add test for sorted test postprocess result

Swift: add `git lfs` installation to codespace configuration

.bazelrc

@@ -10,15 +10,19 @@ 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
-build:macos --cxxopt=-std=c++20 --cpu=darwin_x86_64
-build:windows --cxxopt=/std:c++20 --cxxopt=/Zc:preprocessor
+# we use transitions that break builds of `...`, so for `test` to work with that we need the following
+test --build_tests_only
 
 # this requires developer mode, but is required to have pack installer functioning
 startup --windows_enable_symlinks
 common --enable_runfiles
 
+# with the above, we can avoid building python zips which is the default on windows as that's expensive
+build --nobuild_python_zip
+
 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

@@ -2,3 +2,9 @@
 
 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
+7.2.1

.devcontainer/swift/root.sh

@@ -3,6 +3,16 @@ set -xe
 BAZELISK_VERSION=v1.12.0
 BAZELISK_DOWNLOAD_SHA=6b0bcb2ea15bca16fffabe6fda75803440375354c085480fe361d2cbf32501db
 
+# install git lfs apt source
+curl -s https://packagecloud.io/install/repositories/github/git-lfs/script.deb.sh | bash
+
+# install gh apt source
+(type -p wget >/dev/null || (sudo apt update && sudo apt-get install wget -y)) \
+&& sudo mkdir -p -m 755 /etc/apt/keyrings \
+&& wget -qO- https://cli.github.com/packages/githubcli-archive-keyring.gpg | sudo tee /etc/apt/keyrings/githubcli-archive-keyring.gpg > /dev/null \
+&& sudo chmod go+r /etc/apt/keyrings/githubcli-archive-keyring.gpg \
+&& echo "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/githubcli-archive-keyring.gpg] https://cli.github.com/packages stable main" | sudo tee /etc/apt/sources.list.d/github-cli.list > /dev/null \
+
 apt-get update
 export DEBIAN_FRONTEND=noninteractive
 apt-get -y install --no-install-recommends \
@@ -10,7 +20,9 @@ apt-get -y install --no-install-recommends \
     uuid-dev \
     python3-distutils \
     python3-pip \
-    bash-completion
+    bash-completion \
+    git-lfs \
+    gh
 
 # Install Bazel
 curl -fSsL -o /usr/local/bin/bazelisk https://github.com/bazelbuild/bazelisk/releases/download/${BAZELISK_VERSION}/bazelisk-linux-amd64

.devcontainer/swift/user.sh

@@ -1,5 +1,7 @@
 set -xe
 
+git lfs install
+
 # add the workspace to the codeql search path
 mkdir -p /home/vscode/.config/codeql
 echo "--search-path /workspaces/codeql" > /home/vscode/.config/codeql/config

.gitattributes

@@ -50,31 +50,40 @@
 *.dll -text
 *.pdb -text
 
-java/ql/test/stubs/**/*.java linguist-generated=true
-java/ql/test/experimental/stubs/**/*.java linguist-generated=true
+/java/ql/test/stubs/**/*.java linguist-generated=true
+/java/ql/test/experimental/stubs/**/*.java linguist-generated=true
+/java/kotlin-extractor/deps/*.jar filter=lfs diff=lfs merge=lfs -text
 
 # Force git not to modify line endings for go or html files under the go/ql directory
-go/ql/**/*.go -text
-go/ql/**/*.html -text
+/go/ql/**/*.go -text
+/go/ql/**/*.html -text
 # Force git not to modify line endings for go dbschemes
-go/*.dbscheme -text
+/go/*.dbscheme -text
 # Preserve unusual line ending from codeql-go merge
-go/extractor/opencsv/CSVReader.java -text
+/go/extractor/opencsv/CSVReader.java -text
 
 # For some languages, upgrade script testing references really old dbscheme
 # files from legacy upgrades that have CRLF line endings. Since upgrade
 # resolution relies on object hashes, we must suppress line ending conversion
 # for those testing dbscheme files.
-*/ql/lib/upgrades/initial/*.dbscheme -text
+/*/ql/lib/upgrades/initial/*.dbscheme -text
 
 # Auto-generated modeling for Python
-python/ql/lib/semmle/python/frameworks/data/internal/subclass-capture/*.yml linguist-generated=true
+/python/ql/lib/semmle/python/frameworks/data/internal/subclass-capture/*.yml linguist-generated=true
 
 # auto-generated bazel lock file
-ruby/extractor/cargo-bazel-lock.json linguist-generated=true
-ruby/extractor/cargo-bazel-lock.json -merge
+/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.lock linguist-generated=true
+# needs eol=crlf, as `paket` touches this file and saves it as 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
+
+# ripunzip tool
+/misc/ripunzip/ripunzip-* filter=lfs diff=lfs merge=lfs -text
+
+# swift prebuilt resources
+/swift/third_party/resource-dir/*.zip filter=lfs diff=lfs merge=lfs -text

.github/workflows/build-ripunzip.yml

@@ -0,0 +1,74 @@
+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/csharp-qltest.yml

@@ -65,7 +65,7 @@ jobs:
           key: csharp-qltest-${{ matrix.slice }}
       - name: Run QL tests
         run: |
-          codeql test run --threads=0 --ram 50000 --slice ${{ matrix.slice }} --search-path extractor-pack --check-databases --check-undefined-labels --check-repeated-labels --check-redefined-labels --consistency-queries ql/consistency-queries ql/test --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
+          codeql test run --threads=0 --ram 50000 --slice ${{ matrix.slice }} --search-path "${{ github.workspace }}" --check-databases --check-undefined-labels --check-repeated-labels --check-redefined-labels --consistency-queries ql/consistency-queries ql/test --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
         env:
           GITHUB_TOKEN: ${{ github.token }}
   unit-tests:
@@ -101,6 +101,6 @@ jobs:
           # Update existing stubs in the repo with the freshly generated ones
           mv "$STUBS_PATH/output/stubs/_frameworks" ql/test/resources/stubs/
           git status
-          codeql test run --threads=0 --search-path extractor-pack --check-databases --check-undefined-labels --check-repeated-labels --check-redefined-labels --consistency-queries ql/consistency-queries -- ql/test/library-tests/dataflow/flowsources/aspremote
+          codeql test run --threads=0 --search-path "${{ github.workspace }}" --check-databases --check-undefined-labels --check-repeated-labels --check-redefined-labels --consistency-queries ql/consistency-queries -- ql/test/library-tests/dataflow/flowsources/aspremote
         env:
           GITHUB_TOKEN: ${{ github.token }}

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

@@ -7,6 +7,9 @@ on:
       - .github/workflows/go-tests-other-os.yml
       - .github/actions/**
       - codeql-workspace.yml
+      - MODULE.bazel
+      - .bazelrc
+      - misc/bazel/**
 
 permissions:
   contents: read

.github/workflows/go-tests.yml

@@ -15,6 +15,9 @@ on:
       - .github/workflows/go-tests.yml
       - .github/actions/**
       - codeql-workspace.yml
+      - MODULE.bazel
+      - .bazelrc
+      - misc/bazel/**
 
 permissions:
   contents: read

.github/workflows/kotlin-build.yml

@@ -0,0 +1,28 @@
+name: "Kotlin Build"
+
+on:
+  pull_request:
+    paths:
+      - "java/kotlin-extractor/**"
+      - "misc/bazel/**"
+      - "misc/codegen/**"
+      - "*.bazel*"
+      - .github/workflows/kotlin-build.yml
+    branches:
+      - main
+      - rc/*
+      - codeql-cli-*
+
+permissions:
+  contents: read
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - run: |
+          bazel query //java/kotlin-extractor/...
+          # only build the default version as a quick check that we can build from `codeql`
+          # the full official build will be checked by QLucie
+          bazel build //java/kotlin-extractor

.github/workflows/ql-for-ql-build.yml

@@ -49,20 +49,20 @@ jobs:
           key: ${{ runner.os }}-${{ steps.os_version.outputs.version }}-rust-cargo-${{ hashFiles('ql/**/Cargo.lock') }}
       - name: Release build
         if: steps.cache-extractor.outputs.cache-hit != 'true'
-        run: cd ql; ./scripts/create-extractor-pack.sh       
+        run: cd ql; ./scripts/create-extractor-pack.sh
         env:
-          GH_TOKEN: ${{ github.token }}   
+          GH_TOKEN: ${{ github.token }}
       - name: Cache compilation cache
         id: query-cache
         uses: ./.github/actions/cache-query-compilation
-        with: 
+        with:
           key: run-ql-for-ql
       - name: Make database and analyze
         run: |
           ./ql/target/release/buramu | tee deprecated.blame # Add a blame file for the extractor to parse.
-          ${CODEQL} database create -l=ql --search-path ql/extractor-pack ${DB}
+          ${CODEQL} database create -l=ql ${DB} --search-path "${{ github.workspace }}"
           ${CODEQL} database analyze -j0 --format=sarif-latest --output=ql-for-ql.sarif ${DB} ql/ql/src/codeql-suites/ql-code-scanning.qls  --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
-        env: 
+        env:
           CODEQL: ${{ steps.find-codeql.outputs.codeql-path }}
           DB: ${{ runner.temp }}/DB
           LGTM_INDEX_FILTERS: |

.github/workflows/ql-for-ql-dataset_measure.yml

@@ -53,8 +53,8 @@ jobs:
       - name: Create database
         run: |
           "${CODEQL}" database create \
-            --search-path "ql/extractor-pack" \
-            --threads 4 \
+          --search-path "${{ github.workspace }}"
+          --threads 4 \
             --language ql --source-root "${{ github.workspace }}/repo" \
             "${{ runner.temp }}/database"
         env:

.github/workflows/ql-for-ql-tests.yml

@@ -49,15 +49,15 @@ jobs:
       - name: Cache compilation cache
         id: query-cache
         uses: ./.github/actions/cache-query-compilation
-        with: 
+        with:
           key: ql-for-ql-tests
       - name: Run QL tests
         run: |
-          "${CODEQL}" test run --check-databases --check-unused-labels --check-repeated-labels --check-redefined-labels --check-use-before-definition --search-path "${{ github.workspace }}/ql/extractor-pack" --consistency-queries ql/ql/consistency-queries --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}" ql/ql/test
+          "${CODEQL}" test run --check-databases --check-unused-labels --check-repeated-labels --check-redefined-labels --check-use-before-definition --search-path "${{ github.workspace }}" --consistency-queries ql/ql/consistency-queries --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}" ql/ql/test
         env:
           CODEQL: ${{ steps.find-codeql.outputs.codeql-path }}
 
-  other-os: 
+  other-os:
     strategy:
       matrix:
         os: [macos-latest, windows-latest]
@@ -65,7 +65,7 @@ jobs:
     runs-on: ${{ matrix.os }}
     steps:
       - uses: actions/checkout@v4
-      - name: Install GNU tar 
+      - name: Install GNU tar
         if: runner.os == 'macOS'
         run: |
           brew install gnu-tar
@@ -100,7 +100,7 @@ jobs:
       - name: Run a single QL tests - Unix
         if: runner.os != 'Windows'
         run: |
-          "${CODEQL}" test run --check-databases --search-path "${{ github.workspace }}/ql/extractor-pack" ql/ql/test/queries/style/DeadCode/DeadCode.qlref
+          "${CODEQL}" test run --check-databases --search-path "${{ github.workspace }}" ql/ql/test/queries/style/DeadCode/DeadCode.qlref
         env:
           CODEQL: ${{ steps.find-codeql.outputs.codeql-path }}
       - name: Run a single QL tests - Windows
@@ -108,5 +108,4 @@ jobs:
         shell: pwsh
         run: |
           $Env:PATH += ";$(dirname ${{ steps.find-codeql.outputs.codeql-path }})"
-          codeql test run --check-databases --search-path "${{ github.workspace }}/ql/extractor-pack" ql/ql/test/queries/style/DeadCode/DeadCode.qlref
-      
+          codeql test run --check-databases --search-path "${{ github.workspace }}" ql/ql/test/queries/style/DeadCode/DeadCode.qlref

.github/workflows/ruby-build.yml

@@ -7,6 +7,7 @@ on:
       - .github/workflows/ruby-build.yml
       - .github/actions/fetch-codeql/action.yml
       - codeql-workspace.yml
+      - "shared/tree-sitter-extractor/**"
     branches:
       - main
       - "rc/*"
@@ -16,6 +17,7 @@ on:
       - .github/workflows/ruby-build.yml
       - .github/actions/fetch-codeql/action.yml
       - codeql-workspace.yml
+      - "shared/tree-sitter-extractor/**"
     branches:
       - main
       - "rc/*"

.github/workflows/ruby-dataset-measure.yml

@@ -44,7 +44,7 @@ jobs:
       - name: Create database
         run: |
           codeql database create \
-            --search-path "${{ github.workspace }}/ruby/extractor-pack" \
+            --search-path "${{ github.workspace }}" \
             --threads 4 \
             --language ruby --source-root "${{ github.workspace }}/repo" \
             "${{ runner.temp }}/database"

.github/workflows/ruby-qltest.yml

@@ -64,10 +64,10 @@ jobs:
       - name: Cache compilation cache
         id: query-cache
         uses: ./.github/actions/cache-query-compilation
-        with: 
+        with:
           key: ruby-qltest
       - name: Run QL tests
         run: |
-          codeql test run --threads=0 --ram 50000 --search-path "${{ github.workspace }}/ruby/extractor-pack" --check-databases --check-undefined-labels --check-unused-labels --check-repeated-labels --check-redefined-labels --check-use-before-definition --consistency-queries ql/consistency-queries ql/test --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
+          codeql test run --threads=0 --ram 50000 --search-path "${{ github.workspace }}" --check-databases --check-undefined-labels --check-unused-labels --check-repeated-labels --check-redefined-labels --check-use-before-definition --consistency-queries ql/consistency-queries ql/test --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
         env:
           GITHUB_TOKEN: ${{ github.token }}

.github/workflows/swift.yml

@@ -68,21 +68,6 @@ jobs:
     steps:
       - uses: actions/checkout@v4
       - uses: ./swift/actions/run-ql-tests
-  integration-tests-linux:
-    if: github.repository_owner == 'github'
-    needs: build-and-test-linux
-    runs-on: ubuntu-latest-xl
-    steps:
-      - uses: actions/checkout@v4
-      - uses: ./swift/actions/run-integration-tests
-  integration-tests-macos:
-    if: ${{ github.repository_owner == 'github' && github.event_name == 'pull_request' }}
-    needs: build-and-test-macos
-    runs-on: macos-12-xl
-    timeout-minutes: 60
-    steps:
-      - uses: actions/checkout@v4
-      - uses: ./swift/actions/run-integration-tests
   clang-format:
     if : ${{ github.event_name == 'pull_request' }}
     runs-on: ubuntu-latest

.github/workflows/zipmerge-test.yml

@@ -0,0 +1,23 @@
+name: "Test zipmerge code"
+
+on:
+  pull_request:
+    paths:
+      - "misc/bazel/internal/zipmerge/**"
+      - "MODULE.bazel"
+      - ".bazelrc*"
+    branches:
+      - main
+      - "rc/*"
+
+permissions:
+  contents: read
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+
+    steps:
+      - uses: actions/checkout@v4
+      - run: |
+          bazel test //misc/bazel/internal/zipmerge:test --test_output=all

.gitignore

@@ -62,3 +62,6 @@ node_modules/
 
 # Temporary folders for working with generated models
 .model-temp
+
+# bazel-built in-tree extractor packs
+/*/extractor-pack

.lfsconfig

@@ -2,4 +2,6 @@
 # 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.
+# we go for `fetchinclude` to something not exsiting rather than `fetchexclude = *` because the
+# former is easier to override (with `git -c` or a local git config) to fetch something specific
 fetchinclude = /nothing

BUILD.bazel

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

MODULE.bazel

@@ -13,19 +13,54 @@ 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 = "0.0.10")
+bazel_dep(name = "rules_go", version = "0.48.0")
 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")
-bazel_dep(name = "bazel_skylib", version = "1.5.0")
+bazel_dep(name = "rules_nodejs", version = "6.2.0-codeql.1")
+bazel_dep(name = "rules_python", version = "0.32.2")
+bazel_dep(name = "bazel_skylib", version = "1.6.1")
 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_kotlin", version = "1.9.4-codeql.1")
+bazel_dep(name = "gazelle", version = "0.37.0")
+bazel_dep(name = "rules_dotnet", version = "0.15.1")
+bazel_dep(name = "googletest", version = "1.14.0.bcr.1")
+bazel_dep(name = "rules_rust", version = "0.46.0")
 
 bazel_dep(name = "buildifier_prebuilt", version = "6.4.0", dev_dependency = True)
 
+crate = use_extension(
+    "@rules_rust//crate_universe:extension.bzl",
+    "crate",
+)
+crate.from_cargo(
+    name = "py_deps",
+    cargo_lockfile = "//python/extractor/tsg-python:Cargo.lock",
+    manifests = [
+        "//python/extractor/tsg-python:Cargo.toml",
+        "//python/extractor/tsg-python/tsp:Cargo.toml",
+    ],
+)
+crate.from_cargo(
+    name = "ruby_deps",
+    cargo_lockfile = "//ruby/extractor:Cargo.lock",
+    manifests = [
+        "//ruby/extractor:Cargo.toml",
+        "//ruby/extractor/codeql-extractor-fake-crate:Cargo.toml",
+    ],
+)
+use_repo(crate, "py_deps", "ruby_deps")
+
+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(
     hub_name = "codegen_deps",
@@ -50,13 +85,92 @@ use_repo(
 node = use_extension("@rules_nodejs//nodejs:extensions.bzl", "node")
 node.toolchain(
     name = "nodejs",
+    node_urls = [
+        "https://nodejs.org/dist/v{version}/{filename}",
+        "https://mirrors.dotsrc.org/nodejs/release/v{version}/{filename}",
+    ],
     node_version = "18.15.0",
 )
 use_repo(node, "nodejs", "nodejs_toolchains")
 
+kotlin_extractor_deps = use_extension("//java/kotlin-extractor:deps.bzl", "kotlin_extractor_deps")
+
+# following list can be kept in sync by running `bazel mod tidy` in `codeql`
+use_repo(
+    kotlin_extractor_deps,
+    "codeql_kotlin_defaults",
+    "codeql_kotlin_embeddable",
+    "kotlin-compiler-1.5.0",
+    "kotlin-compiler-1.5.10",
+    "kotlin-compiler-1.5.20",
+    "kotlin-compiler-1.5.30",
+    "kotlin-compiler-1.6.0",
+    "kotlin-compiler-1.6.20",
+    "kotlin-compiler-1.7.0",
+    "kotlin-compiler-1.7.20",
+    "kotlin-compiler-1.8.0",
+    "kotlin-compiler-1.9.0-Beta",
+    "kotlin-compiler-1.9.20-Beta",
+    "kotlin-compiler-2.0.0-RC1",
+    "kotlin-compiler-embeddable-1.5.0",
+    "kotlin-compiler-embeddable-1.5.10",
+    "kotlin-compiler-embeddable-1.5.20",
+    "kotlin-compiler-embeddable-1.5.30",
+    "kotlin-compiler-embeddable-1.6.0",
+    "kotlin-compiler-embeddable-1.6.20",
+    "kotlin-compiler-embeddable-1.7.0",
+    "kotlin-compiler-embeddable-1.7.20",
+    "kotlin-compiler-embeddable-1.8.0",
+    "kotlin-compiler-embeddable-1.9.0-Beta",
+    "kotlin-compiler-embeddable-1.9.20-Beta",
+    "kotlin-compiler-embeddable-2.0.0-RC1",
+    "kotlin-stdlib-1.5.0",
+    "kotlin-stdlib-1.5.10",
+    "kotlin-stdlib-1.5.20",
+    "kotlin-stdlib-1.5.30",
+    "kotlin-stdlib-1.6.0",
+    "kotlin-stdlib-1.6.20",
+    "kotlin-stdlib-1.7.0",
+    "kotlin-stdlib-1.7.20",
+    "kotlin-stdlib-1.8.0",
+    "kotlin-stdlib-1.9.0-Beta",
+    "kotlin-stdlib-1.9.20-Beta",
+    "kotlin-stdlib-2.0.0-RC1",
+)
+
 go_sdk = use_extension("@rules_go//go:extensions.bzl", "go_sdk")
 go_sdk.download(version = "1.22.2")
 
+lfs_files = use_repo_rule("//misc/bazel:lfs.bzl", "lfs_files")
+
+lfs_files(
+    name = "ripunzip-linux",
+    srcs = ["//misc/ripunzip:ripunzip-linux"],
+    executable = True,
+)
+
+lfs_files(
+    name = "ripunzip-windows",
+    srcs = ["//misc/ripunzip:ripunzip-windows.exe"],
+    executable = True,
+)
+
+lfs_files(
+    name = "ripunzip-macos",
+    srcs = ["//misc/ripunzip:ripunzip-macos"],
+    executable = True,
+)
+
+lfs_files(
+    name = "swift-resource-dir-linux",
+    srcs = ["//swift/third_party/resource-dir:resource-dir-linux.zip"],
+)
+
+lfs_files(
+    name = "swift-resource-dir-macos",
+    srcs = ["//swift/third_party/resource-dir:resource-dir-macos.zip"],
+)
+
 register_toolchains(
     "@nodejs_toolchains//:all",
 )

codeql-workspace.yml

@@ -6,19 +6,16 @@ provide:
   - "*/ql/consistency-queries/qlpack.yml"
   - "*/ql/automodel/src/qlpack.yml"
   - "*/ql/automodel/test/qlpack.yml"
+  - "*/extractor-pack/codeql-extractor.yml"
   - "python/extractor/qlpack.yml"
   - "shared/**/qlpack.yml"
   - "cpp/ql/test/query-tests/Security/CWE/CWE-190/semmle/tainted/qlpack.yml"
   - "go/ql/config/legacy-support/qlpack.yml"
-  - "go/build/codeql-extractor-go/codeql-extractor.yml"
   - "csharp/ql/campaigns/Solorigate/lib/qlpack.yml"
   - "csharp/ql/campaigns/Solorigate/src/qlpack.yml"
   - "csharp/ql/campaigns/Solorigate/test/qlpack.yml"
   - "misc/legacy-support/*/qlpack.yml"
   - "misc/suite-helpers/qlpack.yml"
-  - "ruby/extractor-pack/codeql-extractor.yml"
-  - "swift/extractor-pack/codeql-extractor.yml"
-  - "ql/extractor-pack/codeql-extractor.yml"
   - ".github/codeql/extensions/**/codeql-pack.yml"
 
 versionPolicies:

config/dbscheme-fragments.json

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

config/identical-files.json

@@ -61,10 +61,6 @@
     "java/ql/src/utils/modelgenerator/internal/CaptureModels.qll",
     "csharp/ql/src/utils/modelgenerator/internal/CaptureModels.qll"
   ],
-  "Model as Data Generation Java/C# - CaptureModelsPrinting": [
-    "java/ql/src/utils/modelgenerator/internal/CaptureModelsPrinting.qll",
-    "csharp/ql/src/utils/modelgenerator/internal/CaptureModelsPrinting.qll"
-  ],
   "Sign Java/C#": [
     "java/ql/lib/semmle/code/java/dataflow/internal/rangeanalysis/Sign.qll",
     "csharp/ql/lib/semmle/code/csharp/dataflow/internal/rangeanalysis/Sign.qll"
@@ -185,11 +181,6 @@
     "cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/AliasAnalysisImports.qll",
     "cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/AliasAnalysisImports.qll"
   ],
-  "C++ IR ValueNumberingImports": [
-    "cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/gvn/internal/ValueNumberingImports.qll",
-    "cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/gvn/internal/ValueNumberingImports.qll",
-    "cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/gvn/internal/ValueNumberingImports.qll"
-  ],
   "IR SSA SSAConstruction": [
     "cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/SSAConstruction.qll",
     "cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/SSAConstruction.qll"
@@ -364,5 +355,9 @@
   "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"
   ]
 }

cpp/downgrades/BUILD.bazel

@@ -6,7 +6,7 @@ pkg_files(
         ["**"],
         exclude = ["BUILD.bazel"],
     ),
-    prefix = "cpp/downgrades",
+    prefix = "downgrades",
     strip_prefix = strip_prefix.from_pkg(),
     visibility = ["//cpp:__pkg__"],
 )

cpp/ql/lib/BUILD.bazel

@@ -5,11 +5,9 @@ package(default_visibility = ["//cpp:__pkg__"])
 pkg_files(
     name = "dbscheme",
     srcs = ["semmlecode.cpp.dbscheme"],
-    prefix = "cpp",
 )
 
 pkg_files(
     name = "dbscheme-stats",
     srcs = ["semmlecode.cpp.dbscheme.stats"],
-    prefix = "cpp",
 )

cpp/ql/lib/CHANGELOG.md

@@ -1,3 +1,23 @@
+## 1.1.1
+
+No user-facing changes.
+
+## 1.1.0
+
+### New Features
+
+* Data models can now be added with data extensions. In this way source, sink and summary models can be added in extension `.model.yml` files, rather than by writing classes in QL code. New models should be added in the `lib/ext` folder.
+
+### Minor Analysis Improvements
+
+* A partial model for the `Boost.Asio` network library has been added. This includes sources, sinks and summaries for certain functions in `Boost.Asio`, such as `read_until` and `write`.
+
+## 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.

cpp/ql/lib/change-notes/2024-06-10-builtin-expect.md

@@ -0,0 +1,4 @@
+---
+category: minorAnalysis
+---
+* The "Guards" library (`semmle.code.cpp.controlflow.Guards`) now also infers guards from calls to the builtin operation `__builtin_expect`. As a result, some queries may produce fewer false positives.

cpp/ql/lib/change-notes/2024-06-13-double-free.md

@@ -0,0 +1,4 @@
+---
+category: minorAnalysis
+---
+* The queries "Potential double free" (`cpp/double-free`) and "Potential use after free" (`cpp/use-after-free`) now produce fewer false positives.

cpp/ql/lib/change-notes/2024-06-20-extensible-allocation-deallocation.md

@@ -0,0 +1,4 @@
+---
+category: feature
+---
+* It is now possible to extend the classes `AllocationFunction` and `DeallocationFunction` via data extensions. Extensions of these classes should be added to the `lib/ext/allocation` and `lib/ext/deallocation` directories respectively.

cpp/ql/lib/change-notes/2024-07-03-extended-mad-syntax.md

@@ -0,0 +1,4 @@
+---
+category: feature
+---
+* The syntax for models-as-data rows has been extended to make it easier to select sources, sinks, and summaries that involve templated functions and classes. Additionally, the syntax has also been extended to make it easier to specify models with arbitrary levels of indirection. See `dataflow/ExternalFlow.qll` for the updated documentation and specification for the model format.

