Bazel/rust: use patched tree-sitter

.bazelrc

@@ -10,19 +10,15 @@ common --override_module=semmle_code=%workspace%/misc/bazel/semmle_code_stub
 
 build --repo_env=CC=clang --repo_env=CXX=clang++
 
-# we use transitions that break builds of `...`, so for `test` to work with that we need the following
-test --build_tests_only
+build:linux --cxxopt=-std=c++20
+build:macos --cxxopt=-std=c++20 --cpu=darwin_x86_64
+build:windows --cxxopt=/std:c++20 --cxxopt=/Zc:preprocessor
 
 # this requires developer mode, but is required to have pack installer functioning
 startup --windows_enable_symlinks
 common --enable_runfiles
 
-# with the above, we can avoid building python zips which is the default on windows as that's expensive
-build --nobuild_python_zip
-
 common --registry=file:///%workspace%/misc/bazel/registry
 common --registry=https://bcr.bazel.build
 
-common --@rules_dotnet//dotnet/settings:strict_deps=false
-
 try-import %workspace%/local.bazelrc

.bazelrc.internal

@@ -2,9 +2,3 @@
 
 common --registry=file:///%workspace%/ql/misc/bazel/registry
 common --registry=https://bcr.bazel.build
-
-# See bazelbuild/rules_dotnet#413: strict_deps in C# also appliy to 3rd-party deps, and when we pull
-# in (for example) the xunit package, there's no code in this at all, it just depends transitively on
-# its implementation packages without providing any code itself.
-# We either can depend on internal implementation details, or turn of strict deps.
-common --@rules_dotnet//dotnet/settings:strict_deps=false

.bazelversion

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

.gitattributes

@@ -50,40 +50,26 @@
 *.dll -text
 *.pdb -text
 
-/java/ql/test/stubs/**/*.java linguist-generated=true
-/java/ql/test/experimental/stubs/**/*.java linguist-generated=true
-/java/kotlin-extractor/deps/*.jar filter=lfs diff=lfs merge=lfs -text
+java/ql/test/stubs/**/*.java linguist-generated=true
+java/ql/test/experimental/stubs/**/*.java linguist-generated=true
 
 # Force git not to modify line endings for go or html files under the go/ql directory
-/go/ql/**/*.go -text
-/go/ql/**/*.html -text
+go/ql/**/*.go -text
+go/ql/**/*.html -text
 # Force git not to modify line endings for go dbschemes
-/go/*.dbscheme -text
+go/*.dbscheme -text
 # Preserve unusual line ending from codeql-go merge
-/go/extractor/opencsv/CSVReader.java -text
+go/extractor/opencsv/CSVReader.java -text
 
 # For some languages, upgrade script testing references really old dbscheme
 # files from legacy upgrades that have CRLF line endings. Since upgrade
 # resolution relies on object hashes, we must suppress line ending conversion
 # for those testing dbscheme files.
-/*/ql/lib/upgrades/initial/*.dbscheme -text
+*/ql/lib/upgrades/initial/*.dbscheme -text
 
 # Auto-generated modeling for Python
-/python/ql/lib/semmle/python/frameworks/data/internal/subclass-capture/*.yml linguist-generated=true
+python/ql/lib/semmle/python/frameworks/data/internal/subclass-capture/*.yml linguist-generated=true
 
 # auto-generated bazel lock file
-/ruby/extractor/cargo-bazel-lock.json linguist-generated=true
-/ruby/extractor/cargo-bazel-lock.json -merge
-
-# auto-generated files for the C# build
-/csharp/paket.lock linguist-generated=true
-# needs eol=crlf, as `paket` touches this file and saves it as crlf
-/csharp/.paket/Paket.Restore.targets linguist-generated=true eol=crlf
-/csharp/paket.main.bzl linguist-generated=true
-/csharp/paket.main_extension.bzl linguist-generated=true
-
-# ripunzip tool
-/misc/ripunzip/ripunzip-* filter=lfs diff=lfs merge=lfs -text
-
-# swift prebuilt resources
-/swift/third_party/resource-dir/*.zip filter=lfs diff=lfs merge=lfs -text
+ruby/extractor/cargo-bazel-lock.json linguist-generated=true
+ruby/extractor/cargo-bazel-lock.json -merge

