Merge pull request #21790 from github/redsun82/codeql-cli-2.25.3

Fix `build mode` to `build-mode` in docs

.bazelrc

@@ -11,6 +11,8 @@ build --compilation_mode opt
 common --override_module=semmle_code=%workspace%/misc/bazel/semmle_code_stub
 
 build --repo_env=CC=clang --repo_env=CXX=clang++
+# Disable Android SDK auto-detection (we don't use it, and rules_android has Bazel 9 compatibility issues)
+build --repo_env=ANDROID_HOME=
 
 # print test output, like sembuild does.
 # Set to `errors` if this is too verbose.
@@ -34,7 +36,7 @@ common --@rules_dotnet//dotnet/settings:strict_deps=false
 common --@rules_rust//rust/toolchain/channel=nightly
 
 # Reduce this eventually to empty, once we've fixed all our usages of java, and https://github.com/bazel-contrib/rules_go/issues/4193 is fixed
-common --incompatible_autoload_externally="+@rules_java,+@rules_shell"
+common --incompatible_autoload_externally="+@rules_cc,+@rules_java,+@rules_shell"
 
 build --java_language_version=17
 build --tool_java_language_version=17

.bazelversion

@@ -1 +1 @@
-8.4.2
+9.0.0

.github/dependabot.yml

@@ -45,3 +45,5 @@ updates:
     directory: "/"
     schedule:
       interval: weekly
+    exclude-paths:
+      - "misc/bazel/registry/**"

.github/workflows/compile-queries.yml

@@ -1,78 +0,0 @@
-name: "Compile all queries using the latest stable CodeQL CLI"
-
-on:
-  push:
-    branches:  # makes sure the cache gets populated - running on the branches people tend to merge into.
-      - main
-      - "rc/*"
-      - "codeql-cli-*"
-  pull_request:
-    paths:
-      - '**.ql'
-      - '**.qll'
-      - '**/qlpack.yml'
-      - '**.dbscheme'
-
-permissions:
-  contents: read
-
-jobs:
-  detect-changes:
-    if: github.repository_owner == 'github'
-    runs-on: ubuntu-latest
-    outputs:
-      languages: ${{ steps.detect.outputs.languages }}
-    steps:
-      - uses: actions/checkout@v5
-      - name: Detect changed languages
-        id: detect
-        run: |
-          if [[ "${{ github.event_name }}" == "pull_request" ]]; then
-            # For PRs, detect which languages have changes
-            changed_files=$(gh pr view ${{ github.event.pull_request.number }} --json files --jq '.files.[].path')
-            languages=()
-            for lang in actions cpp csharp go java javascript python ql ruby rust swift; do
-              if echo "$changed_files" | grep -qE "^($lang/|shared/)" ; then
-                languages+=("$lang")
-              fi
-            done
-            echo "languages=$(jq -c -n '$ARGS.positional' --args "${languages[@]}")" >> $GITHUB_OUTPUT
-          else
-            # For pushes to main/rc branches, run all languages
-            echo 'languages=["actions","cpp","csharp","go","java","javascript","python","ql","ruby","rust","swift"]' >> $GITHUB_OUTPUT
-          fi
-        env:
-          GH_TOKEN: ${{ github.token }}
-
-  compile-queries:
-    needs: detect-changes
-    if: github.repository_owner == 'github' && needs.detect-changes.outputs.languages != '[]'
-    runs-on: ubuntu-latest-xl
-    strategy:
-      fail-fast: false
-      matrix:
-        language: ${{ fromJson(needs.detect-changes.outputs.languages) }}
-
-    steps:
-      - uses: actions/checkout@v5
-      - name: Setup CodeQL
-        uses: ./.github/actions/fetch-codeql
-        with:
-          channel: 'release'
-      - name: Cache compilation cache
-        id: query-cache
-        uses: ./.github/actions/cache-query-compilation
-        with: 
-          key: ${{ matrix.language }}-queries
-      - name: check formatting
-        run: find shared ${{ matrix.language }}/ql -type f \( -name "*.qll" -o -name "*.ql" \) -print0 | xargs -0 -n 3000 -P 10 codeql query format -q --check-only
-      - name: compile queries - check-only
-        # run with --check-only if running in a PR (github.sha != main)
-        if : ${{ github.event_name == 'pull_request' }}
-        shell: bash
-        run: codeql query compile -q -j0 ${{ matrix.language }}/ql/{src,examples} --keep-going --warnings=error --check-only --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}" --compilation-cache-size=500 --ram=56000
-      - name: compile queries - full
-        # do full compile if running on main - this populates the cache
-        if : ${{ github.event_name != 'pull_request' }}
-        shell: bash
-        run: codeql query compile -q -j0 ${{ matrix.language }}/ql/{src,examples} --keep-going --warnings=error --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}" --compilation-cache-size=500 --ram=56000

.github/workflows/ruby-build.yml

@@ -1,236 +0,0 @@
-name: "Ruby: Build"
-
-on:
-  push:
-    paths:
-      - "ruby/**"
-      - .github/workflows/ruby-build.yml
-      - .github/actions/fetch-codeql/action.yml
-      - codeql-workspace.yml
-      - "shared/tree-sitter-extractor/**"
-    branches:
-      - main
-      - "rc/*"
-  pull_request:
-    paths:
-      - "ruby/**"
-      - .github/workflows/ruby-build.yml
-      - .github/actions/fetch-codeql/action.yml
-      - codeql-workspace.yml
-      - "shared/tree-sitter-extractor/**"
-    branches:
-      - main
-      - "rc/*"
-  workflow_dispatch:
-    inputs:
-      tag:
-        description: "Version tag to create"
-        required: false
-
-env:
-  CARGO_TERM_COLOR: always
-
-defaults:
-  run:
-    working-directory: ruby
-
-permissions:
-  contents: read
-
-jobs:
-  build:
-    strategy:
-      fail-fast: false
-      matrix:
-        os: [ubuntu-latest, macos-latest, windows-latest]
-
-    runs-on: ${{ matrix.os }}
-
-    steps:
-      - uses: actions/checkout@v5
-      - name: Install GNU tar
-        if: runner.os == 'macOS'
-        run: |
-          brew install gnu-tar
-          echo "/usr/local/opt/gnu-tar/libexec/gnubin" >> $GITHUB_PATH
-      - name: Prepare Windows
-        if: runner.os == 'Windows'
-        shell: powershell
-        run: |
-          git config --global core.longpaths true
-      - uses: ./.github/actions/os-version
-        id: os_version
-      - name: Cache entire extractor
-        uses: actions/cache@v3
-        id: cache-extractor
-        with:
-          path: |
-            target/release/codeql-extractor-ruby
-            target/release/codeql-extractor-ruby.exe
-            ruby/extractor/ql/lib/codeql/ruby/ast/internal/TreeSitter.qll
-          key: ${{ runner.os }}-${{ steps.os_version.outputs.version }}-ruby-extractor-${{ hashFiles('ruby/extractor/rust-toolchain.toml', 'ruby/extractor/Cargo.lock') }}-${{ hashFiles('shared/tree-sitter-extractor') }}-${{ hashFiles('ruby/extractor/**/*.rs') }}
-      - uses: actions/cache@v3
-        if: steps.cache-extractor.outputs.cache-hit != 'true'
-        with:
-          path: |
-            ~/.cargo/registry
-            ~/.cargo/git
-            target
-          key: ${{ runner.os }}-${{ steps.os_version.outputs.version }}-ruby-rust-cargo-${{ hashFiles('ruby/extractor/rust-toolchain.toml', 'ruby/extractor/**/Cargo.lock') }}
-      - name: Check formatting
-        if: steps.cache-extractor.outputs.cache-hit != 'true'
-        run: cd extractor && cargo fmt -- --check
-      - name: Build
-        if: steps.cache-extractor.outputs.cache-hit != 'true'
-        run: cd extractor && cargo build --verbose
-      - name: Run tests
-        if: steps.cache-extractor.outputs.cache-hit != 'true'
-        run: cd extractor && cargo test --verbose
-      - name: Release build
-        if: steps.cache-extractor.outputs.cache-hit != 'true'
-        run: cd extractor && cargo build --release
-      - name: Generate dbscheme
-        if: ${{ matrix.os == 'ubuntu-latest' && steps.cache-extractor.outputs.cache-hit != 'true'}}
-        run: ../target/release/codeql-extractor-ruby generate --dbscheme ql/lib/ruby.dbscheme --library ql/lib/codeql/ruby/ast/internal/TreeSitter.qll
-      - uses: actions/upload-artifact@v4
-        if: ${{ matrix.os == 'ubuntu-latest' }}
-        with:
-          name: ruby.dbscheme
-          path: ruby/ql/lib/ruby.dbscheme
-      - uses: actions/upload-artifact@v4
-        if: ${{ matrix.os == 'ubuntu-latest' }}
-        with:
-          name: TreeSitter.qll
-          path: ruby/ql/lib/codeql/ruby/ast/internal/TreeSitter.qll
-      - uses: actions/upload-artifact@v4
-        with:
-          name: extractor-${{ matrix.os }}
-          path: |
-            target/release/codeql-extractor-ruby
-            target/release/codeql-extractor-ruby.exe
-          retention-days: 1
-  compile-queries:
-    if: github.repository_owner == 'github'
-    runs-on: ubuntu-latest-xl
-    steps:
-      - uses: actions/checkout@v5
-      - name: Fetch CodeQL
-        uses: ./.github/actions/fetch-codeql
-      - name: Cache compilation cache
-        id: query-cache
-        uses: ./.github/actions/cache-query-compilation
-        with:
-          key: ruby-build
-      - name: Build Query Pack
-        run: |
-          PACKS=${{ runner.temp }}/query-packs
-          rm -rf $PACKS
-          codeql pack create ../misc/suite-helpers --output "$PACKS"
-          codeql pack create ../shared/regex --output "$PACKS"
-          codeql pack create ../shared/ssa --output "$PACKS"
-          codeql pack create ../shared/tutorial --output "$PACKS"
-          codeql pack create ql/lib --output "$PACKS"
-          codeql pack create -j0 ql/src --output "$PACKS" --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
-          PACK_FOLDER=$(readlink -f "$PACKS"/codeql/ruby-queries/*)
-          codeql generate query-help --format=sarifv2.1.0 --output="${PACK_FOLDER}/rules.sarif" ql/src
-          (cd ql/src; find queries \( -name '*.qhelp' -o -name '*.rb' -o -name '*.erb' \) -exec bash -c 'mkdir -p "'"${PACK_FOLDER}"'/$(dirname "{}")"' \; -exec cp "{}" "${PACK_FOLDER}/{}" \;)
-      - uses: actions/upload-artifact@v4
-        with:
-          name: codeql-ruby-queries
-          path: |
-            ${{ runner.temp }}/query-packs/*
-          retention-days: 1
-          include-hidden-files: true
-
-  package:
-    runs-on: ubuntu-latest
-    needs: [build, compile-queries]
-    steps:
-      - uses: actions/checkout@v5
-      - uses: actions/download-artifact@v4
-        with:
-          name: ruby.dbscheme
-          path: ruby/ruby
-      - uses: actions/download-artifact@v4
-        with:
-          name: extractor-ubuntu-latest
-          path: ruby/linux64
-      - uses: actions/download-artifact@v4
-        with:
-          name: extractor-windows-latest
-          path: ruby/win64
-      - uses: actions/download-artifact@v4
-        with:
-          name: extractor-macos-latest
-          path: ruby/osx64
-      - run: |
-          mkdir -p ruby
-          cp -r codeql-extractor.yml tools ql/lib/ruby.dbscheme.stats ruby/
-          mkdir -p ruby/tools/{linux64,osx64,win64}
-          cp linux64/codeql-extractor-ruby ruby/tools/linux64/extractor
-          cp osx64/codeql-extractor-ruby ruby/tools/osx64/extractor
-          cp win64/codeql-extractor-ruby.exe ruby/tools/win64/extractor.exe
-          chmod +x ruby/tools/{linux64,osx64}/extractor
-          zip -rq codeql-ruby.zip ruby
-      - uses: actions/upload-artifact@v4
-        with:
-          name: codeql-ruby-pack
-          path: ruby/codeql-ruby.zip
-          retention-days: 1
-          include-hidden-files: true
-      - uses: actions/download-artifact@v4
-        with:
-          name: codeql-ruby-queries
-          path: ruby/qlpacks
-      - run: |
-          echo '{
-            "provide": [
-            "ruby/codeql-extractor.yml",
-            "qlpacks/*/*/*/qlpack.yml"
-            ]
-          }' > .codeqlmanifest.json
-          zip -rq codeql-ruby-bundle.zip .codeqlmanifest.json ruby qlpacks
-      - uses: actions/upload-artifact@v4
-        with:
-          name: codeql-ruby-bundle
-          path: ruby/codeql-ruby-bundle.zip
-          retention-days: 1
-          include-hidden-files: true
-
-  test:
-    defaults:
-      run:
-        working-directory: ${{ github.workspace }}
-    strategy:
-      fail-fast: false
-      matrix:
-        os: [ubuntu-latest, macos-latest, windows-latest]
-
-    runs-on: ${{ matrix.os }}
-    needs: [package]
-    steps:
-      - uses: actions/checkout@v5
-      - name: Fetch CodeQL
-        uses: ./.github/actions/fetch-codeql
-
-      - name: Download Ruby bundle
-        uses: actions/download-artifact@v4
-        with:
-          name: codeql-ruby-bundle
-          path: ${{ runner.temp }}
-      - name: Unzip Ruby bundle
-        shell: bash
-        run: unzip -q -d "${{ runner.temp }}/ruby-bundle" "${{ runner.temp }}/codeql-ruby-bundle.zip"
-
-      - name: Run QL test
-        shell: bash
-        run: |
-          codeql test run --search-path "${{ runner.temp }}/ruby-bundle" --additional-packs "${{ runner.temp }}/ruby-bundle" ruby/ql/test/library-tests/ast/constants/
-      - name: Create database
-        shell: bash
-        run: |
-          codeql database create --search-path "${{ runner.temp }}/ruby-bundle" --language ruby --source-root ruby/ql/test/library-tests/ast/constants/ ../database
-      - name: Analyze database
-        shell: bash
-        run: |
-          codeql database analyze --search-path "${{ runner.temp }}/ruby-bundle" --format=sarifv2.1.0 --output=out.sarif ../database ruby-code-scanning.qls

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

@@ -1,75 +0,0 @@
-name: "Ruby: Collect database stats"
-
-on:
-  push:
-    branches:
-      - main
-      - "rc/*"
-    paths:
-      - ruby/ql/lib/ruby.dbscheme
-      - .github/workflows/ruby-dataset-measure.yml
-  pull_request:
-    branches:
-      - main
-      - "rc/*"
-    paths:
-      - ruby/ql/lib/ruby.dbscheme
-      - .github/workflows/ruby-dataset-measure.yml
-  workflow_dispatch:
-
-permissions:
-  contents: read
-
-jobs:
-  measure:
-    env:
-      CODEQL_THREADS: 4 # TODO: remove this once it's set by the CLI
-    strategy:
-      fail-fast: false
-      matrix:
-        repo: [rails/rails, discourse/discourse, spree/spree, ruby/ruby]
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v5
-
-      - uses: ./.github/actions/fetch-codeql
-
-      - uses: ./ruby/actions/create-extractor-pack
-
-      - name: Checkout ${{ matrix.repo }}
-        uses: actions/checkout@v5
-        with:
-          repository: ${{ matrix.repo }}
-          path: ${{ github.workspace }}/repo
-      - name: Create database
-        run: |
-          codeql database create \
-            --search-path "${{ github.workspace }}" \
-            --threads 4 \
-            --language ruby --source-root "${{ github.workspace }}/repo" \
-            "${{ runner.temp }}/database"
-      - name: Measure database
-        run: |
-          mkdir -p "stats/${{ matrix.repo }}"
-          codeql dataset measure --threads 4 --output "stats/${{ matrix.repo }}/stats.xml" "${{ runner.temp }}/database/db-ruby"
-      - uses: actions/upload-artifact@v4
-        with:
-          name: measurements-${{ hashFiles('stats/**') }}
-          path: stats
-          retention-days: 1
-
-  merge:
-    runs-on: ubuntu-latest
-    needs: measure
-    steps:
-      - uses: actions/checkout@v5
-      - uses: actions/download-artifact@v4
-        with:
-          path: stats
-      - run: |
-          python -m pip install --user lxml
-          find stats -name 'stats.xml' | sort | xargs python ruby/scripts/merge_stats.py --output ruby/ql/lib/ruby.dbscheme.stats --normalise ruby_tokeninfo
-      - uses: actions/upload-artifact@v4
-        with:
-          name: ruby.dbscheme.stats
-          path: ruby/ql/lib/ruby.dbscheme.stats

.github/workflows/ruby-qltest-rtjo.yml

@@ -1,40 +0,0 @@
-name: "Ruby: Run RTJO Language Tests"
-
-on:
-  pull_request:
-    types:
-      - opened
-      - synchronize
-      - reopened
-      - labeled
-
-env:
-  CARGO_TERM_COLOR: always
-
-defaults:
-  run:
-    working-directory: ruby
-
-permissions:
-  contents: read
-
-jobs:
-  qltest-rtjo:
-    if: "github.repository_owner == 'github' && github.event.label.name == 'Run: RTJO Language Tests'"
-    runs-on: ubuntu-latest-xl
-    strategy:
-      fail-fast: false
-    steps:
-      - uses: actions/checkout@v5
-      - uses: ./.github/actions/fetch-codeql
-      - uses: ./ruby/actions/create-extractor-pack
-      - name: Cache compilation cache
-        id: query-cache
-        uses: ./.github/actions/cache-query-compilation
-        with:
-          key: ruby-qltest
-      - name: Run QL tests
-        run: |
-          codeql test run --dynamic-join-order-mode=all --threads=0 --ram 50000 --search-path "${{ github.workspace }}" --check-databases --check-diff-informed --check-undefined-labels --check-unused-labels --check-repeated-labels --check-redefined-labels --check-use-before-definition --consistency-queries ql/consistency-queries ql/test --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
-        env:
-          GITHUB_TOKEN: ${{ github.token }}

.github/workflows/ruby-qltest.yml

@@ -1,73 +0,0 @@
-name: "Ruby: Run QL Tests"
-
-on:
-  push:
-    paths:
-      - "ruby/**"
-      - "shared/**"
-      - .github/workflows/ruby-build.yml
-      - .github/actions/fetch-codeql/action.yml
-      - codeql-workspace.yml
-    branches:
-      - main
-      - "rc/*"
-  pull_request:
-    paths:
-      - "ruby/**"
-      - "shared/**"
-      - .github/workflows/ruby-qltest.yml
-      - .github/actions/fetch-codeql/action.yml
-      - codeql-workspace.yml
-    branches:
-      - main
-      - "rc/*"
-
-env:
-  CARGO_TERM_COLOR: always
-
-defaults:
-  run:
-    working-directory: ruby
-
-permissions:
-  contents: read
-
-jobs:
-  qlupgrade:
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v5
-      - uses: ./.github/actions/fetch-codeql
-      - name: Check DB upgrade scripts
-        run: |
-          echo >empty.trap
-          codeql dataset import -S ql/lib/upgrades/initial/ruby.dbscheme testdb empty.trap
-          codeql dataset upgrade testdb --additional-packs ql/lib
-          diff -q testdb/ruby.dbscheme ql/lib/ruby.dbscheme
-      - name: Check DB downgrade scripts
-        run: |
-          echo >empty.trap
-          rm -rf testdb; codeql dataset import -S ql/lib/ruby.dbscheme testdb empty.trap
-          codeql resolve upgrades --format=lines --allow-downgrades --additional-packs downgrades \
-           --dbscheme=ql/lib/ruby.dbscheme --target-dbscheme=downgrades/initial/ruby.dbscheme |
-           xargs codeql execute upgrades testdb
-          diff -q testdb/ruby.dbscheme downgrades/initial/ruby.dbscheme
-  qltest:
-    if: github.repository_owner == 'github'
-    runs-on: ubuntu-latest-xl
-    strategy:
-      fail-fast: false
-    steps:
-      - uses: actions/checkout@v5
-      - uses: ./.github/actions/fetch-codeql
-      - uses: ./ruby/actions/create-extractor-pack
-      - name: Cache compilation cache
-        id: query-cache
-        uses: ./.github/actions/cache-query-compilation
-        with:
-          key: ruby-qltest
-      - name: Run QL tests
-        run: |
-          codeql test run --threads=0 --ram 50000 --search-path "${{ github.workspace }}" --check-databases --check-diff-informed --check-undefined-labels --check-unused-labels --check-repeated-labels --check-redefined-labels --check-use-before-definition --consistency-queries ql/consistency-queries ql/test --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
-        env:
-          GITHUB_TOKEN: ${{ github.token }}

.pre-commit-config.yaml

@@ -7,9 +7,9 @@ repos:
     rev: v3.2.0
     hooks:
       - id: trailing-whitespace
-        exclude: /test/.*$(?<!\.qlref)|.*\.patch$|.*\.qll?$
+        exclude: /test([^/]*)/.*$(?<!\.qlref)|.*\.patch$|.*\.qll?$
       - id: end-of-file-fixer
-        exclude: Cargo.lock$|/test/.*$(?<!\.qlref)|.*\.patch$|.*\.qll?$
+        exclude: Cargo.lock$|/test([^/]*)/.*$(?<!\.qlref)|.*\.patch$|.*\.qll?$
 
   - repo: https://github.com/pre-commit/mirrors-clang-format
     rev: v17.0.6

MODULE.bazel

@@ -15,21 +15,23 @@ local_path_override(
 # see https://registry.bazel.build/ for a list of available packages
 
 bazel_dep(name = "platforms", version = "1.0.0")
-bazel_dep(name = "rules_go", version = "0.56.1")
-bazel_dep(name = "rules_pkg", version = "1.0.1")
-bazel_dep(name = "rules_nodejs", version = "6.2.0-codeql.1")
-bazel_dep(name = "rules_python", version = "0.40.0")
-bazel_dep(name = "rules_shell", version = "0.5.0")
-bazel_dep(name = "bazel_skylib", version = "1.8.1")
-bazel_dep(name = "abseil-cpp", version = "20240116.1", repo_name = "absl")
+bazel_dep(name = "rules_cc", version = "0.2.17")
+bazel_dep(name = "rules_go", version = "0.60.0")
+bazel_dep(name = "rules_java", version = "9.6.1")
+bazel_dep(name = "rules_pkg", version = "1.2.0")
+bazel_dep(name = "rules_nodejs", version = "6.7.3")
+bazel_dep(name = "rules_python", version = "1.9.0")
+bazel_dep(name = "rules_shell", version = "0.7.1")
+bazel_dep(name = "bazel_skylib", version = "1.9.0")
+bazel_dep(name = "abseil-cpp", version = "20260107.1", repo_name = "absl")
 bazel_dep(name = "nlohmann_json", version = "3.11.3", repo_name = "json")
 bazel_dep(name = "fmt", version = "12.1.0-codeql.1")
-bazel_dep(name = "rules_kotlin", version = "2.2.0-codeql.1")
-bazel_dep(name = "gazelle", version = "0.40.0")
+bazel_dep(name = "rules_kotlin", version = "2.2.2-codeql.1")
+bazel_dep(name = "gazelle", version = "0.50.0")
 bazel_dep(name = "rules_dotnet", version = "0.21.5-codeql.1")
-bazel_dep(name = "googletest", version = "1.14.0.bcr.1")
-bazel_dep(name = "rules_rust", version = "0.66.0")
-bazel_dep(name = "zstd", version = "1.5.5.bcr.1")
+bazel_dep(name = "googletest", version = "1.17.0.bcr.2")
+bazel_dep(name = "rules_rust", version = "0.69.0")
+bazel_dep(name = "zstd", version = "1.5.7.bcr.1")
 
 bazel_dep(name = "buildifier_prebuilt", version = "6.4.0", dev_dependency = True)
 