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

@@ -1,4 +1,5 @@
----
-category: breakingChange
----
+## 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/change-notes/released/1.1.0.md

@@ -0,0 +1,9 @@
+## 1.1.0
+
+### New Features
+
+* Data models can now be added with data extensions. In this way source, sink and summary models can be added in extension `.model.yml` files, rather than by writing classes in QL code. New models should be added in the `lib/ext` folder.
+
+### Minor Analysis Improvements
+
+* A partial model for the `Boost.Asio` network library has been added. This includes sources, sinks and summaries for certain functions in `Boost.Asio`, such as `read_until` and `write`.

cpp/ql/lib/change-notes/released/1.1.1.md

@@ -0,0 +1,3 @@
+## 1.1.1
+
+No user-facing changes.

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

@@ -1,2 +1,2 @@
 ---
-lastReleaseVersion: 0.13.1
+lastReleaseVersion: 1.1.1

cpp/ql/lib/ext/Boost.Asio.model.yml

@@ -0,0 +1,25 @@
+  # partial model of the Boost::Asio network library
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: sourceModel
+    data: # namespace, type, subtypes, name, signature, ext, output, kind, provenance
+      - ["boost::asio", "", False, "read", "", "", "Argument[*1]", "remote", "manual"]
+      - ["boost::asio", "", False, "read_at", "", "", "Argument[*2]", "remote", "manual"]
+      - ["boost::asio", "", False, "read_until", "", "", "Argument[*1]", "remote", "manual"]
+      - ["boost::asio", "", False, "async_read", "", "", "Argument[*1]", "remote", "manual"]
+      - ["boost::asio", "", False, "async_read_at", "", "", "Argument[*2]", "remote", "manual"]
+      - ["boost::asio", "", False, "async_read_until", "", "", "Argument[*1]", "remote", "manual"]
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: sinkModel
+    data: # namespace, type, subtypes, name, signature, ext, input, kind, provenance
+      - ["boost::asio", "", False, "write", "", "", "Argument[*1]", "remote-sink", "manual"]
+      - ["boost::asio", "", False, "write_at", "", "", "Argument[*2]", "remote-sink", "manual"]
+      - ["boost::asio", "", False, "async_write", "", "", "Argument[*1]", "remote-sink", "manual"]
+      - ["boost::asio", "", False, "async_write_at", "", "", "Argument[*2]", "remote-sink", "manual"]
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: summaryModel
+    data: # namespace, type, subtypes, name, signature, ext, input, output, kind, provenance
+      - ["boost::asio", "", False, "buffer", "", "", "Argument[*0]", "ReturnValue", "taint", "manual"]

cpp/ql/lib/ext/allocation/Bsd.allocation.model.yml

@@ -0,0 +1,7 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: allocationFunctionModel
+    data:
+      - ["", "", False, "kmem_alloc", "0", "", "", True]
+      - ["", "", False, "kmem_zalloc", "0", "", "", True]

cpp/ql/lib/ext/allocation/Glibc.allocation.model.yml

@@ -0,0 +1,7 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: allocationFunctionModel
+    data:
+      - ["", "", False, "g_malloc", "0", "", "", True]
+      - ["", "", False, "g_try_malloc", "0", "", "", True]

cpp/ql/lib/ext/allocation/OpenSSL.allocation.model.yml

@@ -0,0 +1,10 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: allocationFunctionModel
+    data:
+      - ["", "", False, "CRYPTO_malloc", "0", "", "", True]
+      - ["", "", False, "CRYPTO_zalloc", "0", "", "", True]
+      - ["", "", False, "CRYPTO_secure_malloc", "0", "", "", True]
+      - ["", "", False, "CRYPTO_secure_zalloc", "0", "", "", True]
+

cpp/ql/lib/ext/allocation/Std.allocation.model.yml

@@ -0,0 +1,15 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: allocationFunctionModel
+    data:
+      - ["", "", False, "malloc", "0", "", "", True]
+      - ["std", "", False, "malloc", "0", "", "", True]
+      - ["bsl", "", False, "malloc", "0", "", "", True]
+      - ["", "", False, "alloca", "0", "", "", False]
+      - ["", "", False, "__builtin_alloca", "0", "", "", False]
+      - ["", "", False, "_alloca", "0", "", "", False]
+      - ["", "", False, "_malloca", "0", "", "", False]
+      - ["", "", False, "calloc", "1", "0", "", True]
+      - ["std", "", False, "calloc", "1", "0", "", True]
+      - ["bsl", "", False, "calloc", "1", "0", "", True]

cpp/ql/lib/ext/allocation/Windows.allocation.model.yml

@@ -0,0 +1,29 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: allocationFunctionModel
+    data:
+      - ["", "", False, "MmAllocateContiguousMemory", "0", "", "", True]
+      - ["", "", False, "MmAllocateContiguousNodeMemory", "0", "", "", True]
+      - ["", "", False, "MmAllocateContiguousMemorySpecifyCache", "0", "", "", True]
+      - ["", "", False, "MmAllocateContiguousMemorySpecifyCacheNode", "0", "", "", True]
+      - ["", "", False, "MmAllocateNonCachedMemory", "0", "", "", True]
+      - ["", "", False, "MmAllocateMappingAddress", "0", "", "", True]
+      - ["", "", False, "CoTaskMemAlloc", "0", "", "", True]
+      - ["", "", False, "ExAllocatePool", "1", "", "", True]
+      - ["", "", False, "ExAllocatePool2", "1", "", "", True]
+      - ["", "", False, "ExAllocatePool3", "1", "", "", True]
+      - ["", "", False, "ExAllocatePoolWithTag", "1", "", "", True]
+      - ["", "", False, "ExAllocatePoolWithTagPriority", "1", "", "", True]
+      - ["", "", False, "ExAllocatePoolWithQuota", "1", "", "", True]
+      - ["", "", False, "ExAllocatePoolWithQuotaTag", "1", "", "", True]
+      - ["", "", False, "ExAllocatePoolZero", "1", "", "", True]
+      - ["", "", False, "IoAllocateMdl", "1", "", "", True]
+      - ["", "", False, "IoAllocateErrorLogEntry", "1", "", "", True]
+      - ["", "", False, "LocalAlloc", "1", "", "", True]
+      - ["", "", False, "GlobalAlloc", "1", "", "", True]
+      - ["", "", False, "VirtualAlloc", "1", "", "", True]
+      - ["", "", False, "HeapAlloc", "2", "", "", True]
+      - ["", "", False, "MmAllocatePagesForMdl", "3", "", "", True]
+      - ["", "", False, "MmAllocatePagesForMdlEx", "3", "", "", True]
+      - ["", "", False, "MmAllocateNodePagesForMdlEx", "3", "", "", True]

cpp/ql/lib/ext/allocation/empty.allocation.model.yml

@@ -0,0 +1,5 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: allocationFunctionModel
+    data: []

cpp/ql/lib/ext/bsl.array.model.yml

@@ -0,0 +1,14 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: summaryModel
+    data: # namespace, type, subtypes, name, signature, ext, input, output, kind, provenance
+      - ["bsl", "array", True, "at", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "array", True, "begin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "array", True, "cbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "array", True, "data", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "array", True, "operator[]", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "array", True, "rbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "array", True, "rcbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "array", True, "front", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "array", True, "back", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]

cpp/ql/lib/ext/bsl.deque.model.yml

@@ -0,0 +1,73 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: summaryModel
+    data: # namespace, type, subtypes, name, signature, ext, input, output, kind, provenance
+      - ["bsl", "deque<T,Allocator>", True, "assign", "(size_type,const T &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "assign<InputIt>", "(InputIt,InputIt)", "", "Argument[0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "at", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "back", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "begin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "cbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "deque", "(const deque &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "deque", "(deque &&)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace", "", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace", "", "", "Argument[*@2]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace", "", "", "Argument[*@3]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace", "", "", "Argument[*@4]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace", "", "", "Argument[*@5]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_back", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_back", "", "", "Argument[*@0]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_back", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_back", "", "", "Argument[*@1]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_back", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_back", "", "", "Argument[*@2]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_back", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_back", "", "", "Argument[*@3]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_back", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_back", "", "", "Argument[*@4]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_back", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_back", "", "", "Argument[*@5]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_back", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_front", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_front", "", "", "Argument[*@0]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_front", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_front", "", "", "Argument[*@1]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_front", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_front", "", "", "Argument[*@2]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_front", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_front", "", "", "Argument[*@3]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_front", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_front", "", "", "Argument[*@4]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_front", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_front", "", "", "Argument[*@5]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "emplace_front", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "front", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[1].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "operator=", "", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "operator[]", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "deque", True, "push_back", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "push_front", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "rbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque", True, "rcbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque<T,Allocator>", True, "deque", "(const deque &,const Allocator &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque<T,Allocator>", True, "deque", "(deque &&,const Allocator &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque<T,Allocator>", True, "deque", "(size_type,const T &,const Allocator &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque<T,Allocator>", True, "deque<InputIterator>", "(InputIterator,InputIterator,const Allocator &)", "", "Argument[0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[*@2]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "deque<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "deque<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]

cpp/ql/lib/ext/bsl.forward_list.model.yml

@@ -0,0 +1,56 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: summaryModel
+    data: # namespace, type, subtypes, name, signature, ext, input, output, kind, provenance
+      - ["bsl", "forward_list", True, "insert_after<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list<T>", True, "insert_after", "(const_iterator,const T &)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list<T>", True, "insert_after", "(const_iterator,size_type,const T &)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list<T,Allocator>", True, "assign", "(size_type,const T &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "assign<InputIt>", "(InputIt,InputIt)", "", "Argument[0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "begin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "cbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_after", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_after", "", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_after", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_after", "", "", "Argument[*@2]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_after", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_after", "", "", "Argument[*@3]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_after", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_after", "", "", "Argument[*@4]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_after", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_after", "", "", "Argument[*@5]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_after", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_front", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_front", "", "", "Argument[*@0]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_front", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_front", "", "", "Argument[*@1]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_front", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_front", "", "", "Argument[*@2]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_front", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_front", "", "", "Argument[*@3]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_front", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_front", "", "", "Argument[*@4]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_front", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_front", "", "", "Argument[*@5]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "emplace_front", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "forward_list", "(const forward_list &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "forward_list", "(forward_list &&)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "front", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "insert_after<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[1].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "insert_after<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "operator=", "", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "push_front", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "rbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list", True, "rcbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list<T,Allocator>", True, "forward_list", "(const forward_list &,const Allocator &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list<T,Allocator>", True, "forward_list", "(forward_list &&,const Allocator &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list<T,Allocator>", True, "forward_list", "(InputIterator,InputIterator,const Allocator &)", "", "Argument[0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list<T,Allocator>", True, "forward_list", "(size_type,const T &,const Allocator &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list<T>", True, "insert_after", "(const_iterator,const T &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list<T>", True, "insert_after", "(const_iterator,const T &)", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list<T>", True, "insert_after", "(const_iterator,size_type,const T &)", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list<T>", True, "insert_after", "(const_iterator,size_type,const T &)", "", "Argument[*@2]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list<T>", True, "insert_after", "(const_iterator,T &&)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "forward_list<T>", True, "insert_after", "(const_iterator,T &&)", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "forward_list<T>", True, "insert_after", "(const_iterator,T &&)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]

cpp/ql/lib/ext/bsl.list.model.yml

@@ -0,0 +1,71 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: summaryModel
+    data: # namespace, type, subtypes, name, signature, ext, input, output, kind, provenance
+      - ["bsl", "list<T,Allocator>", True, "assign", "(size_type,const T &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "assign<InputIt>", "(InputIt,InputIt)", "", "Argument[0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "back", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "begin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "cbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace", "", "", "Argument[*@1]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace", "", "", "Argument[*@2]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace", "", "", "Argument[*@3]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace", "", "", "Argument[*@4]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace", "", "", "Argument[*@5]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_back", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_back", "", "", "Argument[*@0]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_back", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_back", "", "", "Argument[*@1]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_back", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_back", "", "", "Argument[*@2]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_back", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_back", "", "", "Argument[*@3]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_back", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_back", "", "", "Argument[*@4]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_back", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_back", "", "", "Argument[*@5]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_back", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_front", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_front", "", "", "Argument[*@0]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_front", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_front", "", "", "Argument[*@1]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_front", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_front", "", "", "Argument[*@2]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_front", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_front", "", "", "Argument[*@3]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_front", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_front", "", "", "Argument[*@4]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_front", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_front", "", "", "Argument[*@5]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "emplace_front", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "front", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "list", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[1].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "list", "(const list &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "list", "(list &&)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "operator=", "", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "push_back", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "push_front", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "rbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "list", True, "rcbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "list<T,Allocator>", True, "list", "(const list &,const Allocator &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list<T,Allocator>", True, "list", "(list &&,const Allocator &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list<T,Allocator>", True, "list", "(size_type,const T &,const Allocator &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list<T,Allocator>", True, "list<InputIterator>", "(InputIterator,InputIterator,const Allocator &)", "", "Argument[0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "list<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "list<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[*@2]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "list<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "list<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "list<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "list<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]

cpp/ql/lib/ext/bsl.vector.model.yml

@@ -0,0 +1,60 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: summaryModel
+    data: # namespace, type, subtypes, name, signature, ext, input, output, kind, provenance
+      - ["bsl", "vector<T,Allocator>", True, "assign", "(size_type,const T &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "assign<InputIt>", "(InputIt,InputIt)", "", "Argument[0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "at", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "vector", True, "back", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "vector", True, "begin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "cbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "data", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace", "", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace", "", "", "Argument[*@2]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace", "", "", "Argument[*@3]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace", "", "", "Argument[*@4]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace", "", "", "Argument[*@5]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace_back", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace_back", "", "", "Argument[*@0]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace_back", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace_back", "", "", "Argument[*@1]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace_back", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace_back", "", "", "Argument[*@2]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace_back", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace_back", "", "", "Argument[*@3]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace_back", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace_back", "", "", "Argument[*@4]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace_back", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace_back", "", "", "Argument[*@5]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "vector", True, "emplace_back", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "vector", True, "front", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "vector", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[1].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "operator=", "", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "operator[]", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["bsl", "vector", True, "push_back", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "rbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "rcbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "vector", "(const vector &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector", True, "vector", "(vector &&)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector<T,Allocator>", True, "vector", "(const vector &,const Allocator &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector<T,Allocator>", True, "vector", "(size_type,const T &,const Allocator &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector<T,Allocator>", True, "vector", "(vector &&,const Allocator &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector<T,Allocator>", True, "vector<InputIterator>", "(InputIterator,InputIterator,const Allocator &)", "", "Argument[0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[*@2]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["bsl", "vector<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["bsl", "vector<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]

cpp/ql/lib/ext/deallocation/Bsd.deallocation.model.yml

@@ -0,0 +1,8 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: deallocationFunctionModel
+    data:
+      - ["", "", False, "pool_put", "1"]
+      - ["", "", False, "pool_cache_put", "1"]
+      - ["", "", False, "kmem_free", "0"]

cpp/ql/lib/ext/deallocation/Std.deallocation.model.yml

@@ -0,0 +1,42 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: deallocationFunctionModel
+    data:
+      - ["", "", False, "free", "0"]
+      - ["std", "", False, "free", "0"]
+      - ["bsl", "", False, "free", "0"]
+      - ["", "", False, "realloc", "0"]
+      - ["std", "", False, "realloc", "0"]
+      - ["bsl", "", False, "realloc", "0"]
+      - ["", "", False, "CRYPTO_free", "0"]
+      - ["", "", False, "CRYPTO_secure_free", "0"]
+      - ["", "", False, "g_free", "0"]
+      - ["", "", False, "ExFreePool", "0"]
+      - ["", "", False, "ExFreePoolWithTag", "0"]
+      - ["", "", False, "ExDeleteTimer", "0"]
+      - ["", "", False, "IoFreeIrp", "0"]
+      - ["", "", False, "IoFreeMdl", "0"]
+      - ["", "", False, "IoFreeErrorLogEntry", "0"]
+      - ["", "", False, "IoFreeWorkItem", "0"]
+      - ["", "", False, "MmFreeContiguousMemory", "0"]
+      - ["", "", False, "MmFreeContiguousMemorySpecifyCache", "0"]
+      - ["", "", False, "MmFreeNonCachedMemory", "0"]
+      - ["", "", False, "MmFreeMappingAddress", "0"]
+      - ["", "", False, "MmFreePagesFromMdl", "0"]
+      - ["", "", False, "MmUnmapReservedMapping", "0"]
+      - ["", "", False, "MmUnmapLockedPages", "0"]
+      - ["", "", False, "NdisFreeGenericObject", "0"]
+      - ["", "", False, "NdisFreeMemory", "0"]
+      - ["", "", False, "NdisFreeMemoryWithTag", "0"]
+      - ["", "", False, "NdisFreeMdl", "0"]
+      - ["", "", False, "NdisFreeNetBufferListPool", "0"]
+      - ["", "", False, "NdisFreeNetBufferPool", "0"]
+      - ["", "", False, "LocalFree", "0"]
+      - ["", "", False, "GlobalFree", "0"]
+      - ["", "", False, "LocalReAlloc", "0"]
+      - ["", "", False, "GlobalReAlloc", "0"]
+      - ["", "", False, "VirtualFree", "0"]
+      - ["", "", False, "CoTaskMemFree", "0"]
+      - ["", "", False, "CoTaskMemRealloc", "0"]
+      - ["", "", False, "SysFreeString", "0"]

cpp/ql/lib/ext/deallocation/Windows.deallocation.model.yml

@@ -0,0 +1,41 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: deallocationFunctionModel
+    data:
+      - ["", "", False, "ExFreePool", "0"]
+      - ["", "", False, "ExFreePoolWithTag", "0"]
+      - ["", "", False, "ExDeleteTimer", "0"]
+      - ["", "", False, "IoFreeIrp", "0"]
+      - ["", "", False, "IoFreeMdl", "0"]
+      - ["", "", False, "IoFreeErrorLogEntry", "0"]
+      - ["", "", False, "IoFreeWorkItem", "0"]
+      - ["", "", False, "MmFreeContiguousMemory", "0"]
+      - ["", "", False, "MmFreeContiguousMemorySpecifyCache", "0"]
+      - ["", "", False, "MmFreeNonCachedMemory", "0"]
+      - ["", "", False, "MmFreeMappingAddress", "0"]
+      - ["", "", False, "MmFreePagesFromMdl", "0"]
+      - ["", "", False, "MmUnmapReservedMapping", "0"]
+      - ["", "", False, "MmUnmapLockedPages", "0"]
+      - ["", "", False, "NdisFreeGenericObject", "0"]
+      - ["", "", False, "NdisFreeMemory", "0"]
+      - ["", "", False, "NdisFreeMemoryWithTag", "0"]
+      - ["", "", False, "NdisFreeMdl", "0"]
+      - ["", "", False, "NdisFreeNetBufferListPool", "0"]
+      - ["", "", False, "NdisFreeNetBufferPool", "0"]
+      - ["", "", False, "LocalFree", "0"]
+      - ["", "", False, "GlobalFree", "0"]
+      - ["", "", False, "LocalReAlloc", "0"]
+      - ["", "", False, "GlobalReAlloc", "0"]
+      - ["", "", False, "VirtualFree", "0"]
+      - ["", "", False, "CoTaskMemFree", "0"]
+      - ["", "", False, "CoTaskMemRealloc", "0"]
+      - ["", "", False, "SysFreeString", "0"]
+      - ["", "", False, "ExFreeToLookasideListEx", "1"]
+      - ["", "", False, "ExFreeToPagedLookasideList", "1"]
+      - ["", "", False, "ExFreeToNPagedLookasideList", "1"]
+      - ["", "", False, "NdisFreeMemoryWithTagPriority", "1"]
+      - ["", "", False, "StorPortFreeMdl", "1"]
+      - ["", "", False, "StorPortFreePool", "1"]
+      - ["", "", False, "HeapFree", "2"]
+      - ["", "", False, "HeapReAlloc", "2"]

cpp/ql/lib/ext/deallocation/empty.deallocation.model.yml

@@ -0,0 +1,5 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: deallocationFunctionModel
+    data: []

cpp/ql/lib/ext/empty.model.yml

@@ -0,0 +1,15 @@
+extensions:
+  # Make sure that the extensible model predicates have at least one definition
+  # to avoid errors about undefined extensionals.
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: sourceModel
+    data: []
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: sinkModel
+    data: []
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: summaryModel
+    data: []

cpp/ql/lib/ext/std.array.model.yml

@@ -0,0 +1,14 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: summaryModel
+    data: # namespace, type, subtypes, name, signature, ext, input, output, kind, provenance
+      - ["std", "array", True, "at", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "array", True, "begin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "array", True, "cbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "array", True, "data", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "array", True, "operator[]", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "array", True, "rbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "array", True, "rcbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "array", True, "front", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "array", True, "back", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]

cpp/ql/lib/ext/std.deque.model.yml

@@ -0,0 +1,73 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: summaryModel
+    data: # namespace, type, subtypes, name, signature, ext, input, output, kind, provenance
+      - ["std", "deque<T,Allocator>", True, "assign", "(size_type,const T &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "assign<InputIt>", "(InputIt,InputIt)", "", "Argument[0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "at", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "back", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "begin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque", True, "cbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque", True, "deque", "(const deque &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "deque", "(deque &&)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace", "", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace", "", "", "Argument[*@2]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace", "", "", "Argument[*@3]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace", "", "", "Argument[*@4]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace", "", "", "Argument[*@5]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_back", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_back", "", "", "Argument[*@0]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_back", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_back", "", "", "Argument[*@1]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_back", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_back", "", "", "Argument[*@2]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_back", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_back", "", "", "Argument[*@3]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_back", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_back", "", "", "Argument[*@4]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_back", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_back", "", "", "Argument[*@5]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_back", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_front", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_front", "", "", "Argument[*@0]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_front", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_front", "", "", "Argument[*@1]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_front", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_front", "", "", "Argument[*@2]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_front", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_front", "", "", "Argument[*@3]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_front", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_front", "", "", "Argument[*@4]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_front", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_front", "", "", "Argument[*@5]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "emplace_front", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "front", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[1].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque", True, "operator=", "", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "operator[]", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "deque", True, "push_back", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "push_front", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque", True, "rbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque", True, "rcbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque<T,Allocator>", True, "deque", "(const deque &,const Allocator &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque<T,Allocator>", True, "deque", "(deque &&,const Allocator &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque<T,Allocator>", True, "deque", "(size_type,const T &,const Allocator &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque<T,Allocator>", True, "deque<InputIterator>", "(InputIterator,InputIterator,const Allocator &)", "", "Argument[0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[*@2]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "deque<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "deque<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]

cpp/ql/lib/ext/std.forward_list.model.yml

@@ -0,0 +1,56 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: summaryModel
+    data: # namespace, type, subtypes, name, signature, ext, input, output, kind, provenance
+      - ["std", "forward_list", True, "insert_after<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list<T>", True, "insert_after", "(const_iterator,const T &)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list<T>", True, "insert_after", "(const_iterator,size_type,const T &)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list<T,Allocator>", True, "assign", "(size_type,const T &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "assign<InputIt>", "(InputIt,InputIt)", "", "Argument[0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "begin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "cbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_after", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_after", "", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_after", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_after", "", "", "Argument[*@2]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_after", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_after", "", "", "Argument[*@3]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_after", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_after", "", "", "Argument[*@4]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_after", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_after", "", "", "Argument[*@5]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_after", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_front", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_front", "", "", "Argument[*@0]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_front", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_front", "", "", "Argument[*@1]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_front", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_front", "", "", "Argument[*@2]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_front", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_front", "", "", "Argument[*@3]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_front", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_front", "", "", "Argument[*@4]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_front", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_front", "", "", "Argument[*@5]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "forward_list", True, "emplace_front", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "forward_list", True, "forward_list", "(const forward_list &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "forward_list", "(forward_list &&)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "front", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "forward_list", True, "insert_after<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[1].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "insert_after<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "operator=", "", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "push_front", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "rbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list", True, "rcbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list<T,Allocator>", True, "forward_list", "(const forward_list &,const Allocator &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list<T,Allocator>", True, "forward_list", "(forward_list &&,const Allocator &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list<T,Allocator>", True, "forward_list", "(InputIterator,InputIterator,const Allocator &)", "", "Argument[0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list<T,Allocator>", True, "forward_list", "(size_type,const T &,const Allocator &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list<T>", True, "insert_after", "(const_iterator,const T &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list<T>", True, "insert_after", "(const_iterator,const T &)", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list<T>", True, "insert_after", "(const_iterator,size_type,const T &)", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list<T>", True, "insert_after", "(const_iterator,size_type,const T &)", "", "Argument[*@2]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list<T>", True, "insert_after", "(const_iterator,T &&)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "forward_list<T>", True, "insert_after", "(const_iterator,T &&)", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "forward_list<T>", True, "insert_after", "(const_iterator,T &&)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]