.github/workflows/build-ripunzip.yml

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

.github/workflows/codeql-analysis.yml

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

.github/workflows/csharp-qltest.yml

@@ -65,7 +65,7 @@ jobs:
           key: csharp-qltest-${{ matrix.slice }}
       - name: Run QL tests
         run: |
-          codeql test run --threads=0 --ram 50000 --slice ${{ matrix.slice }} --search-path "${{ github.workspace }}" --check-databases --check-undefined-labels --check-repeated-labels --check-redefined-labels --consistency-queries ql/consistency-queries ql/test --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
+          codeql test run --threads=0 --ram 50000 --slice ${{ matrix.slice }} --search-path extractor-pack --check-databases --check-undefined-labels --check-repeated-labels --check-redefined-labels --consistency-queries ql/consistency-queries ql/test --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
         env:
           GITHUB_TOKEN: ${{ github.token }}
   unit-tests:
@@ -81,11 +81,10 @@ jobs:
           dotnet-version: 8.0.101
       - name: Extractor unit tests
         run: |
-          dotnet tool restore
           dotnet test -p:RuntimeFrameworkVersion=8.0.1 extractor/Semmle.Util.Tests
           dotnet test -p:RuntimeFrameworkVersion=8.0.1 extractor/Semmle.Extraction.Tests
           dotnet test -p:RuntimeFrameworkVersion=8.0.1 autobuilder/Semmle.Autobuild.CSharp.Tests
-          dotnet test -p:RuntimeFrameworkVersion=8.0.1 autobuilder/Semmle.Autobuild.Cpp.Tests
+          dotnet test -p:RuntimeFrameworkVersion=8.0.1 "${{ github.workspace }}/cpp/autobuilder/Semmle.Autobuild.Cpp.Tests"
         shell: bash
   stubgentest:
     runs-on: ubuntu-latest
@@ -101,6 +100,6 @@ jobs:
           # Update existing stubs in the repo with the freshly generated ones
           mv "$STUBS_PATH/output/stubs/_frameworks" ql/test/resources/stubs/
           git status
-          codeql test run --threads=0 --search-path "${{ github.workspace }}" --check-databases --check-undefined-labels --check-repeated-labels --check-redefined-labels --consistency-queries ql/consistency-queries -- ql/test/library-tests/dataflow/flowsources/aspremote
+          codeql test run --threads=0 --search-path extractor-pack --check-databases --check-undefined-labels --check-repeated-labels --check-redefined-labels --consistency-queries ql/consistency-queries -- ql/test/library-tests/dataflow/flowsources/aspremote
         env:
           GITHUB_TOKEN: ${{ github.token }}

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

@@ -7,9 +7,8 @@ on:
       - .github/workflows/go-tests-other-os.yml
       - .github/actions/**
       - codeql-workspace.yml
-      - MODULE.bazel
-      - .bazelrc
-      - misc/bazel/**
+env:
+  GO_VERSION: '~1.22.0'
 
 permissions:
   contents: read
@@ -19,17 +18,72 @@ jobs:
     name: Test MacOS
     runs-on: macos-latest
     steps:
+      - name: Set up Go ${{ env.GO_VERSION }}
+        uses: actions/setup-go@v5
+        with:
+          go-version: ${{ env.GO_VERSION }}
+          cache: false
+        id: go
+
       - name: Check out code
         uses: actions/checkout@v4
-      - name: Run tests
-        uses: ./go/actions/test
+
+      - name: Set up CodeQL CLI
+        uses: ./.github/actions/fetch-codeql
+
+      - name: Enable problem matchers in repository
+        shell: bash
+        run: 'find .github/problem-matchers -name \*.json -exec echo "::add-matcher::{}" \;'
+
+      - name: Build
+        run: |
+          cd go
+          make
+
+      - name: Cache compilation cache
+        id: query-cache
+        uses: ./.github/actions/cache-query-compilation
+        with:
+          key: go-qltest
+      - name: Test
+        run: |
+          cd go
+          make test cache="${{ steps.query-cache.outputs.cache-dir }}"
 