@@ -41,7 +43,7 @@ RUST_EDITION = "2024"
 # a nightly toolchain is required to enable experimental_use_cc_common_link, which we require internally
 # we prefer to run the same version as internally, even if experimental_use_cc_common_link is not really
 # required in this repo
-RUST_VERSION = "nightly/2025-08-01"
+RUST_VERSION = "nightly/2026-01-22"
 
 rust = use_extension("@rules_rust//rust:extensions.bzl", "rust")
 rust.toolchain(
@@ -53,26 +55,26 @@ rust.toolchain(
     ],
     # generated by buildutils-internal/scripts/fill-rust-sha256s.py (internal repo)
     sha256s = {
-        "2025-08-01/rustc-nightly-x86_64-unknown-linux-gnu.tar.xz": "9bbeaf5d3fc7247d31463a9083aa251c995cc50662c8219e7a2254d76a72a9a4",
-        "2025-08-01/rustc-nightly-x86_64-apple-darwin.tar.xz": "c9ea539a8eff0d5d162701f99f9e1aabe14dd0dfb420d62362817a5d09219de7",
-        "2025-08-01/rustc-nightly-aarch64-apple-darwin.tar.xz": "ae83feebbc39cfd982e4ecc8297731fe79c185173aee138467b334c5404b3773",
-        "2025-08-01/rustc-nightly-x86_64-pc-windows-msvc.tar.xz": "9f170c30d802a349be60cf52ec46260802093cb1013ad667fc0d528b7b10152f",
-        "2025-08-01/clippy-nightly-x86_64-unknown-linux-gnu.tar.xz": "9ae5f3cd8f557c4f6df522597c69d14398cf604cfaed2b83e767c4b77a7eaaf6",
-        "2025-08-01/clippy-nightly-x86_64-apple-darwin.tar.xz": "983cb9ee0b6b968188e04ab2d33743d54764b2681ce565e1b3f2b9135c696a3e",
-        "2025-08-01/clippy-nightly-aarch64-apple-darwin.tar.xz": "ed2219dbc49d088225e1b7c5c4390fa295066e071fddaa2714018f6bb39ddbf0",
-        "2025-08-01/clippy-nightly-x86_64-pc-windows-msvc.tar.xz": "911f40ab5cbdd686f40e00965271fe47c4805513a308ed01f30eafb25b448a50",
-        "2025-08-01/cargo-nightly-x86_64-unknown-linux-gnu.tar.xz": "106463c284e48e4904c717471eeec2be5cc83a9d2cae8d6e948b52438cad2e69",
-        "2025-08-01/cargo-nightly-x86_64-apple-darwin.tar.xz": "6ad35c40efc41a8c531ea43235058347b6902d98a9693bf0aed7fc16d5590cef",
-        "2025-08-01/cargo-nightly-aarch64-apple-darwin.tar.xz": "dd28c365e9d298abc3154c797720ad36a0058f131265c9978b4c8e4e37012c8a",
-        "2025-08-01/cargo-nightly-x86_64-pc-windows-msvc.tar.xz": "7b431286e12d6b3834b038f078389a00cac73f351e8c3152b2504a3c06420b3b",
-        "2025-08-01/llvm-tools-nightly-x86_64-unknown-linux-gnu.tar.xz": "e342e305d7927cc288d386983b2bc253cfad3776b113386e903d0b302648ef47",
-        "2025-08-01/llvm-tools-nightly-x86_64-apple-darwin.tar.xz": "e44dd3506524d85c37b3a54bcc91d01378fd2c590b2db5c5974d12f05c1b84d1",
-        "2025-08-01/llvm-tools-nightly-aarch64-apple-darwin.tar.xz": "0c1b5f46dd81be4a9227b10283a0fcaa39c14fea7e81aea6fd6d9887ff6cdc41",
-        "2025-08-01/llvm-tools-nightly-x86_64-pc-windows-msvc.tar.xz": "423e5fd11406adccbc31b8456ceb7375ce055cdf45e90d2c3babeb2d7f58383f",
-        "2025-08-01/rust-std-nightly-x86_64-unknown-linux-gnu.tar.xz": "3c0ceb46a252647a1d4c7116d9ccae684fa5e42aaf3296419febd2c962c3b41d",
-        "2025-08-01/rust-std-nightly-x86_64-apple-darwin.tar.xz": "3be416003cab10f767390a753d1d16ae4d26c7421c03c98992cf1943e5b0efe8",
-        "2025-08-01/rust-std-nightly-aarch64-apple-darwin.tar.xz": "4046ac0ef951cb056b5028a399124f60999fa37792eab69d008d8d7965f389b4",
-        "2025-08-01/rust-std-nightly-x86_64-pc-windows-msvc.tar.xz": "191ed9d8603c3a4fe5a7bbbc2feb72049078dae2df3d3b7d5dedf3abbf823e6e",
+        "2026-01-22/rustc-nightly-x86_64-unknown-linux-gnu.tar.xz": "88db619323cc1321630d124efa51ed02fabc5e020f08cfa0eda2c0ac1afbe69a",
+        "2026-01-22/rustc-nightly-x86_64-apple-darwin.tar.xz": "08484da3fa38db56f93629aeabdc0ae9ff8ed9704c0792d35259cbc849b3f54c",
+        "2026-01-22/rustc-nightly-aarch64-apple-darwin.tar.xz": "a39c0b21b7058e364ea1bd43144e42e4bf1efade036b2e82455f2afce194ee81",
+        "2026-01-22/rustc-nightly-x86_64-pc-windows-msvc.tar.xz": "d00248ee9850dbb6932b2578e32ff74fc7c429854c1aa071066ca31b65385a3b",
+        "2026-01-22/clippy-nightly-x86_64-unknown-linux-gnu.tar.xz": "70656a0ce994ffff16d5a35a7b170a0acd41e9bb54a589c96ed45bf97b094a4d",
+        "2026-01-22/clippy-nightly-x86_64-apple-darwin.tar.xz": "fe242519fa961522734733009705aec3c2d9a20cc57291f2aa614e5e6262c88f",
+        "2026-01-22/clippy-nightly-aarch64-apple-darwin.tar.xz": "38bb226363ec97c9722edf966cd58774a683e19fd2ff2a6030094445d51e06f9",
+        "2026-01-22/clippy-nightly-x86_64-pc-windows-msvc.tar.xz": "6da9b4470beea67abfebf046f141eee0d2a8db7c7a9e4e2294478734fd477228",
+        "2026-01-22/cargo-nightly-x86_64-unknown-linux-gnu.tar.xz": "99004e9d10c43a01499642f53bb3184d41137a95d65bfb217098840a9e79e892",
+        "2026-01-22/cargo-nightly-x86_64-apple-darwin.tar.xz": "6e021394cf8d8400ac6cfdfcef24e4d74f988e91eb8028b36de3a64ce3502990",
+        "2026-01-22/cargo-nightly-aarch64-apple-darwin.tar.xz": "4b2494cb69ab64132cddbc411a38ea9f1105e54d6f986e43168d54f79510c673",
+        "2026-01-22/cargo-nightly-x86_64-pc-windows-msvc.tar.xz": "c36613cf57407212d10d37b76e49a60ff42336e953cdff9e177283f530a83fc1",
+        "2026-01-22/llvm-tools-nightly-x86_64-unknown-linux-gnu.tar.xz": "0b123c5027dbd833aae6845ffe9bd07d309bf798746a7176aadaea68fbcbd05d",
+        "2026-01-22/llvm-tools-nightly-x86_64-apple-darwin.tar.xz": "a47864491ad5619158c950ab7570fb6e487d5117338585c27334d45824b406d8",
+        "2026-01-22/llvm-tools-nightly-aarch64-apple-darwin.tar.xz": "db9bc826d6e2e7e914505d50157682e516ceb90357e83d77abddc32c2d962f41",
+        "2026-01-22/llvm-tools-nightly-x86_64-pc-windows-msvc.tar.xz": "ffaa406932b2fe62e01dad61cf4ed34860a5d2a6f9306ca340d79e630d930039",
+        "2026-01-22/rust-std-nightly-x86_64-unknown-linux-gnu.tar.xz": "e9c0d5e06e18a4b509391b3088f29293e310cdc8ccc865be8fa3f09733326925",
+        "2026-01-22/rust-std-nightly-x86_64-apple-darwin.tar.xz": "25d75995cee679a4828ca9fe48c5a31a67c3b0846018440ef912e5a6208f53f6",
+        "2026-01-22/rust-std-nightly-aarch64-apple-darwin.tar.xz": "e4132bf3f2eed4684c86756a02315bcf481c23e675e3e25630fc604c9cb4594c",
+        "2026-01-22/rust-std-nightly-x86_64-pc-windows-msvc.tar.xz": "961bb535ef95ae8a5fa4e224cb94aff190f155c45a9bcf7a53e184b024aa41b1",
     },
     versions = [RUST_VERSION],
 )
@@ -188,6 +190,15 @@ pip.parse(
 )
 use_repo(pip, "codegen_deps")
 
+python = use_extension("@rules_python//python/extensions:python.bzl", "python")
+python.toolchain(
+    is_default = True,
+    python_version = "3.12",
+)
+use_repo(python, "python_3_12", "python_versions")
+
+register_toolchains("@python_versions//3.12:all")
+
 swift_deps = use_extension("//swift/third_party:load.bzl", "swift_deps")
 
 # following list can be kept in sync with `bazel mod tidy`
@@ -231,6 +242,7 @@ use_repo(
     "kotlin-compiler-2.2.0-Beta1",
     "kotlin-compiler-2.2.20-Beta2",
     "kotlin-compiler-2.3.0",
+    "kotlin-compiler-2.3.20",
     "kotlin-compiler-embeddable-1.8.0",
     "kotlin-compiler-embeddable-1.9.0-Beta",
     "kotlin-compiler-embeddable-1.9.20-Beta",
@@ -241,6 +253,7 @@ use_repo(
     "kotlin-compiler-embeddable-2.2.0-Beta1",
     "kotlin-compiler-embeddable-2.2.20-Beta2",
     "kotlin-compiler-embeddable-2.3.0",
+    "kotlin-compiler-embeddable-2.3.20",
     "kotlin-stdlib-1.8.0",
     "kotlin-stdlib-1.9.0-Beta",
     "kotlin-stdlib-1.9.20-Beta",
@@ -251,14 +264,15 @@ use_repo(
     "kotlin-stdlib-2.2.0-Beta1",
     "kotlin-stdlib-2.2.20-Beta2",
     "kotlin-stdlib-2.3.0",
+    "kotlin-stdlib-2.3.20",
 )
 
 go_sdk = use_extension("@rules_go//go:extensions.bzl", "go_sdk")
-go_sdk.download(version = "1.25.0")
+go_sdk.download(version = "1.26.0")
 
 go_deps = use_extension("@gazelle//:extensions.bzl", "go_deps")
 go_deps.from_file(go_mod = "//go/extractor:go.mod")
-use_repo(go_deps, "org_golang_x_mod", "org_golang_x_tools")
+use_repo(go_deps, "com_github_stretchr_testify", "org_golang_x_mod", "org_golang_x_tools")
 
 ripunzip_archive = use_repo_rule("//misc/ripunzip:ripunzip.bzl", "ripunzip_archive")
 

actions/ql/lib/CHANGELOG.md

@@ -1,3 +1,33 @@
+## 0.4.34
+
+### Minor Analysis Improvements
+
+* Removed false positive injection sink models for the `context` input of `docker/build-push-action` and the `allowed-endpoints` input of `step-security/harden-runner`.
+
+## 0.4.33
+
+No user-facing changes.
+
+## 0.4.32
+
+No user-facing changes.
+
+## 0.4.31
+
+No user-facing changes.
+
+## 0.4.30
+
+No user-facing changes.
+
+## 0.4.29
+
+No user-facing changes.
+
+## 0.4.28
+
+No user-facing changes.
+
 ## 0.4.27
 
 ### Bug Fixes

actions/ql/lib/change-notes/released/0.4.28.md

@@ -0,0 +1,3 @@
+## 0.4.28
+
+No user-facing changes.

actions/ql/lib/change-notes/released/0.4.29.md

@@ -0,0 +1,3 @@
+## 0.4.29
+
+No user-facing changes.

actions/ql/lib/change-notes/released/0.4.30.md

@@ -0,0 +1,3 @@
+## 0.4.30
+
+No user-facing changes.

actions/ql/lib/change-notes/released/0.4.31.md

@@ -0,0 +1,3 @@
+## 0.4.31
+
+No user-facing changes.

actions/ql/lib/change-notes/released/0.4.32.md

@@ -0,0 +1,3 @@
+## 0.4.32
+
+No user-facing changes.

actions/ql/lib/change-notes/released/0.4.33.md

@@ -0,0 +1,3 @@
+## 0.4.33
+
+No user-facing changes.

actions/ql/lib/change-notes/released/0.4.34.md

@@ -0,0 +1,5 @@
+## 0.4.34
+
+### Minor Analysis Improvements
+
+* Removed false positive injection sink models for the `context` input of `docker/build-push-action` and the `allowed-endpoints` input of `step-security/harden-runner`.

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

@@ -1,2 +1,2 @@
 ---
-lastReleaseVersion: 0.4.27
+lastReleaseVersion: 0.4.34

actions/ql/lib/ext/manual/docker_build-push-action.model.yml

@@ -1,6 +0,0 @@
-extensions:
-  - addsTo:
-      pack: codeql/actions-all
-      extensible: actionsSinkModel
-    data:
-      - ["docker/build-push-action", "*", "input.context", "code-injection", "manual"]

actions/ql/lib/ext/manual/step-security_harden-runner.model.yml

@@ -1,6 +0,0 @@
-extensions:
-  - addsTo:
-      pack: codeql/actions-all
-      extensible: actionsSinkModel
-    data:
-      - ["step-security/harden-runner", "*", "input.allowed-endpoints", "command-injection", "manual"]

actions/ql/lib/qlpack.yml

@@ -1,5 +1,5 @@
 name: codeql/actions-all
-version: 0.4.27
+version: 0.4.34
 library: true
 warnOnImplicitThis: true
 dependencies:

actions/ql/src/CHANGELOG.md

@@ -1,3 +1,37 @@
+## 0.6.26
+
+### Major Analysis Improvements
+
+* Fixed alert messages in `actions/artifact-poisoning/critical` and `actions/artifact-poisoning/medium` as they previously included a redundant placeholder in the alert message that would on occasion contain a long block of yml that makes the alert difficult to understand. Also improved the wording to make it clearer that it is not the artifact that is being poisoned, but instead a potentially untrusted artifact that is consumed. Finally, changed the alert location to be the source, to align more with other queries reporting an artifact (e.g. zipslip) which is more useful.
+
+### Minor Analysis Improvements
+
+* The query `actions/missing-workflow-permissions` no longer produces false positive results on reusable workflows where all callers set permissions.
+
+## 0.6.25
+
+No user-facing changes.
+
+## 0.6.24
+
+No user-facing changes.
+
+## 0.6.23
+
+No user-facing changes.
+
+## 0.6.22
+
+No user-facing changes.
+
+## 0.6.21
+
+No user-facing changes.
+
+## 0.6.20
+
+No user-facing changes.
+
 ## 0.6.19
 
 No user-facing changes.
@@ -139,7 +173,7 @@ No user-facing changes.
   * `actions/if-expression-always-true/critical`
   * `actions/if-expression-always-true/high`
   * `actions/unnecessary-use-of-advanced-config`
-  
+
 * The following query has been moved from the `code-scanning` suite to the `security-extended`
   suite. Any existing alerts for this query will be closed automatically unless the analysis is
   configured to use the `security-extended` suite.

actions/ql/src/Security/CWE-275/MissingActionsPermissions.ql

@@ -26,10 +26,23 @@ string permissionsForJob(Job job) {
     "{" + concat(string permission | permission = jobNeedsPermission(job) | permission, ", ") + "}"
 }
 
+predicate jobHasPermissions(Job job) {
+  exists(job.getPermissions())
+  or
+  exists(job.getEnclosingWorkflow().getPermissions())
+  or
+  // The workflow is reusable and cannot be triggered in any other way; check callers
+  exists(ReusableWorkflow r | r = job.getEnclosingWorkflow() |
+    not exists(Event e | e = r.getOn().getAnEvent() | e.getName() != "workflow_call") and
+    forall(Job caller | caller = job.getEnclosingWorkflow().(ReusableWorkflow).getACaller() |
+      jobHasPermissions(caller)
+    )
+  )
+}
+
 from Job job, string permissions
 where
-  not exists(job.getPermissions()) and
-  not exists(job.getEnclosingWorkflow().getPermissions()) and
+  not jobHasPermissions(job) and
   // exists a trigger event that is not a workflow_call
   exists(Event e |
     e = job.getATriggerEvent() and

actions/ql/src/Security/CWE-829/ArtifactPoisoningCritical.ql

@@ -20,6 +20,6 @@ from ArtifactPoisoningFlow::PathNode source, ArtifactPoisoningFlow::PathNode sin
 where
   ArtifactPoisoningFlow::flowPath(source, sink) and
   event = getRelevantEventInPrivilegedContext(sink.getNode())
-select sink.getNode(), source, sink,
-  "Potential artifact poisoning in $@, which may be controlled by an external user ($@).", sink,
-  sink.getNode().toString(), event, event.getName()
+select source.getNode(), source, sink,
+  "Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@).",
+  event, event.getName()

actions/ql/src/Security/CWE-829/ArtifactPoisoningMedium.ql

@@ -20,6 +20,5 @@ from ArtifactPoisoningFlow::PathNode source, ArtifactPoisoningFlow::PathNode sin
 where
   ArtifactPoisoningFlow::flowPath(source, sink) and
   inNonPrivilegedContext(sink.getNode().asExpr())
-select sink.getNode(), source, sink,
-  "Potential artifact poisoning in $@, which may be controlled by an external user.", sink,
-  sink.getNode().toString()
+select source.getNode(), source, sink,
+  "Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user."

actions/ql/src/change-notes/released/0.6.20.md

@@ -0,0 +1,3 @@
+## 0.6.20
+
+No user-facing changes.

actions/ql/src/change-notes/released/0.6.21.md

@@ -0,0 +1,3 @@
+## 0.6.21
+
+No user-facing changes.

actions/ql/src/change-notes/released/0.6.22.md

@@ -0,0 +1,3 @@
+## 0.6.22
+
+No user-facing changes.

actions/ql/src/change-notes/released/0.6.23.md

@@ -0,0 +1,3 @@
+## 0.6.23
+
+No user-facing changes.

actions/ql/src/change-notes/released/0.6.24.md

@@ -0,0 +1,3 @@
+## 0.6.24
+
+No user-facing changes.

actions/ql/src/change-notes/released/0.6.25.md

@@ -0,0 +1,3 @@
+## 0.6.25
+
+No user-facing changes.

actions/ql/src/change-notes/released/0.6.26.md

@@ -0,0 +1,9 @@
+## 0.6.26
+
+### Major Analysis Improvements
+
+* Fixed alert messages in `actions/artifact-poisoning/critical` and `actions/artifact-poisoning/medium` as they previously included a redundant placeholder in the alert message that would on occasion contain a long block of yml that makes the alert difficult to understand. Also improved the wording to make it clearer that it is not the artifact that is being poisoned, but instead a potentially untrusted artifact that is consumed. Finally, changed the alert location to be the source, to align more with other queries reporting an artifact (e.g. zipslip) which is more useful.
+
+### Minor Analysis Improvements
+
+* The query `actions/missing-workflow-permissions` no longer produces false positive results on reusable workflows where all callers set permissions.

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

@@ -1,2 +1,2 @@
 ---
-lastReleaseVersion: 0.6.19
+lastReleaseVersion: 0.6.26

actions/ql/src/qlpack.yml

@@ -1,5 +1,5 @@
 name: codeql/actions-queries
-version: 0.6.19
+version: 0.6.26
 library: false
 warnOnImplicitThis: true
 groups: [actions, queries]

actions/ql/test/query-tests/Security/CWE-275/.github/workflows/perms11.yml

@@ -0,0 +1,9 @@
+on:
+  workflow_call:
+
+jobs:
+  build:
+    name: Build and test
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/deploy-pages

actions/ql/test/query-tests/Security/CWE-275/.github/workflows/perms12.yml

@@ -0,0 +1,11 @@
+on:
+  workflow_dispatch:
+
+permissions:
+  contents: read
+  id-token: write
+  pages: write
+
+jobs:
+  call-workflow:
+    uses: ./.github/workflows/perms11.yml