cpp/ql/lib/ext/std.iterator.model.yml

@@ -0,0 +1,11 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: summaryModel
+    data: # namespace, type, subtypes, name, signature, ext, input, output, kind, provenance
+      - ["std", "iterator", True, "operator*", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "iterator", True, "operator->", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "iterator", True, "iterator", "", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["__gnu_cxx", "__normal_iterator", True, "operator*", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["__gnu_cxx", "__normal_iterator", True, "operator->", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["__gnu_cxx", "__normal_iterator", True, "__normal_iterator", "", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]

cpp/ql/lib/ext/std.list.model.yml

@@ -0,0 +1,71 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: summaryModel
+    data: # namespace, type, subtypes, name, signature, ext, input, output, kind, provenance
+      - ["std", "list<T,Allocator>", True, "assign", "(size_type,const T &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "assign<InputIt>", "(InputIt,InputIt)", "", "Argument[0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "back", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "list", True, "begin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list", True, "cbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace", "", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace", "", "", "Argument[*@2]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace", "", "", "Argument[*@3]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace", "", "", "Argument[*@4]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace", "", "", "Argument[*@5]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace_back", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace_back", "", "", "Argument[*@0]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "list", True, "emplace_back", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace_back", "", "", "Argument[*@1]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "list", True, "emplace_back", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace_back", "", "", "Argument[*@2]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "list", True, "emplace_back", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace_back", "", "", "Argument[*@3]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "list", True, "emplace_back", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace_back", "", "", "Argument[*@4]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "list", True, "emplace_back", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace_back", "", "", "Argument[*@5]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "list", True, "emplace_back", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "list", True, "emplace_front", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace_front", "", "", "Argument[*@0]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "list", True, "emplace_front", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace_front", "", "", "Argument[*@1]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "list", True, "emplace_front", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace_front", "", "", "Argument[*@2]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "list", True, "emplace_front", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace_front", "", "", "Argument[*@3]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "list", True, "emplace_front", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace_front", "", "", "Argument[*@4]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "list", True, "emplace_front", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "emplace_front", "", "", "Argument[*@5]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "list", True, "emplace_front", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "list", True, "front", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "list", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[1].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list", True, "list", "(const list &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "list", "(list &&)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "operator=", "", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "push_back", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "push_front", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list", True, "rbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list", True, "rcbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list<T,Allocator>", True, "list", "(const list &,const Allocator &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list<T,Allocator>", True, "list", "(list &&,const Allocator &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list<T,Allocator>", True, "list", "(size_type,const T &,const Allocator &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list<T,Allocator>", True, "list<InputIterator>", "(InputIterator,InputIterator,const Allocator &)", "", "Argument[0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[*@2]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "list<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "list<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]

cpp/ql/lib/ext/std.vector.model.yml

@@ -0,0 +1,60 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: summaryModel
+    data: # namespace, type, subtypes, name, signature, ext, input, output, kind, provenance
+      - ["std", "vector<T,Allocator>", True, "assign", "(size_type,const T &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "assign<InputIt>", "(InputIt,InputIt)", "", "Argument[0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "at", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "vector", True, "back", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "vector", True, "begin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector", True, "cbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector", True, "data", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "vector", True, "emplace", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace", "", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace", "", "", "Argument[*@2]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace", "", "", "Argument[*@3]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace", "", "", "Argument[*@4]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace", "", "", "Argument[*@5]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace_back", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace_back", "", "", "Argument[*@0]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "vector", True, "emplace_back", "", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace_back", "", "", "Argument[*@1]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "vector", True, "emplace_back", "", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace_back", "", "", "Argument[*@2]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "vector", True, "emplace_back", "", "", "Argument[*@3]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace_back", "", "", "Argument[*@3]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "vector", True, "emplace_back", "", "", "Argument[*@4]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace_back", "", "", "Argument[*@4]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "vector", True, "emplace_back", "", "", "Argument[*@5]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "emplace_back", "", "", "Argument[*@5]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "vector", True, "emplace_back", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "vector", True, "front", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "vector", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[1].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "insert<InputIt>", "(const_iterator,InputIt,InputIt)", "", "Argument[1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector", True, "operator=", "", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "operator[]", "", "", "Argument[-1].Element[@]", "ReturnValue[*@]", "value", "manual"]
+      - ["std", "vector", True, "push_back", "", "", "Argument[*@0]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "rbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector", True, "rcbegin", "", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector", True, "vector", "(const vector &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector", True, "vector", "(vector &&)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector<T,Allocator>", True, "vector", "(const vector &,const Allocator &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector<T,Allocator>", True, "vector", "(size_type,const T &,const Allocator &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector<T,Allocator>", True, "vector", "(vector &&,const Allocator &)", "", "Argument[*0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector<T,Allocator>", True, "vector<InputIterator>", "(InputIterator,InputIterator,const Allocator &)", "", "Argument[0].Element[@]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector<T>", True, "insert", "(const_iterator,const T &)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[*@2]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[*@2]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector<T>", True, "insert", "(const_iterator,size_type,const T &)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[*@1]", "Argument[-1].Element[@]", "value", "manual"]
+      - ["std", "vector<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[*@1]", "ReturnValue.Element[@]", "value", "manual"]
+      - ["std", "vector<T>", True, "insert", "(const_iterator,T &&)", "", "Argument[-1].Element[@]", "ReturnValue.Element[@]", "value", "manual"]

cpp/ql/lib/qlpack.yml

@@ -1,5 +1,5 @@
 name: codeql/cpp-all
-version: 1.0.0-dev
+version: 1.1.2-dev
 groups: cpp
 dbscheme: semmlecode.cpp.dbscheme
 extractor: cpp
@@ -14,4 +14,8 @@ dependencies:
   codeql/tutorial: ${workspace}
   codeql/util: ${workspace}
   codeql/xml: ${workspace}
+dataExtensions:
+  - ext/*.model.yml
+  - ext/deallocation/*.model.yml
+  - ext/allocation/*.model.yml
 warnOnImplicitThis: true

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

@@ -410,6 +410,10 @@ class LocalVariable extends LocalScopeVariable, @localvariable {
     or
     orphaned_variables(underlyingElement(this), unresolveElement(result))
   }
+
+  override predicate isStatic() {
+    super.isStatic() or orphaned_variables(underlyingElement(this), _)
+  }
 }
 
 /**

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

@@ -59,8 +59,7 @@ class MatchValue extends AbstractValue, TMatchValue {
 }
 
 /**
- * A Boolean condition in the AST that guards one or more basic blocks. This includes
- * operands of logical operators but not switch statements.
+ * A Boolean condition in the AST that guards one or more basic blocks.
  */
 cached
 class GuardCondition extends Expr {
@@ -366,15 +365,42 @@ private predicate nonExcludedIRAndBasicBlock(IRBlock irb, BasicBlock controlled)
 }
 
 /**
- * A Boolean condition in the IR that guards one or more basic blocks. This includes
- * operands of logical operators but not switch statements. Note that `&&` and `||`
- * don't have an explicit representation in the IR, and therefore will not appear as
- * IRGuardConditions.
+ * A Boolean condition in the IR that guards one or more basic blocks.
+ *
+ * Note that `&&` and `||` don't have an explicit representation in the IR,
+ * and therefore will not appear as IRGuardConditions.
  */
 cached
 class IRGuardCondition extends Instruction {
   Instruction branch;
 
+  /*
+   * An `IRGuardCondition` supports reasoning about four different kinds of
+   * relations:
+   * 1. A unary equality relation of the form `e == k`
+   * 2. A binary equality relation of the form `e1 == e2 + k`
+   * 3. A unary inequality relation of the form `e < k`
+   * 4. A binary inequality relation of the form `e1 < e2 + k`
+   *
+   * where `k` is a constant.
+   *
+   * Furthermore, the unary relations (i.e., case 1 and case 3) are also
+   * inferred from `switch` statement guards: equality relations are inferred
+   * from the unique `case` statement, if any, and inequality relations are
+   * inferred from the [case range](https://gcc.gnu.org/onlinedocs/gcc/Case-Ranges.html)
+   * gcc extension.
+   *
+   * The implementation of all four follows the same structure: Each relation
+   * has a cached user-facing predicate that. For example,
+   * `GuardCondition::comparesEq` calls `compares_eq`. This predicate has
+   * several cases that recursively decompose the relation to bring it to a
+   * canonical form (i.e., a relation of the form `e1 == e2 + k`). The base
+   * case for this relation (i.e., `simple_comparison_eq`) handles
+   * `CompareEQInstruction`s and `CompareNEInstruction`, and recursive
+   * predicates (e.g., `complex_eq`) rewrites larger expressions such as
+   * `e1 + k1 == e2 + k2` into canonical the form `e1 == e2 + (k2 - k1)`.
+   */
+
   cached
   IRGuardCondition() { branch = getBranchForCondition(this) }
 
@@ -735,6 +761,8 @@ private predicate compares_eq(
   exists(AbstractValue dual | value = dual.getDualValue() |
     compares_eq(test.(LogicalNotInstruction).getUnary(), left, right, k, areEqual, dual)
   )
+  or
+  compares_eq(test.(BuiltinExpectCallInstruction).getCondition(), left, right, k, areEqual, value)
 }
 
 /**
@@ -776,7 +804,9 @@ private predicate unary_compares_eq(
   Instruction test, Operand op, int k, boolean areEqual, boolean inNonZeroCase, 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 |
+    unary_simple_comparison_eq(test, k, inNonZeroCase, v) and op.getDef() = test
+  |
     areEqual = true and value = v
     or
     areEqual = false and value = v.getDualValue()
@@ -802,6 +832,9 @@ private predicate unary_compares_eq(
     int_value(const) = k1 and
     k = k1 + k2
   )
+  or
+  unary_compares_eq(test.(BuiltinExpectCallInstruction).getCondition(), op, k, areEqual,
+    inNonZeroCase, value)
 }
 
 /** Rearrange various simple comparisons into `left == right + k` form. */
@@ -821,45 +854,55 @@ 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
+  Instruction test, int k, boolean inNonZeroCase, 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
+  // Any instruction with an integral type could potentially be part of a
+  // check for nullness when used in a guard. So we include all integral
+  // typed instructions here. However, since some of these instructions are
+  // already included as guards in other cases, we exclude those here.
+  // These are instructions that compute a binary equality or inequality
+  // relation. For example, the following:
+  // ```cpp
+  // if(a == b + 42) { ... }
+  // ```
+  // generates the following IR:
+  // ```
+  // r1(glval<int>) = VariableAddress[a]     :
+  // r2(int)        = Load[a]                : &:r1, m1
+  // r3(glval<int>) = VariableAddress[b]     :
+  // r4(int)        = Load[b]                : &:r3, m2
+  // r5(int)        = Constant[42]           :
+  // r6(int)        = Add                    : r4, r5
+  // r7(bool)       = CompareEQ              : r2, r6
+  // v1(void)       = ConditionalBranch      : r7
+  // ```
+  // and since `r7` is an integral typed instruction this predicate could
+  // include a case for when `r7` evaluates to true (in which case we would
+  // infer that `r6` was non-zero, and a case for when `r7` evaluates to false
+  // (in which case we would infer that `r6` was zero).
+  // However, since `a == b + 42` is already supported when reasoning about
+  // binary equalities we exclude those cases here.
+  not test.isGLValue() and
+  not simple_comparison_eq(test, _, _, _, _) and
+  not simple_comparison_lt(test, _, _, _) and
+  not test = any(SwitchInstruction switch).getExpression() and
+  (
+    test.getResultIRType() instanceof IRAddressType or
+    test.getResultIRType() instanceof IRIntegerType or
+    test.getResultIRType() instanceof IRBooleanType
+  ) and
   (
     k = 1 and
     value.(BooleanValue).getValue() = true and
@@ -871,12 +914,68 @@ private predicate unary_simple_comparison_eq(
   )
 }
 
+/** A call to the builtin operation `__builtin_expect`. */
+private class BuiltinExpectCallInstruction extends CallInstruction {
+  BuiltinExpectCallInstruction() { this.getStaticCallTarget().hasName("__builtin_expect") }
+
+  /** Gets the condition of this call. */
+  Instruction getCondition() {
+    // The first parameter of `__builtin_expect` has type `long`. So we skip
+    // the conversion when inferring guards.
+    result = this.getArgument(0).(ConvertInstruction).getUnary()
+  }
+}
+
+/**
+ * Holds if `left == right + k` is `areEqual` if `cmp` evaluates to `value`,
+ * and `cmp` is an instruction that compares the value of
+ * `__builtin_expect(left == right + k, _)` to `0`.
+ */
+private predicate builtin_expect_eq(
+  CompareInstruction cmp, Operand left, Operand right, int k, boolean areEqual, AbstractValue value
+) {
+  exists(BuiltinExpectCallInstruction call, Instruction const, AbstractValue innerValue |
+    int_value(const) = 0 and
+    cmp.hasOperands(call.getAUse(), const.getAUse()) and
+    compares_eq(call.getCondition(), left, right, k, areEqual, innerValue)
+  |
+    cmp instanceof CompareNEInstruction and
+    value = innerValue
+    or
+    cmp instanceof CompareEQInstruction and
+    value.getDualValue() = innerValue
+  )
+}
+
 private predicate complex_eq(
   CompareInstruction cmp, Operand left, Operand right, int k, boolean areEqual, AbstractValue value
 ) {
   sub_eq(cmp, left, right, k, areEqual, value)
   or
   add_eq(cmp, left, right, k, areEqual, value)
+  or
+  builtin_expect_eq(cmp, left, right, k, areEqual, value)
+}
+
+/**
+ * Holds if `op == k` is `areEqual` if `cmp` evaluates to `value`, and `cmp` is
+ * an instruction that compares the value of `__builtin_expect(op == k, _)` to `0`.
+ */
+private predicate unary_builtin_expect_eq(
+  CompareInstruction cmp, Operand op, int k, boolean areEqual, boolean inNonZeroCase,
+  AbstractValue value
+) {
+  exists(BuiltinExpectCallInstruction call, Instruction const, AbstractValue innerValue |
+    int_value(const) = 0 and
+    cmp.hasOperands(call.getAUse(), const.getAUse()) and
+    unary_compares_eq(call.getCondition(), op, k, areEqual, inNonZeroCase, innerValue)
+  |
+    cmp instanceof CompareNEInstruction and
+    value = innerValue
+    or
+    cmp instanceof CompareEQInstruction and
+    value.getDualValue() = innerValue
+  )
 }
 
 private predicate unary_complex_eq(
@@ -885,6 +984,8 @@ private predicate unary_complex_eq(
   unary_sub_eq(test, op, k, areEqual, inNonZeroCase, value)
   or
   unary_add_eq(test, op, k, areEqual, inNonZeroCase, value)
+  or
+  unary_builtin_expect_eq(test, op, k, areEqual, inNonZeroCase, value)
 }
 
 /*
@@ -913,7 +1014,8 @@ private predicate compares_lt(
 
 /** Holds if `op < k` evaluates to `isLt` given that `test` evaluates to `value`. */
 private predicate compares_lt(Instruction test, Operand op, int k, boolean isLt, AbstractValue value) {
-  simple_comparison_lt(test, op, k, isLt, value)
+  unary_simple_comparison_lt(test, k, isLt, value) and
+  op.getDef() = test
   or
   complex_lt(test, op, k, isLt, value)
   or
@@ -960,12 +1062,11 @@ private predicate simple_comparison_lt(CompareInstruction cmp, Operand left, Ope
 }
 
 /** Rearrange various simple comparisons into `op < k` form. */
-private predicate simple_comparison_lt(
-  Instruction test, Operand op, int k, boolean isLt, AbstractValue value
+private predicate unary_simple_comparison_lt(
+  Instruction test, int k, boolean isLt, 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.getMaxValue() > case.getMinValue()

cpp/ql/lib/semmle/code/cpp/dataflow/ExternalFlow.qll

@@ -14,16 +14,22 @@
  * The interpretation of a row is similar to API-graphs with a left-to-right
  * reading.
  * 1. The `namespace` column selects a namespace.
- * 2. The `type` column selects a type within that namespace.
+ * 2. The `type` column selects a type within that namespace. This column can
+ *    introduce template names that can be mentioned in the `signature` column.
+ *    For example, `vector<T,Allocator>` introduces the template names `T` and
+ *    `Allocator`.
  * 3. The `subtypes` is a boolean that indicates whether to jump to an
  *    arbitrary subtype of that type. Set this to `false` if leaving the `type`
  *    blank (for example, a free function).
  * 4. The `name` column optionally selects a specific named member of the type.
+ *    Like the `type` column, this column can introduce template names that can
+ *    be mentioned in the `signature` column. For example, `insert<InputIt>`
+ *    introduces the template name `InputIt`.
  * 5. The `signature` column optionally restricts the named member. If
  *    `signature` is blank then no such filtering is done. The format of the
  *    signature is a comma-separated list of types enclosed in parentheses. The
- *    types can be short names or fully qualified names (mixing these two options
- *    is not allowed within a single signature).
+ *    types must be stripped of template names. That is, write `const vector &`
+ *    instead of `const vector<T> &`.
  * 6. The `ext` column specifies additional API-graph-like edges. Currently
  *    there is only one valid value: "".
  * 7. The `input` column specifies how data enters the element selected by the
@@ -44,6 +50,9 @@
  *      One or more "*" can be added as an argument to indicate indirection, for
  *      example, "ReturnValue[*]" indicates the first indirection of the return
  *      value.
+ *    The special symbol `@` can be used to specify an arbitrary (but fixed)
+ *    number of indirections. For example, the `input` column `Argument[*@0]`
+ *    indicates one or more indirections of the 0th argument.
  *
  *    An `output` can be either:
  *    - "": Selects a read of a selected field.
@@ -65,6 +74,17 @@
  *      One or more "*" can be added as an argument to indicate indirection, for
  *      example, "ReturnValue[*]" indicates the first indirection of the return
  *      value.
+ *    The special symbol `@` can be used to specify an arbitrary (but fixed)
+ *    number of indirections. For example, the `output` column
+ *    `ReturnValue[*@0]` indicates one or more indirections of the return
+ *    value.
+ *    Note: The symbol `@` only ever takes a single value across a row. Thus,
+ *    the (`input`, `output`) pair `("Argument[*@0]", "ReturnValue[@]")`
+ *    represents:
+ *    - flow from the _first_ indirection of the 0th argument to the return
+ *    value, and
+ *    - flow from the _second_ indirection of the 0th argument to the first
+ *    indirection of the return value, etc.
  * 8. The `kind` column is a tag that can be referenced from QL to determine to
  *    which classes the interpreted elements should be added. For example, for
  *    sources "remote" indicates a default remote flow source, and for summaries
@@ -74,10 +94,13 @@
 
 import cpp
 private import new.DataFlow
+private import semmle.code.cpp.ir.dataflow.internal.DataFlowPrivate as Private
+private import semmle.code.cpp.ir.dataflow.internal.DataFlowUtil
 private import internal.FlowSummaryImpl
 private import internal.FlowSummaryImpl::Public
 private import internal.FlowSummaryImpl::Private
 private import internal.FlowSummaryImpl::Private::External
+private import internal.ExternalFlowExtensions as Extensions
 private import codeql.mad.ModelValidation as SharedModelVal
 private import codeql.util.Unit
 
@@ -138,6 +161,9 @@ predicate sourceModel(
     row.splitAt(";", 7) = kind
   ) and
   provenance = "manual"
+  or
+  Extensions::sourceModel(namespace, type, subtypes, name, signature, ext, output, kind, provenance,
+    _)
 }
 
 /** Holds if a sink model exists for the given parameters. */
@@ -158,10 +184,16 @@ predicate sinkModel(
     row.splitAt(";", 7) = kind
   ) and
   provenance = "manual"
+  or
+  Extensions::sinkModel(namespace, type, subtypes, name, signature, ext, input, kind, provenance, _)
 }
 
-/** Holds if a summary model exists for the given parameters. */
-predicate summaryModel(
+/**
+ * Holds if a summary model exists for the given parameters.
+ *
+ * This predicate does not expand `@` to `*`s.
+ */
+private predicate summaryModel0(
   string namespace, string type, boolean subtypes, string name, string signature, string ext,
   string input, string output, string kind, string provenance
 ) {
@@ -179,6 +211,36 @@ predicate summaryModel(
     row.splitAt(";", 8) = kind
   ) and
   provenance = "manual"
+  or
+  Extensions::summaryModel(namespace, type, subtypes, name, signature, ext, input, output, kind,
+    provenance, _)
+}
+
+/**
+ * Holds if `input` is `input0`, but with all occurrences of `@` replaced
+ * by `n` repetitions of `*` (and similarly for `output` and `output0`).
+ */
+bindingset[input0, output0, n]
+pragma[inline_late]
+private predicate expandInputAndOutput(
+  string input0, string input, string output0, string output, int n
+) {
+  input = input0.replaceAll("@", repeatStars(n)) and
+  output = output0.replaceAll("@", repeatStars(n))
+}
+
+/**
+ * Holds if a summary model exists for the given parameters.
+ */
+predicate summaryModel(
+  string namespace, string type, boolean subtypes, string name, string signature, string ext,
+  string input, string output, string kind, string provenance
+) {
+  exists(string input0, string output0 |
+    summaryModel0(namespace, type, subtypes, name, signature, ext, input0, output0, kind, provenance) and
+    expandInputAndOutput(input0, input, output0, output,
+      [0 .. Private::getMaxElementContentIndirectionIndex() - 1])
+  )
 }
 
 private predicate relevantNamespace(string namespace) {
@@ -203,8 +265,10 @@ private predicate canonicalNamespaceLink(string namespace, string subns) {
 }
 
 /**
- * Holds if CSV framework coverage of `namespace` is `n` api endpoints of the
- * kind `(kind, part)`.
+ * Holds if MaD framework coverage of `namespace` is `n` api endpoints of the
+ * kind `(kind, part)`, and `namespaces` is the number of subnamespaces of
+ * `namespace` which have MaD framework coverage (including `namespace`
+ * itself).
  */
 predicate modelCoverage(string namespace, int namespaces, string kind, string part, int n) {
   namespaces = strictcount(string subns | canonicalNamespaceLink(namespace, subns)) and
@@ -321,10 +385,10 @@ module CsvValidation {
       or
       summaryModel(namespace, type, _, name, signature, ext, _, _, _, _) and pred = "summary"
     |
-      not namespace.regexpMatch("[a-zA-Z0-9_\\.]+") and
+      not namespace.regexpMatch("[a-zA-Z0-9_\\.:]*") and
       result = "Dubious namespace \"" + namespace + "\" in " + pred + " model."
       or
-      not type.regexpMatch("[a-zA-Z0-9_<>,\\+]+") and
+      not type.regexpMatch("[a-zA-Z0-9_<>,\\+]*") and
       result = "Dubious type \"" + type + "\" in " + pred + " model."
       or
       not name.regexpMatch("[a-zA-Z0-9_<>,]*") and
@@ -356,16 +420,155 @@ private predicate elementSpec(
   summaryModel(namespace, type, subtypes, name, signature, ext, _, _, _, _)
 }
 
