less than
+<p>The standard library function <code>strncpy</code> copies a source string to a destination buffer. The third argument defines the maximum number of characters to copy and should be less than 
 or equal to the size of the destination buffer. Calls of the form <code>strncpy(dest, src, strlen(src))</code> or <code>strncpy(dest, src, sizeof(src))</code> incorrectly set the third argument to the size of the source buffer. Executing a call of this type may cause a buffer overflow. Buffer overflows can lead to anything from a segmentation fault to a security vulnerability.</p>
 
 </overview>
@@ -12,20 +12,14 @@ or equal to the size of the destination buffer. Calls of the form <code>strncpy(
 not the source buffer.</p>
 
 </recommendation>
+<example><sample src="StrncpyFlippedArgs.cpp" />
 
-<example>
-<p>In the following examples, the size of the source buffer is incorrectly used as a parameter to <code>strncpy</code>:</p>
 
-<sample src="StrncpyFlippedArgsBad.cpp" />
 
-<p>The corrected version uses the size of the destination buffer, or a variable containing the size of the destination buffer as the size parameter to <code>strncpy</code>:</p>
-
-<sample src="StrncpyFlippedArgsGood.cpp" />
 </example>
-
 <references>
 
-<li>cplusplus.com: <a href="https://cplusplus.com/reference/cstring/strncpy/">strncpy</a>.</li>
+<li>cplusplus.com: <a href="http://www.cplusplus.com/reference/clibrary/cstring/strncpy/">strncpy</a>.</li>
 <li>
   I. Gerg. <em>An Overview and Example of the Buffer-Overflow Exploit</em>. IANewsletter vol 7 no 4. 2005.
 </li>

cpp/ql/src/Likely Bugs/Memory Management/StrncpyFlippedArgsBad.cpp

@@ -1,9 +0,0 @@
-char src[256];
-char dest1[128];
-
-...
-
-strncpy(dest1, src, sizeof(src)); // wrong: size of dest should be used
-
-char *dest2 = (char *)malloc(sz1 + sz2 + sz3);
-strncpy(dest2, src, strlen(src)); // wrong: size of dest should be used

cpp/ql/src/Likely Bugs/Memory Management/StrncpyFlippedArgsGood.cpp

@@ -1,10 +0,0 @@
-char src[256];
-char dest1[128];
-
-...
-
-strncpy(dest1, src, sizeof(dest1)); // correct
-
-size_t destSize = sz1 + sz2 + sz3;
-char *dest2 = (char *)malloc(destSize);
-strncpy(dest2, src, destSize); // correct

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

@@ -0,0 +1,22 @@
+int main(int argc, char** argv) {
+  char *userAndFile = argv[2];
+  
+  {
+    char fileBuffer[FILENAME_MAX] = "/home/";
+    char *fileName = fileBuffer;
+    size_t len = strlen(fileName);
+    strncat(fileName+len, userAndFile, FILENAME_MAX-len-1);
+    // BAD: a string from the user is used in a filename
+    fopen(fileName, "wb+");
+  }
+
+  {
+    char fileBuffer[FILENAME_MAX] = "/home/";
+    char *fileName = fileBuffer;
+    size_t len = strlen(fileName);
+    // GOOD: use a fixed file
+    char* fixed = "jim/file.txt";
+    strncat(fileName+len, fixed, FILENAME_MAX-len-1);
+    fopen(fileName, "wb+");
+  }
+}

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

@@ -3,57 +3,36 @@
   "qhelp.dtd">
 <qhelp>
 <overview>
-<p>Accessing paths controlled by users can allow an attacker to access unexpected resources. This 
+<p>Accessing paths controlled by users can allow an attacker to access unexpected resources. This
 can result in sensitive information being revealed or deleted, or an attacker being able to influence
 behavior by modifying unexpected files.</p>
 
-<p>Paths that are naively constructed from data controlled by a user may be absolute paths, or may contain
-unexpected special characters such as "..". Such a path could point anywhere on the file system.</p>
+<p>Paths that are naively constructed from data controlled by a user may contain unexpected special characters,
+such as "..". Such a path may potentially point to any directory on the filesystem.</p>
 
 </overview>
 <recommendation>
 
-<p>Validate user input before using it to construct a file path.</p>
+<p>Validate user input before using it to construct a filepath. Ideally, follow these rules:</p>
 