actions/ql/test/query-tests/Security/CWE-829/ArtifactPoisoningCritical.expected

@@ -55,21 +55,21 @@ nodes
 | .github/workflows/test25.yml:39:14:40:45 | ./gradlew buildScanPublishPrevious\n | semmle.label | ./gradlew buildScanPublishPrevious\n |
 subpaths
 #select
-| .github/workflows/artifactpoisoning11.yml:38:11:38:77 | ./sonarcloud-data/x.py build -j$(nproc) --compiler gcc --skip-build | .github/workflows/artifactpoisoning11.yml:13:9:32:6 | Uses Step | .github/workflows/artifactpoisoning11.yml:38:11:38:77 | ./sonarcloud-data/x.py build -j$(nproc) --compiler gcc --skip-build | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/artifactpoisoning11.yml:38:11:38:77 | ./sonarcloud-data/x.py build -j$(nproc) --compiler gcc --skip-build | ./sonarcloud-data/x.py build -j$(nproc) --compiler gcc --skip-build | .github/workflows/artifactpoisoning11.yml:4:3:4:14 | workflow_run | workflow_run |
-| .github/workflows/artifactpoisoning12.yml:38:11:38:25 | python foo/x.py | .github/workflows/artifactpoisoning12.yml:13:9:32:6 | Uses Step | .github/workflows/artifactpoisoning12.yml:38:11:38:25 | python foo/x.py | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/artifactpoisoning12.yml:38:11:38:25 | python foo/x.py | python foo/x.py | .github/workflows/artifactpoisoning12.yml:4:3:4:14 | workflow_run | workflow_run |
-| .github/workflows/artifactpoisoning21.yml:19:14:20:21 | sh foo/cmd\n | .github/workflows/artifactpoisoning21.yml:13:9:18:6 | Uses Step | .github/workflows/artifactpoisoning21.yml:19:14:20:21 | sh foo/cmd\n | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/artifactpoisoning21.yml:19:14:20:21 | sh foo/cmd\n | sh foo/cmd\n | .github/workflows/artifactpoisoning21.yml:4:3:4:14 | workflow_run | workflow_run |
-| .github/workflows/artifactpoisoning22.yml:18:14:18:19 | sh cmd | .github/workflows/artifactpoisoning22.yml:13:9:17:6 | Uses Step | .github/workflows/artifactpoisoning22.yml:18:14:18:19 | sh cmd | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/artifactpoisoning22.yml:18:14:18:19 | sh cmd | sh cmd | .github/workflows/artifactpoisoning22.yml:4:3:4:14 | workflow_run | workflow_run |
-| .github/workflows/artifactpoisoning31.yml:19:14:19:22 | ./foo/cmd | .github/workflows/artifactpoisoning31.yml:13:9:15:6 | Run Step | .github/workflows/artifactpoisoning31.yml:19:14:19:22 | ./foo/cmd | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/artifactpoisoning31.yml:19:14:19:22 | ./foo/cmd | ./foo/cmd | .github/workflows/artifactpoisoning31.yml:4:3:4:14 | workflow_run | workflow_run |
-| .github/workflows/artifactpoisoning32.yml:17:14:18:20 | ./bar/cmd\n | .github/workflows/artifactpoisoning32.yml:13:9:16:6 | Run Step | .github/workflows/artifactpoisoning32.yml:17:14:18:20 | ./bar/cmd\n | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/artifactpoisoning32.yml:17:14:18:20 | ./bar/cmd\n | ./bar/cmd\n | .github/workflows/artifactpoisoning32.yml:4:3:4:14 | workflow_run | workflow_run |
-| .github/workflows/artifactpoisoning33.yml:17:14:18:20 | ./bar/cmd\n | .github/workflows/artifactpoisoning33.yml:13:9:16:6 | Run Step | .github/workflows/artifactpoisoning33.yml:17:14:18:20 | ./bar/cmd\n | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/artifactpoisoning33.yml:17:14:18:20 | ./bar/cmd\n | ./bar/cmd\n | .github/workflows/artifactpoisoning33.yml:4:3:4:14 | workflow_run | workflow_run |
-| .github/workflows/artifactpoisoning34.yml:20:14:22:23 | npm install\nnpm run lint\n | .github/workflows/artifactpoisoning34.yml:13:9:16:6 | Run Step | .github/workflows/artifactpoisoning34.yml:20:14:22:23 | npm install\nnpm run lint\n | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/artifactpoisoning34.yml:20:14:22:23 | npm install\nnpm run lint\n | npm install\nnpm run lint\n | .github/workflows/artifactpoisoning34.yml:4:3:4:14 | workflow_run | workflow_run |
-| .github/workflows/artifactpoisoning41.yml:22:14:22:22 | ./foo/cmd | .github/workflows/artifactpoisoning41.yml:13:9:21:6 | Run Step | .github/workflows/artifactpoisoning41.yml:22:14:22:22 | ./foo/cmd | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/artifactpoisoning41.yml:22:14:22:22 | ./foo/cmd | ./foo/cmd | .github/workflows/artifactpoisoning41.yml:4:3:4:14 | workflow_run | workflow_run |
-| .github/workflows/artifactpoisoning42.yml:22:14:22:18 | ./cmd | .github/workflows/artifactpoisoning42.yml:13:9:21:6 | Run Step | .github/workflows/artifactpoisoning42.yml:22:14:22:18 | ./cmd | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/artifactpoisoning42.yml:22:14:22:18 | ./cmd | ./cmd | .github/workflows/artifactpoisoning42.yml:4:3:4:14 | workflow_run | workflow_run |
-| .github/workflows/artifactpoisoning71.yml:17:14:18:40 | sed -f config foo.md > bar.md\n | .github/workflows/artifactpoisoning71.yml:9:9:16:6 | Uses Step | .github/workflows/artifactpoisoning71.yml:17:14:18:40 | sed -f config foo.md > bar.md\n | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/artifactpoisoning71.yml:17:14:18:40 | sed -f config foo.md > bar.md\n | sed -f config foo.md > bar.md\n | .github/workflows/artifactpoisoning71.yml:4:5:4:16 | workflow_run | workflow_run |
-| .github/workflows/artifactpoisoning81.yml:31:14:31:27 | python test.py | .github/workflows/artifactpoisoning81.yml:28:9:31:6 | Uses Step | .github/workflows/artifactpoisoning81.yml:31:14:31:27 | python test.py | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/artifactpoisoning81.yml:31:14:31:27 | python test.py | python test.py | .github/workflows/artifactpoisoning81.yml:3:5:3:23 | pull_request_target | pull_request_target |
-| .github/workflows/artifactpoisoning92.yml:28:9:29:6 | Uses Step | .github/actions/download-artifact-2/action.yaml:6:7:25:4 | Uses Step | .github/workflows/artifactpoisoning92.yml:28:9:29:6 | Uses Step | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/artifactpoisoning92.yml:28:9:29:6 | Uses Step | Uses Step | .github/workflows/artifactpoisoning92.yml:3:3:3:14 | workflow_run | workflow_run |
-| .github/workflows/artifactpoisoning92.yml:29:14:29:26 | make snapshot | .github/actions/download-artifact-2/action.yaml:6:7:25:4 | Uses Step | .github/workflows/artifactpoisoning92.yml:29:14:29:26 | make snapshot | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/artifactpoisoning92.yml:29:14:29:26 | make snapshot | make snapshot | .github/workflows/artifactpoisoning92.yml:3:3:3:14 | workflow_run | workflow_run |
-| .github/workflows/artifactpoisoning96.yml:18:14:18:24 | npm install | .github/workflows/artifactpoisoning96.yml:13:9:18:6 | Uses Step | .github/workflows/artifactpoisoning96.yml:18:14:18:24 | npm install | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/artifactpoisoning96.yml:18:14:18:24 | npm install | npm install | .github/workflows/artifactpoisoning96.yml:2:3:2:14 | workflow_run | workflow_run |
-| .github/workflows/artifactpoisoning101.yml:17:14:19:59 | PR_NUMBER=$(./get_pull_request_number.sh pr_number.txt)\necho "PR_NUMBER=$PR_NUMBER" >> $GITHUB_OUTPUT \n | .github/workflows/artifactpoisoning101.yml:10:9:16:6 | Uses Step | .github/workflows/artifactpoisoning101.yml:17:14:19:59 | PR_NUMBER=$(./get_pull_request_number.sh pr_number.txt)\necho "PR_NUMBER=$PR_NUMBER" >> $GITHUB_OUTPUT \n | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/artifactpoisoning101.yml:17:14:19:59 | PR_NUMBER=$(./get_pull_request_number.sh pr_number.txt)\necho "PR_NUMBER=$PR_NUMBER" >> $GITHUB_OUTPUT \n | PR_NUMBER=$(./get_pull_request_number.sh pr_number.txt)\necho "PR_NUMBER=$PR_NUMBER" >> $GITHUB_OUTPUT \n | .github/workflows/artifactpoisoning101.yml:4:3:4:21 | pull_request_target | pull_request_target |
-| .github/workflows/test18.yml:36:15:40:58 | Uses Step | .github/workflows/test18.yml:12:15:33:12 | Uses Step | .github/workflows/test18.yml:36:15:40:58 | Uses Step | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/test18.yml:36:15:40:58 | Uses Step | Uses Step | .github/workflows/test18.yml:3:5:3:16 | workflow_run | workflow_run |
-| .github/workflows/test25.yml:39:14:40:45 | ./gradlew buildScanPublishPrevious\n | .github/workflows/test25.yml:22:9:32:6 | Uses Step: downloadBuildScan | .github/workflows/test25.yml:39:14:40:45 | ./gradlew buildScanPublishPrevious\n | Potential artifact poisoning in $@, which may be controlled by an external user ($@). | .github/workflows/test25.yml:39:14:40:45 | ./gradlew buildScanPublishPrevious\n | ./gradlew buildScanPublishPrevious\n | .github/workflows/test25.yml:2:3:2:14 | workflow_run | workflow_run |
+| .github/actions/download-artifact-2/action.yaml:6:7:25:4 | Uses Step | .github/actions/download-artifact-2/action.yaml:6:7:25:4 | Uses Step | .github/workflows/artifactpoisoning92.yml:28:9:29:6 | Uses Step | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/artifactpoisoning92.yml:3:3:3:14 | workflow_run | workflow_run |
+| .github/actions/download-artifact-2/action.yaml:6:7:25:4 | Uses Step | .github/actions/download-artifact-2/action.yaml:6:7:25:4 | Uses Step | .github/workflows/artifactpoisoning92.yml:29:14:29:26 | make snapshot | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/artifactpoisoning92.yml:3:3:3:14 | workflow_run | workflow_run |
+| .github/workflows/artifactpoisoning11.yml:13:9:32:6 | Uses Step | .github/workflows/artifactpoisoning11.yml:13:9:32:6 | Uses Step | .github/workflows/artifactpoisoning11.yml:38:11:38:77 | ./sonarcloud-data/x.py build -j$(nproc) --compiler gcc --skip-build | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/artifactpoisoning11.yml:4:3:4:14 | workflow_run | workflow_run |
+| .github/workflows/artifactpoisoning12.yml:13:9:32:6 | Uses Step | .github/workflows/artifactpoisoning12.yml:13:9:32:6 | Uses Step | .github/workflows/artifactpoisoning12.yml:38:11:38:25 | python foo/x.py | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/artifactpoisoning12.yml:4:3:4:14 | workflow_run | workflow_run |
+| .github/workflows/artifactpoisoning21.yml:13:9:18:6 | Uses Step | .github/workflows/artifactpoisoning21.yml:13:9:18:6 | Uses Step | .github/workflows/artifactpoisoning21.yml:19:14:20:21 | sh foo/cmd\n | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/artifactpoisoning21.yml:4:3:4:14 | workflow_run | workflow_run |
+| .github/workflows/artifactpoisoning22.yml:13:9:17:6 | Uses Step | .github/workflows/artifactpoisoning22.yml:13:9:17:6 | Uses Step | .github/workflows/artifactpoisoning22.yml:18:14:18:19 | sh cmd | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/artifactpoisoning22.yml:4:3:4:14 | workflow_run | workflow_run |
+| .github/workflows/artifactpoisoning31.yml:13:9:15:6 | Run Step | .github/workflows/artifactpoisoning31.yml:13:9:15:6 | Run Step | .github/workflows/artifactpoisoning31.yml:19:14:19:22 | ./foo/cmd | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/artifactpoisoning31.yml:4:3:4:14 | workflow_run | workflow_run |
+| .github/workflows/artifactpoisoning32.yml:13:9:16:6 | Run Step | .github/workflows/artifactpoisoning32.yml:13:9:16:6 | Run Step | .github/workflows/artifactpoisoning32.yml:17:14:18:20 | ./bar/cmd\n | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/artifactpoisoning32.yml:4:3:4:14 | workflow_run | workflow_run |
+| .github/workflows/artifactpoisoning33.yml:13:9:16:6 | Run Step | .github/workflows/artifactpoisoning33.yml:13:9:16:6 | Run Step | .github/workflows/artifactpoisoning33.yml:17:14:18:20 | ./bar/cmd\n | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/artifactpoisoning33.yml:4:3:4:14 | workflow_run | workflow_run |
+| .github/workflows/artifactpoisoning34.yml:13:9:16:6 | Run Step | .github/workflows/artifactpoisoning34.yml:13:9:16:6 | Run Step | .github/workflows/artifactpoisoning34.yml:20:14:22:23 | npm install\nnpm run lint\n | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/artifactpoisoning34.yml:4:3:4:14 | workflow_run | workflow_run |
+| .github/workflows/artifactpoisoning41.yml:13:9:21:6 | Run Step | .github/workflows/artifactpoisoning41.yml:13:9:21:6 | Run Step | .github/workflows/artifactpoisoning41.yml:22:14:22:22 | ./foo/cmd | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/artifactpoisoning41.yml:4:3:4:14 | workflow_run | workflow_run |
+| .github/workflows/artifactpoisoning42.yml:13:9:21:6 | Run Step | .github/workflows/artifactpoisoning42.yml:13:9:21:6 | Run Step | .github/workflows/artifactpoisoning42.yml:22:14:22:18 | ./cmd | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/artifactpoisoning42.yml:4:3:4:14 | workflow_run | workflow_run |
+| .github/workflows/artifactpoisoning71.yml:9:9:16:6 | Uses Step | .github/workflows/artifactpoisoning71.yml:9:9:16:6 | Uses Step | .github/workflows/artifactpoisoning71.yml:17:14:18:40 | sed -f config foo.md > bar.md\n | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/artifactpoisoning71.yml:4:5:4:16 | workflow_run | workflow_run |
+| .github/workflows/artifactpoisoning81.yml:28:9:31:6 | Uses Step | .github/workflows/artifactpoisoning81.yml:28:9:31:6 | Uses Step | .github/workflows/artifactpoisoning81.yml:31:14:31:27 | python test.py | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/artifactpoisoning81.yml:3:5:3:23 | pull_request_target | pull_request_target |
+| .github/workflows/artifactpoisoning96.yml:13:9:18:6 | Uses Step | .github/workflows/artifactpoisoning96.yml:13:9:18:6 | Uses Step | .github/workflows/artifactpoisoning96.yml:18:14:18:24 | npm install | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/artifactpoisoning96.yml:2:3:2:14 | workflow_run | workflow_run |
+| .github/workflows/artifactpoisoning101.yml:10:9:16:6 | Uses Step | .github/workflows/artifactpoisoning101.yml:10:9:16:6 | Uses Step | .github/workflows/artifactpoisoning101.yml:17:14:19:59 | PR_NUMBER=$(./get_pull_request_number.sh pr_number.txt)\necho "PR_NUMBER=$PR_NUMBER" >> $GITHUB_OUTPUT \n | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/artifactpoisoning101.yml:4:3:4:21 | pull_request_target | pull_request_target |
+| .github/workflows/test18.yml:12:15:33:12 | Uses Step | .github/workflows/test18.yml:12:15:33:12 | Uses Step | .github/workflows/test18.yml:36:15:40:58 | Uses Step | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/test18.yml:3:5:3:16 | workflow_run | workflow_run |
+| .github/workflows/test25.yml:22:9:32:6 | Uses Step: downloadBuildScan | .github/workflows/test25.yml:22:9:32:6 | Uses Step: downloadBuildScan | .github/workflows/test25.yml:39:14:40:45 | ./gradlew buildScanPublishPrevious\n | Potential artifact poisoning; the artifact being consumed has contents that may be controlled by an external user ($@). | .github/workflows/test25.yml:2:3:2:14 | workflow_run | workflow_run |

config/add-overlay-annotations.py

@@ -199,6 +199,7 @@ def annotate_as_appropriate(filename, lines):
     # as overlay[local?].  It is not clear that these heuristics are exactly what we want,
     # but they seem to work well enough for now (as determined by speed and accuracy numbers).
     if (filename.endswith("Test.qll") or
+        re.search(r"go/ql/lib/semmle/go/security/[^/]+[.]qll$", filename.replace(os.sep, "/")) or
         ((filename.endswith("Query.qll") or filename.endswith("Config.qll")) and
          any("implements DataFlow::ConfigSig" in line for line in lines))):
         return None

config/identical-files.json

@@ -172,10 +172,6 @@
     "cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/reachability/PrintDominance.qll",
     "cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/reachability/PrintDominance.qll"
   ],
-  "C# ControlFlowReachability": [
-    "csharp/ql/lib/semmle/code/csharp/dataflow/internal/ControlFlowReachability.qll",
-    "csharp/ql/lib/semmle/code/csharp/dataflow/internal/rangeanalysis/ControlFlowReachability.qll"
-  ],
   "C++ ExternalAPIs": [
     "cpp/ql/src/Security/CWE/CWE-020/ExternalAPIs.qll",
     "cpp/ql/src/Security/CWE/CWE-020/ir/ExternalAPIs.qll"

cpp/downgrades/770002bb02322e04fa25345838ce6e82af285a0b/in_trap.ql

@@ -0,0 +1,21 @@
+class Element extends @element {
+  string toString() { none() }
+}
+
+class Trap extends @trap {
+  string toString() { none() }
+}
+
+class Tag extends @tag {
+  string toString() { none() }
+}
+
+from Element e, Trap trap
+where
+  in_trap_or_tag(e, trap)
+  or
+  exists(Tag tag |
+    in_trap_or_tag(e, tag) and
+    trap_uses_tag(trap, tag)
+  )
+select e, trap

cpp/downgrades/770002bb02322e04fa25345838ce6e82af285a0b/source_file_uses_trap.ql

@@ -0,0 +1,13 @@
+class SourceFile extends @source_file {
+  string toString() { none() }
+}
+
+class Trap extends @trap {
+  string toString() { none() }
+}
+
+from SourceFile source_file, string name, Trap trap
+where
+  source_file_uses_trap(source_file, trap) and
+  source_file_name(source_file, name)
+select name, trap

cpp/downgrades/770002bb02322e04fa25345838ce6e82af285a0b/upgrade.properties

@@ -0,0 +1,8 @@
+description: Add source_file_name
+compatibility: backwards
+source_file_uses_trap.rel: run source_file_uses_trap.ql
+source_file_name.rel: delete
+tag_name.rel: delete
+trap_uses_tag.rel: delete
+in_trap.rel: run in_trap.ql
+in_trap_or_tag.rel: delete

cpp/downgrades/7e7c2f55670f8123d514cf542ccb1938118ac561/upgrade.properties

@@ -0,0 +1,5 @@
+description: Add trap_filename, source_file_uses_trap and in_trap relations
+compatibility: full
+trap_filename.rel: delete
+source_file_uses_trap.rel: delete
+in_trap.rel: delete

cpp/ql/integration-tests/query-suite/cpp-code-scanning.qls.expected

@@ -7,10 +7,12 @@ ql/cpp/ql/src/Diagnostics/ExtractedFiles.ql
 ql/cpp/ql/src/Diagnostics/ExtractionWarnings.ql
 ql/cpp/ql/src/Diagnostics/FailedExtractorInvocations.ql
 ql/cpp/ql/src/Likely Bugs/Arithmetic/BadAdditionOverflowCheck.ql
+ql/cpp/ql/src/Likely Bugs/Arithmetic/IntMultToLong.ql
 ql/cpp/ql/src/Likely Bugs/Arithmetic/SignedOverflowCheck.ql
 ql/cpp/ql/src/Likely Bugs/Conversion/CastArrayPointerArithmetic.ql
 ql/cpp/ql/src/Likely Bugs/Format/SnprintfOverflow.ql
 ql/cpp/ql/src/Likely Bugs/Format/WrongNumberOfFormatArguments.ql
+ql/cpp/ql/src/Likely Bugs/Format/WrongTypeFormatArguments.ql
 ql/cpp/ql/src/Likely Bugs/Memory Management/AllocaInLoop.ql
 ql/cpp/ql/src/Likely Bugs/Memory Management/PointerOverflow.ql
 ql/cpp/ql/src/Likely Bugs/Memory Management/ReturnStackAllocatedMemory.ql
@@ -28,6 +30,7 @@ ql/cpp/ql/src/Security/CWE/CWE-120/VeryLikelyOverrunWrite.ql
 ql/cpp/ql/src/Security/CWE/CWE-131/NoSpaceForZeroTerminator.ql
 ql/cpp/ql/src/Security/CWE/CWE-134/UncontrolledFormatString.ql
 ql/cpp/ql/src/Security/CWE/CWE-190/ArithmeticUncontrolled.ql
+ql/cpp/ql/src/Security/CWE/CWE-190/ComparisonWithWiderType.ql
 ql/cpp/ql/src/Security/CWE/CWE-191/UnsignedDifferenceExpressionComparedZero.ql
 ql/cpp/ql/src/Security/CWE/CWE-253/HResultBooleanConversion.ql
 ql/cpp/ql/src/Security/CWE/CWE-311/CleartextFileWrite.ql
@@ -40,6 +43,7 @@ ql/cpp/ql/src/Security/CWE/CWE-367/TOCTOUFilesystemRace.ql
 ql/cpp/ql/src/Security/CWE/CWE-416/IteratorToExpiredContainer.ql
 ql/cpp/ql/src/Security/CWE/CWE-416/UseOfStringAfterLifetimeEnds.ql
 ql/cpp/ql/src/Security/CWE/CWE-416/UseOfUniquePointerAfterLifetimeEnds.ql
+ql/cpp/ql/src/Security/CWE/CWE-468/SuspiciousAddWithSizeof.ql
 ql/cpp/ql/src/Security/CWE/CWE-497/ExposedSystemData.ql
 ql/cpp/ql/src/Security/CWE/CWE-611/XXE.ql
 ql/cpp/ql/src/Security/CWE/CWE-676/DangerousFunctionOverflow.ql
@@ -52,5 +56,6 @@ ql/cpp/ql/src/Summary/LinesOfUserCode.ql
 ql/cpp/ql/src/Telemetry/CompilerErrors.ql
 ql/cpp/ql/src/Telemetry/DatabaseQuality.ql
 ql/cpp/ql/src/Telemetry/ExtractionMetrics.ql
+ql/cpp/ql/src/Telemetry/ExtractorInformation.ql
 ql/cpp/ql/src/Telemetry/MissingIncludes.ql
 ql/cpp/ql/src/Telemetry/SucceededIncludes.ql

cpp/ql/integration-tests/query-suite/cpp-security-and-quality.qls.expected

@@ -160,6 +160,7 @@ ql/cpp/ql/src/Summary/LinesOfUserCode.ql
 ql/cpp/ql/src/Telemetry/CompilerErrors.ql
 ql/cpp/ql/src/Telemetry/DatabaseQuality.ql
 ql/cpp/ql/src/Telemetry/ExtractionMetrics.ql
+ql/cpp/ql/src/Telemetry/ExtractorInformation.ql
 ql/cpp/ql/src/Telemetry/MissingIncludes.ql
 ql/cpp/ql/src/Telemetry/SucceededIncludes.ql
 ql/cpp/ql/src/jsf/4.06 Pre-Processing Directives/AV Rule 32.ql

cpp/ql/integration-tests/query-suite/cpp-security-extended.qls.expected

@@ -93,5 +93,6 @@ ql/cpp/ql/src/Summary/LinesOfUserCode.ql
 ql/cpp/ql/src/Telemetry/CompilerErrors.ql
 ql/cpp/ql/src/Telemetry/DatabaseQuality.ql
 ql/cpp/ql/src/Telemetry/ExtractionMetrics.ql
+ql/cpp/ql/src/Telemetry/ExtractorInformation.ql
 ql/cpp/ql/src/Telemetry/MissingIncludes.ql
 ql/cpp/ql/src/Telemetry/SucceededIncludes.ql

cpp/ql/lib/CHANGELOG.md

@@ -1,3 +1,69 @@
+## 10.0.0
+
+### Breaking Changes
+
+* The deprecated `NonThrowingFunction` class has been removed, use `NonCppThrowingFunction` instead.
+* The deprecated `ThrowingFunction` class has been removed, use `AlwaysSehThrowingFunction` instead.
+
+### New Features
+
+* Added a subclass `AutoconfConfigureTestFile` of `ConfigurationTestFile` that represents files created by GNU autoconf configure scripts to test the build configuration.
+
+## 9.0.0
+
+### Breaking Changes
+
+* The `SourceModelCsv`, `SinkModelCsv`, and `SummaryModelCsv` classes and the associated CSV parsing infrastructure have been removed from `ExternalFlow.qll`. New models should be added as `.model.yml` files in the `ext/` directory.
+
+### New Features
+
+* Added a subclass `MesonPrivateTestFile` of `ConfigurationTestFile` that represents files created by Meson to test the build configuration.
+* Added a class `ConstructorDirectFieldInit` to represent field initializations that occur in member initializer lists.
+* Added a class `ConstructorDefaultFieldInit` to represent default field initializations.
+* Added a class `DataFlow::IndirectParameterNode` to represent the indirection of a parameter as a dataflow node.
+* Added a predicate `Node::asIndirectInstruction` which returns the `Instruction` that defines the indirect dataflow node, if any.
+* Added a class `IndirectUninitializedNode` to represent the indirection of an uninitialized local variable as a dataflow node.
+
+### Minor Analysis Improvements
+
+* Added `HttpReceiveHttpRequest`, `HttpReceiveRequestEntityBody`, and `HttpReceiveClientCertificate` from Win32's `http.h` as remote flow sources.
+* Added dataflow through members initialized via non-static data member initialization (NSDMI).
+
+## 8.0.3
+
+No user-facing changes.
+
+## 8.0.2
+
+No user-facing changes.
+
+## 8.0.1
+
+### Minor Analysis Improvements
+
+* Inline expectations test comments, which are of the form `// $ tag` or `// $ tag=value`, are now parsed more strictly and will not be recognized if there isn't a space after the `$` symbol.
+
+## 8.0.0
+
+### Breaking Changes
+
+* CodeQL version 2.24.2 accidentally introduced a syntactical breaking change to `BarrierGuard<...>::getAnIndirectBarrierNode` and `InstructionBarrierGuard<...>::getAnIndirectBarrierNode`. These breaking changes have now been reverted so that the original code compiles again.
+* `MustFlow`, the inter-procedural must-flow data flow analysis library, has been re-worked to use parameterized modules. Like in the case of data flow and taint tracking, instead of extending the `MustFlowConfiguration` class, the user should now implement a module with the `MustFlow::ConfigSig` signature, and instantiate the `MustFlow::Global` parameterized module with the implemented module.
+
+### Minor Analysis Improvements
+
+* Refactored the "Year field changed using an arithmetic operation without checking for leap year" query (`cpp/leap-year/unchecked-after-arithmetic-year-modification`) to address large numbers of false positive results.
+
+### Bug Fixes
+
+* The `allowInterproceduralFlow` predicate of must-flow data flow configurations now correctly handles direct recursion.
+
+## 7.1.1
+
+### Minor Analysis Improvements
+
+* Added remote flow source models for the `winhttp.h` windows header and the Azure SDK core library for C/C++.
+
 ## 7.1.0
 
 ### New Features

cpp/ql/lib/change-notes/released/10.0.0.md

@@ -0,0 +1,10 @@
+## 10.0.0
+
+### Breaking Changes
+
+* The deprecated `NonThrowingFunction` class has been removed, use `NonCppThrowingFunction` instead.
+* The deprecated `ThrowingFunction` class has been removed, use `AlwaysSehThrowingFunction` instead.
+
+### New Features
+
+* Added a subclass `AutoconfConfigureTestFile` of `ConfigurationTestFile` that represents files created by GNU autoconf configure scripts to test the build configuration.

cpp/ql/lib/change-notes/released/7.1.1.md

@@ -0,0 +1,5 @@
+## 7.1.1
+
+### Minor Analysis Improvements
+
+* Added remote flow source models for the `winhttp.h` windows header and the Azure SDK core library for C/C++.

cpp/ql/lib/change-notes/released/8.0.0.md

@@ -0,0 +1,14 @@
+## 8.0.0
+
+### Breaking Changes
+
+* CodeQL version 2.24.2 accidentally introduced a syntactical breaking change to `BarrierGuard<...>::getAnIndirectBarrierNode` and `InstructionBarrierGuard<...>::getAnIndirectBarrierNode`. These breaking changes have now been reverted so that the original code compiles again.
+* `MustFlow`, the inter-procedural must-flow data flow analysis library, has been re-worked to use parameterized modules. Like in the case of data flow and taint tracking, instead of extending the `MustFlowConfiguration` class, the user should now implement a module with the `MustFlow::ConfigSig` signature, and instantiate the `MustFlow::Global` parameterized module with the implemented module.
+
+### Minor Analysis Improvements
+
+* Refactored the "Year field changed using an arithmetic operation without checking for leap year" query (`cpp/leap-year/unchecked-after-arithmetic-year-modification`) to address large numbers of false positive results.
+
+### Bug Fixes
+
+* The `allowInterproceduralFlow` predicate of must-flow data flow configurations now correctly handles direct recursion.

cpp/ql/lib/change-notes/released/8.0.1.md

@@ -0,0 +1,5 @@
+## 8.0.1
+
+### Minor Analysis Improvements
+
+* Inline expectations test comments, which are of the form `// $ tag` or `// $ tag=value`, are now parsed more strictly and will not be recognized if there isn't a space after the `$` symbol.

cpp/ql/lib/change-notes/released/8.0.2.md

@@ -0,0 +1,3 @@
+## 8.0.2
+
+No user-facing changes.

cpp/ql/lib/change-notes/released/8.0.3.md

@@ -0,0 +1,3 @@
+## 8.0.3
+
+No user-facing changes.

cpp/ql/lib/change-notes/released/9.0.0.md

@@ -0,0 +1,19 @@
+## 9.0.0
+
+### Breaking Changes
+
+* The `SourceModelCsv`, `SinkModelCsv`, and `SummaryModelCsv` classes and the associated CSV parsing infrastructure have been removed from `ExternalFlow.qll`. New models should be added as `.model.yml` files in the `ext/` directory.
+
+### New Features
+
+* Added a subclass `MesonPrivateTestFile` of `ConfigurationTestFile` that represents files created by Meson to test the build configuration.
+* Added a class `ConstructorDirectFieldInit` to represent field initializations that occur in member initializer lists.
+* Added a class `ConstructorDefaultFieldInit` to represent default field initializations.
+* Added a class `DataFlow::IndirectParameterNode` to represent the indirection of a parameter as a dataflow node.
+* Added a predicate `Node::asIndirectInstruction` which returns the `Instruction` that defines the indirect dataflow node, if any.
+* Added a class `IndirectUninitializedNode` to represent the indirection of an uninitialized local variable as a dataflow node.
+
+### Minor Analysis Improvements
+
+* Added `HttpReceiveHttpRequest`, `HttpReceiveRequestEntityBody`, and `HttpReceiveClientCertificate` from Win32's `http.h` as remote flow sources.
+* Added dataflow through members initialized via non-static data member initialization (NSDMI).