-private string paramsStringPart(Function c, int i) {
-  i = -1 and result = "(" and exists(c)
-  or
-  exists(int n, string p | c.getParameter(n).getType().toString() = p |
-    i = 2 * n and result = p
-    or
-    i = 2 * n - 1 and result = "," and n != 0
+/** Gets the fully templated version of `f`. */
+private Function getFullyTemplatedMemberFunction(Function f) {
+  not f.isFromUninstantiatedTemplate(_) and
+  exists(Class c, Class templateClass, int i |
+    c.isConstructedFrom(templateClass) and
+    f = c.getAMember(i) and
+    result = templateClass.getCanonicalMember(i)
+  )
+}
+
+/**
+ * Gets the type name of the `n`'th parameter of `f` without any template
+ * arguments.
+ */
+bindingset[f]
+pragma[inline_late]
+string getParameterTypeWithoutTemplateArguments(Function f, int n) {
+  exists(string s, string base, string specifiers |
+    s = f.getParameter(n).getType().getName() and
+    parseAngles(s, base, _, specifiers) and
+    result = base + specifiers
+  )
+}
+
+/**
+ * Normalize the `n`'th parameter of `f` by replacing template names
+ * with `func:N` (where `N` is the index of the template).
+ */
+private string getTypeNameWithoutFunctionTemplates(Function f, int n, int remaining) {
+  exists(Function templateFunction |
+    templateFunction = getFullyTemplatedMemberFunction(f) and
+    remaining = templateFunction.getNumberOfTemplateArguments() and
+    result = getParameterTypeWithoutTemplateArguments(templateFunction, n)
   )
   or
-  i = 2 * c.getNumberOfParameters() and result = ")"
+  exists(string mid, TemplateParameter tp, Function templateFunction |
+    mid = getTypeNameWithoutFunctionTemplates(f, n, remaining + 1) and
+    templateFunction = getFullyTemplatedMemberFunction(f) and
+    tp = templateFunction.getTemplateArgument(remaining) and
+    result = mid.replaceAll(tp.getName(), "func:" + remaining.toString())
+  )
+}
+
+/**
+ * Normalize the `n`'th parameter of `f` by replacing template names
+ * with `class:N` (where `N` is the index of the template).
+ */
+private string getTypeNameWithoutClassTemplates(Function f, int n, int remaining) {
+  exists(Class template |
+    f.getDeclaringType().isConstructedFrom(template) and
+    remaining = template.getNumberOfTemplateArguments() and
+    result = getTypeNameWithoutFunctionTemplates(f, n, 0)
+  )
+  or
+  exists(string mid, TemplateParameter tp, Class template |
+    mid = getTypeNameWithoutClassTemplates(f, n, remaining + 1) and
+    f.getDeclaringType().isConstructedFrom(template) and
+    tp = template.getTemplateArgument(remaining) and
+    result = mid.replaceAll(tp.getName(), "class:" + remaining.toString())
+  )
+}
+
+/** Gets the string representation of the `i`'th parameter of `c`. */
+private string getParameterTypeName(Function c, int i) {
+  result = getTypeNameWithoutClassTemplates(c, i, 0)
+}
+
+/** Splits `s` by `,` and gets the `i`'th element. */
+bindingset[s]
+pragma[inline_late]
+private string getAtIndex(string s, int i) {
+  result = s.splitAt(",", i) and
+  // when `s` is `""` and `i` is `0` we get `result = ""` which we don't want.
+  not (s = "" and i = 0)
+}
+
+/**
+ * Normalizes `partiallyNormalizedSignature` by replacing the `remaining`
+ * number of template arguments in `partiallyNormalizedSignature` with their
+ * index in `typeArgs`.
+ */
+private string getSignatureWithoutClassTemplateNames(
+  string partiallyNormalizedSignature, string typeArgs, string nameArgs, int remaining
+) {
+  elementSpecWithArguments0(_, _, _, partiallyNormalizedSignature, typeArgs, nameArgs) and
+  remaining = count(partiallyNormalizedSignature.indexOf(",")) + 1 and
+  result = partiallyNormalizedSignature
+  or
+  exists(string mid |
+    mid =
+      getSignatureWithoutClassTemplateNames(partiallyNormalizedSignature, typeArgs, nameArgs,
+        remaining + 1)
+  |
+    exists(string typeArg |
+      typeArg = getAtIndex(typeArgs, remaining) and
+      result = mid.replaceAll(typeArg, "class:" + remaining.toString())
+    )
+    or
+    // Make sure `remaining` is properly bound
+    remaining = [0 .. count(partiallyNormalizedSignature.indexOf(",")) + 1] and
+    not exists(getAtIndex(typeArgs, remaining)) and
+    result = mid
+  )
+}
+
+/**
+ * Normalizes `partiallyNormalizedSignature` by replacing:
+ * - _All_ the template arguments in `partiallyNormalizedSignature` that refer to
+ * template parameters in `typeArgs` with their index in `typeArgs`, and
+ * - The `remaining` number of template arguments in `partiallyNormalizedSignature`
+ * with their index in `nameArgs`.
+ */
+private string getSignatureWithoutFunctionTemplateNames(
+  string partiallyNormalizedSignature, string typeArgs, string nameArgs, int remaining
+) {
+  remaining = count(partiallyNormalizedSignature.indexOf(",")) + 1 and
+  result =
+    getSignatureWithoutClassTemplateNames(partiallyNormalizedSignature, typeArgs, nameArgs, 0)
+  or
+  exists(string mid |
+    mid =
+      getSignatureWithoutFunctionTemplateNames(partiallyNormalizedSignature, typeArgs, nameArgs,
+        remaining + 1)
+  |
+    exists(string nameArg |
+      nameArg = getAtIndex(nameArgs, remaining) and
+      result = mid.replaceAll(nameArg, "func:" + remaining.toString())
+    )
+    or
+    // Make sure `remaining` is properly bound
+    remaining = [0 .. count(partiallyNormalizedSignature.indexOf(",")) + 1] and
+    not exists(getAtIndex(nameArgs, remaining)) and
+    result = mid
+  )
+}
+
+private string paramsStringPart(Function c, int i) {
+  not c.isFromUninstantiatedTemplate(_) and
+  (
+    i = -1 and result = "(" and exists(c)
+    or
+    exists(int n, string p | getParameterTypeName(c, n) = p |
+      i = 2 * n and result = p
+      or
+      i = 2 * n - 1 and result = "," and n != 0
+    )
+    or
+    i = 2 * c.getNumberOfParameters() and result = ")"
+  )
 }
 
 /**
@@ -385,6 +588,193 @@ private predicate matchesSignature(Function func, string signature) {
   paramsString(func) = signature
 }
 
+/**
+ * Holds if `elementSpec(_, type, _, name, signature, _)` holds and
+ * - `typeArgs` represents the named template parameters supplied to `type`, and
+ * - `nameArgs` represents the named template parameters supplied to `name`, and
+ * - `normalizedSignature` is `signature`, except with
+ *    - template parameter names replaced by `func:i` if the template name is
+ *      the `i`'th entry in `nameArgs`, and
+ *    - template parameter names replaced by `class:i` if the template name is
+ *      the `i`'th entry in `typeArgs`.
+ *
+ * In other words, the string `normalizedSignature` represents a "normalized"
+ * signature with no mention of any free template parameters.
+ *
+ * For example, consider a summary row such as:
+ * ```
+ * elementSpec(_, "MyClass<B, C>", _, myFunc<A>, "(const A &,int,C,B *)", _)
+ * ```
+ * In this case, `normalizedSignature` will be `"(const func:0 &,int,class:1,class:0 *)"`.
+ */
+private predicate elementSpecWithArguments(
+  string signature, string type, string name, string normalizedSignature, string typeArgs,
+  string nameArgs
+) {
+  exists(string signatureWithoutParens |
+    elementSpecWithArguments0(signature, type, name, signatureWithoutParens, typeArgs, nameArgs) and
+    normalizedSignature =
+      getSignatureWithoutFunctionTemplateNames(signatureWithoutParens, typeArgs, nameArgs, 0)
+  )
+}
+
+/** Gets the `n`'th normalized signature parameter for the function `name` in class `type`. */
+private string getSignatureParameterName(string signature, string type, string name, int n) {
+  exists(string normalizedSignature |
+    elementSpecWithArguments(signature, type, name, normalizedSignature, _, _) and
+    result = getAtIndex(normalizedSignature, n)
+  )
+}
+
+/**
+ * Holds if the suffix containing the entries in `signature` starting at entry
+ * `i` matches the suffix containing the parameters of `func` starting at entry `i`.
+ *
+ * For example, consider the signature `(int,bool,char)` and a function:
+ * ```
+ * void f(int a, bool b, char c);
+ * ```
+ * 1. The predicate holds for `i = 2` because the suffix containing all the entries
+ * in `signature` starting at `2` is `char`, and suffix containing all the parameters
+ * of `func` starting at `2` is `char`.
+ * 2. The predicate holds for `i = 1` because the suffix containing all the entries
+ * in `signature` starting at `1` is `bool,char`, and the suffix containing all the
+ * parameters of `func` starting at `1` is `bool, char`.
+ * 3. The predicate holds for `i = 0` because the suffix containing all the entries
+ * in `signature` starting at `0` is `int,bool,char` and the suffix containing all
+ * the parameters of `func` starting at `0` is `int, bool, char`.
+ *
+ * When `paramsString(func)[i]` is `class:n` then the signature name is
+ * compared with the `n`'th name in `type`, and when `paramsString(func)[i]`
+ * is `func:n` then the signature name is compared with the `n`'th name
+ * in `name`.
+ */
+private predicate signatureMatches(Function func, string signature, string type, string name, int i) {
+  exists(string s |
+    s = getSignatureParameterName(signature, type, name, i) and
+    s = getParameterTypeName(func, i)
+  ) and
+  if exists(getParameterTypeName(func, i + 1))
+  then signatureMatches(func, signature, type, name, i + 1)
+  else i = count(signature.indexOf(","))
+}
+
+/**
+ * Internal: Do not use.
+ *
+ * This module only exists to expose internal predicates for testing purposes.
+ */
+module ExternalFlowDebug {
+  /**
+   * INTERNAL: Do not use.
+   *
+   * Exposed for testing purposes.
+   */
+  predicate signatureMatches_debug = signatureMatches/5;
+
+  /**
+   * INTERNAL: Do not use.
+   *
+   * Exposed for testing purposes.
+   */
+  predicate getSignatureParameterName_debug = getSignatureParameterName/4;
+
+  /**
+   * INTERNAL: Do not use.
+   *
+   * Exposed for testing purposes.
+   */
+  predicate getParameterTypeName_debug = getParameterTypeName/2;
+}
+
+/**
+ * Holds if `s` can be broken into a string of the form
+ * `beforeAngles<betweenAngles>`,
+ * or `s = beforeAngles` where `beforeAngles` does not have any brackets.
+ */
+bindingset[s]
+pragma[inline_late]
+private predicate parseAngles(
+  string s, string beforeAngles, string betweenAngles, string afterAngles
+) {
+  beforeAngles = s.regexpCapture("([^<]+)(?:<([^>]+)>(.*))?", 1) and
+  (
+    betweenAngles = s.regexpCapture("([^<]+)(?:<([^>]+)>(.*))?", 2) and
+    afterAngles = s.regexpCapture("([^<]+)(?:<([^>]+)>(.*))?", 3)
+    or
+    not exists(s.regexpCapture("([^<]+)(?:<([^>]+)>(.*))?", 2)) and
+    betweenAngles = "" and
+    afterAngles = ""
+  )
+}
+
+/** Holds if `s` can be broken into a string of the form `(betweenParens)`. */
+bindingset[s]
+pragma[inline_late]
+private predicate parseParens(string s, string betweenParens) { s = "(" + betweenParens + ")" }
+
+/**
+ * Holds if `elementSpec(_, type, _, name, signature, _)` and:
+ * - `type` introduces template parameters `typeArgs`, and
+ * - `name` introduces template parameters `nameArgs`, and
+ * - `signatureWithoutParens` equals `signature`, but with the surrounding
+ *    parentheses removed.
+ */
+private predicate elementSpecWithArguments0(
+  string signature, string type, string name, string signatureWithoutParens, string typeArgs,
+  string nameArgs
+) {
+  elementSpec(_, type, _, name, signature, _) and
+  parseAngles(name, _, nameArgs, "") and
+  (
+    type = "" and typeArgs = ""
+    or
+    parseAngles(type, _, typeArgs, "")
+  ) and
+  parseParens(signature, signatureWithoutParens)
+}
+
+/**
+ * Holds if `elementSpec(namespace, type, subtypes, name, signature, _)` and
+ * `method`'s signature matches `signature`.
+ *
+ * `signature` may contain template parameter names that are bound by `type` and `name`.
+ */
+pragma[nomagic]
+private predicate elementSpecMatchesSignature(
+  Function method, string namespace, string type, boolean subtypes, string name, string signature
+) {
+  elementSpec(namespace, pragma[only_bind_into](type), subtypes, pragma[only_bind_into](name),
+    pragma[only_bind_into](signature), _) and
+  signatureMatches(method, signature, type, name, 0)
+}
+
+/**
+ * Holds if `classWithMethod` has `method` named `name` (excluding any
+ * template parameters).
+ */
+bindingset[name]
+pragma[inline_late]
+private predicate hasClassAndName(Class classWithMethod, Function method, string name) {
+  exists(string nameWithoutArgs |
+    parseAngles(name, nameWithoutArgs, _, "") and
+    classWithMethod = method.getClassAndName(nameWithoutArgs)
+  )
+}
+
+/**
+ * Holds if `namedClass` is in namespace `namespace` and has
+ * name `type` (excluding any template parameters).
+ */
+bindingset[type, namespace]
+pragma[inline_late]
+private predicate hasQualifiedName(Class namedClass, string namespace, string type) {
+  exists(string typeWithoutArgs |
+    parseAngles(type, typeWithoutArgs, _, "") and
+    namedClass.hasQualifiedName(namespace, typeWithoutArgs)
+  )
+}
+
 /**
  * Gets the element in module `namespace` that satisfies the following properties:
  * 1. If the element is a member of a class-like type, then the class-like type has name `type`
@@ -399,8 +789,8 @@ pragma[nomagic]
 private Element interpretElement0(
   string namespace, string type, boolean subtypes, string name, string signature
 ) {
-  elementSpec(namespace, type, subtypes, name, signature, _) and
   (
+    elementSpec(namespace, type, subtypes, name, signature, _) and
     // Non-member functions
     exists(Function func |
       func.hasQualifiedName(namespace, name) and
@@ -412,21 +802,28 @@ private Element interpretElement0(
     )
     or
     // Member functions
-    exists(Class namedClass, Class classWithMethod, Function method |
-      classWithMethod = method.getClassAndName(name) and
-      namedClass.hasQualifiedName(namespace, type) and
-      matchesSignature(method, signature) and
-      result = method
-    |
-      // member declared in the named type or a subtype of it
-      subtypes = true and
-      classWithMethod = namedClass.getADerivedClass*()
-      or
-      // member declared directly in the named type
-      subtypes = false and
-      classWithMethod = namedClass
+    exists(Class namedClass, Class classWithMethod |
+      (
+        elementSpecMatchesSignature(result, namespace, type, subtypes, name, signature) and
+        hasClassAndName(classWithMethod, result, name)
+        or
+        signature = "" and
+        elementSpec(namespace, type, subtypes, name, "", _) and
+        hasClassAndName(classWithMethod, result, name)
+      ) and
+      hasQualifiedName(namedClass, namespace, type) and
+      (
+        // member declared in the named type or a subtype of it
+        subtypes = true and
+        classWithMethod = namedClass.getADerivedClass*()
+        or
+        // member declared directly in the named type
+        subtypes = false and
+        classWithMethod = namedClass
+      )
     )
     or
+    elementSpec(namespace, type, subtypes, name, signature, _) and
     // Member variables
     signature = "" and
     exists(Class namedClass, Class classWithMember, MemberVariable member |
@@ -445,6 +842,7 @@ private Element interpretElement0(
     )
     or
     // Global or namespace variables
+    elementSpec(namespace, type, subtypes, name, signature, _) and
     signature = "" and
     type = "" and
     subtypes = false and

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`.
  */
-Function viableCallable(DataFlowCall call) {
+DataFlowCallable 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

@@ -216,7 +216,7 @@ predicate localMustFlowStep(Node node1, Node node2) { none() }
 
 /** Gets the type of `n` used for type pruning. */
 Type getNodeType(Node n) {
-  suppressUnusedNode(n) and
+  exists(n) and
   result instanceof VoidType // stub implementation
 }
 
@@ -227,13 +227,10 @@ string ppReprType(Type t) { none() } // stub implementation
  * Holds if `t1` and `t2` are compatible, that is, whether data can flow from
  * a node of type `t1` to a node of type `t2`.
  */
-pragma[inline]
 predicate compatibleTypes(Type t1, Type t2) {
-  any() // stub implementation
+  t1 instanceof VoidType and t2 instanceof VoidType // stub implementation
 }
 
-private predicate suppressUnusedNode(Node n) { any() }
-
 //////////////////////////////////////////////////////////////////////////////
 // Java QL library compatibility wrappers
 //////////////////////////////////////////////////////////////////////////////
@@ -242,7 +239,17 @@ class CastNode extends Node {
   CastNode() { none() } // stub implementation
 }
 
-class DataFlowCallable = Function;
+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 DataFlowExpr = Expr;
 
@@ -261,10 +268,28 @@ class DataFlowCall extends Expr instanceof Call {
   ExprNode getNode() { result.getExpr() = this }
 
   /** Gets the enclosing callable of this call. */
-  Function getEnclosingCallable() { result = this.getEnclosingFunction() }
+  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
+      )
+  }
 }
 
-predicate isUnreachableInCall(Node n, DataFlowCall call) { none() } // stub implementation
+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
 
 /**
  * Holds if access paths with `c` at their head always should be tracked at high

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

@@ -0,0 +1,27 @@
+/**
+ * This module provides extensible predicates for defining MaD models.
+ */
+
+/**
+ * Holds if an external source model exists for the given parameters.
+ */
+extensible predicate sourceModel(
+  string namespace, string type, boolean subtypes, string name, string signature, string ext,
+  string output, string kind, string provenance, QlBuiltins::ExtensionId madId
+);
+
+/**
+ * Holds if an external sink model exists for the given parameters.
+ */
+extensible predicate sinkModel(
+  string namespace, string type, boolean subtypes, string name, string signature, string ext,
+  string input, string kind, string provenance, QlBuiltins::ExtensionId madId
+);
+
+/**
+ * Holds if an external summary model exists for the given parameters.
+ */
+extensible predicate summaryModel(
+  string namespace, string type, boolean subtypes, string name, string signature, string ext,
+  string input, string output, string kind, string provenance, QlBuiltins::ExtensionId madId
+);

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

@@ -35,16 +35,22 @@ module Input implements InputSig<Location, DataFlowImplSpecific::CppDataFlow> {
       result = "Field" and
       arg = repeatStars(c.getIndirectionIndex() - 1) + c.getField().getName()
     )
+    or
+    exists(ElementContent ec |
+      cs.isSingleton(ec) and
+      result = "Element" and
+      arg = repeatStars(ec.getIndirectionIndex() - 1)
+    )
   }
 
   string encodeWithoutContent(ContentSet c, string arg) {
     // used for type tracking, not currently used in C/C++.
-    result = "WithoutContent" + c and arg = ""
+    none()
   }
 
   string encodeWithContent(ContentSet c, string arg) {
     // used for type tracking, not currently used in C/C++.
-    result = "WithContent" + c and arg = ""
+    none()
   }
 
   /**
@@ -79,25 +85,6 @@ module Input implements InputSig<Location, DataFlowImplSpecific::CppDataFlow> {
     token.getName() = "Parameter" and
     result = decodePosition(token.getAnArgument())
   }
-
-  bindingset[token]
-  ContentSet decodeUnknownContent(AccessPath::AccessPathTokenBase token) {
-    // field content (no indirection support)
-    exists(FieldContent c |
-      result.isSingleton(c) and
-      token.getName() = c.getField().getName() and
-      not exists(token.getArgumentList()) and
-      c.getIndirectionIndex() = 1
-    )
-    or
-    // field content (with indirection support)
-    exists(FieldContent c |
-      result.isSingleton(c) and
-      token.getName() = c.getField().getName() and
-      // FieldContent indices have 0 for the address, 1 for content, so we need to subtract one.
-      token.getAnArgument() = repeatStars(c.getIndirectionIndex() - 1)
-    )
-  }
 }
 
 private import Make<Location, DataFlowImplSpecific::CppDataFlow, Input> as Impl

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

@@ -412,6 +412,8 @@ class ArgumentPosition = Position;
 
 abstract class Position extends TPosition {
   abstract string toString();
+
+  abstract int getIndirectionIndex();
 }
 
 class DirectPosition extends Position, TDirectPosition {
@@ -421,13 +423,15 @@ class DirectPosition extends Position, TDirectPosition {
 
   override string toString() {
     index = -1 and
-    result = "this"
+    result = "this pointer"
     or
     index != -1 and
     result = index.toString()
   }
 
   int getIndex() { result = index }
+
+  final override int getIndirectionIndex() { result = 0 }
 }
 
 class IndirectionPosition extends Position, TIndirectionPosition {
@@ -438,16 +442,13 @@ class IndirectionPosition extends Position, TIndirectionPosition {
 
   override string toString() {
     if argumentIndex = -1
-    then if indirectionIndex > 0 then result = "this indirection" else result = "this"
-    else
-      if indirectionIndex > 0
-      then result = argumentIndex.toString() + " indirection"
-      else result = argumentIndex.toString()
+    then result = repeatStars(indirectionIndex - 1) + "this"
+    else result = repeatStars(indirectionIndex) + argumentIndex.toString()
   }
 
   int getArgumentIndex() { result = argumentIndex }
 
-  int getIndirectionIndex() { result = indirectionIndex }
+  final override int getIndirectionIndex() { result = indirectionIndex }
 }
 
 newtype TPosition =
@@ -988,7 +989,7 @@ predicate localMustFlowStep(Node node1, Node node2) { none() }
 
 /** Gets the type of `n` used for type pruning. */
 DataFlowType getNodeType(Node n) {
-  suppressUnusedNode(n) and
+  exists(n) and
   result instanceof VoidType // stub implementation
 }
 
@@ -999,13 +1000,10 @@ string ppReprType(DataFlowType t) { none() } // stub implementation
  * Holds if `t1` and `t2` are compatible, that is, whether data can flow from
  * a node of type `t1` to a node of type `t2`.
  */
-pragma[inline]
 predicate compatibleTypes(DataFlowType t1, DataFlowType t2) {
-  any() // stub implementation
+  t1 instanceof VoidType and t2 instanceof VoidType // stub implementation
 }
 
-private predicate suppressUnusedNode(Node n) { any() }
-
 //////////////////////////////////////////////////////////////////////////////
 // Java QL library compatibility wrappers
 //////////////////////////////////////////////////////////////////////////////
@@ -1062,6 +1060,16 @@ 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
+      )
+  }
 }
 
 /**
@@ -1159,6 +1167,16 @@ 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
+      )
+  }
 }
 
 /**
@@ -1247,43 +1265,53 @@ module IsUnreachableInCall {
     any(G::IRGuardCondition guard).ensuresLt(left, right, k, block, areEqual)
   }
 
-  predicate isUnreachableInCall(Node n, DataFlowCall call) {
+  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) {
     exists(
       InstructionDirectParameterNode paramNode, ConstantIntegralTypeArgumentNode arg,
-      IntegerConstantInstruction constant, int k, Operand left, Operand right, IRBlock block
+      IntegerConstantInstruction constant, int k, Operand left, Operand right, int argval
     |
       // arg flows into `paramNode`
-      DataFlowImplCommon::viableParamArg(call, paramNode, arg) and
+      DataFlowImplCommon::viableParamArg(call, pragma[only_bind_into](paramNode),
+        pragma[only_bind_into](arg)) and
       left = constant.getAUse() and
       right = valueNumber(paramNode.getInstruction()).getAUse() and
-      block = n.getBasicBlock()
+      argval = arg.getValue()
     |
       // and there's a guard condition which ensures that the result of `left == right + k` is `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)
-      |
+      exists(boolean areEqual | ensuresEq(left, right, k, block, areEqual) |
         // this block ensures that left = right + k, but it holds that `left != right + k`
         areEqual = true and
-        constant.getValue().toInt() != arg.getValue() + k
+        constant.getValue().toInt() != argval + k
         or
         // this block ensures that or `left != right + k`, but it holds that `left = right + k`
         areEqual = false and
-        constant.getValue().toInt() = arg.getValue() + k
+        constant.getValue().toInt() = argval + k
       )
       or
       // or there's a guard condition which ensures that the result of `left < right + k` is `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)
-      |
+      exists(boolean isLessThan | ensuresLt(left, right, k, block, isLessThan) |
         isLessThan = true and
         // this block ensures that `left < right + k`, but it holds that `left >= right + k`
-        constant.getValue().toInt() >= arg.getValue() + k
+        constant.getValue().toInt() >= argval + k
         or
         // this block ensures that `left >= right + k`, but it holds that `left < right + k`
         isLessThan = false and
-        constant.getValue().toInt() < arg.getValue() + k
+        constant.getValue().toInt() < argval + k
       )
     )
   }
@@ -1295,7 +1323,7 @@ import IsUnreachableInCall
  * Holds if access paths with `c` at their head always should be tracked at high
  * precision. This disables adaptive access path precision for such access paths.
  */
-predicate forceHighPrecision(Content c) { none() }
+predicate forceHighPrecision(Content c) { c instanceof ElementContent }
 
 /** Holds if `n` should be hidden from path explanations. */
 predicate nodeIsHidden(Node n) {
@@ -1306,6 +1334,8 @@ predicate nodeIsHidden(Node n) {
   n instanceof FinalGlobalValue
   or
   n instanceof InitialGlobalValue
+  or
+  n instanceof SsaPhiInputNode
 }
 
 predicate neverSkipInPathGraph(Node n) {
@@ -1364,7 +1394,8 @@ private predicate unionHasApproxName(Cpp::Union u, string s) { s = u.getName().c
 cached
 private newtype TContentApprox =
   TFieldApproxContent(string s) { fieldHasApproxName(_, s) } or
-  TUnionApproxContent(string s) { unionHasApproxName(_, s) }
+  TUnionApproxContent(string s) { unionHasApproxName(_, s) } or
+  TElementApproxContent()
 
 /** An approximated `Content`. */
 class ContentApprox extends TContentApprox {
@@ -1395,6 +1426,10 @@ private class UnionApproxContent extends ContentApprox, TUnionApproxContent {
   final override string toString() { result = s }
 }
 
+private class ElementApproxContent extends ContentApprox, TElementApproxContent {
+  final override string toString() { result = "ElementApprox" }
+}
+
 /** Gets an approximated value for content `c`. */
 pragma[inline]
 ContentApprox getContentApprox(Content c) {
@@ -1409,6 +1444,9 @@ ContentApprox getContentApprox(Content c) {
     u = c.(UnionContent).getUnion() and
     unionHasApproxName(u, prefix)
   )
+  or
+  c instanceof ElementContent and
+  result instanceof ElementApproxContent
 }
 
 /**
@@ -1604,6 +1642,8 @@ private Instruction getAnInstruction(Node n) {
   or
   result = n.(SsaPhiNode).getPhiNode().getBasicBlock().getFirstInstruction()
   or
+  result = n.(SsaPhiInputNode).getBasicBlock().getFirstInstruction()
+  or
   n.(IndirectInstruction).hasInstructionAndIndirectionIndex(result, _)
   or
   not n instanceof IndirectInstruction and
@@ -1666,6 +1706,14 @@ class DataFlowSecondLevelScope extends TDataFlowSecondLevelScope {
 /** Gets the second-level scope containing the node `n`, if any. */
 DataFlowSecondLevelScope getSecondLevelScope(Node n) { result.getANode() = n }
 