-<p>Common validation methods include checking that the normalized path is relative and does not contain
-any ".." components, or checking that the path is contained within a safe folder. The method you should use depends 
-on how the path is used in the application, and whether the path should be a single path component.
-</p>
-
-<p>If the path should be a single path component (such as a file name), you can check for the existence 
-of any path separators ("/" or "\"), or ".." sequences in the input, and reject the input if any are found.
-</p>
-
-<p>
-Note that removing "../" sequences is <i>not</i> sufficient, since the input could still contain a path separator
-followed by "..". For example, the input ".../...//" would still result in the string "../" if only "../" sequences
-are removed.
-</p>
-
-<p>Finally, the simplest (but most restrictive) option is to use an allow list of safe patterns and make sure that
-the user input matches one of these patterns.</p>
+<ul>
+<li>Do not allow more than a single "." character.</li>
+<li>Do not allow directory separators such as "/" or "\" (depending on the filesystem).</li>
+<li>Do not rely on simply replacing problematic sequences such as "../". For example, after applying this filter to
+".../...//" the resulting string would still be "../".</li>
+<li>Ideally use a whitelist of known good patterns.</li>
+</ul>
 
 </recommendation>
 <example>
 
-<p>In this example, a file name is read from a user and then used to access a file. 
-However, a malicious user could enter a file name anywhere on the file system,
-such as "/etc/passwd" or "../../../etc/passwd".</p>
+<p>In this example, a username and file are read from the arguments to main and then used to access a file in the
+user's home directory. However, a malicious user could enter a filename which contains special
+characters. For example, the string "../../etc/passwd" will result in the code reading the file located at
+"/home/[user]/../../etc/passwd", which is the system's password file. This could potentially allow them to
+access all the system's passwords.</p>
 
-<sample src="examples/TaintedPath.c" />
-
-<p>
-If the input should only be a file name, you can check that it doesn't contain any path separators or ".." sequences.
-</p>
-
-<sample src="examples/TaintedPathNormalize.c" />
-
-<p>
-If the input should be within a specific directory, you can check that the resolved path 
-is still contained within that directory.
-</p>
-
-<sample src="examples/TaintedPathFolder.c" />
+<sample src="TaintedPath.c" />
 
 </example>
 <references>
@@ -62,7 +41,6 @@ is still contained within that directory.
 OWASP:
 <a href="https://owasp.org/www-community/attacks/Path_Traversal">Path Traversal</a>.
 </li>
-<li>Linux man pages: <a href="https://man7.org/linux/man-pages/man3/realpath.3.html">realpath(3)</a>.</li>
 
 </references>
 </qhelp>

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

@@ -88,11 +88,6 @@ module TaintedPathConfig implements DataFlow::ConfigSig {
       hasUpperBoundsCheck(checkedVar)
     )
   }
-
-  predicate isBarrierOut(DataFlow::Node node) {
-    // make sinks barriers so that we only report the closest instance
-    isSink(node)
-  }
 }
 
 module TaintedPath = TaintTracking::Global<TaintedPathConfig>;

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

@@ -1,10 +0,0 @@
-int main(int argc, char** argv) {
-  char *userAndFile = argv[2];
-  
-  {
-    char fileBuffer[PATH_MAX];
-    snprintf(fileBuffer, sizeof(fileBuffer), "/home/%s", userAndFile);
-    // BAD: a string from the user is used in a filename
-    fopen(fileBuffer, "wb+");
-  }
-}

cpp/ql/src/Security/CWE/CWE-022/examples/TaintedPathFolder.c

@@ -1,28 +0,0 @@
-#include <stdio.h>
-#include <string.h>
-
-int main(int argc, char** argv) {
-    char *userAndFile = argv[2];
-    const char *baseDir = "/home/user/public/";
-    char fullPath[PATH_MAX];
-
-    // Attempt to concatenate the base directory and the user-supplied path
-    snprintf(fullPath, sizeof(fullPath), "%s%s", baseDir, userAndFile);
-
-    // Resolve the absolute path, normalizing any ".." or "."
-    char *resolvedPath = realpath(fullPath, NULL);
-    if (resolvedPath == NULL) {
-        perror("Error resolving path");
-        return 1;
-    }
-
-    // Check if the resolved path starts with the base directory
-    if (strncmp(baseDir, resolvedPath, strlen(baseDir)) != 0) {
-        free(resolvedPath);
-        return 1;
-    }
-
-    // GOOD: Path is within the intended directory
-    FILE *file = fopen(resolvedPath, "wb+");
-    free(resolvedPath);
-}

cpp/ql/src/Security/CWE/CWE-022/examples/TaintedPathNormalize.c

@@ -1,16 +0,0 @@
-#include <stdio.h>
-#include <string.h>
-
-int main(int argc, char** argv) {
-    char *fileName = argv[2];
-    // Check for invalid sequences in the user input
-    if (strstr(fileName , "..") || strchr(fileName , '/') || strchr(fileName , '\\')) {
-        printf("Invalid filename.\n");
-        return 1;
-    }
-
-    char fileBuffer[PATH_MAX];
-    snprintf(fileBuffer, sizeof(fileBuffer), "/home/user/files/%s", fileName);
-    // GOOD: We know that the filename is safe and stays within the public folder
-    FILE *file = fopen(fileBuffer, "wb+");
-}