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

@@ -1,2 +1,2 @@
 ---
-lastReleaseVersion: 7.1.0
+lastReleaseVersion: 10.0.0

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

@@ -24,6 +24,16 @@ extensions:
       - ["", "", False, "MapViewOfFileNuma2", "", "", "ReturnValue[*]", "local", "manual"]
       # ntifs.h
       - ["", "", False, "NtReadFile", "", "", "Argument[*5]", "local", "manual"]
+      # winhttp.h
+      - ["", "", False, "WinHttpReadData", "", "", "Argument[*1]", "remote", "manual"]
+      - ["", "", False, "WinHttpReadDataEx", "", "", "Argument[*1]", "remote", "manual"]
+      - ["", "", False, "WinHttpQueryHeaders", "", "", "Argument[*3]", "remote", "manual"]
+      - ["", "", False, "WinHttpQueryHeadersEx", "", "", "Argument[*5]", "remote", "manual"]
+      - ["", "", False, "WinHttpQueryHeadersEx", "", "", "Argument[*6]", "remote", "manual"]
+      - ["", "", False, "WinHttpQueryHeadersEx", "", "", "Argument[**8]", "remote", "manual"]
+      - ["", "", False, "HttpReceiveHttpRequest", "", "", "Argument[*3]", "remote", "manual"]
+      - ["", "", False, "HttpReceiveRequestEntityBody", "", "", "Argument[*3]", "remote", "manual"]
+      - ["", "", False, "HttpReceiveClientCertificate", "", "", "Argument[*3]", "remote", "manual"]
   - addsTo:
       pack: codeql/cpp-all
       extensible: summaryModel
@@ -46,4 +56,6 @@ extensions:
       - ["", "", False, "RtlMoveMemory", "", "", "Argument[*@1]", "Argument[*@0]", "value", "manual"]
       - ["", "", False, "RtlMoveVolatileMemory", "", "", "Argument[*@1]", "Argument[*@0]", "value", "manual"]
       # winternl.h
-      - ["", "", False, "RtlInitUnicodeString", "", "", "Argument[*1]", "Argument[*0].Field[*Buffer]", "value", "manual"]
+      - ["", "", False, "RtlInitUnicodeString", "", "", "Argument[*1]", "Argument[*0].Field[*Buffer]", "value", "manual"]
+      # winhttp.h
+      - ["", "", False, "WinHttpCrackUrl", "", "", "Argument[*0]", "Argument[*3]", "taint", "manual"]

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

@@ -0,0 +1,22 @@
+# ZeroMQ networking library models
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: sourceModel
+    data: # namespace, type, subtypes, name, signature, ext, output, kind, provenance
+      - ["", "", False, "zmq_recv", "", "", "Argument[*1]", "remote", "manual"]
+      - ["", "", False, "zmq_recvmsg", "", "", "Argument[*1]", "remote", "manual"]
+      - ["", "", False, "zmq_msg_recv", "", "", "Argument[*0]", "remote", "manual"]
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: sinkModel
+    data: # namespace, type, subtypes, name, signature, ext, input, kind, provenance
+      - ["", "", False, "zmq_send", "", "", "Argument[*1]", "remote-sink", "manual"]
+      - ["", "", False, "zmq_sendmsg", "", "", "Argument[*1]", "remote-sink", "manual"]
+      - ["", "", False, "zmq_msg_send", "", "", "Argument[*0]", "remote-sink", "manual"]
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: summaryModel
+    data: # namespace, type, subtypes, name, signature, ext, input, output, kind, provenance
+      - ["", "", False, "zmq_msg_init_data", "", "", "Argument[*1]", "Argument[*0]", "taint", "manual"]
+      - ["", "", False, "zmq_msg_data", "", "", "Argument[*0]", "ReturnValue[*]", "taint", "manual"]

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

@@ -12,4 +12,7 @@ extensions:
       - ["", "", False, "_malloca", "0", "", "", False]
       - ["", "", False, "calloc", "1", "0", "", True]
       - ["std", "", False, "calloc", "1", "0", "", True]
-      - ["bsl", "", False, "calloc", "1", "0", "", True]
+      - ["bsl", "", False, "calloc", "1", "0", "", True]
+      - ["", "", False, "aligned_alloc", "1", "", "", True]
+      - ["std", "", False, "aligned_alloc", "1", "", "", True]
+      - ["bsl", "", False, "aligned_alloc", "1", "", "", True]

cpp/ql/lib/ext/azure.core.model.yml

@@ -0,0 +1,41 @@
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: sourceModel
+    data: # namespace, type, subtypes, name, signature, ext, output, kind, provenance
+      - ["Azure::Core::Http", "RawResponse", True, "GetHeaders", "", "", "ReturnValue[*]", "remote", "manual"]
+      - ["Azure::Core::Http", "RawResponse", True, "GetBody", "", "", "ReturnValue[*]", "remote", "manual"]
+      - ["Azure::Core::Http", "RawResponse", True, "ExtractBodyStream", "", "", "ReturnValue[*]", "remote", "manual"]
+      - ["Azure::Core::Http", "Request", True, "GetHeaders", "", "", "ReturnValue", "remote", "manual"]
+      - ["Azure::Core::Http", "Request", True, "GetHeader", "", "", "ReturnValue", "remote", "manual"]
+      - ["Azure::Core::Http", "Request", True, "GetBodyStream", "", "", "ReturnValue[*]", "remote", "manual"]
+
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: summaryModel
+    data: # namespace, type, subtypes, name, signature, ext, input, output, kind, provenance
+      - ["Azure::Core", "Url", True, "Url", "", "", "Argument[*0]", "Argument[-1]", "taint", "manual"]
+      - ["Azure::Core", "Url", True, "SetScheme", "", "", "Argument[*0]", "Argument[-1]", "taint", "manual"]
+      - ["Azure::Core", "Url", True, "SetHost", "", "", "Argument[*0]", "Argument[-1]", "taint", "manual"]
+      - ["Azure::Core", "Url", True, "SetPort", "", "", "Argument[0]", "Argument[-1]", "taint", "manual"]
+      - ["Azure::Core", "Url", True, "SetPath", "", "", "Argument[*0]", "Argument[-1]", "taint", "manual"]
+      - ["Azure::Core", "Url", True, "SetQueryParameters", "", "", "Argument[0]", "Argument[-1]", "taint", "manual"]
+      - ["Azure::Core", "Url", True, "AppendPath", "", "", "Argument[*0]", "Argument[-1]", "taint", "manual"]
+      - ["Azure::Core", "Url", True, "AppendQueryParameter", "", "", "Argument[*1]", "Argument[-1]", "taint", "manual"]
+      - ["Azure::Core", "Url", True, "GetHost", "", "", "Argument[-1]", "ReturnValue[*]", "taint", "manual"]
+      - ["Azure::Core", "Url", True, "GetPath", "", "", "Argument[-1]", "ReturnValue[*]", "taint", "manual"]
+      - ["Azure::Core", "Url", True, "GetPort", "", "", "Argument[-1]", "ReturnValue", "taint", "manual"]
+      - ["Azure::Core", "Url", True, "GetQueryParameters", "", "", "Argument[-1]", "ReturnValue", "taint", "manual"]
+      - ["Azure::Core", "Url", True, "GetScheme", "", "", "Argument[-1]", "ReturnValue[*]", "taint", "manual"]
+      - ["Azure::Core", "Url", True, "GetRelativeUrl", "", "", "Argument[-1]", "ReturnValue", "taint", "manual"]
+      - ["Azure::Core", "Url", True, "GetAbsoluteUrl", "", "", "Argument[-1]", "ReturnValue", "taint", "manual"]
+      - ["Azure::Core", "Url", True, "Decode", "", "", "Argument[*0]", "ReturnValue", "taint", "manual"]
+      - ["Azure::Core", "Url", True, "Encode", "", "", "Argument[*0]", "ReturnValue", "taint", "manual"]
+      - ["Azure::Core::IO", "BodyStream", True, "Read", "", "", "Argument[-1]", "Argument[*0]", "taint", "manual"]
+      - ["Azure::Core::IO", "BodyStream", True, "ReadToCount", "", "", "Argument[-1]", "Argument[*0]", "taint", "manual"]
+      - ["Azure::Core::IO", "BodyStream", True, "ReadToEnd", "", "", "Argument[-1]", "ReturnValue.Element", "taint", "manual"]
+      - ["Azure", "Nullable", True, "Nullable", "", "", "Argument[0]", "Argument[-1]", "taint", "manual"]
+      - ["Azure", "Nullable", True, "operator=", "", "", "Argument[*0]", "Argument[-1]", "value", "manual"]
+      - ["Azure", "Nullable", True, "Value", "", "", "Argument[-1]", "ReturnValue[*]", "taint", "manual"]
+      - ["Azure", "Nullable", True, "operator->", "", "", "Argument[-1]", "ReturnValue[*]", "taint", "manual"]
+      - ["Azure", "Nullable", True, "operator*", "", "", "Argument[-1]", "ReturnValue[*]", "taint", "manual"]

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

@@ -0,0 +1,19 @@
+# Models for getc and similar character-reading functions
+extensions:
+  - addsTo:
+      pack: codeql/cpp-all
+      extensible: sourceModel
+    data: # namespace, type, subtypes, name, signature, ext, output, kind, provenance
+      - ["", "", False, "getc", "", "", "ReturnValue", "remote", "manual"]
+      - ["", "", False, "getwc", "", "", "ReturnValue", "remote", "manual"]
+      - ["", "", False, "_getc_nolock", "", "", "ReturnValue", "remote", "manual"]
+      - ["", "", False, "_getwc_nolock", "", "", "ReturnValue", "remote", "manual"]
+      - ["", "", False, "getch", "", "", "ReturnValue", "local", "manual"]
+      - ["", "", False, "_getch", "", "", "ReturnValue", "local", "manual"]
+      - ["", "", False, "_getwch", "", "", "ReturnValue", "local", "manual"]
+      - ["", "", False, "_getch_nolock", "", "", "ReturnValue", "local", "manual"]
+      - ["", "", False, "_getwch_nolock", "", "", "ReturnValue", "local", "manual"]
+      - ["", "", False, "getchar", "", "", "ReturnValue", "local", "manual"]
+      - ["", "", False, "getwchar", "", "", "ReturnValue", "local", "manual"]
+      - ["", "", False, "_getchar_nolock", "", "", "ReturnValue", "local", "manual"]
+      - ["", "", False, "_getwchar_nolock", "", "", "ReturnValue", "local", "manual"]

cpp/ql/lib/qlpack.yml

@@ -1,5 +1,5 @@
 name: codeql/cpp-all
-version: 7.1.0
+version: 10.0.0
 groups: cpp
 dbscheme: semmlecode.cpp.dbscheme
 extractor: cpp

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

@@ -26,3 +26,26 @@ class CmakeTryCompileFile extends ConfigurationTestFile {
     )
   }
 }
+
+/**
+ * A file created by Meson to test the system configuration.
+ */
+class MesonPrivateTestFile extends ConfigurationTestFile {
+  MesonPrivateTestFile() {
+    this.getBaseName() = "testfile.c" and
+    exists(Folder folder, Folder parent |
+      folder = this.getParentContainer() and
+      parent = folder.getParentContainer()
+    |
+      folder.getBaseName().matches("tmp%") and
+      parent.getBaseName() = "meson-private"
+    )
+  }
+}
+
+/**
+ * A file created by a GNU autoconf configure script to test the system configuration.
+ */
+class AutoconfConfigureTestFile extends ConfigurationTestFile {
+  AutoconfConfigureTestFile() { this.getBaseName().regexpMatch("conftest[0-9]*\\.c(pp)?") }
+}

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

@@ -524,6 +524,12 @@ class Function extends Declaration, ControlFlowNode, AccessHolder, @function {
       not exists(NewOrNewArrayExpr new | e = new.getAllocatorCall().getArgument(0))
     )
   }
+
+  /**
+   * Holds if this function has an ambiguous return type, meaning that zero or multiple return
+   * types for this function are present in the database (this can occur in `build-mode: none`).
+   */
+  predicate hasAmbiguousReturnType() { count(this.getType()) != 1 }
 }
 
 pragma[noinline]

cpp/ql/lib/semmle/code/cpp/commons/DateTime.qll

@@ -14,7 +14,9 @@ class PackedTimeType extends Type {
   }
 }
 
-private predicate timeType(string typeName) { typeName = ["_SYSTEMTIME", "SYSTEMTIME", "tm"] }
+private predicate timeType(string typeName) {
+  typeName = ["_SYSTEMTIME", "SYSTEMTIME", "tm", "TIME_FIELDS", "_TIME_FIELDS", "PTIME_FIELDS"]
+}
 
 /**
  * A type that is used to represent times and dates in an 'unpacked' form, that is,
@@ -95,3 +97,24 @@ class StructTmMonthFieldAccess extends MonthFieldAccess {
 class StructTmYearFieldAccess extends YearFieldAccess {
   StructTmYearFieldAccess() { this.getTarget().getName() = "tm_year" }
 }
+
+/**
+ * A `DayFieldAccess` for the `TIME_FIELDS` struct.
+ */
+class TimeFieldsDayFieldAccess extends DayFieldAccess {
+  TimeFieldsDayFieldAccess() { this.getTarget().getName() = "Day" }
+}
+
+/**
+ * A `MonthFieldAccess` for the `TIME_FIELDS` struct.
+ */
+class TimeFieldsMonthFieldAccess extends MonthFieldAccess {
+  TimeFieldsMonthFieldAccess() { this.getTarget().getName() = "Month" }
+}
+
+/**
+ * A `YearFieldAccess` for the `TIME_FIELDS` struct.
+ */
+class TimeFieldsYearFieldAccess extends YearFieldAccess {
+  TimeFieldsYearFieldAccess() { this.getTarget().getName() = "Year" }
+}

cpp/ql/lib/semmle/code/cpp/commons/Printf.qll

@@ -163,12 +163,23 @@ predicate primitiveVariadicFormatter(
   )
 }
 
+/**
+ * Gets a function call whose target is a variadic formatter with the given
+ * `type`, `format` parameter index and `output` parameter index.
+ *
+ * Join-order helper for `callsVariadicFormatter`.
+ */
+pragma[nomagic]
+private predicate callsVariadicFormatterCall(FunctionCall fc, string type, int format, int output) {
+  variadicFormatter(fc.getTarget(), type, format, output)
+}
+
 private predicate callsVariadicFormatter(
   Function f, string type, int formatParamIndex, int outputParamIndex
 ) {
   // calls a variadic formatter with `formatParamIndex`, `outputParamIndex` linked
   exists(FunctionCall fc, int format, int output |
-    variadicFormatter(pragma[only_bind_into](fc.getTarget()), type, format, output) and
+    callsVariadicFormatterCall(fc, type, format, output) and
     fc.getEnclosingFunction() = f and
     fc.getArgument(format) = f.getParameter(formatParamIndex).getAnAccess() and
     fc.getArgument(output) = f.getParameter(outputParamIndex).getAnAccess()
@@ -176,7 +187,7 @@ private predicate callsVariadicFormatter(
   or
   // calls a variadic formatter with only `formatParamIndex` linked
   exists(FunctionCall fc, string calledType, int format, int output |
-    variadicFormatter(pragma[only_bind_into](fc.getTarget()), calledType, format, output) and
+    callsVariadicFormatterCall(fc, calledType, format, output) and
     fc.getEnclosingFunction() = f and
     fc.getArgument(format) = f.getParameter(formatParamIndex).getAnAccess() and
     not fc.getArgument(output) = f.getParameter(_).getAnAccess() and
@@ -448,6 +459,13 @@ class FormatLiteral extends Literal instanceof StringLiteral {
    */
   int getConvSpecOffset(int n) { result = this.getFormat().indexOf("%", n, 0) }
 
+  /**
+   * Gets the nth conversion specifier string.
+   */
+  private string getConvSpecString(int n) {
+    n >= 0 and result = "%" + this.getFormat().splitAt("%", n + 1)
+  }
+
   /*
    * Each of these predicates gets a regular expressions to match each individual
    * parts of a conversion specifier.
@@ -513,22 +531,20 @@ class FormatLiteral extends Literal instanceof StringLiteral {
     int n, string spec, string params, string flags, string width, string prec, string len,
     string conv
   ) {
-    exists(int offset, string fmt, string rst, string regexp |
-      offset = this.getConvSpecOffset(n) and
-      fmt = this.getFormat() and
-      rst = fmt.substring(offset, fmt.length()) and
+    exists(string convSpec, string regexp |
+      convSpec = this.getConvSpecString(n) and
       regexp = this.getConvSpecRegexp() and
       (
-        spec = rst.regexpCapture(regexp, 1) and
-        params = rst.regexpCapture(regexp, 2) and
-        flags = rst.regexpCapture(regexp, 3) and
-        width = rst.regexpCapture(regexp, 4) and
-        prec = rst.regexpCapture(regexp, 5) and
-        len = rst.regexpCapture(regexp, 6) and
-        conv = rst.regexpCapture(regexp, 7)
+        spec = convSpec.regexpCapture(regexp, 1) and
+        params = convSpec.regexpCapture(regexp, 2) and
+        flags = convSpec.regexpCapture(regexp, 3) and
+        width = convSpec.regexpCapture(regexp, 4) and
+        prec = convSpec.regexpCapture(regexp, 5) and
+        len = convSpec.regexpCapture(regexp, 6) and
+        conv = convSpec.regexpCapture(regexp, 7)
         or
-        spec = rst.regexpCapture(regexp, 1) and
-        not exists(rst.regexpCapture(regexp, 2)) and
+        spec = convSpec.regexpCapture(regexp, 1) and
+        not exists(convSpec.regexpCapture(regexp, 2)) and
         params = "" and
         flags = "" and
         width = "" and
@@ -543,12 +559,10 @@ class FormatLiteral extends Literal instanceof StringLiteral {
    * Gets the nth conversion specifier (including the initial `%`).
    */
   string getConvSpec(int n) {
-    exists(int offset, string fmt, string rst, string regexp |
-      offset = this.getConvSpecOffset(n) and
-      fmt = this.getFormat() and
-      rst = fmt.substring(offset, fmt.length()) and
+    exists(string convSpec, string regexp |
+      convSpec = this.getConvSpecString(n) and
       regexp = this.getConvSpecRegexp() and
-      result = rst.regexpCapture(regexp, 1)
+      result = convSpec.regexpCapture(regexp, 1)
     )
   }
 

cpp/ql/lib/semmle/code/cpp/commons/Scanf.qll

@@ -194,6 +194,13 @@ class ScanfFormatLiteral extends Expr {
     )
   }
 
+  /**
+   * Gets the nth conversion specifier string.
+   */
+  private string getConvSpecString(int n) {
+    n >= 0 and result = "%" + this.getFormat().splitAt("%", n + 1)
+  }
+
   /**
    * Gets the regular expression to match each individual part of a conversion specifier.
    */
@@ -227,16 +234,14 @@ class ScanfFormatLiteral extends Expr {
    * specifier.
    */
   predicate parseConvSpec(int n, string spec, string width, string len, string conv) {
-    exists(int offset, string fmt, string rst, string regexp |
-      offset = this.getConvSpecOffset(n) and
-      fmt = this.getFormat() and
-      rst = fmt.substring(offset, fmt.length()) and
+    exists(string convSpec, string regexp |
+      convSpec = this.getConvSpecString(n) and
       regexp = this.getConvSpecRegexp() and
       (
-        spec = rst.regexpCapture(regexp, 1) and
-        width = rst.regexpCapture(regexp, 2) and
-        len = rst.regexpCapture(regexp, 3) and
-        conv = rst.regexpCapture(regexp, 4)
+        spec = convSpec.regexpCapture(regexp, 1) and
+        width = convSpec.regexpCapture(regexp, 2) and
+        len = convSpec.regexpCapture(regexp, 3) and
+        conv = convSpec.regexpCapture(regexp, 4)
       )
     )
   }

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

@@ -1663,7 +1663,7 @@ private module Cached {
   private predicate compares_ge(
     ValueNumber test, Operand left, Operand right, int k, boolean isGe, GuardValue value
   ) {
-    exists(int onemk | k = 1 - onemk | compares_lt(test, right, left, onemk, isGe, value))
+    compares_lt(test, right, left, 1 - k, isGe, value)
   }
 
   /** Rearrange various simple comparisons into `left < right + k` form. */

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

@@ -1,15 +1,20 @@
 /**
  * INTERNAL use only. This is an experimental API subject to change without notice.
  *
- * Provides classes and predicates for dealing with flow models specified in CSV format.
+ * Provides classes and predicates for dealing with flow models specified
+ * in data extension files.
  *
- * The CSV specification has the following columns:
+ * The extensible relations have the following columns:
  * - Sources:
- *   `namespace; type; subtypes; name; signature; ext; output; kind`
+ *   `namespace; type; subtypes; name; signature; ext; output; kind; provenance`
  * - Sinks:
- *   `namespace; type; subtypes; name; signature; ext; input; kind`
+ *   `namespace; type; subtypes; name; signature; ext; input; kind; provenance`
  * - Summaries:
- *   `namespace; type; subtypes; name; signature; ext; input; output; kind`
+ *   `namespace; type; subtypes; name; signature; ext; input; output; kind; provenance`
+ * - Barriers:
+ *   `namespace; type; subtypes; name; signature; ext; output; kind; provenance`
+ * - BarrierGuards:
+ *   `namespace; type; subtypes; name; signature; ext; input; acceptingValue; kind; provenance`
  *
  * The interpretation of a row is similar to API-graphs with a left-to-right
  * reading.
@@ -86,11 +91,23 @@
  *    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
+ * 8. The `acceptingValue` column of barrier guard models specifies the condition
+ *    under which the guard blocks flow. It can be one of "true" or "false". In
+ *    the future "no-exception", "not-zero", "null", "not-null" may be supported.
+ * 9. 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
  *    "taint" indicates a default additional taint step and "value" indicates a
  *    globally applicable value-preserving step.
+ * 10. The `provenance` column is a tag to indicate the origin and verification of a model.
+ *    The format is {origin}-{verification} or just "manual" where the origin describes
+ *    the origin of the model and verification describes how the model has been verified.
+ *    Some examples are:
+ *    - "df-generated": The model has been generated by the model generator tool.
+ *    - "df-manual": The model has been generated by the model generator and verified by a human.
+ *    - "manual": The model has been written by hand.
+ *    This information is used in a heuristic for dataflow analysis to determine, if a
+ *    model or source code should be used for determining flow.
  */
 
 import cpp
@@ -104,117 +121,9 @@ private import internal.FlowSummaryImpl::Private
 private import internal.FlowSummaryImpl::Private::External
 private import internal.ExternalFlowExtensions::Extensions as Extensions
 private import codeql.mad.ModelValidation as SharedModelVal
-private import codeql.util.Unit
 private import codeql.mad.static.ModelsAsData as SharedMaD
 
-/**
- * A unit class for adding additional source model rows.
- *
- * Extend this class to add additional source definitions.
- */
-class SourceModelCsv extends Unit {
-  /** Holds if `row` specifies a source definition. */
-  abstract predicate row(string row);
-}
-
-/**
- * A unit class for adding additional sink model rows.
- *
- * Extend this class to add additional sink definitions.
- */
-class SinkModelCsv extends Unit {
-  /** Holds if `row` specifies a sink definition. */
-  abstract predicate row(string row);
-}
-
-/**
- * A unit class for adding additional summary model rows.
- *
- * Extend this class to add additional flow summary definitions.
- */
-class SummaryModelCsv extends Unit {
-  /** Holds if `row` specifies a summary definition. */
-  abstract predicate row(string row);
-}
-
-/** Holds if `row` is a source model. */
-predicate sourceModel(string row) { any(SourceModelCsv s).row(row) }
-
-/** Holds if `row` is a sink model. */
-predicate sinkModel(string row) { any(SinkModelCsv s).row(row) }
-
-/** Holds if `row` is a summary model. */
-predicate summaryModel(string row) { any(SummaryModelCsv s).row(row) }
-
 private module MadInput implements SharedMaD::InputSig {
-  /** Holds if a source model exists for the given parameters. */
-  predicate additionalSourceModel(
-    string namespace, string type, boolean subtypes, string name, string signature, string ext,
-    string output, string kind, string provenance, string model
-  ) {
-    exists(string row |
-      sourceModel(row) and
-      row.splitAt(";", 0) = namespace and
-      row.splitAt(";", 1) = type and
-      row.splitAt(";", 2) = subtypes.toString() and
-      subtypes = [true, false] and
-      row.splitAt(";", 3) = name and
-      row.splitAt(";", 4) = signature and
-      row.splitAt(";", 5) = ext and
-      row.splitAt(";", 6) = output and
-      row.splitAt(";", 7) = kind
-    ) and
-    provenance = "manual" and
-    model = ""
-  }
-
-  /** Holds if a sink model exists for the given parameters. */
-  predicate additionalSinkModel(
-    string namespace, string type, boolean subtypes, string name, string signature, string ext,
-    string input, string kind, string provenance, string model
-  ) {
-    exists(string row |
-      sinkModel(row) and
-      row.splitAt(";", 0) = namespace and
-      row.splitAt(";", 1) = type and
-      row.splitAt(";", 2) = subtypes.toString() and
-      subtypes = [true, false] and
-      row.splitAt(";", 3) = name and
-      row.splitAt(";", 4) = signature and
-      row.splitAt(";", 5) = ext and
-      row.splitAt(";", 6) = input and
-      row.splitAt(";", 7) = kind
-    ) and
-    provenance = "manual" and
-    model = ""
-  }
-
-  /**
-   * Holds if a summary model exists for the given parameters.
-   *
-   * This predicate does not expand `@` to `*`s.
-   */
-  predicate additionalSummaryModel(
-    string namespace, string type, boolean subtypes, string name, string signature, string ext,
-    string input, string output, string kind, string provenance, string model
-  ) {
-    exists(string row |
-      summaryModel(row) and
-      row.splitAt(";", 0) = namespace and
-      row.splitAt(";", 1) = type and
-      row.splitAt(";", 2) = subtypes.toString() and
-      subtypes = [true, false] and
-      row.splitAt(";", 3) = name and
-      row.splitAt(";", 4) = signature and
-      row.splitAt(";", 5) = ext and
-      row.splitAt(";", 6) = input and
-      row.splitAt(";", 7) = output and
-      row.splitAt(";", 8) = kind
-    ) and
-    provenance = "manual" and
-    model = ""
-  }
-
   string namespaceSegmentSeparator() { result = "::" }
 }
 