   test-win:
     if: github.repository_owner == 'github'
     name: Test Windows
     runs-on: windows-latest-xl
     steps:
+      - name: Set up Go ${{ env.GO_VERSION }}
+        uses: actions/setup-go@v5
+        with:
+          go-version: ${{ env.GO_VERSION }}
+          cache: false
+        id: go
+
       - name: Check out code
         uses: actions/checkout@v4
-      - name: Run tests
-        uses: ./go/actions/test
+
+      - name: Set up CodeQL CLI
+        uses: ./.github/actions/fetch-codeql
+
+      - name: Enable problem matchers in repository
+        shell: bash
+        run: 'find .github/problem-matchers -name \*.json -exec echo "::add-matcher::{}" \;'
+
+      - name: Build
+        run: |
+          cd go
+          make
+
+      - name: Cache compilation cache
+        id: query-cache
+        uses: ./.github/actions/cache-query-compilation
+        with:
+          key: go-qltest
+
+      - name: Test
+        run: |
+          cd go
+          make test cache="${{ steps.query-cache.outputs.cache-dir }}"

.github/workflows/go-tests.yml

@@ -15,9 +15,9 @@ on:
       - .github/workflows/go-tests.yml
       - .github/actions/**
       - codeql-workspace.yml
-      - MODULE.bazel
-      - .bazelrc
-      - misc/bazel/**
+
+env:
+  GO_VERSION: '~1.22.0'
 
 permissions:
   contents: read
@@ -28,9 +28,51 @@ jobs:
     name: Test Linux (Ubuntu)
     runs-on: ubuntu-latest-xl
     steps:
+      - name: Set up Go ${{ env.GO_VERSION }}
+        uses: actions/setup-go@v5
+        with:
+          go-version: ${{ env.GO_VERSION }}
+          cache: false
+        id: go
+
       - name: Check out code
         uses: actions/checkout@v4
-      - name: Run tests
-        uses: ./go/actions/test
+
+      - name: Set up CodeQL CLI
+        uses: ./.github/actions/fetch-codeql
+
+      - name: Enable problem matchers in repository
+        shell: bash
+        run: 'find .github/problem-matchers -name \*.json -exec echo "::add-matcher::{}" \;'
+
+      - name: Build
+        run: |
+          cd go
+          make
+
+      - name: Check that all Go code is autoformatted
+        run: |
+          cd go
+          make check-formatting
+
+      - name: Compile qhelp files to markdown
+        run: |
+          cd go
+          env QHELP_OUT_DIR=qhelp-out make qhelp-to-markdown
+
+      - name: Upload qhelp markdown
+        uses: actions/upload-artifact@v3
         with:
-          run-code-checks: true
+          name: qhelp-markdown
+          path: go/qhelp-out/**/*.md
+
+      - name: Cache compilation cache
+        id: query-cache
+        uses: ./.github/actions/cache-query-compilation
+        with:
+          key: go-qltest
+
+      - name: Test
+        run: |
+          cd go
+          make test cache="${{ steps.query-cache.outputs.cache-dir }}"

.github/workflows/kotlin-build.yml

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

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

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

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

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

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

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

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

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

.github/workflows/ruby-qltest.yml

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

.github/workflows/swift.yml

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

.github/workflows/zipmerge-test.yml

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

.gitignore

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

.lfsconfig

@@ -2,6 +2,4 @@
 # codeql is publicly forked by many users, and we don't want any LFS file polluting their working
 # copies. We therefore exclude everything by default.
 # For files required by bazel builds, use rules in `misc/bazel/lfs.bzl` to download them on demand.
-# we go for `fetchinclude` to something not exsiting rather than `fetchexclude = *` because the
-# former is easier to override (with `git -c` or a local git config) to fetch something specific
 fetchinclude = /nothing

.pre-commit-config.yaml

@@ -29,13 +29,12 @@ repos:
         entry: bazel run //misc/bazel:buildifier
         pass_filenames: false
 
-#      DISABLED: can be enabled by copying this config and installing `pre-commit` with `--config` on the copy
-#      - id: go-gen
-#        name: Check checked in generated files in go
-#        files: ^go/.*
-#        language: system
-#        entry: bazel run //go:gen
-#        pass_filenames: false
+      - id: go-gen
+        name: Check checked in generated files in go
+        files: ^go/.*
+        language: system
+        entry: bazel run //go:gen
+        pass_filenames: false
 