+/**
+ * Gets the maximum number of indirections to use for `ElementContent`.
+ *
+ * This should be equal to the largest number of stars (i.e., `*`s) in any
+ * `Element` content across all of our MaD summaries, sources, and sinks.
+ */
+int getMaxElementContentIndirectionIndex() { result = 5 }
+
 /**
  * Module that defines flow through iterators.
  * For example,
@@ -1733,7 +1781,7 @@ module IteratorFlow {
         crementCall = def.getValue().asInstruction().(StoreInstruction).getSourceValue() and
         sv = def.getSourceVariable() and
         bb.getInstruction(i) = crementCall and
-        Ssa::ssaDefReachesRead(sv, result.asDef(), bb, i)
+        Ssa::ssaDefReachesReadExt(sv, result.asDef(), bb, i)
       )
     }
 
@@ -1767,7 +1815,7 @@ module IteratorFlow {
         isIteratorWrite(writeToDeref, address) and
         operandForFullyConvertedCall(address, starCall) and
         bbStar.getInstruction(iStar) = starCall and
-        Ssa::ssaDefReachesRead(_, def.asDef(), bbStar, iStar) and
+        Ssa::ssaDefReachesReadExt(_, def.asDef(), bbStar, iStar) and
         ultimate = getAnUltimateDefinition*(def) and
         beginStore = ultimate.getValue().asInstruction() and
         operandForFullyConvertedCall(beginStore.getSourceValueOperand(), beginCall)
@@ -1778,7 +1826,7 @@ module IteratorFlow {
      * 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) {
+    predicate variableWrite(BasicBlock bb, int i, SourceVariable v, boolean certain) {
       certain = false and
       exists(GetsIteratorCall beginCall, Instruction writeToDeref, IRBlock bbQual, int iQual |
         isIteratorStoreInstruction(beginCall, writeToDeref) and
@@ -1789,7 +1837,7 @@ module IteratorFlow {
     }
 
     /** Holds if `(bb, i)` reads the container variable `v`. */
-    predicate variableRead(IRBlock bb, int i, SourceVariable v, boolean certain) {
+    predicate variableRead(BasicBlock bb, int i, SourceVariable v, boolean certain) {
       Ssa::variableRead(bb, i, v, certain)
     }
   }

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

@@ -17,6 +17,7 @@ private import SsaInternals as Ssa
 private import DataFlowImplCommon as DataFlowImplCommon
 private import codeql.util.Unit
 private import Node0ToString
+import ExprNodes
 
 /**
  * The IR dataflow graph consists of the following nodes:
@@ -45,6 +46,7 @@ private newtype TIRDataFlowNode =
     or
     Ssa::isModifiableByCall(operand, indirectionIndex)
   } or
+  TSsaPhiInputNode(Ssa::PhiNode phi, IRBlock input) { phi.hasInputFromBlock(_, _, _, _, input) } or
   TSsaPhiNode(Ssa::PhiNode phi) or
   TSsaIteratorNode(IteratorFlow::IteratorFlowNode n) or
   TRawIndirectOperand0(Node0Impl node, int indirectionIndex) {
@@ -114,6 +116,13 @@ predicate conversionFlow(
       instrTo.(CheckedConvertOrNullInstruction).getUnaryOperand() = opFrom
       or
       instrTo.(InheritanceConversionInstruction).getUnaryOperand() = opFrom
+      or
+      exists(BuiltInInstruction builtIn |
+        builtIn = instrTo and
+        // __builtin_bit_cast
+        builtIn.getBuiltInOperation() instanceof BuiltInBitCast and
+        opFrom = builtIn.getAnOperand()
+      )
     )
     or
     additional = true and
@@ -158,6 +167,12 @@ class Node extends TIRDataFlowNode {
   /** Gets the operands corresponding to this node, if any. */
   Operand asOperand() { result = this.(OperandNode).getOperand() }
 
+  /**
+   * Gets the operand that is indirectly tracked by this node behind `index`
+   * number of indirections.
+   */
+  Operand asIndirectOperand(int index) { hasOperandAndIndex(this, result, index) }
+
   /**
    * Holds if this node is at index `i` in basic block `block`.
    *
@@ -170,6 +185,9 @@ class Node extends TIRDataFlowNode {
     or
     this.(SsaPhiNode).getPhiNode().getBasicBlock() = block and i = -1
     or
+    this.(SsaPhiInputNode).getBlock() = block and
+    i = block.getInstructionCount()
+    or
     this.(RawIndirectOperand).getOperand().getUse() = block.getInstruction(i)
     or
     this.(RawIndirectInstruction).getInstruction() = block.getInstruction(i)
@@ -622,7 +640,7 @@ class SsaPhiNode extends Node, TSsaPhiNode {
 
   final override Location getLocationImpl() { result = phi.getBasicBlock().getLocation() }
 
-  override string toStringImpl() { result = "Phi" }
+  override string toStringImpl() { result = phi.toString() }
 
   /**
    * Gets a node that is used as input to this phi node.
@@ -631,7 +649,7 @@ class SsaPhiNode extends Node, TSsaPhiNode {
    */
   cached
   final Node getAnInput(boolean fromBackEdge) {
-    localFlowStep(result, this) and
+    result.(SsaPhiInputNode).getPhiNode() = phi and
     exists(IRBlock bPhi, IRBlock bResult |
       bPhi = phi.getBasicBlock() and bResult = result.getBasicBlock()
     |
@@ -654,6 +672,58 @@ class SsaPhiNode extends Node, TSsaPhiNode {
   predicate isPhiRead() { phi.isPhiRead() }
 }
 
+/**
+ * INTERNAL: Do not use.
+ *
+ * A node that is used as an input to a phi node.
+ *
+ * This class exists to allow more powerful barrier guards. Consider this
+ * example:
+ *
+ * ```cpp
+ * int x = source();
+ * if(!safe(x)) {
+ *   x = clear();
+ * }
+ * // phi node for x here
+ * sink(x);
+ * ```
+ *
+ * At the phi node for `x` it is neither the case that `x` is dominated by
+ * `safe(x)`, or is the case that the phi is dominated by a clearing of `x`.
+ *
+ * By inserting a "phi input" node as the last entry in the basic block that
+ * defines the inputs to the phi we can conclude that each of those inputs are
+ * safe to pass to `sink`.
+ */
+class SsaPhiInputNode extends Node, TSsaPhiInputNode {
+  Ssa::PhiNode phi;
+  IRBlock block;
+
+  SsaPhiInputNode() { this = TSsaPhiInputNode(phi, block) }
+
+  /** Gets the phi node associated with this node. */
+  Ssa::PhiNode getPhiNode() { result = phi }
+
+  /** Gets the basic block in which this input originates. */
+  IRBlock getBlock() { result = block }
+
+  override Declaration getEnclosingCallable() { result = this.getFunction() }
+
+  override Declaration getFunction() { result = phi.getBasicBlock().getEnclosingFunction() }
+
+  override DataFlowType getType() { result = this.getSourceVariable().getType() }
+
+  override predicate isGLValue() { phi.getSourceVariable().isGLValue() }
+
+  final override Location getLocationImpl() { result = block.getLastInstruction().getLocation() }
+
+  override string toStringImpl() { result = "Phi input" }
+
+  /** Gets the source variable underlying this phi node. */
+  Ssa::SourceVariable getSourceVariable() { result = phi.getSourceVariable() }
+}
+
 /**
  * INTERNAL: do not use.
  *
@@ -1227,466 +1297,6 @@ class UninitializedNode extends Node {
   LocalVariable getLocalVariable() { result = v }
 }
 
-private module GetConvertedResultExpression {
-  private import semmle.code.cpp.ir.implementation.raw.internal.TranslatedExpr
-  private import semmle.code.cpp.ir.implementation.raw.internal.InstructionTag
-
-  private Operand getAnInitializeDynamicAllocationInstructionAddress() {
-    result = any(InitializeDynamicAllocationInstruction init).getAllocationAddressOperand()
-  }
-
-  /**
-   * Gets the expression that should be returned as the result expression from `instr`.
-   *
-   * Note that this predicate may return multiple results in cases where a conversion belongs to a
-   * different AST element than its operand.
-   */
-  Expr getConvertedResultExpression(Instruction instr, int n) {
-    // Only fully converted instructions have a result for `asConvertedExpr`
-    not conversionFlow(unique(Operand op |
-        // The address operand of a `InitializeDynamicAllocationInstruction` is
-        // special: we need to handle it during dataflow (since it's
-        // effectively a store to an indirection), but it doesn't appear in
-        // source syntax, so dataflow node <-> expression conversion shouldn't
-        // care about it.
-        op = getAUse(instr) and not op = getAnInitializeDynamicAllocationInstructionAddress()
-      |
-        op
-      ), _, false, false) and
-    result = getConvertedResultExpressionImpl(instr) and
-    n = 0
-    or
-    // If the conversion also has a result then we return multiple results
-    exists(Operand operand | conversionFlow(operand, instr, false, false) |
-      n = 1 and
-      result = getConvertedResultExpressionImpl(operand.getDef())
-      or
-      result = getConvertedResultExpression(operand.getDef(), n - 1)
-    )
-  }
-
-  private Expr getConvertedResultExpressionImpl0(Instruction instr) {
-    // IR construction inserts an additional cast to a `size_t` on the extent
-    // of a `new[]` expression. The resulting `ConvertInstruction` doesn't have
-    // a result for `getConvertedResultExpression`. We remap this here so that
-    // this `ConvertInstruction` maps to the result of the expression that
-    // represents the extent.
-    exists(TranslatedNonConstantAllocationSize tas |
-      result = tas.getExtent().getExpr() and
-      instr = tas.getInstruction(AllocationExtentConvertTag())
-    )
-    or
-    // There's no instruction that returns `ParenthesisExpr`, but some queries
-    // expect this
-    exists(TranslatedTransparentConversion ttc |
-      result = ttc.getExpr().(ParenthesisExpr) and
-      instr = ttc.getResult()
-    )
-    or
-    // Certain expressions generate `CopyValueInstruction`s only when they
-    // are needed. Examples of this include crement operations and compound
-    // assignment operations. For example:
-    // ```cpp
-    // int x = ...
-    // int y = x++;
-    // ```
-    // this generate IR like:
-    // ```
-    // r1(glval<int>) = VariableAddress[x] :
-    // r2(int)        = Constant[0]        :
-    // m3(int)        = Store[x]           : &:r1, r2
-    // r4(glval<int>) = VariableAddress[y] :
-    // r5(glval<int>) = VariableAddress[x] :
-    // r6(int)        = Load[x]            : &:r5, m3
-    // r7(int)        = Constant[1]        :
-    // r8(int)        = Add                : r6, r7
-    // m9(int)        = Store[x]           : &:r5, r8
-    // r11(int)       = CopyValue         : r6
-    // m12(int)       = Store[y]          : &:r4, r11
-    // ```
-    // When the `CopyValueInstruction` is not generated there is no instruction
-    // whose `getConvertedResultExpression` maps back to the expression. When
-    // such an instruction doesn't exist it means that the old value is not
-    // needed, and in that case the only value that will propagate forward in
-    // the program is the value that's been updated. So in those cases we just
-    // use the result of `node.asDefinition()` as the result of `node.asExpr()`.
-    exists(TranslatedCoreExpr tco |
-      tco.getInstruction(_) = instr and
-      tco.producesExprResult() and
-      result = asDefinitionImpl0(instr)
-    )
-  }
-
-  private Expr getConvertedResultExpressionImpl(Instruction instr) {
-    result = getConvertedResultExpressionImpl0(instr)
-    or
-    not exists(getConvertedResultExpressionImpl0(instr)) and
-    result = instr.getConvertedResultExpression()
-  }
-
-  /**
-   * Gets the result for `node.asDefinition()` (when `node` is the instruction
-   * node that wraps `store`) in the cases where `store.getAst()` should not be
-   * used to define the result of `node.asDefinition()`.
-   */
-  private Expr asDefinitionImpl0(StoreInstruction store) {
-    // For an expression such as `i += 2` we pretend that the generated
-    // `StoreInstruction` contains the result of the expression even though
-    // this isn't totally aligned with the C/C++ standard.
-    exists(TranslatedAssignOperation tao |
-      store = tao.getInstruction(AssignmentStoreTag()) and
-      result = tao.getExpr()
-    )
-    or
-    // Similarly for `i++` and `++i` we pretend that the generated
-    // `StoreInstruction` is contains the result of the expression even though
-    // this isn't totally aligned with the C/C++ standard.
-    exists(TranslatedCrementOperation tco |
-      store = tco.getInstruction(CrementStoreTag()) and
-      result = tco.getExpr()
-    )
-  }
-
-  /**
-   * Holds if the expression returned by `store.getAst()` should not be
-   * returned as the result of `node.asDefinition()` when `node` is the
-   * instruction node that wraps `store`.
-   */
-  private predicate excludeAsDefinitionResult(StoreInstruction store) {
-    // Exclude the store to the temporary generated by a ternary expression.
-    exists(TranslatedConditionalExpr tce |
-      store = tce.getInstruction(ConditionValueFalseStoreTag())
-      or
-      store = tce.getInstruction(ConditionValueTrueStoreTag())
-    )
-  }
-
-  /**
-   * Gets the expression that represents the result of `StoreInstruction` for
-   * dataflow purposes.
-   *
-   * For example, consider the following example
-   * ```cpp
-   * int x = 42;     // 1
-   * x = 34;         // 2
-   * ++x;            // 3
-   * x++;            // 4
-   * x += 1;         // 5
-   * int y = x += 2; // 6
-   * ```
-   * For (1) the result is `42`.
-   * For (2) the result is `x = 34`.
-   * For (3) the result is `++x`.
-   * For (4) the result is `x++`.
-   * For (5) the result is `x += 1`.
-   * For (6) there are two results:
-   *   - For the `StoreInstruction` generated by `x += 2` the result
-   *     is `x += 2`
-   *   - For the `StoreInstruction` generated by `int y = ...` the result
-   *     is also `x += 2`
-   */
-  Expr asDefinitionImpl(StoreInstruction store) {
-    not exists(asDefinitionImpl0(store)) and
-    not excludeAsDefinitionResult(store) and
-    result = store.getAst().(Expr).getUnconverted()
-    or
-    result = asDefinitionImpl0(store)
-  }
-}
-
-private import GetConvertedResultExpression
-
-/** Holds if `node` is an `OperandNode` that should map `node.asExpr()` to `e`. */
-predicate exprNodeShouldBeOperand(OperandNode node, Expr e, int n) {
-  not exprNodeShouldBeIndirectOperand(_, e, n) and
-  exists(Instruction def |
-    unique( | | getAUse(def)) = node.getOperand() and
-    e = getConvertedResultExpression(def, n)
-  )
-}
-
-/** Holds if `node` should be an `IndirectOperand` that maps `node.asIndirectExpr()` to `e`. */
-private predicate indirectExprNodeShouldBeIndirectOperand(
-  IndirectOperand node, Expr e, int n, int indirectionIndex
-) {
-  exists(Instruction def |
-    node.hasOperandAndIndirectionIndex(unique( | | getAUse(def)), indirectionIndex) and
-    e = getConvertedResultExpression(def, n)
-  )
-}
-
-/** Holds if `node` should be an `IndirectOperand` that maps `node.asExpr()` to `e`. */
-private predicate exprNodeShouldBeIndirectOperand(IndirectOperand node, Expr e, int n) {
-  exists(ArgumentOperand operand |
-    // When an argument (qualifier or positional) is a prvalue and the
-    // parameter (qualifier or positional) is a (const) reference, IR
-    // construction introduces a temporary `IRVariable`. The `VariableAddress`
-    // instruction has the argument as its `getConvertedResultExpression`
-    // result. However, the instruction actually represents the _address_ of
-    // the argument. So to fix this mismatch, we have the indirection of the
-    // `VariableAddressInstruction` map to the expression.
-    node.hasOperandAndIndirectionIndex(operand, 1) and
-    e = getConvertedResultExpression(operand.getDef(), n) and
-    operand.getDef().(VariableAddressInstruction).getIRVariable() instanceof IRTempVariable
-  )
-}
-
-private predicate exprNodeShouldBeIndirectOutNode(IndirectArgumentOutNode node, Expr e, int n) {
-  exists(CallInstruction call |
-    call.getStaticCallTarget() instanceof Constructor and
-    e = getConvertedResultExpression(call, n) and
-    call.getThisArgumentOperand() = node.getAddressOperand()
-  )
-}
-
-/** Holds if `node` should be an instruction node that maps `node.asExpr()` to `e`. */
-predicate exprNodeShouldBeInstruction(Node node, Expr e, int n) {
-  not exprNodeShouldBeOperand(_, e, n) and
-  not exprNodeShouldBeIndirectOutNode(_, e, n) and
-  not exprNodeShouldBeIndirectOperand(_, e, n) and
-  e = getConvertedResultExpression(node.asInstruction(), n)
-}
-
-/** Holds if `node` should be an `IndirectInstruction` that maps `node.asIndirectExpr()` to `e`. */
-predicate indirectExprNodeShouldBeIndirectInstruction(
-  IndirectInstruction node, Expr e, int n, int indirectionIndex
-) {
-  not indirectExprNodeShouldBeIndirectOperand(_, e, n, indirectionIndex) and
-  exists(Instruction instr |
-    node.hasInstructionAndIndirectionIndex(instr, indirectionIndex) and
-    e = getConvertedResultExpression(instr, n)
-  )
-}
-
-abstract private class ExprNodeBase extends Node {
-  /**
-   * Gets the expression corresponding to this node, if any. The returned
-   * expression may be a `Conversion`.
-   */
-  abstract Expr getConvertedExpr(int n);
-
-  /** Gets the non-conversion expression corresponding to this node, if any. */
-  final Expr getExpr(int n) { result = this.getConvertedExpr(n).getUnconverted() }
-}
-
-/**
- * Holds if there exists a dataflow node whose `asExpr(n)` should evaluate
- * to `e`.
- */
-private predicate exprNodeShouldBe(Expr e, int n) {
-  exprNodeShouldBeInstruction(_, e, n) or
-  exprNodeShouldBeOperand(_, e, n) or
-  exprNodeShouldBeIndirectOutNode(_, e, n) or
-  exprNodeShouldBeIndirectOperand(_, e, n)
-}
-
-private class InstructionExprNode extends ExprNodeBase, InstructionNode {
-  InstructionExprNode() {
-    exists(Expr e, int n |
-      exprNodeShouldBeInstruction(this, e, n) and
-      not exists(Expr conv |
-        exprNodeShouldBe(conv, n + 1) and
-        conv.getUnconverted() = e.getUnconverted()
-      )
-    )
-  }
-
-  final override Expr getConvertedExpr(int n) { exprNodeShouldBeInstruction(this, result, n) }
-}
-
-private class OperandExprNode extends ExprNodeBase, OperandNode {
-  OperandExprNode() {
-    exists(Expr e, int n |
-      exprNodeShouldBeOperand(this, e, n) and
-      not exists(Expr conv |
-        exprNodeShouldBe(conv, n + 1) and
-        conv.getUnconverted() = e.getUnconverted()
-      )
-    )
-  }
-
-  final override Expr getConvertedExpr(int n) { exprNodeShouldBeOperand(this, result, n) }
-}
-
-abstract private class IndirectExprNodeBase extends Node {
-  /**
-   * Gets the expression corresponding to this node, if any. The returned
-   * expression may be a `Conversion`.
-   */
-  abstract Expr getConvertedExpr(int n, int indirectionIndex);
-
-  /** Gets the non-conversion expression corresponding to this node, if any. */
-  final Expr getExpr(int n, int indirectionIndex) {
-    result = this.getConvertedExpr(n, indirectionIndex).getUnconverted()
-  }
-}
-
-/** A signature for converting an indirect node to an expression. */
-private signature module IndirectNodeToIndirectExprSig {
-  /** The indirect node class to be converted to an expression */
-  class IndirectNode;
-
-  /**
-   * Holds if the indirect expression at indirection index `indirectionIndex`
-   * of `node` is `e`. The integer `n` specifies how many conversions has been
-   * applied to `node`.
-   */
-  predicate indirectNodeHasIndirectExpr(IndirectNode node, Expr e, int n, int indirectionIndex);
-}
-
-/**
- * A module that implements the logic for deciding whether an indirect node
- * should be an `IndirectExprNode`.
- */
-private module IndirectNodeToIndirectExpr<IndirectNodeToIndirectExprSig Sig> {
-  import Sig
-
-  /**
-   * This predicate shifts the indirection index by one when `conv` is a
-   * `ReferenceDereferenceExpr`.
-   *
-   * This is necessary because `ReferenceDereferenceExpr` is a conversion
-   * in the AST, but appears as a `LoadInstruction` in the IR.
-   */
-  bindingset[e, indirectionIndex]
-  private predicate adjustForReference(
-    Expr e, int indirectionIndex, Expr conv, int adjustedIndirectionIndex
-  ) {
-    conv.(ReferenceDereferenceExpr).getExpr() = e and
-    adjustedIndirectionIndex = indirectionIndex - 1
-    or
-    not conv instanceof ReferenceDereferenceExpr and
-    conv = e and
-    adjustedIndirectionIndex = indirectionIndex
-  }
-
-  /** Holds if `node` should be an `IndirectExprNode`. */
-  predicate charpred(IndirectNode node) {
-    exists(Expr e, int n, int indirectionIndex |
-      indirectNodeHasIndirectExpr(node, e, n, indirectionIndex) and
-      not exists(Expr conv, int adjustedIndirectionIndex |
-        adjustForReference(e, indirectionIndex, conv, adjustedIndirectionIndex) and
-        indirectExprNodeShouldBe(conv, n + 1, adjustedIndirectionIndex)
-      )
-    )
-  }
-}
-
-private predicate indirectExprNodeShouldBe(Expr e, int n, int indirectionIndex) {
-  indirectExprNodeShouldBeIndirectOperand(_, e, n, indirectionIndex) or
-  indirectExprNodeShouldBeIndirectInstruction(_, e, n, indirectionIndex)
-}
-
-private module IndirectOperandIndirectExprNodeImpl implements IndirectNodeToIndirectExprSig {
-  class IndirectNode = IndirectOperand;
-
-  predicate indirectNodeHasIndirectExpr = indirectExprNodeShouldBeIndirectOperand/4;
-}
-
-module IndirectOperandToIndirectExpr =
-  IndirectNodeToIndirectExpr<IndirectOperandIndirectExprNodeImpl>;
-
-private class IndirectOperandIndirectExprNode extends IndirectExprNodeBase instanceof IndirectOperand
-{
-  IndirectOperandIndirectExprNode() { IndirectOperandToIndirectExpr::charpred(this) }
-
-  final override Expr getConvertedExpr(int n, int index) {
-    IndirectOperandToIndirectExpr::indirectNodeHasIndirectExpr(this, result, n, index)
-  }
-}
-
-private module IndirectInstructionIndirectExprNodeImpl implements IndirectNodeToIndirectExprSig {
-  class IndirectNode = IndirectInstruction;
-
-  predicate indirectNodeHasIndirectExpr = indirectExprNodeShouldBeIndirectInstruction/4;
-}
-
-module IndirectInstructionToIndirectExpr =
-  IndirectNodeToIndirectExpr<IndirectInstructionIndirectExprNodeImpl>;
-
-private class IndirectInstructionIndirectExprNode extends IndirectExprNodeBase instanceof IndirectInstruction
-{
-  IndirectInstructionIndirectExprNode() { IndirectInstructionToIndirectExpr::charpred(this) }
-
-  final override Expr getConvertedExpr(int n, int index) {
-    IndirectInstructionToIndirectExpr::indirectNodeHasIndirectExpr(this, result, n, index)
-  }
-}
-
-private class IndirectArgumentOutExprNode extends ExprNodeBase, IndirectArgumentOutNode {
-  IndirectArgumentOutExprNode() { exprNodeShouldBeIndirectOutNode(this, _, _) }
-
-  final override Expr getConvertedExpr(int n) { exprNodeShouldBeIndirectOutNode(this, result, n) }
-}
-
-private class IndirectOperandExprNode extends ExprNodeBase instanceof IndirectOperand {
-  IndirectOperandExprNode() { exprNodeShouldBeIndirectOperand(this, _, _) }
-
-  final override Expr getConvertedExpr(int n) { exprNodeShouldBeIndirectOperand(this, result, n) }
-}
-
-/**
- * An expression, viewed as a node in a data flow graph.
- */
-class ExprNode extends Node instanceof ExprNodeBase {
-  /**
-   * INTERNAL: Do not use.
-   */
-  Expr getExpr(int n) { result = super.getExpr(n) }
-
-  /**
-   * Gets the non-conversion expression corresponding to this node, if any. If
-   * this node strictly (in the sense of `getConvertedExpr`) corresponds to a
-   * `Conversion`, then the result is that `Conversion`'s non-`Conversion` base
-   * expression.
-   */
-  final Expr getExpr() { result = this.getExpr(_) }
-
-  /**
-   * INTERNAL: Do not use.
-   */
-  Expr getConvertedExpr(int n) { result = super.getConvertedExpr(n) }
-
-  /**
-   * Gets the expression corresponding to this node, if any. The returned
-   * expression may be a `Conversion`.
-   */
-  final Expr getConvertedExpr() { result = this.getConvertedExpr(_) }
-}
-
-/**
- * An indirect expression, viewed as a node in a data flow graph.
- */
-class IndirectExprNode extends Node instanceof IndirectExprNodeBase {
-  /**
-   * Gets the non-conversion expression corresponding to this node, if any. If
-   * this node strictly (in the sense of `getConvertedExpr`) corresponds to a
-   * `Conversion`, then the result is that `Conversion`'s non-`Conversion` base
-   * expression.
-   */
-  final Expr getExpr(int indirectionIndex) { result = this.getExpr(_, indirectionIndex) }
-
-  /**
-   * INTERNAL: Do not use.
-   */
-  Expr getExpr(int n, int indirectionIndex) { result = super.getExpr(n, indirectionIndex) }
-
-  /**
-   * INTERNAL: Do not use.
-   */
-  Expr getConvertedExpr(int n, int indirectionIndex) {
-    result = super.getConvertedExpr(n, indirectionIndex)
-  }
-
-  /**
-   * Gets the expression corresponding to this node, if any. The returned
-   * expression may be a `Conversion`.
-   */
-  Expr getConvertedExpr(int indirectionIndex) {
-    result = this.getConvertedExpr(_, indirectionIndex)
-  }
-}
-
 abstract private class AbstractParameterNode extends Node {
   /**
    * Holds if this node is the parameter of `f` at the specified position. The
@@ -2176,6 +1786,9 @@ private module Cached {
       // Def-use/Use-use flow
       Ssa::ssaFlow(nodeFrom, nodeTo)
       or
+      // Phi input -> Phi
+      nodeFrom.(SsaPhiInputNode).getPhiNode() = nodeTo.(SsaPhiNode).getPhiNode()
+      or
       IteratorFlow::localFlowStep(nodeFrom, nodeTo)
       or
       // Operand -> Instruction flow
@@ -2470,6 +2083,9 @@ private newtype TContent =
       indirectionIndex =
         [1 .. max(Ssa::getMaxIndirectionsForType(getAFieldWithSize(u, bytes).getUnspecifiedType()))]
     )
+  } or
+  TElementContent(int indirectionIndex) {
+    indirectionIndex = [1 .. getMaxElementContentIndirectionIndex()]
   }
 