@@ -250,8 +159,8 @@ predicate summaryModel(
   )
 }
 
-/** Provides a query predicate to check the CSV data for validation errors. */
-module CsvValidation {
+/** Provides a query predicate to check the data for validation errors. */
+module ModelValidation {
   private string getInvalidModelInput() {
     exists(string pred, AccessPath input, string part |
       sinkModel(_, _, _, _, _, _, input, _, _, _) and pred = "sink"
@@ -294,40 +203,6 @@ module CsvValidation {
 
   private module KindVal = SharedModelVal::KindValidation<KindValConfig>;
 
-  private string getInvalidModelSubtype() {
-    exists(string pred, string row |
-      sourceModel(row) and pred = "source"
-      or
-      sinkModel(row) and pred = "sink"
-      or
-      summaryModel(row) and pred = "summary"
-    |
-      exists(string b |
-        b = row.splitAt(";", 2) and
-        not b = ["true", "false"] and
-        result = "Invalid boolean \"" + b + "\" in " + pred + " model."
-      )
-    )
-  }
-
-  private string getInvalidModelColumnCount() {
-    exists(string pred, string row, int expect |
-      sourceModel(row) and expect = 8 and pred = "source"
-      or
-      sinkModel(row) and expect = 8 and pred = "sink"
-      or
-      summaryModel(row) and expect = 9 and pred = "summary"
-    |
-      exists(int cols |
-        cols = 1 + max(int n | exists(row.splitAt(";", n))) and
-        cols != expect and
-        result =
-          "Wrong number of columns in " + pred + " model row, expected " + expect + ", got " + cols +
-            "."
-      )
-    )
-  }
-
   private string getInvalidModelSignature() {
     exists(string pred, string namespace, string type, string name, string signature, string ext |
       sourceModel(namespace, type, _, name, signature, ext, _, _, _, _) and pred = "source"
@@ -353,12 +228,25 @@ module CsvValidation {
     )
   }
 
-  /** Holds if some row in a CSV-based flow model appears to contain typos. */
+  private string getIncorrectConstructorSummaryOutput() {
+    exists(string namespace, string type, string name, string output |
+      type = name or
+      type = name + "<" + any(string s)
+    |
+      summaryModel(namespace, type, _, name, _, _, _, output, _, _, _) and
+      output.matches("ReturnValue%") and
+      result =
+        "Constructor model for " + namespace + "." + type +
+          " should use `Argument[this]` in the output, not `ReturnValue`."
+    )
+  }
+
+  /** Holds if some row in a MaD flow model appears to contain typos. */
   query predicate invalidModelRow(string msg) {
     msg =
       [
         getInvalidModelSignature(), getInvalidModelInput(), getInvalidModelOutput(),
-        getInvalidModelSubtype(), getInvalidModelColumnCount(), KindVal::getInvalidModelKind()
+        KindVal::getInvalidModelKind(), getIncorrectConstructorSummaryOutput()
       ]
   }
 }
@@ -555,6 +443,7 @@ private Locatable getSupportedFunctionTemplateArgument(Function templateFunction
  * Normalize the `n`'th parameter of `f` by replacing template names
  * with `func:N` (where `N` is the index of the template).
  */
+pragma[nomagic]
 private string getTypeNameWithoutFunctionTemplates(Function f, int n, int remaining) {
   exists(Function templateFunction |
     templateFunction = getFullyTemplatedFunction(f) and
@@ -1011,7 +900,7 @@ private module Cached {
   }
 
   /**
-   * Holds if `node` is specified as a source with the given kind in a CSV flow
+   * Holds if `node` is specified as a source with the given kind in a MaD flow
    * model.
    */
   cached
@@ -1022,7 +911,7 @@ private module Cached {
   }
 
   /**
-   * Holds if `node` is specified as a sink with the given kind in a CSV flow
+   * Holds if `node` is specified as a sink with the given kind in a MaD flow
    * model.
    */
   cached
@@ -1058,13 +947,13 @@ private module Cached {
 
   private predicate barrierGuardChecks(IRGuardCondition g, Expr e, boolean gv, TKindModelPair kmp) {
     exists(
-      SourceSinkInterpretationInput::InterpretNode n, Public::AcceptingValue acceptingvalue,
+      SourceSinkInterpretationInput::InterpretNode n, Public::AcceptingValue acceptingValue,
       string kind, string model
     |
-      isBarrierGuardNode(n, acceptingvalue, kind, model) and
+      isBarrierGuardNode(n, acceptingValue, kind, model) and
       n.asNode().asExpr() = e and
       kmp = TMkPair(kind, model) and
-      gv = convertAcceptingValue(acceptingvalue).asBooleanValue() and
+      gv = convertAcceptingValue(acceptingValue).asBooleanValue() and
       n.asNode().(Private::ArgumentNode).getCall().asCallInstruction() = g
     )
   }
@@ -1081,14 +970,14 @@ private module Cached {
   ) {
     exists(
       SourceSinkInterpretationInput::InterpretNode interpretNode,
-      Public::AcceptingValue acceptingvalue, string kind, string model, int indirectionIndex,
+      Public::AcceptingValue acceptingValue, string kind, string model, int indirectionIndex,
       Private::ArgumentNode arg
     |
-      isBarrierGuardNode(interpretNode, acceptingvalue, kind, model) and
+      isBarrierGuardNode(interpretNode, acceptingValue, kind, model) and
       arg = interpretNode.asNode() and
       arg.asIndirectExpr(indirectionIndex) = e and
       kmp = MkKindModelPairIntPair(TMkPair(kind, model), indirectionIndex) and
-      gv = convertAcceptingValue(acceptingvalue).asBooleanValue() and
+      gv = convertAcceptingValue(acceptingValue).asBooleanValue() and
       arg.getCall().asCallInstruction() = g
     )
   }

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

@@ -33,7 +33,7 @@ extensible predicate barrierModel(
  */
 extensible predicate barrierGuardModel(
   string namespace, string type, boolean subtypes, string name, string signature, string ext,
-  string input, string acceptingvalue, string kind, string provenance, QlBuiltins::ExtensionId madId
+  string input, string acceptingValue, string kind, string provenance, QlBuiltins::ExtensionId madId
 );
 
 /**

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

@@ -162,13 +162,13 @@ module SourceSinkInterpretationInput implements
   }
 
   predicate barrierGuardElement(
-    Element e, string input, Public::AcceptingValue acceptingvalue, string kind,
+    Element e, string input, Public::AcceptingValue acceptingValue, string kind,
     Public::Provenance provenance, string model
   ) {
     exists(
       string package, string type, boolean subtypes, string name, string signature, string ext
     |
-      barrierGuardModel(package, type, subtypes, name, signature, ext, input, acceptingvalue, kind,
+      barrierGuardModel(package, type, subtypes, name, signature, ext, input, acceptingValue, kind,
         provenance, model) and
       e = interpretElement(package, type, subtypes, name, signature, ext)
     )
@@ -201,7 +201,7 @@ module SourceSinkInterpretationInput implements
     string toString() {
       result = this.asElement().toString()
       or
-      result = this.asNode().toString()
+      result = this.asNode().toStringImpl()
       or
       result = this.asCall().toString()
     }

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

@@ -585,12 +585,15 @@ class ConstructorDelegationInit extends ConstructorBaseInit, @ctordelegatinginit
 
 /**
  * An initialization of a member variable performed as part of a
- * constructor's explicit initializer list or implicit actions.
+ * constructor's initializer list or by default initialization.
+ *
  * In the example below, member variable `b` is being initialized by
- * constructor parameter `a`:
+ * constructor parameter `a`, and `c` is initialized by default
+ * initialization:
  * ```
  * struct S {
  *   int b;
+ *   int c = 3;
  *   S(int a): b(a) {}
  * } s(2);
  * ```
@@ -616,6 +619,28 @@ class ConstructorFieldInit extends ConstructorInit, @ctorfieldinit {
   override predicate mayBeGloballyImpure() { this.getExpr().mayBeGloballyImpure() }
 }
 
+/**
+ * An initialization of a member variable performed as part of a
+ * constructor's explicit initializer list.
+ */
+class ConstructorDirectFieldInit extends ConstructorFieldInit {
+  ConstructorDirectFieldInit() { exists(this.getChild(0)) }
+
+  override string getAPrimaryQlClass() { result = "ConstructorDirectFieldInit" }
+}
+
+/**
+ * An initialization of a member variable performed by default
+ * initialization.
+ */
+class ConstructorDefaultFieldInit extends ConstructorFieldInit {
+  ConstructorDefaultFieldInit() {
+    not exists(this.getChild(0)) and exists(this.getTarget().getInitializer())
+  }
+
+  override string getAPrimaryQlClass() { result = "ConstructorDefaultFieldInit" }
+}
+
 /**
  * A call to a destructor of a base class or field as part of a destructor's
  * compiler-generated actions.

cpp/ql/lib/semmle/code/cpp/internal/Overlay.qll

@@ -6,75 +6,67 @@ private import OverlayXml
 
 /**
  * Holds always for the overlay variant and never for the base variant.
- * This local predicate is used to define local predicates that behave
- * differently for the base and overlay variant.
  */
 overlay[local]
 predicate isOverlay() { databaseMetadata("isOverlay", "true") }
 
-overlay[local]
-private string getLocationFilePath(@location_default loc) {
-  exists(@file file | locations_default(loc, file, _, _, _, _) | files(file, result))
-}
-
 /**
- * Gets the file path for an element with a single location.
+ * Holds if the TRAP file or tag `t` is reachable from source file `sourceFile`
+ * in the base (isOverlayVariant=false) or overlay (isOverlayVariant=true) variant.
  */
 overlay[local]
-private string getSingleLocationFilePath(@element e) {
-  exists(@location_default loc |
-    var_decls(e, _, _, _, loc)
-    or
-    fun_decls(e, _, _, _, loc)
-    or
-    type_decls(e, _, loc)
-    or
-    namespace_decls(e, _, loc, _)
-    or
-    macroinvocations(e, _, loc, _)
-    or
-    preprocdirects(e, _, loc)
-  |
-    result = getLocationFilePath(loc)
+private predicate locallyReachableTrapOrTag(
+  boolean isOverlayVariant, string sourceFile, @trap_or_tag t
+) {
+  exists(@source_file sf, @trap trap |
+    (if isOverlay() then isOverlayVariant = true else isOverlayVariant = false) and
+    source_file_uses_trap(sf, trap) and
+    source_file_name(sf, sourceFile) and
+    (t = trap or trap_uses_tag(trap, t))
   )
 }
 
 /**
- * Gets the file path for an element with potentially multiple locations.
+ * Holds if element `e` is in TRAP file or tag `t`
+ * in the base (isOverlayVariant=false) or overlay (isOverlayVariant=true) variant.
  */
 overlay[local]
-private string getMultiLocationFilePath(@element e) {
-  exists(@location_default loc |
-    var_decls(_, e, _, _, loc)
-    or
-    fun_decls(_, e, _, _, loc)
-    or
-    type_decls(_, e, loc)
-    or
-    namespace_decls(_, e, loc, _)
-  |
-    result = getLocationFilePath(loc)
-  )
+private predicate locallyInTrapOrTag(boolean isOverlayVariant, @element e, @trap_or_tag t) {
+  (if isOverlay() then isOverlayVariant = true else isOverlayVariant = false) and
+  in_trap_or_tag(e, t)
 }
 
-/**
- * A local helper predicate that holds in the base variant and never in the
- * overlay variant.
- */
-overlay[local]
-private predicate isBase() { not isOverlay() }
-
 /**
  * Discards an element from the base variant if:
- * - It has a single location in a changed file, or
- * - All of its locations are in changed files.
+ * - We have knowledge about what TRAP file or tag it is in (in the base).
+ * - It is not in any overlay TRAP file or tag that is reachable from an overlay source file.
+ * - For every base TRAP file or tag that contains it and is reachable from a base source file,
+ *   either the source file has changed, or the overlay has redefined the TRAP file or tag,
+ *   or the overlay runner has re-extracted the same source file.
  */
 overlay[discard_entity]
 private predicate discardElement(@element e) {
-  isBase() and
-  (
-    overlayChangedFiles(getSingleLocationFilePath(e))
-    or
-    forex(string path | path = getMultiLocationFilePath(e) | overlayChangedFiles(path))
+  // If we don't have any knowledge about what TRAP file something
+  // is in, then we don't want to discard it, so we only consider
+  // entities that are known to be in a base TRAP file or tag.
+  locallyInTrapOrTag(false, e, _) and
+  // Anything that is reachable from an overlay source file should
+  // not be discarded.
+  not exists(@trap_or_tag t | locallyInTrapOrTag(true, e, t) |
+    locallyReachableTrapOrTag(true, _, t)
+  ) and
+  // Finally, we have to make sure the base variant does not retain it.
+  // If it is reachable from a base source file, then that is
+  // sufficient unless either the base source file has changed (in
+  // particular, been deleted), or the overlay has redefined the TRAP
+  // file or tag it is in, or the overlay runner has re-extracted the same
+  // source file (e.g. because a header it includes has changed).
+  forall(@trap_or_tag t, string sourceFile |
+    locallyInTrapOrTag(false, e, t) and
+    locallyReachableTrapOrTag(false, sourceFile, t)
+  |
+    overlayChangedFiles(sourceFile) or
+    locallyReachableTrapOrTag(true, _, t) or
+    locallyReachableTrapOrTag(true, sourceFile, _)
   )
 }

cpp/ql/lib/semmle/code/cpp/ir/dataflow/MustFlow.qll

@@ -8,81 +8,143 @@ private import cpp
 private import semmle.code.cpp.ir.IR
 
 /**
- * A configuration of a data flow analysis that performs must-flow analysis. This is different
- * from `DataFlow.qll` which performs may-flow analysis (i.e., it finds paths where the source _may_
- * flow to the sink).
- *
- * Like in `DataFlow.qll`, each use of the `MustFlow.qll` library must define its own unique extension
- * of this abstract class. To create a configuration, extend this class with a subclass whose
- * characteristic predicate is a unique singleton string and override `isSource`, `isSink` (and
- * `isAdditionalFlowStep` if additional steps are required).
+ * Provides an inter-procedural must-flow data flow analysis.
  */
-abstract class MustFlowConfiguration extends string {
-  bindingset[this]
-  MustFlowConfiguration() { any() }
-
+module MustFlow {
   /**
-   * Holds if `source` is a relevant data flow source.
+   * An input configuration of a data flow analysis that performs must-flow analysis. This is different
+   * from `DataFlow.qll` which performs may-flow analysis (i.e., it finds paths where the source _may_
+   * flow to the sink).
    */
-  abstract predicate isSource(Instruction source);
+  signature module ConfigSig {
+    /**
+     * Holds if `source` is a relevant data flow source.
+     */
+    predicate isSource(Instruction source);
 
-  /**
-   * Holds if `sink` is a relevant data flow sink.
-   */
-  abstract predicate isSink(Operand sink);
+    /**
+     * Holds if `sink` is a relevant data flow sink.
+     */
+    predicate isSink(Operand sink);
 
-  /**
-   * Holds if data flow through `instr` is prohibited.
-   */
-  predicate isBarrier(Instruction instr) { none() }
+    /**
+     * Holds if data flow through `instr` is prohibited.
+     */
+    default predicate isBarrier(Instruction instr) { none() }
 
-  /**
-   * Holds if the additional flow step from `node1` to `node2` must be taken
-   * into account in the analysis.
-   */
-  predicate isAdditionalFlowStep(Operand node1, Instruction node2) { none() }
+    /**
+     * Holds if the additional flow step from `node1` to `node2` must be taken
+     * into account in the analysis.
+     */
+    default predicate isAdditionalFlowStep(Operand node1, Instruction node2) { none() }
 
-  /** Holds if this configuration allows flow from arguments to parameters. */
-  predicate allowInterproceduralFlow() { any() }
-
-  /**
-   * Holds if data must flow from `source` to `sink` for this configuration.
-   *
-   * The corresponding paths are generated from the end-points and the graph
-   * included in the module `PathGraph`.
-   */
-  final predicate hasFlowPath(MustFlowPathNode source, MustFlowPathSink sink) {
-    this.isSource(source.getInstruction()) and
-    source.getASuccessor*() = sink
+    /** Holds if this configuration allows flow from arguments to parameters. */
+    default predicate allowInterproceduralFlow() { any() }
   }
-}
 
-/** Holds if `node` flows from a source. */
-pragma[nomagic]
-private predicate flowsFromSource(Instruction node, MustFlowConfiguration config) {
-  not config.isBarrier(node) and
-  (
-    config.isSource(node)
-    or
-    exists(Instruction mid |
-      step(mid, node, config) and
-      flowsFromSource(mid, pragma[only_bind_into](config))
-    )
-  )
-}
+  /**
+   * Constructs a global must-flow computation.
+   */
+  module Global<ConfigSig Config> {
+    import Config
 
-/** Holds if `node` flows to a sink. */
-pragma[nomagic]
-private predicate flowsToSink(Instruction node, MustFlowConfiguration config) {
-  flowsFromSource(node, pragma[only_bind_into](config)) and
-  (
-    config.isSink(node.getAUse())
-    or
-    exists(Instruction mid |
-      step(node, mid, config) and
-      flowsToSink(mid, pragma[only_bind_into](config))
-    )
-  )
+    /**
+     * Holds if data must flow from `source` to `sink`.
+     *
+     * The corresponding paths are generated from the end-points and the graph
+     * included in the module `PathGraph`.
+     */
+    predicate flowPath(PathNode source, PathSink sink) {
+      isSource(source.getInstruction()) and
+      source.getASuccessor*() = sink
+    }
+
+    /** Holds if `node` flows from a source. */
+    pragma[nomagic]
+    private predicate flowsFromSource(Instruction node) {
+      not isBarrier(node) and
+      (
+        isSource(node)
+        or
+        exists(Instruction mid |
+          step(mid, node) and
+          flowsFromSource(mid)
+        )
+      )
+    }
+
+    /** Holds if `node` flows to a sink. */
+    pragma[nomagic]
+    private predicate flowsToSink(Instruction node) {
+      flowsFromSource(node) and
+      (
+        isSink(node.getAUse())
+        or
+        exists(Instruction mid |
+          step(node, mid) and
+          flowsToSink(mid)
+        )
+      )
+    }
+
+    /** Holds if `nodeFrom` flows to `nodeTo`. */
+    private predicate step(Instruction nodeFrom, Instruction nodeTo) {
+      Cached::localStep(nodeFrom, nodeTo)
+      or
+      allowInterproceduralFlow() and
+      Cached::flowThroughCallable(nodeFrom, nodeTo)
+      or
+      isAdditionalFlowStep(nodeFrom.getAUse(), nodeTo)
+    }
+
+    private newtype TLocalPathNode =
+      MkLocalPathNode(Instruction n) {
+        flowsToSink(n) and
+        (
+          isSource(n)
+          or
+          exists(PathNode mid | step(mid.getInstruction(), n))
+        )
+      }
+
+    /** A `Node` that is in a path from a source to a sink. */
+    class PathNode extends TLocalPathNode {
+      Instruction n;
+
+      PathNode() { this = MkLocalPathNode(n) }
+
+      /** Gets the underlying node. */
+      Instruction getInstruction() { result = n }
+
+      /** Gets a textual representation of this node. */
+      string toString() { result = n.getAst().toString() }
+
+      /** Gets the location of this element. */
+      Location getLocation() { result = n.getLocation() }
+
+      /** Gets a successor node, if any. */
+      PathNode getASuccessor() { step(this.getInstruction(), result.getInstruction()) }
+    }
+
+    private class PathSink extends PathNode {
+      PathSink() { isSink(this.getInstruction().getAUse()) }
+    }
+
+    /**
+     * Provides the query predicates needed to include a graph in a path-problem query.
+     */
+    module PathGraph {
+      private predicate reach(PathNode n) { n instanceof PathSink or reach(n.getASuccessor()) }
+
+      /** Holds if `(a,b)` is an edge in the graph of data flow path explanations. */
+      query predicate edges(PathNode a, PathNode b) { a.getASuccessor() = b and reach(b) }
+
+      /** Holds if `n` is a node in the graph of data flow path explanations. */
+      query predicate nodes(PathNode n, string key, string val) {
+        reach(n) and key = "semmle.label" and val = n.toString()
+      }
+    }
+  }
 }
 
 cached
@@ -102,7 +164,7 @@ private module Cached {
     not f.isVirtual() and
     call.getPositionalArgument(n) = instr and
     f = call.getStaticCallTarget() and
-    getEnclosingNonVirtualFunctionInitializeParameter(init, f) and
+    isEnclosingNonVirtualFunctionInitializeParameter(init, f) and
     init.getParameter().getIndex() = pragma[only_bind_into](pragma[only_bind_out](n))
   }
 
@@ -111,7 +173,7 @@ private module Cached {
    * corresponding initialization instruction that receives the value of `instr` in `f`.
    */
   pragma[noinline]
-  private predicate getPositionalArgumentInitParam(
+  private predicate isPositionalArgumentInitParam(
     CallInstruction call, Instruction instr, InitializeParameterInstruction init, Function f
   ) {
     exists(int n |
@@ -126,18 +188,18 @@ private module Cached {
    * `instr` in `f`.
    */
   pragma[noinline]
-  private predicate getThisArgumentInitParam(
+  private predicate isThisArgumentInitParam(
     CallInstruction call, Instruction instr, InitializeParameterInstruction init, Function f
   ) {
     not f.isVirtual() and
     call.getStaticCallTarget() = f and
-    getEnclosingNonVirtualFunctionInitializeParameter(init, f) and
+    isEnclosingNonVirtualFunctionInitializeParameter(init, f) and
     call.getThisArgument() = instr and
     init.getIRVariable() instanceof IRThisVariable
   }
 
   /** Holds if `f` is the enclosing non-virtual function of `init`. */
-  private predicate getEnclosingNonVirtualFunctionInitializeParameter(
+  private predicate isEnclosingNonVirtualFunctionInitializeParameter(
     InitializeParameterInstruction init, Function f
   ) {
     not f.isVirtual() and
@@ -145,7 +207,7 @@ private module Cached {
   }
 
   /** Holds if `f` is the enclosing non-virtual function of `init`. */
-  private predicate getEnclosingNonVirtualFunctionInitializeIndirection(
+  private predicate isEnclosingNonVirtualFunctionInitializeIndirection(
     InitializeIndirectionInstruction init, Function f
   ) {
     not f.isVirtual() and
@@ -153,15 +215,16 @@ private module Cached {
   }
 
   /**
-   * Holds if `instr` is an argument (or argument indirection) to a call, and
-   * `succ` is the corresponding initialization instruction in the call target.
+   * Holds if `argument` is an argument (or argument indirection) to a call, and
+   * `parameter` is the corresponding initialization instruction in the call target.
    */
-  private predicate flowThroughCallable(Instruction argument, Instruction parameter) {
+  cached
+  predicate flowThroughCallable(Instruction argument, Instruction parameter) {
     // Flow from an argument to a parameter
     exists(CallInstruction call, InitializeParameterInstruction init | init = parameter |
-      getPositionalArgumentInitParam(call, argument, init, call.getStaticCallTarget())
+      isPositionalArgumentInitParam(call, argument, init, call.getStaticCallTarget())
       or
-      getThisArgumentInitParam(call, argument, init, call.getStaticCallTarget())
+      isThisArgumentInitParam(call, argument, init, call.getStaticCallTarget())
     )
     or
     // Flow from argument indirection to parameter indirection
@@ -170,7 +233,7 @@ private module Cached {
     |
       init = parameter and
       read.getPrimaryInstruction() = call and
-      getEnclosingNonVirtualFunctionInitializeIndirection(init, call.getStaticCallTarget())
+      isEnclosingNonVirtualFunctionInitializeIndirection(init, call.getStaticCallTarget())
     |
       exists(int n |
         read.getSideEffectOperand().getAnyDef() = argument and
@@ -205,92 +268,10 @@ private module Cached {
   }
 
   cached
-  predicate step(Instruction nodeFrom, Instruction nodeTo) {
+  predicate localStep(Instruction nodeFrom, Instruction nodeTo) {
     exists(Operand mid |
       instructionToOperandStep(nodeFrom, mid) and
       operandToInstructionStep(mid, nodeTo)
     )
-    or
-    flowThroughCallable(nodeFrom, nodeTo)
-  }
-}
-
-/**
- * Gets the enclosing callable of `n`. Unlike `n.getEnclosingCallable()`, this
- * predicate ensures that joins go from `n` to the result instead of the other
- * way around.
- */
-pragma[inline]
-private IRFunction getEnclosingCallable(Instruction n) {
-  pragma[only_bind_into](result) = pragma[only_bind_out](n).getEnclosingIRFunction()
-}
-
-/** Holds if `nodeFrom` flows to `nodeTo`. */
-private predicate step(Instruction nodeFrom, Instruction nodeTo, MustFlowConfiguration config) {
-  exists(config) and
-  Cached::step(pragma[only_bind_into](nodeFrom), pragma[only_bind_into](nodeTo)) and
-  (
-    config.allowInterproceduralFlow()
-    or
-    getEnclosingCallable(nodeFrom) = getEnclosingCallable(nodeTo)
-  )
-  or
-  config.isAdditionalFlowStep(nodeFrom.getAUse(), nodeTo)
-}
-
-private newtype TLocalPathNode =
-  MkLocalPathNode(Instruction n, MustFlowConfiguration config) {
-    flowsToSink(n, config) and
-    (
-      config.isSource(n)
-      or
-      exists(MustFlowPathNode mid | step(mid.getInstruction(), n, config))
-    )
-  }
-
-/** A `Node` that is in a path from a source to a sink. */
-class MustFlowPathNode extends TLocalPathNode {
-  Instruction n;
-
-  MustFlowPathNode() { this = MkLocalPathNode(n, _) }
-
-  /** Gets the underlying node. */
-  Instruction getInstruction() { result = n }
-
-  /** Gets a textual representation of this node. */
-  string toString() { result = n.getAst().toString() }
-
-  /** Gets the location of this element. */
-  Location getLocation() { result = n.getLocation() }
-
-  /** Gets a successor node, if any. */
-  MustFlowPathNode getASuccessor() {
-    step(this.getInstruction(), result.getInstruction(), this.getConfiguration())
-  }
-
-  /** Gets the associated configuration. */
-  MustFlowConfiguration getConfiguration() { this = MkLocalPathNode(_, result) }
-}
-
-private class MustFlowPathSink extends MustFlowPathNode {
-  MustFlowPathSink() { this.getConfiguration().isSink(this.getInstruction().getAUse()) }
-}
-
-/**
- * Provides the query predicates needed to include a graph in a path-problem query.
- */
-module PathGraph {
-  private predicate reach(MustFlowPathNode n) {
-    n instanceof MustFlowPathSink or reach(n.getASuccessor())
-  }
-
-  /** Holds if `(a,b)` is an edge in the graph of data flow path explanations. */
-  query predicate edges(MustFlowPathNode a, MustFlowPathNode b) {
-    a.getASuccessor() = b and reach(b)
-  }
-
-  /** Holds if `n` is a node in the graph of data flow path explanations. */
-  query predicate nodes(MustFlowPathNode n, string key, string val) {
-    reach(n) and key = "semmle.label" and val = n.toString()
   }
 }