       - id: codeql-format
         name: Fix QL file formatting

BUILD.bazel

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

CODEOWNERS

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

CONTRIBUTING.md

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

MODULE.bazel

@@ -22,22 +22,10 @@ bazel_dep(name = "bazel_skylib", version = "1.5.0")
 bazel_dep(name = "abseil-cpp", version = "20240116.0", repo_name = "absl")
 bazel_dep(name = "nlohmann_json", version = "3.11.3", repo_name = "json")
 bazel_dep(name = "fmt", version = "10.0.0")
-bazel_dep(name = "rules_kotlin", version = "1.9.4-codeql.1")
 bazel_dep(name = "gazelle", version = "0.36.0")
-bazel_dep(name = "rules_dotnet", version = "0.15.1")
-bazel_dep(name = "googletest", version = "1.14.0.bcr.1")
 
 bazel_dep(name = "buildifier_prebuilt", version = "6.4.0", dev_dependency = True)
 
-dotnet = use_extension("@rules_dotnet//dotnet:extensions.bzl", "dotnet")
-dotnet.toolchain(dotnet_version = "8.0.101")
-use_repo(dotnet, "dotnet_toolchains")
-
-register_toolchains("@dotnet_toolchains//:all")
-
-csharp_main_extension = use_extension("//csharp:paket.main_extension.bzl", "main_extension")
-use_repo(csharp_main_extension, "paket.main")
-
 pip = use_extension("@rules_python//python/extensions:pip.bzl", "pip")
 pip.parse(
     hub_name = "codegen_deps",
@@ -66,84 +54,9 @@ node.toolchain(
 )
 use_repo(node, "nodejs", "nodejs_toolchains")
 
-kotlin_extractor_deps = use_extension("//java/kotlin-extractor:deps.bzl", "kotlin_extractor_deps")
-
-# following list can be kept in sync by running `bazel mod tidy` in `codeql`
-use_repo(
-    kotlin_extractor_deps,
-    "codeql_kotlin_defaults",
-    "codeql_kotlin_embeddable",
-    "kotlin-compiler-1.5.0",
-    "kotlin-compiler-1.5.10",
-    "kotlin-compiler-1.5.20",
-    "kotlin-compiler-1.5.30",
-    "kotlin-compiler-1.6.0",
-    "kotlin-compiler-1.6.20",
-    "kotlin-compiler-1.7.0",
-    "kotlin-compiler-1.7.20",
-    "kotlin-compiler-1.8.0",
-    "kotlin-compiler-1.9.0-Beta",
-    "kotlin-compiler-1.9.20-Beta",
-    "kotlin-compiler-2.0.0-RC1",
-    "kotlin-compiler-embeddable-1.5.0",
-    "kotlin-compiler-embeddable-1.5.10",
-    "kotlin-compiler-embeddable-1.5.20",
-    "kotlin-compiler-embeddable-1.5.30",
-    "kotlin-compiler-embeddable-1.6.0",
-    "kotlin-compiler-embeddable-1.6.20",
-    "kotlin-compiler-embeddable-1.7.0",
-    "kotlin-compiler-embeddable-1.7.20",
-    "kotlin-compiler-embeddable-1.8.0",
-    "kotlin-compiler-embeddable-1.9.0-Beta",
-    "kotlin-compiler-embeddable-1.9.20-Beta",
-    "kotlin-compiler-embeddable-2.0.0-RC1",
-    "kotlin-stdlib-1.5.0",
-    "kotlin-stdlib-1.5.10",
-    "kotlin-stdlib-1.5.20",
-    "kotlin-stdlib-1.5.30",
-    "kotlin-stdlib-1.6.0",
-    "kotlin-stdlib-1.6.20",
-    "kotlin-stdlib-1.7.0",
-    "kotlin-stdlib-1.7.20",
-    "kotlin-stdlib-1.8.0",
-    "kotlin-stdlib-1.9.0-Beta",
-    "kotlin-stdlib-1.9.20-Beta",
-    "kotlin-stdlib-2.0.0-RC1",
-)
-
 go_sdk = use_extension("@rules_go//go:extensions.bzl", "go_sdk")
 go_sdk.download(version = "1.22.2")
 