 /**
@@ -2580,6 +2196,25 @@ class UnionContent extends Content, TUnionContent {
   }
 }
 
+/**
+ * A `Content` that represents one of the elements of a
+ * container (e.g., `std::vector`).
+ */
+class ElementContent extends Content, TElementContent {
+  int indirectionIndex;
+
+  ElementContent() { this = TElementContent(indirectionIndex) }
+
+  pragma[inline]
+  override int getIndirectionIndex() {
+    pragma[only_bind_into](result) = pragma[only_bind_out](indirectionIndex)
+  }
+
+  override predicate impliesClearOf(Content c) { none() }
+
+  override string toString() { result = contentStars(this) + "element" }
+}
+
 /**
  * An entity that represents a set of `Content`s.
  *
@@ -2614,6 +2249,22 @@ class ContentSet instanceof Content {
   }
 }
 
+pragma[nomagic]
+private predicate guardControlsPhiInput(
+  IRGuardCondition g, boolean branch, Ssa::Definition def, IRBlock input, Ssa::PhiNode phi
+) {
+  phi.hasInputFromBlock(def, _, _, _, input) and
+  (
+    g.controls(input, branch)
+    or
+    exists(EdgeKind kind |
+      g.getBlock() = input and
+      kind = getConditionalEdge(branch) and
+      input.getSuccessor(kind) = phi.getBasicBlock()
+    )
+  )
+}
+
 /**
  * Holds if the guard `g` validates the expression `e` upon evaluating to `branch`.
  *
@@ -2662,13 +2313,21 @@ module BarrierGuard<guardChecksSig/3 guardChecks> {
    *
    * NOTE: If an indirect expression is tracked, use `getAnIndirectBarrierNode` instead.
    */
-  ExprNode getABarrierNode() {
+  Node getABarrierNode() {
     exists(IRGuardCondition g, Expr e, ValueNumber value, boolean edge |
       e = value.getAnInstruction().getConvertedResultExpression() and
-      result.getConvertedExpr() = e and
+      result.asConvertedExpr() = e and
       guardChecks(g, value.getAnInstruction().getConvertedResultExpression(), edge) and
       g.controls(result.getBasicBlock(), edge)
     )
+    or
+    exists(
+      IRGuardCondition g, boolean branch, Ssa::DefinitionExt def, IRBlock input, Ssa::PhiNode phi
+    |
+      guardChecks(g, def.getARead().asOperand().getDef().getConvertedResultExpression(), branch) and
+      guardControlsPhiInput(g, branch, def, input, phi) and
+      result = TSsaPhiInputNode(phi, input)
+    )
   }
 
   /**
@@ -2704,7 +2363,7 @@ module BarrierGuard<guardChecksSig/3 guardChecks> {
    *
    * NOTE: If a non-indirect expression is tracked, use `getABarrierNode` instead.
    */
-  IndirectExprNode getAnIndirectBarrierNode() { result = getAnIndirectBarrierNode(_) }
+  Node getAnIndirectBarrierNode() { result = getAnIndirectBarrierNode(_) }
 
   /**
    * Gets an indirect expression node with indirection index `indirectionIndex` that is
@@ -2740,13 +2399,23 @@ module BarrierGuard<guardChecksSig/3 guardChecks> {
    *
    * NOTE: If a non-indirect expression is tracked, use `getABarrierNode` instead.
    */
-  IndirectExprNode getAnIndirectBarrierNode(int indirectionIndex) {
+  Node getAnIndirectBarrierNode(int indirectionIndex) {
     exists(IRGuardCondition g, Expr e, ValueNumber value, boolean edge |
       e = value.getAnInstruction().getConvertedResultExpression() and
-      result.getConvertedExpr(indirectionIndex) = e and
+      result.asIndirectConvertedExpr(indirectionIndex) = e and
       guardChecks(g, value.getAnInstruction().getConvertedResultExpression(), edge) and
       g.controls(result.getBasicBlock(), edge)
     )
+    or
+    exists(
+      IRGuardCondition g, boolean branch, Ssa::DefinitionExt def, IRBlock input, Ssa::PhiNode phi
+    |
+      guardChecks(g,
+        def.getARead().asIndirectOperand(indirectionIndex).getDef().getConvertedResultExpression(),
+        branch) and
+      guardControlsPhiInput(g, branch, def, input, phi) and
+      result = TSsaPhiInputNode(phi, input)
+    )
   }
 }
 
@@ -2755,6 +2424,14 @@ module BarrierGuard<guardChecksSig/3 guardChecks> {
  */
 signature predicate instructionGuardChecksSig(IRGuardCondition g, Instruction instr, boolean branch);
 
+private EdgeKind getConditionalEdge(boolean branch) {
+  branch = true and
+  result instanceof TrueEdge
+  or
+  branch = false and
+  result instanceof FalseEdge
+}
+
 /**
  * Provides a set of barrier nodes for a guard that validates an instruction.
  *
@@ -2763,12 +2440,20 @@ signature predicate instructionGuardChecksSig(IRGuardCondition g, Instruction in
  */
 module InstructionBarrierGuard<instructionGuardChecksSig/3 instructionGuardChecks> {
   /** Gets a node that is safely guarded by the given guard check. */
-  ExprNode getABarrierNode() {
+  Node getABarrierNode() {
     exists(IRGuardCondition g, ValueNumber value, boolean edge, Operand use |
       instructionGuardChecks(g, value.getAnInstruction(), edge) and
       use = value.getAnInstruction().getAUse() and
       result.asOperand() = use and
-      g.controls(use.getDef().getBlock(), edge)
+      g.controls(result.getBasicBlock(), edge)
+    )
+    or
+    exists(
+      IRGuardCondition g, boolean branch, Ssa::DefinitionExt def, IRBlock input, Ssa::PhiNode phi
+    |
+      instructionGuardChecks(g, def.getARead().asOperand().getDef(), branch) and
+      guardControlsPhiInput(g, branch, def, input, phi) and
+      result = TSsaPhiInputNode(phi, input)
     )
   }
 }

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

@@ -0,0 +1,518 @@
+/**
+ * Provides the classes `ExprNode` and `IndirectExprNode` for converting between `Expr` and `Node`.
+ */
+
+private import cpp
+private import semmle.code.cpp.ir.IR
+private import DataFlowUtil
+private import DataFlowPrivate
+private import semmle.code.cpp.ir.implementation.raw.internal.TranslatedExpr
+private import semmle.code.cpp.ir.implementation.raw.internal.InstructionTag
+
+cached
+private module Cached {
+  private Operand getAnInitializeDynamicAllocationInstructionAddress() {
+    result = any(InitializeDynamicAllocationInstruction init).getAllocationAddressOperand()
+  }
+
+  /**
+   * Gets the expression that should be returned as the result expression from `instr`.
+   *
+   * Note that this predicate may return multiple results in cases where a conversion belongs to a
+   * different AST element than its operand.
+   */
+  private Expr getConvertedResultExpression(Instruction instr, int n) {
+    // Only fully converted instructions have a result for `asConvertedExpr`
+    not conversionFlow(unique(Operand op |
+        // The address operand of a `InitializeDynamicAllocationInstruction` is
+        // special: we need to handle it during dataflow (since it's
+        // effectively a store to an indirection), but it doesn't appear in
+        // source syntax, so dataflow node <-> expression conversion shouldn't
+        // care about it.
+        op = getAUse(instr) and not op = getAnInitializeDynamicAllocationInstructionAddress()
+      |
+        op
+      ), _, false, false) and
+    result = getConvertedResultExpressionImpl(instr) and
+    n = 0
+    or
+    // If the conversion also has a result then we return multiple results
+    exists(Operand operand | conversionFlow(operand, instr, false, false) |
+      n = 1 and
+      result = getConvertedResultExpressionImpl(operand.getDef())
+      or
+      result = getConvertedResultExpression(operand.getDef(), n - 1)
+    )
+  }
+
+  private Expr getConvertedResultExpressionImpl0(Instruction instr) {
+    // IR construction inserts an additional cast to a `size_t` on the extent
+    // of a `new[]` expression. The resulting `ConvertInstruction` doesn't have
+    // a result for `getConvertedResultExpression`. We remap this here so that
+    // this `ConvertInstruction` maps to the result of the expression that
+    // represents the extent.
+    exists(TranslatedNonConstantAllocationSize tas |
+      result = tas.getExtent().getExpr() and
+      instr = tas.getInstruction(AllocationExtentConvertTag())
+    )
+    or
+    // There's no instruction that returns `ParenthesisExpr`, but some queries
+    // expect this
+    exists(TranslatedTransparentConversion ttc |
+      result = ttc.getExpr().(ParenthesisExpr) and
+      instr = ttc.getResult()
+    )
+    or
+    // Certain expressions generate `CopyValueInstruction`s only when they
+    // are needed. Examples of this include crement operations and compound
+    // assignment operations. For example:
+    // ```cpp
+    // int x = ...
+    // int y = x++;
+    // ```
+    // this generate IR like:
+    // ```
+    // r1(glval<int>) = VariableAddress[x] :
+    // r2(int)        = Constant[0]        :
+    // m3(int)        = Store[x]           : &:r1, r2
+    // r4(glval<int>) = VariableAddress[y] :
+    // r5(glval<int>) = VariableAddress[x] :
+    // r6(int)        = Load[x]            : &:r5, m3
+    // r7(int)        = Constant[1]        :
+    // r8(int)        = Add                : r6, r7
+    // m9(int)        = Store[x]           : &:r5, r8
+    // r11(int)       = CopyValue         : r6
+    // m12(int)       = Store[y]          : &:r4, r11
+    // ```
+    // When the `CopyValueInstruction` is not generated there is no instruction
+    // whose `getConvertedResultExpression` maps back to the expression. When
+    // such an instruction doesn't exist it means that the old value is not
+    // needed, and in that case the only value that will propagate forward in
+    // the program is the value that's been updated. So in those cases we just
+    // use the result of `node.asDefinition()` as the result of `node.asExpr()`.
+    exists(TranslatedCoreExpr tco |
+      tco.getInstruction(_) = instr and
+      tco.producesExprResult() and
+      result = asDefinitionImpl0(instr)
+    )
+  }
+
+  private Expr getConvertedResultExpressionImpl(Instruction instr) {
+    result = getConvertedResultExpressionImpl0(instr)
+    or
+    not exists(getConvertedResultExpressionImpl0(instr)) and
+    result = instr.getConvertedResultExpression()
+  }
+
+  /**
+   * Gets the result for `node.asDefinition()` (when `node` is the instruction
+   * node that wraps `store`) in the cases where `store.getAst()` should not be
+   * used to define the result of `node.asDefinition()`.
+   */
+  private Expr asDefinitionImpl0(StoreInstruction store) {
+    // For an expression such as `i += 2` we pretend that the generated
+    // `StoreInstruction` contains the result of the expression even though
+    // this isn't totally aligned with the C/C++ standard.
+    exists(TranslatedAssignOperation tao |
+      store = tao.getInstruction(AssignmentStoreTag()) and
+      result = tao.getExpr()
+    )
+    or
+    // Similarly for `i++` and `++i` we pretend that the generated
+    // `StoreInstruction` is contains the result of the expression even though
+    // this isn't totally aligned with the C/C++ standard.
+    exists(TranslatedCrementOperation tco |
+      store = tco.getInstruction(CrementStoreTag()) and
+      result = tco.getExpr()
+    )
+  }
+
+  /**
+   * Holds if the expression returned by `store.getAst()` should not be
+   * returned as the result of `node.asDefinition()` when `node` is the
+   * instruction node that wraps `store`.
+   */
+  private predicate excludeAsDefinitionResult(StoreInstruction store) {
+    // Exclude the store to the temporary generated by a ternary expression.
+    exists(TranslatedConditionalExpr tce |
+      store = tce.getInstruction(ConditionValueFalseStoreTag())
+      or
+      store = tce.getInstruction(ConditionValueTrueStoreTag())
+    )
+  }
+
+  /**
+   * Gets the expression that represents the result of `StoreInstruction` for
+   * dataflow purposes.
+   *
+   * For example, consider the following example
+   * ```cpp
+   * int x = 42;     // 1
+   * x = 34;         // 2
+   * ++x;            // 3
+   * x++;            // 4
+   * x += 1;         // 5
+   * int y = x += 2; // 6
+   * ```
+   * For (1) the result is `42`.
+   * For (2) the result is `x = 34`.
+   * For (3) the result is `++x`.
+   * For (4) the result is `x++`.
+   * For (5) the result is `x += 1`.
+   * For (6) there are two results:
+   *   - For the `StoreInstruction` generated by `x += 2` the result
+   *     is `x += 2`
+   *   - For the `StoreInstruction` generated by `int y = ...` the result
+   *     is also `x += 2`
+   */
+  cached
+  Expr asDefinitionImpl(StoreInstruction store) {
+    not exists(asDefinitionImpl0(store)) and
+    not excludeAsDefinitionResult(store) and
+    result = store.getAst().(Expr).getUnconverted()
+    or
+    result = asDefinitionImpl0(store)
+  }
+
+  /** Holds if `node` is an `OperandNode` that should map `node.asExpr()` to `e`. */
+  private predicate exprNodeShouldBeOperand(OperandNode node, Expr e, int n) {
+    not exprNodeShouldBeIndirectOperand(_, e, n) and
+    exists(Instruction def |
+      unique( | | getAUse(def)) = node.getOperand() and
+      e = getConvertedResultExpression(def, n)
+    )
+  }
+
+  /** Holds if `node` should be an `IndirectOperand` that maps `node.asIndirectExpr()` to `e`. */
+  private predicate indirectExprNodeShouldBeIndirectOperand(
+    IndirectOperand node, Expr e, int n, int indirectionIndex
+  ) {
+    exists(Instruction def |
+      node.hasOperandAndIndirectionIndex(unique( | | getAUse(def)), indirectionIndex) and
+      e = getConvertedResultExpression(def, n)
+    )
+  }
+
+  /** Holds if `operand`'s definition is a `VariableAddressInstruction` whose variable is a temporary */
+  private predicate isIRTempVariable(Operand operand) {
+    operand.getDef().(VariableAddressInstruction).getIRVariable() instanceof IRTempVariable
+  }
+
+  /**
+   * Holds if `node` is an indirect operand whose operand is an argument, and
+   * the `n`'th expression associated with the operand is `e`.
+   */
+  private predicate isIndirectOperandOfArgument(
+    IndirectOperand node, ArgumentOperand operand, Expr e, int n
+  ) {
+    node.hasOperandAndIndirectionIndex(operand, 1) and
+    e = getConvertedResultExpression(operand.getDef(), n)
+  }
+
+  /**
+   * Holds if `opFrom` is an operand to a conversion, and `opTo` is the unique
+   * use of the conversion.
+   */
+  private predicate isConversionStep(Operand opFrom, Operand opTo) {
+    exists(Instruction mid |
+      conversionFlow(opFrom, mid, false, false) and
+      opTo = unique( | | getAUse(mid))
+    )
+  }
+
+  /**
+   * Holds if an operand that satisfies `isIRTempVariable` flows to `op`
+   * through a (possibly empty) sequence of conversions.
+   */
+  private predicate irTempOperandConversionFlows(Operand op) {
+    isIRTempVariable(op)
+    or
+    exists(Operand mid |
+      irTempOperandConversionFlows(mid) and
+      isConversionStep(mid, op)
+    )
+  }
+
+  /** Holds if `node` should be an `IndirectOperand` that maps `node.asExpr()` to `e`. */
+  private predicate exprNodeShouldBeIndirectOperand(IndirectOperand node, Expr e, int n) {
+    exists(ArgumentOperand operand |
+      // When an argument (qualifier or positional) is a prvalue and the
+      // parameter (qualifier or positional) is a (const) reference, IR
+      // construction introduces a temporary `IRVariable`. The `VariableAddress`
+      // instruction has the argument as its `getConvertedResultExpression`
+      // result. However, the instruction actually represents the _address_ of
+      // the argument. So to fix this mismatch, we have the indirection of the
+      // `VariableAddressInstruction` map to the expression.
+      isIndirectOperandOfArgument(node, operand, e, n) and
+      irTempOperandConversionFlows(operand)
+    )
+  }
+
+  private predicate exprNodeShouldBeIndirectOutNode(IndirectArgumentOutNode node, Expr e, int n) {
+    exists(CallInstruction call |
+      call.getStaticCallTarget() instanceof Constructor and
+      e = getConvertedResultExpression(call, n) and
+      call.getThisArgumentOperand() = node.getAddressOperand()
+    )
+  }
+
+  /** Holds if `node` should be an instruction node that maps `node.asExpr()` to `e`. */
+  private predicate exprNodeShouldBeInstruction(Node node, Expr e, int n) {
+    not exprNodeShouldBeOperand(_, e, n) and
+    not exprNodeShouldBeIndirectOutNode(_, e, n) and
+    not exprNodeShouldBeIndirectOperand(_, e, n) and
+    e = getConvertedResultExpression(node.asInstruction(), n)
+  }
+
+  /** Holds if `node` should be an `IndirectInstruction` that maps `node.asIndirectExpr()` to `e`. */
+  private predicate indirectExprNodeShouldBeIndirectInstruction(
+    IndirectInstruction node, Expr e, int n, int indirectionIndex
+  ) {
+    not indirectExprNodeShouldBeIndirectOperand(_, e, n, indirectionIndex) and
+    exists(Instruction instr |
+      node.hasInstructionAndIndirectionIndex(instr, indirectionIndex) and
+      e = getConvertedResultExpression(instr, n)
+    )
+  }
+
+  abstract private class ExprNodeBase extends Node {
+    /**
+     * Gets the expression corresponding to this node, if any. The returned
+     * expression may be a `Conversion`.
+     */
+    abstract Expr getConvertedExpr(int n);
+
+    /** Gets the non-conversion expression corresponding to this node, if any. */
+    final Expr getExpr(int n) { result = this.getConvertedExpr(n).getUnconverted() }
+  }
+
+  /**
+   * Holds if there exists a dataflow node whose `asExpr(n)` should evaluate
+   * to `e`.
+   */
+  private predicate exprNodeShouldBe(Expr e, int n) {
+    exprNodeShouldBeInstruction(_, e, n) or
+    exprNodeShouldBeOperand(_, e, n) or
+    exprNodeShouldBeIndirectOutNode(_, e, n) or
+    exprNodeShouldBeIndirectOperand(_, e, n)
+  }
+
+  private class InstructionExprNode extends ExprNodeBase, InstructionNode {
+    InstructionExprNode() {
+      exists(Expr e, int n |
+        exprNodeShouldBeInstruction(this, e, n) and
+        not exists(Expr conv |
+          exprNodeShouldBe(conv, n + 1) and
+          conv.getUnconverted() = e.getUnconverted()
+        )
+      )
+    }
+
+    final override Expr getConvertedExpr(int n) { exprNodeShouldBeInstruction(this, result, n) }
+  }
+
+  private class OperandExprNode extends ExprNodeBase, OperandNode {
+    OperandExprNode() {
+      exists(Expr e, int n |
+        exprNodeShouldBeOperand(this, e, n) and
+        not exists(Expr conv |
+          exprNodeShouldBe(conv, n + 1) and
+          conv.getUnconverted() = e.getUnconverted()
+        )
+      )
+    }
+
+    final override Expr getConvertedExpr(int n) { exprNodeShouldBeOperand(this, result, n) }
+  }
+
+  abstract private class IndirectExprNodeBase extends Node {
+    /**
+     * Gets the expression corresponding to this node, if any. The returned
+     * expression may be a `Conversion`.
+     */
+    abstract Expr getConvertedExpr(int n, int indirectionIndex);
+
+    /** Gets the non-conversion expression corresponding to this node, if any. */
+    final Expr getExpr(int n, int indirectionIndex) {
+      result = this.getConvertedExpr(n, indirectionIndex).getUnconverted()
+    }
+  }
+
+  /** A signature for converting an indirect node to an expression. */
+  private signature module IndirectNodeToIndirectExprSig {
+    /** The indirect node class to be converted to an expression */
+    class IndirectNode;
+
+    /**
+     * Holds if the indirect expression at indirection index `indirectionIndex`
+     * of `node` is `e`. The integer `n` specifies how many conversions has been
+     * applied to `node`.
+     */
+    predicate indirectNodeHasIndirectExpr(IndirectNode node, Expr e, int n, int indirectionIndex);
+  }
+
+  /**
+   * A module that implements the logic for deciding whether an indirect node
+   * should be an `IndirectExprNode`.
+   */
+  private module IndirectNodeToIndirectExpr<IndirectNodeToIndirectExprSig Sig> {
+    import Sig
+
+    /**
+     * This predicate shifts the indirection index by one when `conv` is a
+     * `ReferenceDereferenceExpr`.
+     *
+     * This is necessary because `ReferenceDereferenceExpr` is a conversion
+     * in the AST, but appears as a `LoadInstruction` in the IR.
+     */
+    bindingset[e, indirectionIndex]
+    private predicate adjustForReference(
+      Expr e, int indirectionIndex, Expr conv, int adjustedIndirectionIndex
+    ) {
+      conv.(ReferenceDereferenceExpr).getExpr() = e and
+      adjustedIndirectionIndex = indirectionIndex - 1
+      or
+      not conv instanceof ReferenceDereferenceExpr and
+      conv = e and
+      adjustedIndirectionIndex = indirectionIndex
+    }
+
+    /** Holds if `node` should be an `IndirectExprNode`. */
+    predicate charpred(IndirectNode node) {
+      exists(Expr e, int n, int indirectionIndex |
+        indirectNodeHasIndirectExpr(node, e, n, indirectionIndex) and
+        not exists(Expr conv, int adjustedIndirectionIndex |
+          adjustForReference(e, indirectionIndex, conv, adjustedIndirectionIndex) and
+          indirectExprNodeShouldBe(conv, n + 1, adjustedIndirectionIndex)
+        )
+      )
+    }
+  }
+
+  private predicate indirectExprNodeShouldBe(Expr e, int n, int indirectionIndex) {
+    indirectExprNodeShouldBeIndirectOperand(_, e, n, indirectionIndex) or
+    indirectExprNodeShouldBeIndirectInstruction(_, e, n, indirectionIndex)
+  }
+
+  private module IndirectOperandIndirectExprNodeImpl implements IndirectNodeToIndirectExprSig {
+    class IndirectNode = IndirectOperand;
+
+    predicate indirectNodeHasIndirectExpr = indirectExprNodeShouldBeIndirectOperand/4;
+  }
+
+  module IndirectOperandToIndirectExpr =
+    IndirectNodeToIndirectExpr<IndirectOperandIndirectExprNodeImpl>;
+
+  private class IndirectOperandIndirectExprNode extends IndirectExprNodeBase instanceof IndirectOperand
+  {
+    IndirectOperandIndirectExprNode() { IndirectOperandToIndirectExpr::charpred(this) }
+
+    final override Expr getConvertedExpr(int n, int index) {
+      IndirectOperandToIndirectExpr::indirectNodeHasIndirectExpr(this, result, n, index)
+    }
+  }
+
+  private module IndirectInstructionIndirectExprNodeImpl implements IndirectNodeToIndirectExprSig {
+    class IndirectNode = IndirectInstruction;
+
+    predicate indirectNodeHasIndirectExpr = indirectExprNodeShouldBeIndirectInstruction/4;
+  }
+
+  module IndirectInstructionToIndirectExpr =
+    IndirectNodeToIndirectExpr<IndirectInstructionIndirectExprNodeImpl>;
+
+  private class IndirectInstructionIndirectExprNode extends IndirectExprNodeBase instanceof IndirectInstruction
+  {
+    IndirectInstructionIndirectExprNode() { IndirectInstructionToIndirectExpr::charpred(this) }
+
+    final override Expr getConvertedExpr(int n, int index) {
+      IndirectInstructionToIndirectExpr::indirectNodeHasIndirectExpr(this, result, n, index)
+    }
+  }
+
+  private class IndirectArgumentOutExprNode extends ExprNodeBase, IndirectArgumentOutNode {
+    IndirectArgumentOutExprNode() { exprNodeShouldBeIndirectOutNode(this, _, _) }
+
+    final override Expr getConvertedExpr(int n) { exprNodeShouldBeIndirectOutNode(this, result, n) }
+  }
+
+  private class IndirectOperandExprNode extends ExprNodeBase instanceof IndirectOperand {
+    IndirectOperandExprNode() { exprNodeShouldBeIndirectOperand(this, _, _) }
+
+    final override Expr getConvertedExpr(int n) { exprNodeShouldBeIndirectOperand(this, result, n) }
+  }
+
+  /**
+   * An expression, viewed as a node in a data flow graph.
+   */
+  cached
+  class ExprNode extends Node instanceof ExprNodeBase {
+    /**
+     * INTERNAL: Do not use.
+     */
+    cached
+    Expr getExpr(int n) { result = super.getExpr(n) }
+
+    /**
+     * Gets the non-conversion expression corresponding to this node, if any. If
+     * this node strictly (in the sense of `getConvertedExpr`) corresponds to a
+     * `Conversion`, then the result is that `Conversion`'s non-`Conversion` base
+     * expression.
+     */
+    cached
+    final Expr getExpr() { result = this.getExpr(_) }
+
+    /**
+     * INTERNAL: Do not use.
+     */
+    cached
+    Expr getConvertedExpr(int n) { result = super.getConvertedExpr(n) }
+
+    /**
+     * Gets the expression corresponding to this node, if any. The returned
+     * expression may be a `Conversion`.
+     */
+    cached
+    final Expr getConvertedExpr() { result = this.getConvertedExpr(_) }
+  }
+
+  /**
+   * An indirect expression, viewed as a node in a data flow graph.
+   */
+  cached
+  class IndirectExprNode extends Node instanceof IndirectExprNodeBase {
+    /**
+     * Gets the non-conversion expression corresponding to this node, if any. If
+     * this node strictly (in the sense of `getConvertedExpr`) corresponds to a
+     * `Conversion`, then the result is that `Conversion`'s non-`Conversion` base
+     * expression.
+     */
+    cached
+    final Expr getExpr(int indirectionIndex) { result = this.getExpr(_, indirectionIndex) }
+
+    /**
+     * INTERNAL: Do not use.
+     */
+    cached
+    Expr getExpr(int n, int indirectionIndex) { result = super.getExpr(n, indirectionIndex) }
+
+    /**
+     * INTERNAL: Do not use.
+     */
+    cached
+    Expr getConvertedExpr(int n, int indirectionIndex) {
+      result = super.getConvertedExpr(n, indirectionIndex)
+    }
+
+    /**
+     * Gets the expression corresponding to this node, if any. The returned
+     * expression may be a `Conversion`.
+     */
+    cached
+    Expr getConvertedExpr(int indirectionIndex) {
+      result = this.getConvertedExpr(_, indirectionIndex)
+    }
+  }
+}
+
+import Cached