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

@@ -238,7 +238,12 @@ private module TrackVirtualDispatch<methodDispatchSig/1 virtualDispatch0> {
 
   private import TypeTracking<Location, TtInput>::TypeTrack<qualifierSource/1>::Graph<qualifierOfVirtualCall/1>
 
-  private predicate edgePlus(PathNode n1, PathNode n2) = fastTC(edges/2)(n1, n2)
+  private predicate isSource(PathNode n) { n.isSource() }
+
+  private predicate isSink(PathNode n) { n.isSink() }
+
+  private predicate edgePlus(PathNode n1, PathNode n2) =
+    doublyBoundedFastTC(edges/2, isSource/1, isSink/1)(n1, n2)
 
   /**
    * Gets the most specific implementation of `mf` that may be called when the
@@ -255,6 +260,15 @@ private module TrackVirtualDispatch<methodDispatchSig/1 virtualDispatch0> {
     )
   }
 
+  pragma[nomagic]
+  private MemberFunction mostSpecificForSource(PathNode p1, MemberFunction mf) {
+    p1.isSource() and
+    exists(Class derived |
+      qualifierSourceImpl(p1.getNode(), derived) and
+      result = mostSpecific(mf, derived)
+    )
+  }
+
   /**
    * Gets a possible pair of end-points `(p1, p2)` where:
    * - `p1` is a derived-to-base conversion that converts from some
@@ -264,16 +278,16 @@ private module TrackVirtualDispatch<methodDispatchSig/1 virtualDispatch0> {
    * - `callable` is the most specific implementation that may be called when
    * the qualifier has type `derived`.
    */
+  bindingset[p1, p2]
+  pragma[inline_late]
   private predicate pairCand(
     PathNode p1, PathNode p2, DataFlowPrivate::DataFlowCallable callable,
     DataFlowPrivate::DataFlowCall call
   ) {
-    exists(Class derived, MemberFunction mf |
-      qualifierSourceImpl(p1.getNode(), derived) and
+    p2.isSink() and
+    exists(MemberFunction mf |
       qualifierOfVirtualCallImpl(p2.getNode(), call.asCallInstruction(), mf) and
-      p1.isSource() and
-      p2.isSink() and
-      callable.asSourceCallable() = mostSpecific(mf, derived)
+      callable.asSourceCallable() = mostSpecificForSource(p1, mf)
     )
   }
 

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

@@ -1,5 +1,6 @@
 private import cpp as Cpp
 private import DataFlowUtil
+private import DataFlowNodes
 private import semmle.code.cpp.ir.IR
 private import DataFlowDispatch
 private import semmle.code.cpp.ir.internal.IRCppLanguage
@@ -16,28 +17,42 @@ private import semmle.code.cpp.dataflow.ExternalFlow as External
 cached
 private module Cached {
   cached
-  module Nodes0 {
-    cached
-    newtype TIRDataFlowNode0 =
-      TInstructionNode0(Instruction i) {
-        not Ssa::ignoreInstruction(i) and
-        not exists(Operand op |
-          not Ssa::ignoreOperand(op) and i = Ssa::getIRRepresentationOfOperand(op)
-        ) and
-        // We exclude `void`-typed instructions because they cannot contain data.
-        // However, if the instruction is a glvalue, and their type is `void`, then the result
-        // type of the instruction is really `void*`, and thus we still want to have a dataflow
-        // node for it.
-        (not i.getResultType() instanceof VoidType or i.isGLValue())
-      } or
-      TMultipleUseOperandNode0(Operand op) {
-        not Ssa::ignoreOperand(op) and not exists(Ssa::getIRRepresentationOfOperand(op))
-      } or
-      TSingleUseOperandNode0(Operand op) {
-        not Ssa::ignoreOperand(op) and exists(Ssa::getIRRepresentationOfOperand(op))
-      }
+  newtype TIRDataFlowNode0 =
+    TInstructionNode0(Instruction i) {
+      not Ssa::ignoreInstruction(i) and
+      not exists(Operand op |
+        not Ssa::ignoreOperand(op) and i = Ssa::getIRRepresentationOfOperand(op)
+      ) and
+      // We exclude `void`-typed instructions because they cannot contain data.
+      // However, if the instruction is a glvalue, and their type is `void`, then the result
+      // type of the instruction is really `void*`, and thus we still want to have a dataflow
+      // node for it.
+      (not i.getResultType() instanceof VoidType or i.isGLValue())
+    } or
+    TMultipleUseOperandNode0(Operand op) {
+      not Ssa::ignoreOperand(op) and not exists(Ssa::getIRRepresentationOfOperand(op))
+    } or
+    TSingleUseOperandNode0(Operand op) {
+      not Ssa::ignoreOperand(op) and exists(Ssa::getIRRepresentationOfOperand(op))
+    }
+
+  cached
+  string toStringCached(Node n) {
+    result = toExprString(n)
+    or
+    not exists(toExprString(n)) and
+    result = n.toStringImpl()
   }
 
+  cached
+  Location getLocationCached(Node n) { result = n.getLocationImpl() }
+
+  cached
+  newtype TContentApprox =
+    TFieldApproxContent(string s) { fieldHasApproxName(_, s) } or
+    TUnionApproxContent(string s) { unionHasApproxName(_, s) } or
+    TElementApproxContent()
+
   /**
    * Gets an additional term that is added to the `join` and `branch` computations to reflect
    * an additional forward or backwards branching factor that is not taken into account
@@ -59,38 +74,174 @@ private module Cached {
       result = countNumberOfBranchesUsingParameter(switch, p)
     )
   }
-}
 
-import Cached
-private import Nodes0
+  cached
+  newtype TDataFlowCallable =
+    TSourceCallable(Cpp::Declaration decl) or
+    TSummarizedCallable(FlowSummaryImpl::Public::SummarizedCallable c)
 
-/**
- * A module for calculating the number of stars (i.e., `*`s) needed for various
- * dataflow node `toString` predicates.
- */
-module NodeStars {
-  private int getNumberOfIndirections(Node n) {
-    result = n.(RawIndirectOperand).getIndirectionIndex()
+  cached
+  newtype TDataFlowCall =
+    TNormalCall(CallInstruction call) or
+    TSummaryCall(
+      FlowSummaryImpl::Public::SummarizedCallable c, FlowSummaryImpl::Private::SummaryNode receiver
+    ) {
+      FlowSummaryImpl::Private::summaryCallbackRange(c, receiver)
+    }
+
+  /**
+   * Holds if data can flow from `node1` to `node2` in a way that loses the
+   * calling context. For example, this would happen with flow through a
+   * global or static variable.
+   */
+  cached
+  predicate jumpStep(Node n1, Node n2) {
+    exists(GlobalLikeVariable v |
+      exists(Ssa::GlobalUse globalUse |
+        v = globalUse.getVariable() and
+        n1.(FinalGlobalValue).getGlobalUse() = globalUse
+      |
+        globalUse.getIndirection() = getMinIndirectionForGlobalUse(globalUse) and
+        v = n2.asVariable()
+        or
+        v = n2.asIndirectVariable(globalUse.getIndirection())
+      )
+      or
+      exists(Ssa::GlobalDef globalDef |
+        v = globalDef.getVariable() and
+        n2.(InitialGlobalValue).getGlobalDef() = globalDef
+      |
+        globalDef.getIndirection() = getMinIndirectionForGlobalDef(globalDef) and
+        v = n1.asVariable()
+        or
+        v = n1.asIndirectVariable(globalDef.getIndirection())
+      )
+    )
     or
-    result = n.(RawIndirectInstruction).getIndirectionIndex()
-    or
-    result = n.(VariableNode).getIndirectionIndex()
-    or
-    result = n.(PostUpdateNodeImpl).getIndirectionIndex()
-    or
-    result = n.(FinalParameterNode).getIndirectionIndex()
-    or
-    result = n.(BodyLessParameterNodeImpl).getIndirectionIndex()
+    // models-as-data summarized flow
+    FlowSummaryImpl::Private::Steps::summaryJumpStep(n1.(FlowSummaryNode).getSummaryNode(),
+      n2.(FlowSummaryNode).getSummaryNode())
   }
 
   /**
-   * Gets the number of stars (i.e., `*`s) needed to produce the `toString`
-   * output for `n`.
+   * Holds if data can flow from `node1` to `node2` via an assignment to `f`.
+   * Thus, `node2` references an object with a field `f` that contains the
+   * value of `node1`.
+   *
+   * The boolean `certain` is true if the destination address does not involve
+   * any pointer arithmetic, and false otherwise.
    */
-  string stars(Node n) { result = repeatStars(getNumberOfIndirections(n)) }
+  cached
+  predicate storeStepImpl(Node node1, Content c, Node node2, boolean certain) {
+    exists(
+      PostFieldUpdateNode postFieldUpdate, int indirectionIndex1, int numberOfLoads,
+      StoreInstruction store, FieldContent fc
+    |
+      postFieldUpdate = node2 and
+      fc = c and
+      nodeHasInstruction(node1, pragma[only_bind_into](store),
+        pragma[only_bind_into](indirectionIndex1)) and
+      postFieldUpdate.getIndirectionIndex() = 1 and
+      numberOfLoadsFromOperand(postFieldUpdate.getFieldAddress(),
+        store.getDestinationAddressOperand(), numberOfLoads, certain) and
+      fc.getAField() = postFieldUpdate.getUpdatedField() and
+      getIndirectionIndexLate(fc) = 1 + indirectionIndex1 + numberOfLoads
+    )
+    or
+    // models-as-data summarized flow
+    FlowSummaryImpl::Private::Steps::summaryStoreStep(node1.(FlowSummaryNode).getSummaryNode(), c,
+      node2.(FlowSummaryNode).getSummaryNode()) and
+    certain = true
+  }
+
+  /**
+   * Holds if data can flow from `node1` to `node2` via an assignment to `f`.
+   * Thus, `node2` references an object with a field `f` that contains the
+   * value of `node1`.
+   */
+  cached
+  predicate storeStep(Node node1, ContentSet c, Node node2) { storeStepImpl(node1, c, node2, _) }
+
+  /**
+   * Holds if data can flow from `node1` to `node2` via a read of `f`.
+   * Thus, `node1` references an object with a field `f` whose value ends up in
+   * `node2`.
+   */
+  cached
+  predicate readStep(Node node1, ContentSet c, Node node2) {
+    exists(
+      FieldAddress fa1, Operand operand, int numberOfLoads, int indirectionIndex2, FieldContent fc
+    |
+      fc = c and
+      nodeHasOperand(node2, operand, indirectionIndex2) and
+      // The `1` here matches the `node2.getIndirectionIndex() = 1` conjunct
+      // in `storeStep`.
+      nodeHasOperand(node1, fa1.getObjectAddressOperand(), 1) and
+      numberOfLoadsFromOperand(fa1, operand, numberOfLoads, _) and
+      fc.getAField() = fa1.getField() and
+      getIndirectionIndexLate(fc) = indirectionIndex2 + numberOfLoads
+    )
+    or
+    // models-as-data summarized flow
+    FlowSummaryImpl::Private::Steps::summaryReadStep(node1.(FlowSummaryNode).getSummaryNode(), c,
+      node2.(FlowSummaryNode).getSummaryNode())
+  }
+
+  /**
+   * Holds if values stored inside content `c` are cleared at node `n`.
+   */
+  cached
+  predicate clearsContent(Node n, ContentSet c) {
+    n =
+      any(PostUpdateNode pun, Content d |
+        d.impliesClearOf(c) and storeStepImpl(_, d, pun, true)
+      |
+        pun
+      ).getPreUpdateNode() and
+    (
+      not exists(Operand op, Cpp::Operation p |
+        n.(IndirectOperand).hasOperandAndIndirectionIndex(op, _) and
+        (
+          p instanceof Cpp::AssignPointerAddExpr or
+          p instanceof Cpp::AssignPointerSubExpr or
+          p instanceof Cpp::CrementOperation
+        )
+      |
+        p.getAnOperand() = op.getUse().getAst()
+      )
+      or
+      forex(PostUpdateNode pun, Content d |
+        pragma[only_bind_into](d).impliesClearOf(pragma[only_bind_into](c)) and
+        storeStepImpl(_, d, pun, true) and
+        pun.getPreUpdateNode() = n
+      |
+        c.(Content).getIndirectionIndex() = d.getIndirectionIndex()
+      )
+    )
+  }
 }
 
-import NodeStars
+import Cached
+
+private int getNumberOfIndirections(Node n) {
+  result = n.(RawIndirectOperand).getIndirectionIndex()
+  or
+  result = n.(RawIndirectInstruction).getIndirectionIndex()
+  or
+  result = n.(VariableNode).getIndirectionIndex()
+  or
+  result = n.(PostUpdateNodeImpl).getIndirectionIndex()
+  or
+  result = n.(FinalParameterNode).getIndirectionIndex()
+  or
+  result = n.(BodyLessParameterNodeImpl).getIndirectionIndex()
+}
+
+/**
+ * Gets the number of stars (i.e., `*`s) needed to produce the `toString`
+ * output for `n`.
+ */
+string stars(Node n) { result = repeatStars(getNumberOfIndirections(n)) }
 
 /**
  * A cut-down `DataFlow::Node` class that does not depend on the output of SSA.
@@ -828,85 +979,10 @@ private int getMinIndirectionForGlobalDef(Ssa::GlobalDef def) {
   result = getMinIndirectionsForType(def.getUnspecifiedType())
 }
 
-/**
- * Holds if data can flow from `node1` to `node2` in a way that loses the
- * calling context. For example, this would happen with flow through a
- * global or static variable.
- */
-predicate jumpStep(Node n1, Node n2) {
-  exists(GlobalLikeVariable v |
-    exists(Ssa::GlobalUse globalUse |
-      v = globalUse.getVariable() and
-      n1.(FinalGlobalValue).getGlobalUse() = globalUse
-    |
-      globalUse.getIndirection() = getMinIndirectionForGlobalUse(globalUse) and
-      v = n2.asVariable()
-      or
-      v = n2.asIndirectVariable(globalUse.getIndirection())
-    )
-    or
-    exists(Ssa::GlobalDef globalDef |
-      v = globalDef.getVariable() and
-      n2.(InitialGlobalValue).getGlobalDef() = globalDef
-    |
-      globalDef.getIndirection() = getMinIndirectionForGlobalDef(globalDef) and
-      v = n1.asVariable()
-      or
-      v = n1.asIndirectVariable(globalDef.getIndirection())
-    )
-  )
-  or
-  // models-as-data summarized flow
-  FlowSummaryImpl::Private::Steps::summaryJumpStep(n1.(FlowSummaryNode).getSummaryNode(),
-    n2.(FlowSummaryNode).getSummaryNode())
-}
-
 bindingset[c]
 pragma[inline_late]
 private int getIndirectionIndexLate(Content c) { result = c.getIndirectionIndex() }
 
-/**
- * Holds if data can flow from `node1` to `node2` via an assignment to `f`.
- * Thus, `node2` references an object with a field `f` that contains the
- * value of `node1`.
- *
- * The boolean `certain` is true if the destination address does not involve
- * any pointer arithmetic, and false otherwise. This has to do with whether a
- * store step can be used to clear a field (see `clearsContent`).
- */
-predicate storeStepImpl(Node node1, Content c, Node node2, boolean certain) {
-  exists(
-    PostFieldUpdateNode postFieldUpdate, int indirectionIndex1, int numberOfLoads,
-    StoreInstruction store, FieldContent fc
-  |
-    postFieldUpdate = node2 and
-    fc = c and
-    nodeHasInstruction(node1, pragma[only_bind_into](store),
-      pragma[only_bind_into](indirectionIndex1)) and
-    postFieldUpdate.getIndirectionIndex() = 1 and
-    numberOfLoadsFromOperand(postFieldUpdate.getFieldAddress(),
-      store.getDestinationAddressOperand(), numberOfLoads, certain) and
-    fc.getAField() = postFieldUpdate.getUpdatedField() and
-    getIndirectionIndexLate(fc) = 1 + indirectionIndex1 + numberOfLoads
-  )
-  or
-  // models-as-data summarized flow
-  FlowSummaryImpl::Private::Steps::summaryStoreStep(node1.(FlowSummaryNode).getSummaryNode(), c,
-    node2.(FlowSummaryNode).getSummaryNode()) and
-  certain = true
-}
-
-/**
- * Holds if data can flow from `node1` to `node2` via an assignment to `f`.
- * Thus, `node2` references an object with a field `f` that contains the
- * value of `node1`.
- */
-predicate storeStep(Node node1, ContentSet c, Node node2) { storeStepImpl(node1, c, node2, _) }
-
-/**
- * Holds if `operandFrom` flows to `operandTo` using a sequence of conversion-like
- * operations and exactly `n` `LoadInstruction` operations.
- */
 private predicate numberOfLoadsFromOperandRec(
   Operand operandFrom, Operand operandTo, int ind, boolean certain
 ) {
@@ -957,63 +1033,6 @@ predicate nodeHasInstruction(Node node, Instruction instr, int indirectionIndex)
   hasInstructionAndIndex(node, instr, indirectionIndex)
 }
 
-/**
- * Holds if data can flow from `node1` to `node2` via a read of `f`.
- * Thus, `node1` references an object with a field `f` whose value ends up in
- * `node2`.
- */
-predicate readStep(Node node1, ContentSet c, Node node2) {
-  exists(
-    FieldAddress fa1, Operand operand, int numberOfLoads, int indirectionIndex2, FieldContent fc
-  |
-    fc = c and
-    nodeHasOperand(node2, operand, indirectionIndex2) and
-    // The `1` here matches the `node2.getIndirectionIndex() = 1` conjunct
-    // in `storeStep`.
-    nodeHasOperand(node1, fa1.getObjectAddressOperand(), 1) and
-    numberOfLoadsFromOperand(fa1, operand, numberOfLoads, _) and
-    fc.getAField() = fa1.getField() and
-    getIndirectionIndexLate(fc) = indirectionIndex2 + numberOfLoads
-  )
-  or
-  // models-as-data summarized flow
-  FlowSummaryImpl::Private::Steps::summaryReadStep(node1.(FlowSummaryNode).getSummaryNode(), c,
-    node2.(FlowSummaryNode).getSummaryNode())
-}
-
-/**
- * Holds if values stored inside content `c` are cleared at node `n`.
- */
-predicate clearsContent(Node n, ContentSet c) {
-  n =
-    any(PostUpdateNode pun, Content d | d.impliesClearOf(c) and storeStepImpl(_, d, pun, true) | pun)
-        .getPreUpdateNode() and
-  (
-    // The crement operations and pointer addition and subtraction self-assign. We do not
-    // want to clear the contents if it is indirectly pointed at by any of these operations,
-    // as part of the contents might still be accessible afterwards. If there is no such
-    // indirection clearing the contents is safe.
-    not exists(Operand op, Cpp::Operation p |
-      n.(IndirectOperand).hasOperandAndIndirectionIndex(op, _) and
-      (
-        p instanceof Cpp::AssignPointerAddExpr or
-        p instanceof Cpp::AssignPointerSubExpr or
-        p instanceof Cpp::CrementOperation
-      )
-    |
-      p.getAnOperand() = op.getUse().getAst()
-    )
-    or
-    forex(PostUpdateNode pun, Content d |
-      pragma[only_bind_into](d).impliesClearOf(pragma[only_bind_into](c)) and
-      storeStepImpl(_, d, pun, true) and
-      pun.getPreUpdateNode() = n
-    |
-      c.(Content).getIndirectionIndex() = d.getIndirectionIndex()
-    )
-  )
-}
-
 /**
  * Holds if the value that is being tracked is expected to be stored inside content `c`
  * at node `n`.
@@ -1046,11 +1065,6 @@ class CastNode extends Node {
   CastNode() { none() } // stub implementation
 }
 
-cached
-private newtype TDataFlowCallable =
-  TSourceCallable(Cpp::Declaration decl) or
-  TSummarizedCallable(FlowSummaryImpl::Public::SummarizedCallable c)
-
 /**
  * A callable, which may be:
  *  - a function (that may contain code)
@@ -1134,15 +1148,6 @@ class DataFlowType extends TypeFinal {
   string toString() { result = "" }
 }
 
-cached
-private newtype TDataFlowCall =
-  TNormalCall(CallInstruction call) or
-  TSummaryCall(
-    FlowSummaryImpl::Public::SummarizedCallable c, FlowSummaryImpl::Private::SummaryNode receiver
-  ) {
-    FlowSummaryImpl::Private::summaryCallbackRange(c, receiver)
-  }
-
 private predicate summarizedCallableIsManual(SummarizedCallable sc) {
   sc.asSummarizedCallable().hasManualModel()
 }
@@ -1165,7 +1170,7 @@ class DataFlowCall extends TDataFlowCall {
   /**
    * Gets the `Function` that the call targets, if this is statically known.
    */
-  Function getStaticCallSourceTarget() { none() }
+  Declaration getStaticCallSourceTarget() { none() }
 
   /**
    * Gets the target of this call. We use the following strategy for deciding
@@ -1177,7 +1182,7 @@ class DataFlowCall extends TDataFlowCall {
    * whether is it manual or generated.
    */
   final DataFlowCallable getStaticCallTarget() {
-    exists(Function target | target = this.getStaticCallSourceTarget() |
+    exists(Declaration target | target = this.getStaticCallSourceTarget() |
       // Don't use the source callable if there is a manual model for the
       // target
       not exists(SummarizedCallable sc |
@@ -1237,7 +1242,7 @@ private class NormalCall extends DataFlowCall, TNormalCall {
 
   override CallTargetOperand getCallTargetOperand() { result = call.getCallTargetOperand() }
 
-  override Function getStaticCallSourceTarget() { result = call.getStaticCallTarget() }
+  override Declaration getStaticCallSourceTarget() { result = call.getStaticCallTarget() }
 
   override ArgumentOperand getArgumentOperand(int index) { result = call.getArgumentOperand(index) }
 
@@ -1523,12 +1528,6 @@ private predicate fieldHasApproxName(Field f, string s) {
 
 private predicate unionHasApproxName(Cpp::Union u, string s) { s = u.getName().charAt(0) }
 
-cached
-private newtype TContentApprox =
-  TFieldApproxContent(string s) { fieldHasApproxName(_, s) } or
-  TUnionApproxContent(string s) { unionHasApproxName(_, s) } or
-  TElementApproxContent()
-
 /** An approximated `Content`. */
 class ContentApprox extends TContentApprox {
   string toString() { none() } // overridden in subclasses

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

@@ -6,8 +6,8 @@ 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
+private import DataFlowNodes
+private import semmle.code.cpp.ir.implementation.raw.internal.IRConstruction as IRConstruction
 
 cached
 private module Cached {
@@ -73,17 +73,9 @@ private module Cached {
     // 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())
-    )
+    result = IRConstruction::Raw::getAllocationExtentConvertExpr(instr)
     or
-    // There's no instruction that returns `ParenthesisExpr`, but some queries
-    // expect this
-    exists(TranslatedTransparentConversion ttc |
-      result = ttc.getExpr().(ParenthesisExpr) and
-      instr = ttc.getResult()
-    )
+    result = IRConstruction::Raw::getTransparentConversionParenthesisExpr(instr)
     or
     // Certain expressions generate `CopyValueInstruction`s only when they
     // are needed. Examples of this include crement operations and compound
@@ -112,10 +104,10 @@ private module Cached {
     // 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)
+    exists(StoreInstruction store |
+      store = instr and
+      IRConstruction::Raw::instructionProducesExprResult(store) and
+      result = asDefinitionImpl0(store)
     )
     or
     // IR construction breaks an array aggregate literal `{1, 2, 3}` into a
@@ -145,18 +137,9 @@ private module Cached {
     // 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()
-    )
+    result = IRConstruction::Raw::getAssignOperationStoreExpr(store)
     or
-    // Similarly for `i++` and `++i` 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(TranslatedCrementOperation tco |
-      store = tco.getInstruction(CrementStoreTag()) and
-      result = tco.getExpr()
-    )
+    result = IRConstruction::Raw::getCrementOperationStoreExpr(store)
   }
 