cpp/ql/src/Security/CWE/CWE-131/NoSpaceForZeroTerminator.qhelp

@@ -12,8 +12,8 @@ the required buffer size, but do not allocate space for the zero terminator.
 </overview>
 <recommendation>
 <p>
-The highlighted code segment creates a buffer without ensuring it's large enough to accommodate the copied data.
-This leaves the code susceptible to a buffer overflow attack, which could lead to anything from program crashes to malicious code execution.
+The expression highlighted by this rule creates a buffer that is of insufficient size to contain
+the data being copied. This makes the code vulnerable to buffer overflow which can result in anything from a segmentation fault to a security vulnerability (particularly if the array is on stack-allocated memory).
 </p>
 
 <p>

cpp/ql/src/Security/CWE/CWE-416/IteratorToExpiredContainerExtendedLifetime-fixed.cpp

@@ -1,6 +0,0 @@
-void fixed_lifetime_of_temp_not_extended() {
-  auto&& v = get_vector();
-  for(auto x : log_and_return_argument(v)) {
-    use(x); // GOOD: The lifetime of the container returned by `get_vector()` has been extended to the lifetime of `v`.
-  }
-}

cpp/ql/src/Security/CWE/CWE-416/UseOfStringAfterLifetimeEnds.qhelp

@@ -8,12 +8,6 @@
 When the <code>std::string</code> object is destroyed, the pointer returned by <code>c_str</code> is no
 longer valid. If the pointer is used after the <code>std::string</code> object is destroyed, then the behavior is undefined.
 </p>
-
-<p>Typically, this problem occurs when a <code>std::string</code> is returned by a function call (or overloaded operator)
-by value, and the result is not immediately stored in a variable by value or reference in a way that extends the lifetime of
-the temporary object. The resulting temporary <code>std::string</code> object is destroyed at the end of the containing expression
-statement, along with any memory returned by a call to <code>c_str</code>.
-</p>
 </overview>
 
 <recommendation>
@@ -45,8 +39,6 @@ points to valid memory.
 <references>
 
 <li><a href="https://wiki.sei.cmu.edu/confluence/display/cplusplus/MEM50-CPP.+Do+not+access+freed+memory">MEM50-CPP. Do not access freed memory</a>.</li>
-<li>Microsoft Learn: <a href="https://learn.microsoft.com/en-us/cpp/cpp/temporary-objects?view=msvc-170">Temporary objects</a>.</li>
-<li>cppreference.com: <a href="https://en.cppreference.com/w/cpp/language/reference_initialization#Lifetime_of_a_temporary">Lifetime of a temporary</a>.</li>
 
 </references>
 </qhelp>

cpp/ql/src/Security/CWE/CWE-416/UseOfStringAfterLifetimeEnds.ql

@@ -23,5 +23,4 @@ where
   (c.getTarget() instanceof StdStringCStr or c.getTarget() instanceof StdStringData) and
   isTemporary(c.getQualifier().getFullyConverted())
 select c,
-  "The underlying temporary string object is destroyed after the call to '" + c.getTarget() +
-    "' returns."
+  "The underlying string object is destroyed after the call to '" + c.getTarget() + "' returns."

cpp/ql/src/Security/CWE/CWE-416/UseOfUniquePointerAfterLifetimeEnds.ql

@@ -30,8 +30,6 @@ where
   outlivesFullExpr(c) and
   not c.isFromUninstantiatedTemplate(_) and
   isUniquePointerDerefFunction(c.getTarget()) and
-  // Exclude cases where the pointer is implicitly converted to a non-pointer type
-  not c.getActualType() instanceof IntegralType and
   isTemporary(c.getQualifier().getFullyConverted())
 select c,
   "The underlying unique pointer object is destroyed after the call to '" + c.getTarget() +

cpp/ql/src/change-notes/2024-04-09-reduce-FP.md

@@ -1,7 +1,5 @@
-## 0.9.11
-
-### Minor Analysis Improvements
-
-* The "Uncontrolled data used in path expression" query (`cpp/path-injection`) query produces fewer near-duplicate results.
+---
+category: minorAnalysis
+---
 * The "Global variable may be used before initialization" query (`cpp/global-use-before-init`) no longer raises an alert on global variables that are initialized when they are declared.
-* The "Inconsistent null check of pointer" query (`cpp/inconsistent-nullness-testing`) query no longer raises an alert when the guarded check is in a macro expansion.
+* The "Inconsistent null check of pointer" query (`cpp/inconsistent-nullness-testing`) query no longer raises an alert when the guarded check is in a macro expansion.