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

@@ -657,19 +657,9 @@ class GlobalDefImpl extends DefImpl, TGlobalDefImpl {
  */
 predicate adjacentDefRead(IRBlock bb1, int i1, SourceVariable sv, IRBlock bb2, int i2) {
   adjacentDefReadExt(_, sv, bb1, i1, bb2, i2)
-  or
-  exists(PhiNode phi |
-    lastRefRedefExt(_, sv, bb1, i1, phi) and
-    phi.definesAt(sv, bb2, i2, _)
-  )
 }
 
 predicate useToNode(IRBlock bb, int i, SourceVariable sv, Node nodeTo) {
-  exists(Phi phi |
-    phi.asPhi().definesAt(sv, bb, i, _) and
-    nodeTo = phi.getNode()
-  )
-  or
   exists(UseImpl use |
     use.hasIndexInBlock(bb, i, sv) and
     nodeTo = use.getNode()
@@ -723,46 +713,26 @@ predicate nodeToDefOrUse(Node node, SourceVariable sv, IRBlock bb, int i, boolea
  */
 private predicate indirectConversionFlowStep(Node nFrom, Node nTo) {
   not exists(SourceVariable sv, IRBlock bb2, int i2 |
-    nodeToDefOrUse(nTo, sv, bb2, i2, _) and
+    useToNode(bb2, i2, sv, nTo) and
     adjacentDefRead(bb2, i2, sv, _, _)
   ) and
-  (
-    exists(Operand op1, Operand op2, int indirectionIndex, Instruction instr |
-      hasOperandAndIndex(nFrom, op1, pragma[only_bind_into](indirectionIndex)) and
-      hasOperandAndIndex(nTo, op2, pragma[only_bind_into](indirectionIndex)) and
-      instr = op2.getDef() and
-      conversionFlow(op1, instr, _, _)
-    )
-    or
-    exists(Operand op1, Operand op2, int indirectionIndex, Instruction instr |
-      hasOperandAndIndex(nFrom, op1, pragma[only_bind_into](indirectionIndex)) and
-      hasOperandAndIndex(nTo, op2, indirectionIndex - 1) and
-      instr = op2.getDef() and
-      isDereference(instr, op1, _)
-    )
+  exists(Operand op1, Operand op2, int indirectionIndex, Instruction instr |
+    hasOperandAndIndex(nFrom, op1, pragma[only_bind_into](indirectionIndex)) and
+    hasOperandAndIndex(nTo, op2, pragma[only_bind_into](indirectionIndex)) and
+    instr = op2.getDef() and
+    conversionFlow(op1, instr, _, _)
   )
 }
 
 /**
- * The reason for this predicate is a bit annoying:
- * We cannot mark a `PointerArithmeticInstruction` that computes an offset based on some SSA
- * variable `x` as a use of `x` since this creates taint-flow in the following example:
- * ```c
- * int x = array[source]
- * sink(*array)
- * ```
- * This is because `source` would flow from the operand of `PointerArithmeticInstruction` to the
- * result of the instruction, and into the `IndirectOperand` that represents the value of `*array`.
- * Then, via use-use flow, flow will arrive at `*array` in `sink(*array)`.
- *
- * So this predicate recurses back along conversions and `PointerArithmeticInstruction`s to find the
- * first use that has provides use-use flow, and uses that target as the target of the `nodeFrom`.
+ * Holds if `node` is a phi input node that should receive flow from the
+ * definition to (or use of) `sv` at `(bb1, i1)`.
  */
-private predicate adjustForPointerArith(PostUpdateNode pun, SourceVariable sv, IRBlock bb2, int i2) {
-  exists(IRBlock bb1, int i1, Node adjusted |
-    indirectConversionFlowStep*(adjusted, pun.getPreUpdateNode()) and
-    nodeToDefOrUse(adjusted, sv, bb1, i1, _) and
-    adjacentDefRead(bb1, i1, sv, bb2, i2)
+private predicate phiToNode(SsaPhiInputNode node, SourceVariable sv, IRBlock bb1, int i1) {
+  exists(PhiNode phi, IRBlock input |
+    phi.hasInputFromBlock(_, sv, bb1, i1, input) and
+    node.getPhiNode() = phi and
+    node.getBlock() = input
   )
 }
 
@@ -777,10 +747,14 @@ private predicate adjustForPointerArith(PostUpdateNode pun, SourceVariable sv, I
 private predicate ssaFlowImpl(
   IRBlock bb1, int i1, SourceVariable sv, Node nodeFrom, Node nodeTo, boolean uncertain
 ) {
-  exists(IRBlock bb2, int i2 |
-    nodeToDefOrUse(nodeFrom, sv, bb1, i1, uncertain) and
-    adjacentDefRead(bb1, i1, sv, bb2, i2) and
-    useToNode(bb2, i2, sv, nodeTo)
+  nodeToDefOrUse(nodeFrom, sv, bb1, i1, uncertain) and
+  (
+    exists(IRBlock bb2, int i2 |
+      adjacentDefRead(bb1, i1, sv, bb2, i2) and
+      useToNode(bb2, i2, sv, nodeTo)
+    )
+    or
+    phiToNode(nodeTo, sv, bb1, i1)
   ) and
   nodeFrom != nodeTo
 }
@@ -789,7 +763,7 @@ private predicate ssaFlowImpl(
 private Node getAPriorDefinition(DefinitionExt next) {
   exists(IRBlock bb, int i, SourceVariable sv |
     lastRefRedefExt(_, pragma[only_bind_into](sv), pragma[only_bind_into](bb),
-      pragma[only_bind_into](i), next) and
+      pragma[only_bind_into](i), _, next) and
     nodeToDefOrUse(result, sv, bb, i, _)
   )
 }
@@ -896,9 +870,31 @@ private predicate isArgumentOfCallable(DataFlowCall call, Node n) {
  * Holds if there is use-use flow from `pun`'s pre-update node to `n`.
  */
 private predicate postUpdateNodeToFirstUse(PostUpdateNode pun, Node n) {
-  exists(SourceVariable sv, IRBlock bb2, int i2 |
-    adjustForPointerArith(pun, sv, bb2, i2) and
-    useToNode(bb2, i2, sv, n)
+  // We cannot mark a `PointerArithmeticInstruction` that computes an offset
+  // based on some SSA
+  // variable `x` as a use of `x` since this creates taint-flow in the
+  // following example:
+  // ```c
+  // int x = array[source]
+  // sink(*array)
+  // ```
+  // This is because `source` would flow from the operand of `PointerArithmetic`
+  // instruction to the result of the instruction, and into the `IndirectOperand`
+  // that represents the value of `*array`. Then, via use-use flow, flow will
+  // arrive at `*array` in `sink(*array)`.
+  // So this predicate recurses back along conversions and `PointerArithmetic`
+  // instructions to find the first use that has provides use-use flow, and
+  // uses that target as the target of the `nodeFrom`.
+  exists(Node adjusted, IRBlock bb1, int i1, SourceVariable sv |
+    indirectConversionFlowStep*(adjusted, pun.getPreUpdateNode()) and
+    useToNode(bb1, i1, sv, adjusted)
+  |
+    exists(IRBlock bb2, int i2 |
+      adjacentDefRead(bb1, i1, sv, bb2, i2) and
+      useToNode(bb2, i2, sv, n)
+    )
+    or
+    phiToNode(n, sv, bb1, i1)
   )
 }
 
@@ -953,11 +949,16 @@ predicate postUpdateFlow(PostUpdateNode pun, Node nodeTo) {
 
 /** Holds if `nodeTo` receives flow from the phi node `nodeFrom`. */
 predicate fromPhiNode(SsaPhiNode nodeFrom, Node nodeTo) {
-  exists(PhiNode phi, SourceVariable sv, IRBlock bb1, int i1, IRBlock bb2, int i2 |
+  exists(PhiNode phi, SourceVariable sv, IRBlock bb1, int i1 |
     phi = nodeFrom.getPhiNode() and
-    phi.definesAt(sv, bb1, i1, _) and
-    adjacentDefRead(bb1, i1, sv, bb2, i2) and
-    useToNode(bb2, i2, sv, nodeTo)
+    phi.definesAt(sv, bb1, i1, _)
+  |
+    exists(IRBlock bb2, int i2 |
+      adjacentDefRead(bb1, i1, sv, bb2, i2) and
+      useToNode(bb2, i2, sv, nodeTo)
+    )
+    or
+    phiToNode(nodeTo, sv, bb1, i1)
   )
 }
 
@@ -980,7 +981,7 @@ private module SsaInput implements SsaImplCommon::InputSig<Location> {
    * Holds if the `i`'th write in block `bb` writes to the variable `v`.
    * `certain` is `true` if the write is guaranteed to overwrite the entire variable.
    */
-  predicate variableWrite(IRBlock bb, int i, SourceVariable v, boolean certain) {
+  predicate variableWrite(BasicBlock bb, int i, SourceVariable v, boolean certain) {
     DataFlowImplCommon::forceCachingInSameStage() and
     (
       exists(DefImpl def | def.hasIndexInBlock(bb, i, v) |
@@ -998,7 +999,7 @@ private module SsaInput implements SsaImplCommon::InputSig<Location> {
    * Holds if the `i`'th read in block `bb` reads to the variable `v`.
    * `certain` is `true` if the read is guaranteed. For C++, this is always the case.
    */
-  predicate variableRead(IRBlock bb, int i, SourceVariable v, boolean certain) {
+  predicate variableRead(BasicBlock bb, int i, SourceVariable v, boolean certain) {
     exists(UseImpl use | use.hasIndexInBlock(bb, i, v) |
       if use.isCertain() then certain = true else certain = false
     )
@@ -1031,22 +1032,26 @@ module SsaCached {
    * Holds if the node at index `i` in `bb` is a last reference to SSA definition
    * `def`. The reference is last because it can reach another write `next`,
    * without passing through another read or write.
+   *
+   * The path from node `i` in `bb` to `next` goes via basic block `input`,
+   * which is either a predecessor of the basic block of `next`, or `input` =
+   * `bb` in case `next` occurs in basic block `bb`.
    */
   cached
   predicate lastRefRedefExt(
-    DefinitionExt def, SourceVariable sv, IRBlock bb, int i, DefinitionExt next
+    DefinitionExt def, SourceVariable sv, IRBlock bb, int i, IRBlock input, DefinitionExt next
   ) {
-    SsaImpl::lastRefRedefExt(def, sv, bb, i, next)
+    SsaImpl::lastRefRedefExt(def, sv, bb, i, input, next)
   }
 
   cached
-  Definition phiHasInputFromBlock(PhiNode phi, IRBlock bb) {
-    SsaImpl::phiHasInputFromBlock(phi, result, bb)
+  Definition phiHasInputFromBlockExt(PhiNode phi, IRBlock bb) {
+    SsaImpl::phiHasInputFromBlockExt(phi, result, bb)
   }
 
   cached
-  predicate ssaDefReachesRead(SourceVariable v, Definition def, IRBlock bb, int i) {
-    SsaImpl::ssaDefReachesRead(v, def, bb, i)
+  predicate ssaDefReachesReadExt(SourceVariable v, DefinitionExt def, IRBlock bb, int i) {
+    SsaImpl::ssaDefReachesReadExt(v, def, bb, i)
   }
 
   predicate variableRead = SsaInput::variableRead/4;
@@ -1198,11 +1203,11 @@ class Phi extends TPhi, SsaDef {
 
   final override Location getLocation() { result = phi.getBasicBlock().getLocation() }
 
-  override string toString() { result = "Phi" }
+  override string toString() { result = phi.toString() }
 
-  SsaPhiNode getNode() { result.getPhiNode() = phi }
+  SsaPhiInputNode getNode(IRBlock block) { result.getPhiNode() = phi and result.getBlock() = block }
 
-  predicate hasInputFromBlock(Definition inp, IRBlock bb) { inp = phiHasInputFromBlock(phi, bb) }
+  predicate hasInputFromBlock(Definition inp, IRBlock bb) { inp = phiHasInputFromBlockExt(phi, bb) }
 
   final Definition getAnInput() { this.hasInputFromBlock(result, _) }
 }
@@ -1228,13 +1233,21 @@ class PhiNode extends SsaImpl::DefinitionExt {
    */
   predicate isPhiRead() { this instanceof SsaImpl::PhiReadNode }
 
-  /** Holds if `inp` is an input to this phi node along the edge originating in `bb`. */
-  predicate hasInputFromBlock(Definition inp, IRBlock bb) {
-    inp = SsaCached::phiHasInputFromBlock(this, bb)
+  /**
+   * Holds if the node at index `i` in `bb` is a last reference to SSA
+   * definition `def` of `sv`. The reference is last because it can reach
+   * this phi node, without passing through another read or write.
+   *
+   * The path from node `i` in `bb` to this phi node goes via basic block
+   * `input`, which is either a predecessor of the basic block of this phi
+   * node, or `input` = `bb` in case this phi node occurs in basic block `bb`.
+   */
+  predicate hasInputFromBlock(DefinitionExt def, SourceVariable sv, IRBlock bb, int i, IRBlock input) {
+    SsaCached::lastRefRedefExt(def, sv, bb, i, input, this)
   }
 
   /** Gets a definition that is an input to this phi node. */
-  final Definition getAnInput() { this.hasInputFromBlock(result, _) }
+  final Definition getAnInput() { this.hasInputFromBlock(result, _, _, _, _) }
 }
 
 /** An static single assignment (SSA) definition. */
@@ -1249,6 +1262,15 @@ class DefinitionExt extends SsaImpl::DefinitionExt {
     result = this.getAPhiInputOrPriorDefinition*() and
     not result instanceof PhiNode
   }
+
+  /** Gets a node that represents a read of this SSA definition. */
+  Node getARead() {
+    exists(SourceVariable sv, IRBlock bb, int i | SsaCached::ssaDefReachesReadExt(sv, this, bb, i) |
+      useToNode(bb, i, sv, result)
+      or
+      phiToNode(result, sv, bb, i)
+    )
+  }
 }
 
 class Definition = SsaImpl::Definition;

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

@@ -757,13 +757,19 @@ import Cached
  * between the SSA pruning stage, and the final SSA stage.
  */
 module InputSigCommon {
-  class BasicBlock = IRBlock;
+  class BasicBlock extends IRBlock {
+    ControlFlowNode getNode(int i) { result = this.getInstruction(i) }
+
+    int length() { result = this.getInstructionCount() }
+  }
+
+  class ControlFlowNode = Instruction;
 
   BasicBlock getImmediateBasicBlockDominator(BasicBlock bb) { result.immediatelyDominates(bb) }
 
   BasicBlock getABasicBlockSuccessor(BasicBlock bb) { result = bb.getASuccessor() }
 
-  class ExitBasicBlock extends IRBlock {
+  class ExitBasicBlock extends BasicBlock {
     ExitBasicBlock() { this.getLastInstruction() instanceof ExitFunctionInstruction }
   }
 }

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

@@ -147,7 +147,10 @@ predicate defaultAdditionalTaintStep(DataFlow::Node src, DataFlow::Node sink, st
  * of `c` at sinks and inputs to additional taint steps.
  */
 bindingset[node]
-predicate defaultImplicitTaintRead(DataFlow::Node node, DataFlow::ContentSet c) { none() }
+predicate defaultImplicitTaintRead(DataFlow::Node node, DataFlow::ContentSet c) {
+  node instanceof ArgumentNode and
+  c.isSingleton(any(ElementContent ec))
+}
 
 /**
  * Holds if `node` should be a sanitizer in all global taint flow configurations

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

@@ -1,3 +1,3 @@
-import semmle.code.cpp.ir.implementation.aliased_ssa.IR
+import semmle.code.cpp.ir.implementation.raw.IR
 import semmle.code.cpp.ir.internal.Overlap
 import semmle.code.cpp.ir.internal.IRCppLanguage as Language

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

@@ -3208,9 +3208,20 @@ class TranslatedBuiltInOperation extends TranslatedNonConstantExpr {
 
   final override Instruction getResult() { result = this.getInstruction(OnlyInstructionTag()) }
 
+  /**
+   * Gets the rnk'th (0-indexed) child for which a `TranslatedElement` exists.
+   *
+   * We use this predicate to filter out `TypeName` expressions that sometimes
+   * occur in builtin operations since the IR doesn't have an instruction to
+   * represent a reference to a type.
+   */
+  private TranslatedElement getRankedChild(int rnk) {
+    result = rank[rnk + 1](int id, TranslatedElement te | te = this.getChild(id) | te order by id)
+  }
+
   final override Instruction getFirstInstruction(EdgeKind kind) {
-    if exists(this.getChild(0))
-    then result = this.getChild(0).getFirstInstruction(kind)
+    if exists(this.getRankedChild(0))
+    then result = this.getRankedChild(0).getFirstInstruction(kind)
     else (
       kind instanceof GotoEdge and result = this.getInstruction(OnlyInstructionTag())
     )
@@ -3230,11 +3241,11 @@ class TranslatedBuiltInOperation extends TranslatedNonConstantExpr {
   }
 
   final override Instruction getChildSuccessorInternal(TranslatedElement child, EdgeKind kind) {
-    exists(int id | child = this.getChild(id) |
-      result = this.getChild(id + 1).getFirstInstruction(kind)
+    exists(int id | child = this.getRankedChild(id) |
+      result = this.getRankedChild(id + 1).getFirstInstruction(kind)
       or
       kind instanceof GotoEdge and
-      not exists(this.getChild(id + 1)) and
+      not exists(this.getRankedChild(id + 1)) and
       result = this.getInstruction(OnlyInstructionTag())
     )
   }
@@ -3249,7 +3260,7 @@ class TranslatedBuiltInOperation extends TranslatedNonConstantExpr {
     tag = OnlyInstructionTag() and
     exists(int index |
       operandTag = positionalArgumentOperand(index) and
-      result = this.getChild(index).(TranslatedExpr).getResult()
+      result = this.getRankedChild(index).(TranslatedExpr).getResult()
     )
   }
 

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

@@ -1,3 +1,3 @@
-import semmle.code.cpp.ir.implementation.aliased_ssa.IR
+import semmle.code.cpp.ir.implementation.unaliased_ssa.IR
 import semmle.code.cpp.ir.internal.Overlap
 import semmle.code.cpp.ir.internal.IRCppLanguage as Language

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

@@ -7,119 +7,6 @@
 import semmle.code.cpp.models.interfaces.Allocation
 import semmle.code.cpp.models.interfaces.Taint
 
-/**
- * An allocation function (such as `malloc`) that has an argument for the size
- * in bytes.
- */
-private class MallocAllocationFunction extends AllocationFunction {
-  int sizeArg;
-
-  MallocAllocationFunction() {
-    // --- C library allocation
-    this.hasGlobalOrStdOrBslName("malloc") and // malloc(size)
-    sizeArg = 0
-    or
-    this.hasGlobalName([
-        // --- Windows Memory Management for Windows Drivers
-        "MmAllocateContiguousMemory", // MmAllocateContiguousMemory(size, maxaddress)
-        "MmAllocateContiguousNodeMemory", // MmAllocateContiguousNodeMemory(size, minaddress, maxaddress, bound, flag, prefer)
-        "MmAllocateContiguousMemorySpecifyCache", // MmAllocateContiguousMemorySpecifyCache(size, minaddress, maxaddress, bound, type)
-        "MmAllocateContiguousMemorySpecifyCacheNode", // MmAllocateContiguousMemorySpecifyCacheNode(size, minaddress, maxaddress, bound, type, prefer)
-        "MmAllocateNonCachedMemory", // MmAllocateNonCachedMemory(size)
-        "MmAllocateMappingAddress", // MmAllocateMappingAddress(size, tag)
-        // --- Windows COM allocation
-        "CoTaskMemAlloc", // CoTaskMemAlloc(size)
-        // --- Solaris/BSD kernel memory allocator
-        "kmem_alloc", // kmem_alloc(size, flags)
-        "kmem_zalloc", // kmem_zalloc(size, flags)
-        // --- OpenSSL memory allocation
-        "CRYPTO_malloc", // CRYPTO_malloc(size_t num, const char *file, int line)
-        "CRYPTO_zalloc", // CRYPTO_zalloc(size_t num, const char *file, int line)
-        "CRYPTO_secure_malloc", // CRYPTO_secure_malloc(size_t num, const char *file, int line)
-        "CRYPTO_secure_zalloc", // CRYPTO_secure_zalloc(size_t num, const char *file, int line)
-        "g_malloc", // g_malloc (n_bytes);
-        "g_try_malloc" // g_try_malloc(n_bytes);
-      ]) and
-    sizeArg = 0
-    or
-    this.hasGlobalName([
-        // --- Windows Memory Management for Windows Drivers
-        "ExAllocatePool", // ExAllocatePool(type, size)
-        "ExAllocatePool2", // ExAllocatePool2(flags, size, tag)
-        "ExAllocatePool3", // ExAllocatePool3(flags, size, tag, extparams, extparamscount)
-        "ExAllocatePoolWithTag", // ExAllocatePool(type, size, tag)
-        "ExAllocatePoolWithTagPriority", // ExAllocatePoolWithTagPriority(type, size, tag, priority)
-        "ExAllocatePoolWithQuota", // ExAllocatePoolWithQuota(type, size)
-        "ExAllocatePoolWithQuotaTag", // ExAllocatePoolWithQuotaTag(type, size, tag)
-        "ExAllocatePoolZero", // ExAllocatePoolZero(type, size, tag)
-        "IoAllocateMdl", // IoAllocateMdl(address, size, flag, flag, irp)
-        "IoAllocateErrorLogEntry", // IoAllocateErrorLogEntry(object, size)
-        // --- Windows Global / Local legacy allocation
-        "LocalAlloc", // LocalAlloc(flags, size)
-        "GlobalAlloc", // GlobalAlloc(flags, size)
-        // --- Windows System Services allocation
-        "VirtualAlloc" // VirtualAlloc(address, size, type, flag)
-      ]) and
-    sizeArg = 1
-    or
-    this.hasGlobalName("HeapAlloc") and // HeapAlloc(heap, flags, size)
-    sizeArg = 2
-    or
-    this.hasGlobalName([
-        // --- Windows Memory Management for Windows Drivers
-        "MmAllocatePagesForMdl", // MmAllocatePagesForMdl(minaddress, maxaddress, skip, size)
-        "MmAllocatePagesForMdlEx", // MmAllocatePagesForMdlEx(minaddress, maxaddress, skip, size, type, flags)
-        "MmAllocateNodePagesForMdlEx" // MmAllocateNodePagesForMdlEx(minaddress, maxaddress, skip, size, type, prefer, flags)
-      ]) and
-    sizeArg = 3
-  }
-
-  override int getSizeArg() { result = sizeArg }
-}
-
-/**
- * An allocation function (such as `alloca`) that does not require a
- * corresponding free (and has an argument for the size in bytes).
- */
-private class AllocaAllocationFunction extends AllocationFunction {
-  int sizeArg;
-
-  AllocaAllocationFunction() {
-    this.hasGlobalName([
-        // --- stack allocation
-        "alloca", // // alloca(size)
-        "__builtin_alloca", // __builtin_alloca(size)
-        "_alloca", // _alloca(size)
-        "_malloca" // _malloca(size)
-      ]) and
-    sizeArg = 0
-  }
-
-  override int getSizeArg() { result = sizeArg }
-
-  override predicate requiresDealloc() { none() }
-}
-
-/**
- * An allocation function (such as `calloc`) that has an argument for the size
- * and another argument for the size of those units (in bytes).
- */
-private class CallocAllocationFunction extends AllocationFunction {
-  int sizeArg;
-  int multArg;
-
-  CallocAllocationFunction() {
-    // --- C library allocation
-    this.hasGlobalOrStdOrBslName("calloc") and // calloc(num, size)
-    sizeArg = 1 and
-    multArg = 0
-  }
-
-  override int getSizeArg() { result = sizeArg }
-
-  override int getSizeMult() { result = multArg }
-}
-
 /**
  * An allocation function (such as `realloc`) that has an argument for the size
  * in bytes, and an argument for an existing pointer that is to be reallocated.
@@ -373,6 +260,63 @@ private class NewArrayAllocationExpr extends AllocationExpr, NewArrayExpr {
   override predicate requiresDealloc() { not exists(this.getPlacementPointer()) }
 }
 