   /**
@@ -166,11 +149,7 @@ private module Cached {
    */
   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())
-    )
+    IRConstruction::Raw::isConditionalExprTempStore(store)
   }
 
   /**

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

@@ -6,6 +6,7 @@
 private import semmle.code.cpp.ir.IR
 private import semmle.code.cpp.models.interfaces.FunctionInputsAndOutputs
 private import DataFlowUtil
+private import DataFlowNodes
 private import DataFlowPrivate
 private import SsaImpl as Ssa
 

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

@@ -6,6 +6,7 @@ private import cpp
 private import semmle.code.cpp.ir.IR
 private import semmle.code.cpp.ir.dataflow.internal.DataFlowUtil
 private import semmle.code.cpp.ir.dataflow.internal.DataFlowPrivate
+private import semmle.code.cpp.ir.dataflow.internal.DataFlowNodes
 private import PrintIRUtilities
 
 /** A property provider for local IR dataflow store steps. */

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

@@ -2,6 +2,7 @@ private import cpp
 private import semmle.code.cpp.ir.IR
 private import semmle.code.cpp.ir.dataflow.internal.DataFlowUtil
 private import semmle.code.cpp.ir.dataflow.internal.DataFlowPrivate
+private import semmle.code.cpp.ir.dataflow.internal.DataFlowNodes
 private import SsaImpl as Ssa
 private import PrintIRUtilities
 

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

@@ -6,6 +6,7 @@ private import cpp
 private import semmle.code.cpp.ir.IR
 private import semmle.code.cpp.ir.dataflow.internal.DataFlowUtil
 private import semmle.code.cpp.ir.dataflow.internal.DataFlowPrivate
+private import semmle.code.cpp.ir.dataflow.internal.DataFlowNodes
 
 private Instruction getInstruction(Node n, string stars) {
   result = [n.asInstruction(), n.(RawIndirectInstruction).getInstruction()] and

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

@@ -10,8 +10,9 @@ private import semmle.code.cpp.models.interfaces.PartialFlow as PartialFlow
 private import semmle.code.cpp.models.interfaces.FunctionInputsAndOutputs as FIO
 private import semmle.code.cpp.ir.internal.IRCppLanguage
 private import semmle.code.cpp.ir.dataflow.internal.ModelUtil
-private import semmle.code.cpp.ir.implementation.raw.internal.TranslatedInitialization
+private import semmle.code.cpp.ir.implementation.raw.internal.IRConstruction as IRConstruction
 private import DataFlowPrivate
+private import DataFlowNodes
 import SsaImplCommon
 
 private module SourceVariables {
@@ -438,10 +439,7 @@ private predicate sourceVariableHasBaseAndIndex(SourceVariable v, BaseSourceVari
  * initialize `v`.
  */
 private Instruction getInitializationTargetAddress(IRVariable v) {
-  exists(TranslatedVariableInitialization init |
-    init.getIRVariable() = v and
-    result = init.getTargetAddress()
-  )
+  result = IRConstruction::Raw::getInitializationTargetAddress(v)
 }
 
 /** An initial definition of an SSA variable address. */

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

@@ -4,75 +4,25 @@ import semmle.code.cpp.ir.internal.IRCppLanguage
 private import semmle.code.cpp.ir.implementation.raw.internal.SideEffects as SideEffects
 private import DataFlowImplCommon as DataFlowImplCommon
 private import DataFlowUtil
+private import DataFlowNodes
 private import semmle.code.cpp.models.interfaces.PointerWrapper
 private import DataFlowPrivate
 private import TypeFlow
 private import semmle.code.cpp.ir.ValueNumbering
 
 /**
- * Holds if `operand` is an operand that is not used by the dataflow library.
- * Ignored operands are not recognized as uses by SSA, and they don't have a
- * corresponding `(Indirect)OperandNode`.
- */
-predicate ignoreOperand(Operand operand) {
-  operand = any(Instruction instr | ignoreInstruction(instr)).getAnOperand() or
-  operand = any(Instruction instr | ignoreInstruction(instr)).getAUse() or
-  operand instanceof MemoryOperand
-}
-
-/**
- * Holds if `instr` is an instruction that is not used by the dataflow library.
- * Ignored instructions are not recognized as reads/writes by SSA, and they
- * don't have a corresponding `(Indirect)InstructionNode`.
- */
-predicate ignoreInstruction(Instruction instr) {
-  DataFlowImplCommon::forceCachingInSameStage() and
-  (
-    instr instanceof CallSideEffectInstruction or
-    instr instanceof CallReadSideEffectInstruction or
-    instr instanceof ExitFunctionInstruction or
-    instr instanceof EnterFunctionInstruction or
-    instr instanceof WriteSideEffectInstruction or
-    instr instanceof PhiInstruction or
-    instr instanceof ReadSideEffectInstruction or
-    instr instanceof ChiInstruction or
-    instr instanceof InitializeIndirectionInstruction or
-    instr instanceof AliasedDefinitionInstruction or
-    instr instanceof AliasedUseInstruction or
-    instr instanceof InitializeNonLocalInstruction or
-    instr instanceof ReturnIndirectionInstruction or
-    instr instanceof UninitializedGroupInstruction
-  )
-}
-
-/**
- * Gets the C++ type of `this` in the member function `f`.
+ * Gets the C++ type of `this` in an `IRFunction` generated from `f`.
  * The result is a glvalue if `isGLValue` is true, and
  * a prvalue if `isGLValue` is false.
  */
 bindingset[isGLValue]
-private CppType getThisType(Cpp::MemberFunction f, boolean isGLValue) {
-  result.hasType(f.getTypeOfThis(), isGLValue)
-}
-
-/**
- * Gets the C++ type of the instruction `i`.
- *
- * This is equivalent to `i.getResultLanguageType()` with the exception
- * of instructions that directly references a `this` IRVariable. In this
- * case, `i.getResultLanguageType()` gives an unknown type, whereas the
- * predicate gives the expected type (i.e., a potentially cv-qualified
- * type `A*` where `A` is the declaring type of the member function that
- * contains `i`).
- */
-cached
-CppType getResultLanguageType(Instruction i) {
-  if i.(VariableAddressInstruction).getIRVariable() instanceof IRThisVariable
-  then
-    if i.isGLValue()
-    then result = getThisType(i.getEnclosingFunction(), true)
-    else result = getThisType(i.getEnclosingFunction(), false)
-  else result = i.getResultLanguageType()
+private CppType getThisType(Cpp::Declaration f, boolean isGLValue) {
+  result.hasType(f.(Cpp::MemberFunction).getTypeOfThis(), isGLValue)
+  or
+  exists(Cpp::PointerType pt |
+    pt.getBaseType() = f.(Cpp::Field).getDeclaringType() and
+    result.hasType(pt, isGLValue)
+  )
 }
 
 /**
@@ -230,7 +180,8 @@ private class PointerWrapperTypeIndirection extends Indirection instanceof Point
   override predicate isAdditionalDereference(Instruction deref, Operand address) {
     exists(CallInstruction call |
       operandForFullyConvertedCall(getAUse(deref), call) and
-      this = call.getStaticCallTarget().getClassAndName(["operator*", "operator->", "get"]) and
+      this =
+        call.getStaticCallTarget().(Function).getClassAndName(["operator*", "operator->", "get"]) and
       address = call.getThisArgumentOperand()
     )
   }
@@ -249,7 +200,7 @@ private module IteratorIndirections {
 
     override predicate isAdditionalWrite(Node0Impl value, Operand address, boolean certain) {
       exists(CallInstruction call | call.getArgumentOperand(0) = value.asOperand() |
-        this = call.getStaticCallTarget().getClassAndName("operator=") and
+        this = call.getStaticCallTarget().(Function).getClassAndName("operator=") and
         address = call.getThisArgumentOperand() and
         certain = false
       )
@@ -347,10 +298,6 @@ predicate isWrite(Node0Impl value, Operand address, boolean certain) {
   )
 }
 
-predicate isAdditionalConversionFlow(Operand opFrom, Instruction instrTo) {
-  any(Indirection ind).isAdditionalConversionFlow(opFrom, instrTo)
-}
-
 newtype TBaseSourceVariable =
   // Each IR variable gets its own source variable
   TBaseIRVariable(IRVariable var) or
@@ -572,6 +519,69 @@ private class BaseCallInstruction extends BaseSourceVariableInstruction, CallIns
 
 cached
 private module Cached {
+  /**
+   * Holds if `operand` is an operand that is not used by the dataflow library.
+   * Ignored operands are not recognized as uses by SSA, and they don't have a
+   * corresponding `(Indirect)OperandNode`.
+   */
+  cached
+  predicate ignoreOperand(Operand operand) {
+    operand = any(Instruction instr | ignoreInstruction(instr)).getAnOperand() or
+    operand = any(Instruction instr | ignoreInstruction(instr)).getAUse() or
+    operand instanceof MemoryOperand
+  }
+
+  /**
+   * Holds if `instr` is an instruction that is not used by the dataflow library.
+   * Ignored instructions are not recognized as reads/writes by SSA, and they
+   * don't have a corresponding `(Indirect)InstructionNode`.
+   */
+  cached
+  predicate ignoreInstruction(Instruction instr) {
+    DataFlowImplCommon::forceCachingInSameStage() and
+    (
+      instr instanceof CallSideEffectInstruction or
+      instr instanceof CallReadSideEffectInstruction or
+      instr instanceof ExitFunctionInstruction or
+      instr instanceof EnterFunctionInstruction or
+      instr instanceof WriteSideEffectInstruction or
+      instr instanceof PhiInstruction or
+      instr instanceof ReadSideEffectInstruction or
+      instr instanceof ChiInstruction or
+      instr instanceof InitializeIndirectionInstruction or
+      instr instanceof AliasedDefinitionInstruction or
+      instr instanceof AliasedUseInstruction or
+      instr instanceof InitializeNonLocalInstruction or
+      instr instanceof ReturnIndirectionInstruction or
+      instr instanceof UninitializedGroupInstruction
+    )
+  }
+
+  cached
+  predicate isAdditionalConversionFlow(Operand opFrom, Instruction instrTo) {
+    any(Indirection ind).isAdditionalConversionFlow(opFrom, instrTo)
+  }
+
+  /**
+   * Gets the C++ type of the instruction `i`.
+   *
+   * This is equivalent to `i.getResultLanguageType()` with the exception
+   * of instructions that directly references a `this` IRVariable. In this
+   * case, `i.getResultLanguageType()` gives an unknown type, whereas the
+   * predicate gives the expected type (i.e., a potentially cv-qualified
+   * type `A*` where `A` is the declaring type of the member function that
+   * contains `i`).
+   */
+  cached
+  CppType getResultLanguageType(Instruction i) {
+    if i.(VariableAddressInstruction).getIRVariable() instanceof IRThisVariable
+    then
+      if i.isGLValue()
+      then result = getThisType(i.getEnclosingFunction(), true)
+      else result = getThisType(i.getEnclosingFunction(), false)
+    else result = i.getResultLanguageType()
+  }
+
   /** Holds if `op` is the only use of its defining instruction, and that op is used in a conversation */
   private predicate isConversion(Operand op) {
     exists(Instruction def, Operand use |

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

@@ -5,64 +5,81 @@ private import semmle.code.cpp.models.interfaces.DataFlow
 private import semmle.code.cpp.models.interfaces.SideEffect
 private import DataFlowUtil
 private import DataFlowPrivate
+private import DataFlowNodes
 private import SsaImpl as Ssa
 private import semmle.code.cpp.dataflow.internal.FlowSummaryImpl as FlowSummaryImpl
 private import semmle.code.cpp.ir.dataflow.FlowSteps
 
-/**
- * Holds if taint propagates from `nodeFrom` to `nodeTo` in exactly one local
- * (intra-procedural) step. This relation is only used for local taint flow
- * (for example `TaintTracking::localTaint(source, sink)`) so it may contain
- * special cases that should only apply to local taint flow.
- */
-predicate localTaintStep(DataFlow::Node nodeFrom, DataFlow::Node nodeTo) {
-  // dataflow step
-  DataFlow::localFlowStep(nodeFrom, nodeTo)
-  or
-  // taint flow step
-  localAdditionalTaintStep(nodeFrom, nodeTo, _)
-  or
-  // models-as-data summarized flow for local data flow (i.e. special case for flow
-  // through calls to modeled functions, without relying on global dataflow to join
-  // the dots).
-  FlowSummaryImpl::Private::Steps::summaryThroughStepTaint(nodeFrom, nodeTo, _)
+cached
+private module Cached {
+  private import DataFlowImplCommon as DataFlowImplCommon
+
+  /**
+   * This predicate exists to collapse the `cached` predicates in this module with the
+   * `cached` predicates in other C/C++ dataflow files, which is then collapsed
+   * with the `cached` predicates in `DataFlowImplCommon.qll`.
+   */
+  cached
+  predicate forceCachingInSameStage() { DataFlowImplCommon::forceCachingInSameStage() }
+
+  /**
+   * Holds if taint propagates from `nodeFrom` to `nodeTo` in exactly one local
+   * (intra-procedural) step. This relation is only used for local taint flow
+   * (for example `TaintTracking::localTaint(source, sink)`) so it may contain
+   * special cases that should only apply to local taint flow.
+   */
+  cached
+  predicate localTaintStep(DataFlow::Node nodeFrom, DataFlow::Node nodeTo) {
+    // dataflow step
+    DataFlow::localFlowStep(nodeFrom, nodeTo)
+    or
+    // taint flow step
+    localAdditionalTaintStep(nodeFrom, nodeTo, _)
+    or
+    // models-as-data summarized flow for local data flow (i.e. special case for flow
+    // through calls to modeled functions, without relying on global dataflow to join
+    // the dots).
+    FlowSummaryImpl::Private::Steps::summaryThroughStepTaint(nodeFrom, nodeTo, _)
+  }
+
+  /**
+   * Holds if taint can flow in one local step from `nodeFrom` to `nodeTo` excluding
+   * local data flow steps. That is, `nodeFrom` and `nodeTo` are likely to represent
+   * different objects.
+   */
+  cached
+  predicate localAdditionalTaintStep(DataFlow::Node nodeFrom, DataFlow::Node nodeTo, string model) {
+    operandToInstructionTaintStep(nodeFrom.asOperand(), nodeTo.asInstruction()) and
+    model = ""
+    or
+    modeledTaintStep(nodeFrom, nodeTo, model)
+    or
+    // Flow from (the indirection of) an operand of a pointer arithmetic instruction to the
+    // indirection of the pointer arithmetic instruction. This provides flow from `source`
+    // in `x[source]` to the result of the associated load instruction.
+    exists(PointerArithmeticInstruction pai, int indirectionIndex |
+      nodeHasOperand(nodeFrom, pai.getAnOperand(), pragma[only_bind_into](indirectionIndex)) and
+      hasInstructionAndIndex(nodeTo, pai, indirectionIndex + 1)
+    ) and
+    model = ""
+    or
+    any(Ssa::Indirection ind).isAdditionalTaintStep(nodeFrom, nodeTo) and
+    model = ""
+    or
+    // models-as-data summarized flow
+    FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom.(FlowSummaryNode).getSummaryNode(),
+      nodeTo.(FlowSummaryNode).getSummaryNode(), false, model)
+    or
+    // object->field conflation for content that is a `TaintInheritingContent`.
+    exists(DataFlow::ContentSet f |
+      readStep(nodeFrom, f, nodeTo) and
+      f.getAReadContent() instanceof TaintInheritingContent
+    ) and
+    model = ""
+  }
 }
 
-/**
- * Holds if taint can flow in one local step from `nodeFrom` to `nodeTo` excluding
- * local data flow steps. That is, `nodeFrom` and `nodeTo` are likely to represent
- * different objects.
- */
-cached
-predicate localAdditionalTaintStep(DataFlow::Node nodeFrom, DataFlow::Node nodeTo, string model) {
-  operandToInstructionTaintStep(nodeFrom.asOperand(), nodeTo.asInstruction()) and
-  model = ""
-  or
-  modeledTaintStep(nodeFrom, nodeTo, model)
-  or
-  // Flow from (the indirection of) an operand of a pointer arithmetic instruction to the
-  // indirection of the pointer arithmetic instruction. This provides flow from `source`
-  // in `x[source]` to the result of the associated load instruction.
-  exists(PointerArithmeticInstruction pai, int indirectionIndex |
-    nodeHasOperand(nodeFrom, pai.getAnOperand(), pragma[only_bind_into](indirectionIndex)) and
-    hasInstructionAndIndex(nodeTo, pai, indirectionIndex + 1)
-  ) and
-  model = ""
-  or
-  any(Ssa::Indirection ind).isAdditionalTaintStep(nodeFrom, nodeTo) and
-  model = ""
-  or
-  // models-as-data summarized flow
-  FlowSummaryImpl::Private::Steps::summaryLocalStep(nodeFrom.(FlowSummaryNode).getSummaryNode(),
-    nodeTo.(FlowSummaryNode).getSummaryNode(), false, model)
-  or
-  // object->field conflation for content that is a `TaintInheritingContent`.
-  exists(DataFlow::ContentSet f |
-    readStep(nodeFrom, f, nodeTo) and
-    f.getAReadContent() instanceof TaintInheritingContent
-  ) and
-  model = ""
-}
+import Cached
 
 /**
  * Holds if taint propagates from `nodeFrom` to `nodeTo` in exactly one local
@@ -196,7 +213,7 @@ predicate modeledTaintStep(DataFlow::Node nodeIn, DataFlow::Node nodeOut, string
   // Taint flow from a pointer argument to an output, when the model specifies flow from the deref
   // to that output, but the deref is not modeled in the IR for the caller.
   exists(
-    CallInstruction call, DataFlow::SideEffectOperandNode indirectArgument, Function func,
+    CallInstruction call, SideEffectOperandNode indirectArgument, Function func,
     FunctionInput modelIn, FunctionOutput modelOut
   |
     indirectArgument = callInput(call, modelIn) and

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

@@ -495,7 +495,7 @@ class FieldInstruction extends Instruction {
  * `FunctionAddress` instruction.
  */
 class FunctionInstruction extends Instruction {
-  Language::Function funcSymbol;
+  Language::Declaration funcSymbol;
 
   FunctionInstruction() { funcSymbol = Raw::getInstructionFunction(this) }
 
@@ -504,7 +504,7 @@ class FunctionInstruction extends Instruction {
   /**
    * Gets the function that this instruction references.
    */
-  final Language::Function getFunctionSymbol() { result = funcSymbol }
+  final Language::Declaration getFunctionSymbol() { result = funcSymbol }
 }
 
 /**
@@ -1678,7 +1678,7 @@ class CallInstruction extends Instruction {
   /**
    * Gets the `Function` that the call targets, if this is statically known.
    */
-  final Language::Function getStaticCallTarget() {
+  final Language::Declaration getStaticCallTarget() {
     result = this.getCallTarget().(FunctionAddressInstruction).getFunctionSymbol()
   }
 

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

@@ -495,7 +495,7 @@ class FieldInstruction extends Instruction {
  * `FunctionAddress` instruction.
  */
 class FunctionInstruction extends Instruction {
-  Language::Function funcSymbol;
+  Language::Declaration funcSymbol;
 
   FunctionInstruction() { funcSymbol = Raw::getInstructionFunction(this) }
 
@@ -504,7 +504,7 @@ class FunctionInstruction extends Instruction {
   /**
    * Gets the function that this instruction references.
    */
-  final Language::Function getFunctionSymbol() { result = funcSymbol }
+  final Language::Declaration getFunctionSymbol() { result = funcSymbol }
 }
 
 /**
@@ -1678,7 +1678,7 @@ class CallInstruction extends Instruction {
   /**
    * Gets the `Function` that the call targets, if this is statically known.
    */
-  final Language::Function getStaticCallTarget() {
+  final Language::Declaration getStaticCallTarget() {
     result = this.getCallTarget().(FunctionAddressInstruction).getFunctionSymbol()
   }
 

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

@@ -15,6 +15,8 @@ private import TranslatedCall
 private import TranslatedStmt
 private import TranslatedFunction
 private import TranslatedGlobalVar
+private import TranslatedNonStaticDataMember
+private import TranslatedInitialization
 
 TranslatedElement getInstructionTranslatedElement(Instruction instruction) {
   instruction = TRawInstruction(result, _)
@@ -44,6 +46,9 @@ module Raw {
       or
       not var.isFromUninstantiatedTemplate(_) and
       var instanceof StaticInitializedStaticLocalVariable
+      or
+      not var.isFromUninstantiatedTemplate(_) and
+      var instanceof Field
     ) and
     var.hasInitializer() and
     (
@@ -63,6 +68,8 @@ module Raw {
     getTranslatedFunction(decl).hasUserVariable(var, type)
     or
     getTranslatedVarInit(decl).hasUserVariable(var, type)
+    or
+    getTranslatedFieldInit(decl).hasUserVariable(var, type)
   }
 
   cached
@@ -109,7 +116,7 @@ module Raw {
   }
 
   cached
-  Function getInstructionFunction(Instruction instruction) {
+  Declaration getInstructionFunction(Instruction instruction) {
     result =
       getInstructionTranslatedElement(instruction)
           .getInstructionFunction(getInstructionTag(instruction))
@@ -194,6 +201,89 @@ module Raw {
   Expr getInstructionUnconvertedResultExpression(Instruction instruction) {
     result = getInstructionConvertedResultExpression(instruction).getUnconverted()
   }
+
+  /**
+   * Gets the expression associated with the instruction `instr` that computes
+   * the `Convert` instruction on the extent expression of an allocation.
+   */
+  cached
+  Expr getAllocationExtentConvertExpr(Instruction instr) {
+    exists(TranslatedNonConstantAllocationSize tas |
+      instr = tas.getInstruction(AllocationExtentConvertTag()) and
+      result = tas.getExtent().getExpr()
+    )
+  }
+
+  /**
+   * Gets the `ParenthesisExpr` associated with a transparent conversion
+   * instruction, if any.
+   */
+  cached
+  ParenthesisExpr getTransparentConversionParenthesisExpr(Instruction instr) {
+    exists(TranslatedTransparentConversion ttc |
+      result = ttc.getExpr() and
+      instr = ttc.getResult()
+    )
+  }
+
+  /**
+   * Holds if `instr` belongs to a `TranslatedCoreExpr` that produces an
+   * expression result. This indicates that the instruction represents a
+   * definition whose result should be mapped back to the expression.
+   */
+  cached
+  predicate instructionProducesExprResult(Instruction instr) {
+    exists(TranslatedCoreExpr tco |
+      tco.getInstruction(_) = instr and
+      tco.producesExprResult()
+    )
+  }
+
+  /**
+   * Gets the expression associated with a `StoreInstruction` generated
+   * by an `TranslatedAssignOperation`.
+   */
+  cached
+  Expr getAssignOperationStoreExpr(StoreInstruction store) {
+    exists(TranslatedAssignOperation tao |
+      store = tao.getInstruction(AssignmentStoreTag()) and
+      result = tao.getExpr()
+    )
+  }
+
+  /**
+   * Gets the expression associated with a `StoreInstruction` generated
+   * by an `TranslatedCrementOperation`.
+   */
+  cached
+  Expr getCrementOperationStoreExpr(StoreInstruction store) {
+    exists(TranslatedCrementOperation tco |
+      store = tco.getInstruction(CrementStoreTag()) and
+      result = tco.getExpr()
+    )
+  }
+
+  /**
+   * Holds if `store` is a `StoreInstruction` that defines the temporary
+   * `IRVariable` generated as part of the translation of a ternary expression.
+   */
+  cached
+  predicate isConditionalExprTempStore(StoreInstruction store) {
+    exists(TranslatedConditionalExpr tce |
+      store = tce.getInstruction(ConditionValueFalseStoreTag())
+      or
+      store = tce.getInstruction(ConditionValueTrueStoreTag())
+    )
+  }
+
+  /** Gets the instruction that computes the address used to initialize `v`. */
+  cached
+  Instruction getInitializationTargetAddress(IRVariable v) {
+    exists(TranslatedVariableInitialization init |
+      init.getIRVariable() = v and
+      result = init.getTargetAddress()
+    )
+  }
 }
 
 class TStageInstruction = TRawInstruction or TRawUnreachedInstruction;

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

@@ -130,27 +130,31 @@ private predicate hasDefaultSideEffect(Call call, ParameterIndex i, boolean buff
 }
 
 /**
- * A `Call` or `NewOrNewArrayExpr` or `DeleteOrDeleteArrayExpr`.
+ * An expression that can have call side effects.
  *
- * All kinds of expression invoke a function as part of their evaluation. This class provides a
- * way to treat both kinds of function similarly, and to get the invoked `Function`.
+ * All kinds of expressions invoke a function as part of their evaluation. This class provides a
+ * way to treat those expressions similarly, and to get the invoked `Declaration`.
  */
-class CallOrAllocationExpr extends Expr {
-  CallOrAllocationExpr() {
+class ExprWithCallSideEffects extends Expr {
+  ExprWithCallSideEffects() {
     this instanceof Call
     or
     this instanceof NewOrNewArrayExpr
     or
     this instanceof DeleteOrDeleteArrayExpr
+    or
+    this instanceof ConstructorDefaultFieldInit
   }
 
-  /** Gets the `Function` invoked by this expression, if known. */
-  final Function getTarget() {
+  /** Gets the `Declaration` invoked by this expression, if known. */
+  final Declaration getTarget() {
     result = this.(Call).getTarget()
     or
     result = this.(NewOrNewArrayExpr).getAllocator()
     or
     result = this.(DeleteOrDeleteArrayExpr).getDeallocator()
+    or
+    result = this.(ConstructorDefaultFieldInit).getTarget()
   }
 }
 