+/**
+ * Holds if `f` is an allocation function according to the
+ * extensible `allocationFunctionModel` predicate.
+ */
+private predicate isAllocationFunctionFromModel(
+  Function f, string namespace, string type, string name
+) {
+  exists(boolean subtypes | allocationFunctionModel(namespace, type, subtypes, name, _, _, _, _) |
+    if type = ""
+    then f.hasQualifiedName(namespace, "", name)
+    else
+      exists(Class c |
+        c.hasQualifiedName(namespace, type) and f.hasQualifiedName(namespace, _, name)
+      |
+        if subtypes = true
+        then f = c.getADerivedClass*().getAMemberFunction()
+        else f = c.getAMemberFunction()
+      )
+  )
+}
+
+/**
+ * An allocation function modeled via the extensible `allocationFunctionModel` predicate.
+ */
+private class AllocationFunctionFromModel extends AllocationFunction {
+  string namespace;
+  string type;
+  string name;
+
+  AllocationFunctionFromModel() { isAllocationFunctionFromModel(this, namespace, type, name) }
+
+  final override int getSizeArg() {
+    exists(string sizeArg |
+      allocationFunctionModel(namespace, type, _, name, sizeArg, _, _, _) and
+      result = sizeArg.toInt()
+    )
+  }
+
+  final override int getSizeMult() {
+    exists(string sizeMult |
+      allocationFunctionModel(namespace, type, _, name, _, sizeMult, _, _) and
+      result = sizeMult.toInt()
+    )
+  }
+
+  final override int getReallocPtrArg() {
+    exists(string reallocPtrArg |
+      allocationFunctionModel(namespace, type, _, name, _, _, reallocPtrArg, _) and
+      result = reallocPtrArg.toInt()
+    )
+  }
+
+  final override predicate requiresDealloc() {
+    allocationFunctionModel(namespace, type, _, name, _, _, _, true)
+  }
+}
+
 private module HeuristicAllocation {
   /** A class that maps an `AllocationExpr` to an `HeuristicAllocationExpr`. */
   private class HeuristicAllocationModeled extends HeuristicAllocationExpr instanceof AllocationExpr

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

@@ -7,61 +7,42 @@
 import semmle.code.cpp.models.interfaces.Deallocation
 
 /**
- * A deallocation function such as `free`.
+ * Holds if `f` is an deallocation function according to the
+ * extensible `deallocationFunctionModel` predicate.
  */
-private class StandardDeallocationFunction extends DeallocationFunction {
-  int freedArg;
+private predicate isDeallocationFunctionFromModel(
+  Function f, string namespace, string type, string name
+) {
+  exists(boolean subtypes | deallocationFunctionModel(namespace, type, subtypes, name, _) |
+    if type = ""
+    then f.hasQualifiedName(namespace, "", name)
+    else
+      exists(Class c |
+        c.hasQualifiedName(namespace, type) and f.hasQualifiedName(namespace, _, name)
+      |
+        if subtypes = true
+        then f = c.getADerivedClass*().getAMemberFunction()
+        else f = c.getAMemberFunction()
+      )
+  )
+}
 
-  StandardDeallocationFunction() {
-    this.hasGlobalOrStdOrBslName([
-        // --- C library allocation
-        "free", "realloc"
-      ]) and
-    freedArg = 0
-    or
-    this.hasGlobalName([
-        // --- OpenSSL memory deallocation
-        "CRYPTO_free", "CRYPTO_secure_free",
-        // --- glib memory deallocation
-        "g_free"
-      ]) and
-    freedArg = 0
-    or
-    this.hasGlobalOrStdName([
-        // --- Windows Memory Management for Windows Drivers
-        "ExFreePool", "ExFreePoolWithTag", "ExDeleteTimer", "IoFreeIrp", "IoFreeMdl",
-        "IoFreeErrorLogEntry", "IoFreeWorkItem", "MmFreeContiguousMemory",
-        "MmFreeContiguousMemorySpecifyCache", "MmFreeNonCachedMemory", "MmFreeMappingAddress",
-        "MmFreePagesFromMdl", "MmUnmapReservedMapping", "MmUnmapLockedPages",
-        "NdisFreeGenericObject", "NdisFreeMemory", "NdisFreeMemoryWithTag", "NdisFreeMdl",
-        "NdisFreeNetBufferListPool", "NdisFreeNetBufferPool",
-        // --- Windows Global / Local legacy allocation
-        "LocalFree", "GlobalFree", "LocalReAlloc", "GlobalReAlloc",
-        // --- Windows System Services allocation
-        "VirtualFree",
-        // --- Windows COM allocation
-        "CoTaskMemFree", "CoTaskMemRealloc",
-        // --- Windows Automation
-        "SysFreeString",
-        // --- Solaris/BSD kernel memory allocator
-        "kmem_free"
-      ]) and
-    freedArg = 0
-    or
-    this.hasGlobalOrStdName([
-        // --- Windows Memory Management for Windows Drivers
-        "ExFreeToLookasideListEx", "ExFreeToPagedLookasideList", "ExFreeToNPagedLookasideList",
-        "NdisFreeMemoryWithTagPriority", "StorPortFreeMdl", "StorPortFreePool",
-        // --- NetBSD pool manager
-        "pool_put", "pool_cache_put"
-      ]) and
-    freedArg = 1
-    or
-    this.hasGlobalOrStdName(["HeapFree", "HeapReAlloc"]) and
-    freedArg = 2
+/**
+ * A deallocation function modeled via the extensible `deallocationFunctionModel` predicate.
+ */
+private class DeallocationFunctionFromModel extends DeallocationFunction {
+  string namespace;
+  string type;
+  string name;
+
+  DeallocationFunctionFromModel() { isDeallocationFunctionFromModel(this, namespace, type, name) }
+
+  final override int getFreedArg() {
+    exists(string freedArg |
+      deallocationFunctionModel(namespace, type, _, name, freedArg) and
+      result = freedArg.toInt()
+    )
   }
-
-  override int getFreedArg() { result = freedArg }
 }
 
 /**

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

@@ -2,7 +2,7 @@
  * Provides models for C++ containers `std::array`, `std::vector`, `std::deque`, `std::list` and `std::forward_list`.
  */
 
-import semmle.code.cpp.models.interfaces.Taint
+import semmle.code.cpp.models.interfaces.FlowSource
 import semmle.code.cpp.models.interfaces.Iterator
 
 /**
@@ -55,73 +55,6 @@ private class Vector extends StdSequenceContainer {
   Vector() { this.hasQualifiedName(["std", "bsl"], "vector") }
 }
 
-/**
- * Additional model for standard container constructors that reference the
- * value type of the container (that is, the `T` in `std::vector<T>`).  For
- * example the fill constructor:
- * ```
- * std::vector<std::string> v(100, potentially_tainted_string);
- * ```
- */
-private class StdSequenceContainerConstructor extends Constructor, TaintFunction {
-  StdSequenceContainerConstructor() {
-    this.getDeclaringType() instanceof Vector or
-    this.getDeclaringType() instanceof Deque or
-    this.getDeclaringType() instanceof List or
-    this.getDeclaringType() instanceof ForwardList
-  }
-
-  /**
-   * Gets the index of a parameter to this function that is a reference to the
-   * value type of the container.
-   */
-  int getAValueTypeParameterIndex() {
-    this.getParameter(result).getUnspecifiedType().(ReferenceType).getBaseType() =
-      this.getDeclaringType().getTemplateArgument(0).(Type).getUnspecifiedType() // i.e. the `T` of this `std::vector<T>`
-  }
-
-  /**
-   * Gets the index of a parameter to this function that is an iterator.
-   */
-  int getAnIteratorParameterIndex() { this.getParameter(result).getType() instanceof Iterator }
-
-  override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
-    // taint flow from any parameter of the value type to the returned object
-    (
-      input.isParameterDeref(this.getAValueTypeParameterIndex()) or
-      input.isParameter(this.getAnIteratorParameterIndex())
-    ) and
-    (
-      output.isReturnValue() // TODO: this is only needed for AST data flow, which treats constructors as returning the new object
-      or
-      output.isQualifierObject()
-    )
-  }
-}
-
-/**
- * The standard container function `data`.
- */
-private class StdSequenceContainerData extends TaintFunction {
-  StdSequenceContainerData() {
-    this.getClassAndName("data") instanceof Array or
-    this.getClassAndName("data") instanceof Vector
-  }
-
-  override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
-    // flow from container itself (qualifier) to return value
-    input.isQualifierObject() and
-    output.isReturnValueDeref()
-    or
-    // reverse flow from returned reference to the qualifier (for writes to
-    // `data`)
-    input.isReturnValueDeref() and
-    output.isQualifierObject()
-  }
-
-  override predicate isPartialWrite(FunctionOutput output) { output.isQualifierObject() }
-}
-
 /**
  * The standard container functions `push_back` and `push_front`.
  */
@@ -143,35 +76,6 @@ class StdSequenceContainerPush extends MemberFunction {
   }
 }
 
-private class StdSequenceContainerPushModel extends StdSequenceContainerPush, TaintFunction {
-  override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
-    // flow from parameter to qualifier
-    input.isParameterDeref(0) and
-    output.isQualifierObject()
-  }
-
-  override predicate isPartialWrite(FunctionOutput output) { output.isQualifierObject() }
-}
-
-/**
- * The standard container functions `front` and `back`.
- */
-private class StdSequenceContainerFrontBack extends TaintFunction {
-  StdSequenceContainerFrontBack() {
-    this.getClassAndName(["front", "back"]) instanceof Array or
-    this.getClassAndName(["front", "back"]) instanceof Deque or
-    this.getClassAndName("front") instanceof ForwardList or
-    this.getClassAndName(["front", "back"]) instanceof List or
-    this.getClassAndName(["front", "back"]) instanceof Vector
-  }
-
-  override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
-    // flow from object to returned reference
-    input.isQualifierObject() and
-    output.isReturnValueDeref()
-  }
-}
-
 /**
  * The standard container functions `insert` and `insert_after`.
  */
@@ -198,58 +102,6 @@ class StdSequenceContainerInsert extends MemberFunction {
   int getAnIteratorParameterIndex() { this.getParameter(result).getType() instanceof Iterator }
 }
 
-private class StdSequenceContainerInsertModel extends StdSequenceContainerInsert, TaintFunction {
-  override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
-    // flow from parameter to container itself (qualifier) and return value
-    (
-      input.isQualifierObject() or
-      input.isParameterDeref(this.getAValueTypeParameterIndex()) or
-      input.isParameter(this.getAnIteratorParameterIndex())
-    ) and
-    (
-      output.isQualifierObject() or
-      output.isReturnValue()
-    )
-  }
-
-  override predicate isPartialWrite(FunctionOutput output) { output.isQualifierObject() }
-}
-
-/**
- * The standard container function `assign`.
- */
-private class StdSequenceContainerAssign extends TaintFunction {
-  StdSequenceContainerAssign() {
-    this.getClassAndName("assign") instanceof Deque or
-    this.getClassAndName("assign") instanceof ForwardList or
-    this.getClassAndName("assign") instanceof List or
-    this.getClassAndName("assign") instanceof Vector
-  }
-
-  /**
-   * Gets the index of a parameter to this function that is a reference to the
-   * value type of the container.
-   */
-  int getAValueTypeParameterIndex() {
-    this.getParameter(result).getUnspecifiedType().(ReferenceType).getBaseType() =
-      this.getDeclaringType().getTemplateArgument(0).(Type).getUnspecifiedType() // i.e. the `T` of this `std::vector<T>`
-  }
-
-  /**
-   * Gets the index of a parameter to this function that is an iterator.
-   */
-  int getAnIteratorParameterIndex() { this.getParameter(result).getType() instanceof Iterator }
-
-  override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
-    // flow from parameter to container itself (qualifier)
-    (
-      input.isParameterDeref(this.getAValueTypeParameterIndex()) or
-      input.isParameter(this.getAnIteratorParameterIndex())
-    ) and
-    output.isQualifierObject()
-  }
-}
-
 /**
  * The standard container functions `at` and `operator[]`.
  */
@@ -261,20 +113,6 @@ class StdSequenceContainerAt extends MemberFunction {
   }
 }
 
-private class StdSequenceContainerAtModel extends StdSequenceContainerAt, TaintFunction {
-  override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
-    // flow from qualifier to referenced return value
-    input.isQualifierObject() and
-    output.isReturnValueDeref()
-    or
-    // reverse flow from returned reference to the qualifier
-    input.isReturnValueDeref() and
-    output.isQualifierObject()
-  }
-
-  override predicate isPartialWrite(FunctionOutput output) { output.isQualifierObject() }
-}
-
 /**
  * The standard `emplace` function.
  */
@@ -297,20 +135,6 @@ class StdSequenceEmplace extends MemberFunction {
   }
 }
 
-private class StdSequenceEmplaceModel extends StdSequenceEmplace, TaintFunction {
-  override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
-    // flow from any parameter except the position iterator to qualifier and return value
-    // (here we assume taint flow from any constructor parameter to the constructed object)
-    input.isParameterDeref([1 .. this.getNumberOfParameters() - 1]) and
-    (
-      output.isQualifierObject() or
-      output.isReturnValue()
-    )
-  }
-
-  override predicate isPartialWrite(FunctionOutput output) { output.isQualifierObject() }
-}
-
 /**
  * The standard vector `emplace` function.
  */
@@ -340,17 +164,6 @@ class StdSequenceEmplaceBack extends MemberFunction {
   }
 }
 
-private class StdSequenceEmplaceBackModel extends StdSequenceEmplaceBack, TaintFunction {
-  override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
-    // flow from any parameter to qualifier
-    // (here we assume taint flow from any constructor parameter to the constructed object)
-    input.isParameterDeref([0 .. this.getNumberOfParameters() - 1]) and
-    output.isQualifierObject()
-  }
-
-  override predicate isPartialWrite(FunctionOutput output) { output.isQualifierObject() }
-}
-
 /**
  * The standard vector `emplace_back` function.
  */

cpp/ql/lib/semmle/code/cpp/models/interfaces/Allocation.qll

@@ -89,6 +89,14 @@ abstract class AllocationFunction extends Function {
   predicate requiresDealloc() { any() }
 }
 
+/**
+ * Holds if an external allocation model exists for the given parameters.
+ */
+extensible predicate allocationFunctionModel(
+  string namespace, string type, boolean subtypes, string name, string sizeArg, string multArg,
+  string reallocPtrArg, boolean requiresDealloc
+);
+
 /**
  * An `operator new` or `operator new[]` function that may be associated with
  * `new` or `new[]` expressions.  Note that `new` and `new[]` are not function

cpp/ql/lib/semmle/code/cpp/models/interfaces/Deallocation.qll

@@ -34,6 +34,13 @@ abstract class DeallocationFunction extends Function {
   int getFreedArg() { none() }
 }
 
+/**
+ * Holds if an external deallocation model exists for the given parameters.
+ */
+extensible predicate deallocationFunctionModel(
+  string namespace, string type, boolean subtypes, string name, string freedArg
+);
+
 /**
  * An `operator delete` or `operator delete[]` function that may be associated
  * with `delete` or `delete[]` expressions.  Note that `delete` and `delete[]`

cpp/ql/lib/semmle/code/cpp/security/flowafterfree/FlowAfterFree.qll

@@ -95,7 +95,7 @@ module FlowFromFree<FlowFromFreeParamSig P> {
         e = any(StoreInstruction store).getDestinationAddress().getUnconvertedResultExpression()
       )
       or
-      n.asExpr() instanceof ArrayExpr
+      [n.asExpr(), n.asIndirectExpr()] instanceof ArrayExpr
     }
   }
 

cpp/ql/src/CHANGELOG.md

@@ -1,3 +1,24 @@
+## 1.0.2
+
+No user-facing changes.
+
+## 1.0.1
+
+### Minor Analysis Improvements
+
+* The `cpp/dangerous-function-overflow` no longer produces a false positive alert when the `gets` function does not have exactly one parameter.
+
+## 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

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

@@ -209,6 +209,7 @@ class LoopWithAlloca extends Stmt {
       DataFlow::localFlow(result, DataFlow::exprNode(va)) and
       // Phi nodes will be preceded by nodes that represent actual definitions
       not result instanceof DataFlow::SsaPhiNode and
+      not result instanceof DataFlow::SsaPhiInputNode and
       // A source is outside the loop if it's not inside the loop
       not exists(Expr e | e = getExpr(result) | this = getAnEnclosingLoopOfExpr(e))
     )

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 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).
+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.
 </p>
 
 <p>

cpp/ql/src/Security/CWE/CWE-570/IncorrectAllocationErrorHandling.ql

@@ -215,13 +215,18 @@ predicate noThrowInTryBlock(NewOrNewArrayExpr newExpr, BadAllocCatchBlock catchB
  */
 predicate nullCheckInThrowingNew(NewOrNewArrayExpr newExpr, GuardCondition guard) {
   newExpr.getAllocator() instanceof ThrowingAllocator and
-  (
-    // Handles null comparisons.
-    guard.ensuresEq(globalValueNumber(newExpr).getAnExpr(), any(NullValue null), _, _, _)
-    or
-    // Handles `if(ptr)` and `if(!ptr)` cases.
-    guard = globalValueNumber(newExpr).getAnExpr()
-  )
+  // There can be many guard conditions that compares `newExpr` againgst 0.
+  // For example, for `if(!p)` both `p` and `!p` are guard conditions. To not
+  // produce duplicates results we pick the "first" guard condition according
+  // to some arbitrary ordering (i.e., location information). This means `!p` is the
+  // element that we use to construct the alert.
+  guard =
+    min(GuardCondition gc, int startline, int startcolumn, int endline, int endcolumn |
+      gc.comparesEq(globalValueNumber(newExpr).getAnExpr(), 0, _, _) and
+      gc.getLocation().hasLocationInfo(_, startline, startcolumn, endline, endcolumn)
+    |
+      gc order by startline, startcolumn, endline, endcolumn
+    )
 }
 
 from NewOrNewArrayExpr newExpr, Element element, string msg, string elementString

cpp/ql/src/Security/CWE/CWE-676/DangerousFunctionOverflow.ql

@@ -17,5 +17,6 @@ import cpp
 from FunctionCall call, Function target
 where
   call.getTarget() = target and
-  target.hasGlobalOrStdName("gets")
+  target.hasGlobalOrStdName("gets") and
+  target.getNumberOfParameters() = 1
 select call, "'gets' does not guard against buffer overflow."

cpp/ql/src/change-notes/2024-05-19-avoid-reporting-static-variable.md

@@ -1,4 +0,0 @@
----
-category: minorAnalysis
----
-* The "Variable not initialized before use" query (`cpp/not-initialised`) no longer reports an alert on static variables.

cpp/ql/src/change-notes/2024-05-22-use-of-unique-pointer-after-lifetime-ends-fp.md

@@ -1,4 +0,0 @@
----
-category: minorAnalysis
----
-* 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

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

@@ -0,0 +1,10 @@
+## 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.

cpp/ql/src/change-notes/released/1.0.1.md

@@ -0,0 +1,5 @@
+## 1.0.1
+
+### Minor Analysis Improvements
+
+* The `cpp/dangerous-function-overflow` no longer produces a false positive alert when the `gets` function does not have exactly one parameter.

cpp/ql/src/change-notes/released/1.0.2.md

@@ -0,0 +1,3 @@
+## 1.0.2
+
+No user-facing changes.

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

@@ -1,2 +1,2 @@
 ---
-lastReleaseVersion: 0.9.12
+lastReleaseVersion: 1.0.2

cpp/ql/src/qlpack.yml

@@ -1,5 +1,5 @@
 name: codeql/cpp-queries
-version: 1.0.0-dev
+version: 1.0.3-dev
 groups:
   - cpp
   - queries

cpp/ql/test/experimental/query-tests/Security/CWE/CWE-193/constant-size/ConstantSizeArrayOffByOne.expected

@@ -61,7 +61,6 @@ edges
 | test.cpp:289:19:289:25 | buffer3 | test.cpp:277:35:277:35 | p | provenance |  |
 | test.cpp:289:19:289:25 | buffer3 | test.cpp:289:19:289:25 | buffer3 | provenance |  |
 | test.cpp:292:25:292:27 | arr | test.cpp:299:16:299:21 | access to array | provenance | Config |
-| test.cpp:292:25:292:27 | arr | test.cpp:299:16:299:21 | access to array | provenance | Config |
 | test.cpp:306:20:306:23 | arr1 | test.cpp:292:25:292:27 | arr | provenance |  |
 | test.cpp:306:20:306:23 | arr1 | test.cpp:306:20:306:23 | arr1 | provenance |  |
 | test.cpp:309:20:309:23 | arr2 | test.cpp:292:25:292:27 | arr | provenance |  |
@@ -159,7 +158,6 @@ nodes
 | test.cpp:289:19:289:25 | buffer3 | semmle.label | buffer3 |
 | test.cpp:289:19:289:25 | buffer3 | semmle.label | buffer3 |
 | test.cpp:292:25:292:27 | arr | semmle.label | arr |
-| test.cpp:292:25:292:27 | arr | semmle.label | arr |
 | test.cpp:299:16:299:21 | access to array | semmle.label | access to array |
 | test.cpp:306:20:306:23 | arr1 | semmle.label | arr1 |
 | test.cpp:306:20:306:23 | arr1 | semmle.label | arr1 |

cpp/ql/test/header-variant-tests/iquote/iquote.c

@@ -3,4 +3,4 @@
 #include "b.h"
 static int has_angle_b = __has_include(<b.h>);
 
-// semmle-extractor-options: -I${testdir}/dir2 -iquote ${testdir}/dir1 --edg --clang
+// semmle-extractor-options: -I${testdir}/dir2 -iquote ${testdir}/dir1 --clang

cpp/ql/test/library-tests/CPP-207/options

@@ -1 +1 @@
-semmle-extractor-options: --edg --microsoft
+semmle-extractor-options: --microsoft

cpp/ql/test/library-tests/arguments/arguments.expected

@@ -1,22 +1,21 @@
 | arguments.c | 1 | --preprocessArgs |
-| arguments.c | 2 | --force-recompute |
-| arguments.c | 3 | --edg |
-| arguments.c | 4 | --disable_system_macros |
-| arguments.c | 5 | --edg |
-| arguments.c | 6 | --codeql-verbosity |
-| arguments.c | 7 | --edg |
-| arguments.c | 8 | 2 |
-| arguments.c | 9 | --edg |
-| arguments.c | 10 | --target |
-| arguments.c | 11 | --edg |
-| arguments.c | 12 | linux_x86_64 |
-| arguments.c | 13 | --edg |
-| arguments.c | 14 | -D |
-| arguments.c | 15 | --edg |
-| arguments.c | 16 | __CODEQL_TEST__ |
-| arguments.c | 17 | --gcc |
-| arguments.c | 18 | --predefined_macros |
-| arguments.c | 19 | <tools>/qltest/predefined_macros |
-| arguments.c | 20 | -w |
-| arguments.c | 21 | -Werror |
-| arguments.c | 22 | arguments.c |
+| arguments.c | 2 | --edg |
+| arguments.c | 3 | --force-recompute |
+| arguments.c | 4 | --edg |
+| arguments.c | 5 | --disable_system_macros |
+| arguments.c | 6 | --edg |
+| arguments.c | 7 | --codeql-verbosity |
+| arguments.c | 8 | --edg |
+| arguments.c | 9 | 2 |
+| arguments.c | 10 | --edg |
+| arguments.c | 11 | --target |
+| arguments.c | 12 | --edg |
+| arguments.c | 13 | linux_x86_64 |
+| arguments.c | 14 | --edg |
+| arguments.c | 15 | -D |
+| arguments.c | 16 | --edg |
+| arguments.c | 17 | __CODEQL_TEST__ |
+| arguments.c | 18 | --gcc |
+| arguments.c | 19 | -w |
+| arguments.c | 20 | -Werror |
+| arguments.c | 21 | arguments.c |

cpp/ql/test/library-tests/arguments/arguments.ql

@@ -4,8 +4,5 @@ from Compilation c, int i, string s
 // Skip the extractor name; it'll vary depending on platform
 where
   i > 0 and
-  s =
-    c.getArgument(i)
-        .replaceAll("\\", "/")
-        .regexpReplaceAll(".*(/qltest/predefined_macros)", "<tools>$1")
+  s = c.getArgument(i).replaceAll("\\", "/")
 select c.getAFileCompiled().toString(), i, s