@@ -158,7 +162,7 @@ class CallOrAllocationExpr extends Expr {
  * Returns the side effect opcode, if any, that represents any side effects not specifically modeled
  * by an argument side effect.
  */
-Opcode getCallSideEffectOpcode(CallOrAllocationExpr expr) {
+Opcode getCallSideEffectOpcode(ExprWithCallSideEffects expr) {
   not exists(expr.getTarget().(SideEffectFunction)) and result instanceof Opcode::CallSideEffect
   or
   exists(SideEffectFunction sideEffectFunction |
@@ -175,7 +179,7 @@ Opcode getCallSideEffectOpcode(CallOrAllocationExpr expr) {
 /**
  * Returns a side effect opcode for parameter index `i` of the specified call.
  *
- * This predicate will return at most two results: one read side effect, and one write side effect.
+ * This predicate will yield at most two results: one read side effect, and one write side effect.
  */
 Opcode getASideEffectOpcode(Call call, ParameterIndex i) {
   exists(boolean buffer |
@@ -228,3 +232,14 @@ Opcode getASideEffectOpcode(Call call, ParameterIndex i) {
     )
   )
 }
+
+/**
+ * Returns a side effect opcode for a default field initialization.
+ *
+ * This predicate will yield two results: one read side effect, and one write side effect.
+ */
+Opcode getDefaultFieldInitSideEffectOpcode() {
+  result instanceof Opcode::IndirectReadSideEffect
+  or
+  result instanceof Opcode::IndirectMayWriteSideEffect
+}

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

@@ -10,6 +10,7 @@ private import SideEffects
 private import TranslatedElement
 private import TranslatedExpr
 private import TranslatedFunction
+private import TranslatedInitialization
 private import DefaultOptions as DefaultOptions
 
 /**
@@ -348,7 +349,7 @@ class TranslatedExprCall extends TranslatedCallExpr {
 class TranslatedFunctionCall extends TranslatedCallExpr, TranslatedDirectCall {
   override FunctionCall expr;
 
-  override Function getInstructionFunction(InstructionTag tag) {
+  override Declaration getInstructionFunction(InstructionTag tag) {
     tag = CallTargetTag() and result = expr.getTarget()
   }
 
@@ -429,6 +430,9 @@ class TranslatedCallSideEffects extends TranslatedSideEffects, TTranslatedCallSi
     or
     expr instanceof DeleteOrDeleteArrayExpr and
     result = getTranslatedDeleteOrDeleteArray(expr).getInstruction(CallTag())
+    or
+    expr instanceof ConstructorDefaultFieldInit and
+    result = getTranslatedConstructorFieldInitialization(expr).getInstruction(CallTag())
   }
 }
 
@@ -504,11 +508,25 @@ abstract class TranslatedSideEffect extends TranslatedElement {
   abstract predicate sideEffectInstruction(Opcode opcode, CppType type);
 }
 
+private class CallOrDefaultFieldInit extends Expr {
+  CallOrDefaultFieldInit() {
+    this instanceof Call
+    or
+    this instanceof ConstructorDefaultFieldInit
+  }
+
+  Declaration getTarget() {
+    result = this.(Call).getTarget()
+    or
+    result = this.(ConstructorDefaultFieldInit).getTarget()
+  }
+}
+
 /**
  * The IR translation of a single argument side effect for a call.
  */
 abstract class TranslatedArgumentSideEffect extends TranslatedSideEffect {
-  Call call;
+  CallOrDefaultFieldInit callOrInit;
   int index;
   SideEffectOpcode sideEffectOpcode;
 
@@ -524,7 +542,7 @@ abstract class TranslatedArgumentSideEffect extends TranslatedSideEffect {
     result = "(read side effect for " + this.getArgString() + ")"
   }
 
-  override Call getPrimaryExpr() { result = call }
+  override Expr getPrimaryExpr() { result = callOrInit }
 
   override predicate sortOrder(int group, int indexInGroup) {
     indexInGroup = index and
@@ -586,9 +604,10 @@ abstract class TranslatedArgumentSideEffect extends TranslatedSideEffect {
     tag instanceof OnlyInstructionTag and
     operandTag instanceof BufferSizeOperandTag and
     result =
-      getTranslatedExpr(call.getArgument(call.getTarget()
-              .(SideEffectFunction)
-              .getParameterSizeIndex(index)).getFullyConverted()).getResult()
+      getTranslatedExpr(callOrInit
+            .(Call)
+            .getArgument(callOrInit.getTarget().(SideEffectFunction).getParameterSizeIndex(index))
+            .getFullyConverted()).getResult()
   }
 
   /** Holds if this side effect is a write side effect, rather than a read side effect. */
@@ -616,7 +635,7 @@ class TranslatedArgumentExprSideEffect extends TranslatedArgumentSideEffect,
   Expr arg;
 
   TranslatedArgumentExprSideEffect() {
-    this = TTranslatedArgumentExprSideEffect(call, arg, index, sideEffectOpcode)
+    this = TTranslatedArgumentExprSideEffect(callOrInit, arg, index, sideEffectOpcode)
   }
 
   final override Locatable getAst() { result = arg }
@@ -640,28 +659,31 @@ class TranslatedArgumentExprSideEffect extends TranslatedArgumentSideEffect,
  * The IR translation of an argument side effect for `*this` on a call, where there is no `Expr`
  * object that represents the `this` argument.
  *
- * The applies only to constructor calls, as the AST has exploit qualifier `Expr`s for all other
- * calls to non-static member functions.
+ * This applies to constructor calls and default field initializations, as the AST has explicit
+ * qualifier `Expr`s for all other calls to non-static member functions.
  */
-class TranslatedStructorQualifierSideEffect extends TranslatedArgumentSideEffect,
-  TTranslatedStructorQualifierSideEffect
+class TranslatedImplicitThisQualifierSideEffect extends TranslatedArgumentSideEffect,
+  TTranslatedImplicitThisQualifierSideEffect
 {
-  TranslatedStructorQualifierSideEffect() {
-    this = TTranslatedStructorQualifierSideEffect(call, sideEffectOpcode) and
+  TranslatedImplicitThisQualifierSideEffect() {
+    this = TTranslatedImplicitThisQualifierSideEffect(callOrInit, sideEffectOpcode) and
     index = -1
   }
 
-  final override Locatable getAst() { result = call }
+  final override Locatable getAst() { result = callOrInit }
 
-  final override Type getIndirectionType() { result = call.getTarget().getDeclaringType() }
+  final override Type getIndirectionType() { result = callOrInit.getTarget().getDeclaringType() }
 
   final override string getArgString() { result = "this" }
 
   final override Instruction getArgInstruction() {
     exists(TranslatedStructorCall structorCall |
-      structorCall.getExpr() = call and
+      structorCall.getExpr() = callOrInit and
       result = structorCall.getQualifierResult()
     )
+    or
+    callOrInit instanceof ConstructorDefaultFieldInit and
+    result = getTranslatedFunction(callOrInit.getEnclosingFunction()).getLoadThisInstruction()
   }
 }
 

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

@@ -36,7 +36,8 @@ abstract class TranslatedCondition extends TranslatedElement {
   final override Declaration getFunction() {
     result = getEnclosingFunction(expr) or
     result = getEnclosingVariable(expr).(GlobalOrNamespaceVariable) or
-    result = getEnclosingVariable(expr).(StaticInitializedStaticLocalVariable)
+    result = getEnclosingVariable(expr).(StaticInitializedStaticLocalVariable) or
+    result = getEnclosingVariable(expr).(Field)
   }
 
   final Type getResultType() { result = expr.getUnspecifiedType() }

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

@@ -34,8 +34,11 @@ abstract class TranslatedDeclarationEntry extends TranslatedElement, TTranslated
       or
       result = entry.getDeclaration().(GlobalOrNamespaceVariable)
       or
+      result = entry.getDeclaration().(Field)
+      or
       not entry.getDeclaration() instanceof StaticInitializedStaticLocalVariable and
       not entry.getDeclaration() instanceof GlobalOrNamespaceVariable and
+      not entry.getDeclaration() instanceof Field and
       result = stmt.getEnclosingFunction()
     )
   }

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

@@ -767,7 +767,7 @@ newtype TTranslatedElement =
       expr = initList.getFieldExpr(field, position).getFullyConverted()
     )
     or
-    exists(ConstructorFieldInit init |
+    exists(ConstructorDirectFieldInit init |
       not ignoreExpr(init) and
       ast = init and
       field = init.getTarget() and
@@ -775,6 +775,14 @@ newtype TTranslatedElement =
       position = -1
     )
   } or
+  // The initialization of a field via a default member initializer.
+  TTranslatedDefaultFieldInitialization(Expr ast, Field field) {
+    exists(ConstructorDefaultFieldInit init |
+      not ignoreExpr(init) and
+      ast = init and
+      field = init.getTarget()
+    )
+  } or
   // The value initialization of a field due to an omitted member of an
   // initializer list.
   TTranslatedFieldValueInitialization(Expr ast, Field field) {
@@ -871,7 +879,7 @@ newtype TTranslatedElement =
   // The declaration/initialization part of a `ConditionDeclExpr`
   TTranslatedConditionDecl(ConditionDeclExpr expr) { not ignoreExpr(expr) } or
   // The side effects of a `Call`
-  TTranslatedCallSideEffects(CallOrAllocationExpr expr) {
+  TTranslatedCallSideEffects(ExprWithCallSideEffects expr) {
     not ignoreExpr(expr) and
     not ignoreSideEffects(expr)
   } or
@@ -910,15 +918,23 @@ newtype TTranslatedElement =
   } or
   // Constructor calls lack a qualifier (`this`) expression, so we need to handle the side effects
   // on `*this` without an `Expr`.
-  TTranslatedStructorQualifierSideEffect(Call call, SideEffectOpcode opcode) {
+  TTranslatedImplicitThisQualifierSideEffect(ExprWithCallSideEffects call, SideEffectOpcode opcode) {
     not ignoreExpr(call) and
     not ignoreSideEffects(call) and
-    call instanceof ConstructorCall and
-    opcode = getASideEffectOpcode(call, -1)
+    (
+      call instanceof ConstructorCall and
+      opcode = getASideEffectOpcode(call, -1)
+      or
+      call instanceof ConstructorFieldInit and
+      opcode = getDefaultFieldInitSideEffectOpcode()
+    )
   } or
   // The side effect that initializes newly-allocated memory.
   TTranslatedAllocationSideEffect(AllocationExpr expr) { not ignoreSideEffects(expr) } or
-  TTranslatedStaticStorageDurationVarInit(Variable var) { Raw::varHasIRFunc(var) } or
+  TTranslatedStaticStorageDurationVarInit(Variable var) {
+    Raw::varHasIRFunc(var) and not var instanceof Field
+  } or
+  TTranslatedNonStaticDataMemberVarInit(Field var) { Raw::varHasIRFunc(var) } or
   TTranslatedAssertionOperand(MacroInvocation mi, int index) { hasAssertionOperand(mi, index) }
 
 /**
@@ -1179,7 +1195,7 @@ abstract class TranslatedElement extends TTranslatedElement {
    * If the instruction specified by `tag` is a `FunctionInstruction`, gets the
    * `Function` for that instruction.
    */
-  Function getInstructionFunction(InstructionTag tag) { none() }
+  Declaration getInstructionFunction(InstructionTag tag) { none() }
 
   /**
    * If the instruction specified by `tag` is a `VariableInstruction`, gets the
@@ -1297,5 +1313,7 @@ abstract class TranslatedRootElement extends TranslatedElement {
     this instanceof TTranslatedFunction
     or
     this instanceof TTranslatedStaticStorageDurationVarInit
+    or
+    this instanceof TTranslatedNonStaticDataMemberVarInit
   }
 }

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

@@ -14,6 +14,7 @@ private import TranslatedFunction
 private import TranslatedInitialization
 private import TranslatedStmt
 private import TranslatedGlobalVar
+private import TranslatedNonStaticDataMember
 private import IRConstruction
 import TranslatedCall
 
@@ -138,6 +139,8 @@ abstract class TranslatedExpr extends TranslatedElement {
     result = getTranslatedFunction(getEnclosingFunction(expr))
     or
     result = getTranslatedVarInit(getEnclosingVariable(expr))
+    or
+    result = getTranslatedFieldInit(getEnclosingVariable(expr))
   }
 }
 
@@ -153,7 +156,10 @@ Declaration getEnclosingDeclaration0(Expr e) {
     i.getExpr().getFullyConverted() = e and
     v = i.getDeclaration()
   |
-    if v instanceof StaticInitializedStaticLocalVariable or v instanceof GlobalOrNamespaceVariable
+    if
+      v instanceof StaticInitializedStaticLocalVariable or
+      v instanceof GlobalOrNamespaceVariable or
+      v instanceof Field
     then result = v
     else result = e.getEnclosingDeclaration()
   )
@@ -173,7 +179,10 @@ Variable getEnclosingVariable0(Expr e) {
     i.getExpr().getFullyConverted() = e and
     v = i.getDeclaration()
   |
-    if v instanceof StaticInitializedStaticLocalVariable or v instanceof GlobalOrNamespaceVariable
+    if
+      v instanceof StaticInitializedStaticLocalVariable or
+      v instanceof GlobalOrNamespaceVariable or
+      v instanceof Field
     then result = v
     else result = e.getEnclosingVariable()
   )
@@ -826,6 +835,46 @@ class TranslatedPostfixCrementOperation extends TranslatedCrementOperation {
   override Instruction getResult() { result = this.getLoadedOperand().getResult() }
 }
 
+class TranslatedParamAccessForType extends TranslatedNonConstantExpr {
+  override ParamAccessForType expr;
+
+  TranslatedParamAccessForType() {
+    // Currently only needed for this parameter accesses.
+    expr.isThisAccess()
+  }
+
+  final override Instruction getFirstInstruction(EdgeKind kind) {
+    result = this.getInstruction(OnlyInstructionTag()) and
+    kind instanceof GotoEdge
+  }
+
+  override Instruction getALastInstructionInternal() {
+    result = this.getInstruction(OnlyInstructionTag())
+  }
+
+  final override TranslatedElement getChildInternal(int id) { none() }
+
+  override Instruction getInstructionSuccessorInternal(InstructionTag tag, EdgeKind kind) {
+    tag = OnlyInstructionTag() and
+    result = this.getParent().getChildSuccessor(this, kind)
+  }
+
+  override Instruction getResult() { result = this.getInstruction(OnlyInstructionTag()) }
+
+  override predicate hasInstruction(Opcode opcode, InstructionTag tag, CppType resultType) {
+    tag = OnlyInstructionTag() and
+    opcode instanceof Opcode::CopyValue and
+    resultType = getTypeForPRValue(expr.getType())
+  }
+
+  override Instruction getInstructionRegisterOperand(InstructionTag tag, OperandTag operandTag) {
+    tag = OnlyInstructionTag() and
+    operandTag instanceof UnaryOperandTag and
+    result =
+      this.getEnclosingFunction().(TranslatedNonStaticDataMemberVarInit).getLoadThisInstruction()
+  }
+}
+
 /**
  * IR translation of an array access expression (e.g. `a[i]`). The array being accessed will either
  * be a prvalue of pointer type (possibly due to an implicit array-to-pointer conversion), or a
@@ -1215,7 +1264,7 @@ class TranslatedFunctionAccess extends TranslatedNonConstantExpr {
     resultType = this.getResultType()
   }
 
-  override Function getInstructionFunction(InstructionTag tag) {
+  override Declaration getInstructionFunction(InstructionTag tag) {
     tag = OnlyInstructionTag() and
     result = expr.getTarget()
   }
@@ -2498,7 +2547,7 @@ class TranslatedAllocatorCall extends TTranslatedAllocatorCall, TranslatedDirect
     any()
   }
 
-  override Function getInstructionFunction(InstructionTag tag) {
+  override Declaration getInstructionFunction(InstructionTag tag) {
     tag = CallTargetTag() and result = expr.getAllocator()
   }
 
@@ -2581,7 +2630,7 @@ class TranslatedDeleteOrDeleteArrayExpr extends TranslatedNonConstantExpr, Trans
     result = this.getFirstArgumentOrCallInstruction(kind)
   }
 
-  override Function getInstructionFunction(InstructionTag tag) {
+  override Declaration getInstructionFunction(InstructionTag tag) {
     tag = CallTargetTag() and result = expr.getDeallocator()
   }
 

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

@@ -148,7 +148,8 @@ abstract class TranslatedInitialization extends TranslatedElement, TTranslatedIn
   final override Declaration getFunction() {
     result = getEnclosingFunction(expr) or
     result = getEnclosingVariable(expr).(GlobalOrNamespaceVariable) or
-    result = getEnclosingVariable(expr).(StaticInitializedStaticLocalVariable)
+    result = getEnclosingVariable(expr).(StaticInitializedStaticLocalVariable) or
+    result = getEnclosingVariable(expr).(Field)
   }
 
   final override Locatable getAst() { result = expr }
@@ -514,8 +515,8 @@ TranslatedFieldInitialization getTranslatedConstructorFieldInitialization(Constr
 }
 
 /**
- * Represents the IR translation of the initialization of a field from an
- * element of an initializer list.
+ * The IR translation of the initialization of a field from an element of
+ * an initializer list.
  */
 abstract class TranslatedFieldInitialization extends TranslatedElement {
   Expr ast;
@@ -528,13 +529,11 @@ abstract class TranslatedFieldInitialization extends TranslatedElement {
   final override Declaration getFunction() {
     result = getEnclosingFunction(ast) or
     result = getEnclosingVariable(ast).(GlobalOrNamespaceVariable) or
-    result = getEnclosingVariable(ast).(StaticInitializedStaticLocalVariable)
+    result = getEnclosingVariable(ast).(StaticInitializedStaticLocalVariable) or
+    result = getEnclosingVariable(ast).(Field)
   }
 
-  final override Instruction getFirstInstruction(EdgeKind kind) {
-    result = this.getInstruction(this.getFieldAddressTag()) and
-    kind instanceof GotoEdge
-  }
+  final Field getField() { result = field }
 
   /**
    * Gets the zero-based index describing the order in which this field is to be
@@ -542,6 +541,20 @@ abstract class TranslatedFieldInitialization extends TranslatedElement {
    */
   final int getOrder() { result = field.getInitializationOrder() }
 
+  /** Gets the position in the initializer list, or `-1` if the initialization is implicit. */
+  int getPosition() { result = -1 }
+}
+
+/**
+ * The IR translation of the initialization of a field from an element of an initializer
+ * list where default initialization is not used.
+ */
+abstract class TranslatedNonDefaultFieldInitialization extends TranslatedFieldInitialization {
+  final override Instruction getFirstInstruction(EdgeKind kind) {
+    result = this.getInstruction(this.getFieldAddressTag()) and
+    kind instanceof GotoEdge
+  }
+
   override predicate hasInstruction(Opcode opcode, InstructionTag tag, CppType resultType) {
     tag = this.getFieldAddressTag() and
     opcode instanceof Opcode::FieldAddress and
@@ -559,18 +572,13 @@ abstract class TranslatedFieldInitialization extends TranslatedElement {
   }
 
   final InstructionTag getFieldAddressTag() { result = InitializerFieldAddressTag() }
-
-  final Field getField() { result = field }
-
-  /** Gets the position in the initializer list, or `-1` if the initialization is implicit. */
-  int getPosition() { result = -1 }
 }
 
 /**
- * Represents the IR translation of the initialization of a field from an
- * explicit element in an initializer list.
+ * The IR translation of the initialization of a field from an explicit element in
+ * an initializer list.
  */
-class TranslatedExplicitFieldInitialization extends TranslatedFieldInitialization,
+class TranslatedExplicitFieldInitialization extends TranslatedNonDefaultFieldInitialization,
   InitializationContext, TTranslatedExplicitFieldInitialization
 {
   Expr expr;
@@ -610,15 +618,81 @@ class TranslatedExplicitFieldInitialization extends TranslatedFieldInitializatio
   override int getPosition() { result = position }
 }
 
+/**
+ * The IR translation of the initialization of a field from an element of an initializer
+ * list where default initialization is used.
+ */
+class TranslatedDefaultFieldInitialization extends TranslatedFieldInitialization,
+  TTranslatedDefaultFieldInitialization
+{
+  TranslatedDefaultFieldInitialization() {
+    this = TTranslatedDefaultFieldInitialization(ast, field)
+  }
+
+  final override Instruction getFirstInstruction(EdgeKind kind) {
+    result = this.getInstruction(CallTargetTag()) and
+    kind instanceof GotoEdge
+  }
+
+  override Instruction getALastInstructionInternal() {
+    result = this.getSideEffects().getALastInstruction()
+  }
+
+  override TranslatedElement getLastChild() { result = this.getSideEffects() }
+
+  override Instruction getInstructionSuccessorInternal(InstructionTag tag, EdgeKind kind) {
+    tag = CallTargetTag() and
+    result = this.getInstruction(CallTag())
+    or
+    tag = CallTag() and
+    result = this.getSideEffects().getFirstInstruction(kind)
+  }
+
+  override Instruction getChildSuccessorInternal(TranslatedElement child, EdgeKind kind) {
+    child = this.getSideEffects() and
+    result = this.getParent().getChildSuccessor(this, kind)
+  }
+
+  override predicate hasInstruction(Opcode opcode, InstructionTag tag, CppType resultType) {
+    tag = CallTargetTag() and
+    opcode instanceof Opcode::FunctionAddress and
+    resultType = getFunctionGLValueType()
+    or
+    tag = CallTag() and
+    opcode instanceof Opcode::Call and
+    resultType = getVoidType()
+  }
+
+  override Instruction getInstructionRegisterOperand(InstructionTag tag, OperandTag operandTag) {
+    tag = CallTag() and
+    (
+      operandTag instanceof CallTargetOperandTag and
+      result = this.getInstruction(CallTargetTag())
+      or
+      operandTag instanceof ThisArgumentOperandTag and
+      result = getTranslatedFunction(this.getFunction()).getLoadThisInstruction()
+    )
+  }
+
+  override Declaration getInstructionFunction(InstructionTag tag) {
+    tag = CallTargetTag() and
+    result = field
+  }
+
+  override TranslatedElement getChild(int id) { id = 0 and result = this.getSideEffects() }
+
+  final TranslatedSideEffects getSideEffects() { result.getExpr() = ast }
+}
+
 private string getZeroValue(Type type) {
   if type instanceof FloatingPointType then result = "0.0" else result = "0"
 }
 
 /**
- * Represents the IR translation of the initialization of a field without a
- * corresponding element in the initializer list.
+ * The IR translation of the initialization of a field without a corresponding
+ * element in the initializer list.
  */
-class TranslatedFieldValueInitialization extends TranslatedFieldInitialization,
+class TranslatedFieldValueInitialization extends TranslatedNonDefaultFieldInitialization,
   TTranslatedFieldValueInitialization
 {
   TranslatedFieldValueInitialization() { this = TTranslatedFieldValueInitialization(ast, field) }
@@ -628,7 +702,7 @@ class TranslatedFieldValueInitialization extends TranslatedFieldInitialization,
   }
 
   override predicate hasInstruction(Opcode opcode, InstructionTag tag, CppType resultType) {
-    TranslatedFieldInitialization.super.hasInstruction(opcode, tag, resultType)
+    TranslatedNonDefaultFieldInitialization.super.hasInstruction(opcode, tag, resultType)
     or
     tag = this.getFieldDefaultValueTag() and
     opcode instanceof Opcode::Constant and
@@ -659,7 +733,8 @@ class TranslatedFieldValueInitialization extends TranslatedFieldInitialization,
   }
 
   override Instruction getInstructionRegisterOperand(InstructionTag tag, OperandTag operandTag) {
-    result = TranslatedFieldInitialization.super.getInstructionRegisterOperand(tag, operandTag)
+    result =
+      TranslatedNonDefaultFieldInitialization.super.getInstructionRegisterOperand(tag, operandTag)
     or
     tag = this.getFieldDefaultValueStoreTag() and
     (
@@ -683,8 +758,8 @@ class TranslatedFieldValueInitialization extends TranslatedFieldInitialization,
 }
 
 /**
- * Represents the IR translation of the initialization of an array element from
- * an element of an initializer list.
+ * The IR translation of the initialization of an array element from an element
+ * of an initializer list.
  */
 abstract class TranslatedElementInitialization extends TranslatedElement {
   ArrayOrVectorAggregateLiteral initList;
@@ -701,6 +776,8 @@ abstract class TranslatedElementInitialization extends TranslatedElement {
     result = getEnclosingVariable(initList).(GlobalOrNamespaceVariable)
     or
     result = getEnclosingVariable(initList).(StaticInitializedStaticLocalVariable)
+    or
+    result = getEnclosingVariable(initList).(Field)
   }
 
   final override Instruction getFirstInstruction(EdgeKind kind) {
@@ -759,8 +836,8 @@ abstract class TranslatedElementInitialization extends TranslatedElement {
 }
 
 /**
- * Represents the IR translation of the initialization of an array element from
- * an explicit element in an initializer list.
+ * The IR translation of the initialization of an array element from an explicit
+ * element in an initializer list.
  */
 class TranslatedExplicitElementInitialization extends TranslatedElementInitialization,
   TTranslatedExplicitElementInitialization, InitializationContext
@@ -808,8 +885,8 @@ class TranslatedExplicitElementInitialization extends TranslatedElementInitializ
 }
 
 /**
- * Represents the IR translation of the initialization of a range of array
- * elements without corresponding elements in the initializer list.
+ * The IR translation of the initialization of a range of array elements without
+ * corresponding elements in the initializer list.
  */
 class TranslatedElementValueInitialization extends TranslatedElementInitialization,
   TTranslatedElementValueInitialization