-lfs_files = use_repo_rule("//misc/bazel:lfs.bzl", "lfs_files")
-
-lfs_files(
-    name = "ripunzip-linux",
-    srcs = ["//misc/ripunzip:ripunzip-linux"],
-    executable = True,
-)
-
-lfs_files(
-    name = "ripunzip-windows",
-    srcs = ["//misc/ripunzip:ripunzip-windows.exe"],
-    executable = True,
-)
-
-lfs_files(
-    name = "ripunzip-macos",
-    srcs = ["//misc/ripunzip:ripunzip-macos"],
-    executable = True,
-)
-
-lfs_files(
-    name = "swift-resource-dir-linux",
-    srcs = ["//swift/third_party/resource-dir:resource-dir-linux.zip"],
-)
-
-lfs_files(
-    name = "swift-resource-dir-macos",
-    srcs = ["//swift/third_party/resource-dir:resource-dir-macos.zip"],
-)
-
 register_toolchains(
     "@nodejs_toolchains//:all",
 )

README.md

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

codeql-workspace.yml

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

config/dbscheme-fragments.json

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

config/identical-files.json

@@ -364,9 +364,5 @@
   "Python model summaries test extension": [
     "python/ql/test/library-tests/dataflow/model-summaries/InlineTaintTest.ext.yml",
     "python/ql/test/library-tests/dataflow/model-summaries/NormalDataflowTest.ext.yml"
-  ],
-  "shared tree-sitter extractor cargo.toml": [
-    "shared/tree-sitter-extractor/Cargo.toml",
-    "ruby/extractor/codeql-extractor-fake-crate/Cargo.toml"
   ]
 }

cpp/autobuilder/.gitignore

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

cpp/autobuilder/README.md

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

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

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

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

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

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

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

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

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

cpp/downgrades/19887dbd33327fb07d54251786e0cb2578539775/upgrade.properties

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

cpp/downgrades/BUILD.bazel

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

cpp/ql/lib/BUILD.bazel

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

cpp/ql/lib/CHANGELOG.md

@@ -1,27 +1,3 @@
-## 1.1.1
-
-No user-facing changes.
-
-## 1.1.0
-
-### New Features
-
-* Data models can now be added with data extensions. In this way source, sink and summary models can be added in extension `.model.yml` files, rather than by writing classes in QL code. New models should be added in the `lib/ext` folder.
-
-### Minor Analysis Improvements
-
-* A partial model for the `Boost.Asio` network library has been added. This includes sources, sinks and summaries for certain functions in `Boost.Asio`, such as `read_until` and `write`.
-
-## 1.0.0
-
-### Breaking Changes
-
-* CodeQL package management is now generally available, and all GitHub-produced CodeQL packages have had their version numbers increased to 1.0.0.
-
-## 0.13.1
-
-No user-facing changes.
-
 ## 0.13.0
 
 ### Breaking Changes

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

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

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

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

cpp/ql/lib/change-notes/released/1.1.0.md

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

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

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

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

@@ -1,2 +1,2 @@
 ---
-lastReleaseVersion: 1.1.1
+lastReleaseVersion: 0.13.0

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

@@ -1,26 +0,0 @@
-extensions:
-  # partial model of the Boost::Asio network library
-extensions:
-  - addsTo:
-      pack: codeql/cpp-all
-      extensible: sourceModel
-    data: # namespace, type, subtypes, name, signature, ext, output, kind, provenance
-      - ["boost::asio", "", False, "read", "", "", "Argument[*1]", "remote", "manual"]
-      - ["boost::asio", "", False, "read_at", "", "", "Argument[*2]", "remote", "manual"]
-      - ["boost::asio", "", False, "read_until", "", "", "Argument[*1]", "remote", "manual"]
-      - ["boost::asio", "", False, "async_read", "", "", "Argument[*1]", "remote", "manual"]
-      - ["boost::asio", "", False, "async_read_at", "", "", "Argument[*2]", "remote", "manual"]
-      - ["boost::asio", "", False, "async_read_until", "", "", "Argument[*1]", "remote", "manual"]
-  - addsTo:
-      pack: codeql/cpp-all
-      extensible: sinkModel
-    data: # namespace, type, subtypes, name, signature, ext, input, kind, provenance
-      - ["boost::asio", "", False, "write", "", "", "Argument[*1]", "remote-sink", "manual"]
-      - ["boost::asio", "", False, "write_at", "", "", "Argument[*2]", "remote-sink", "manual"]
-      - ["boost::asio", "", False, "async_write", "", "", "Argument[*1]", "remote-sink", "manual"]
-      - ["boost::asio", "", False, "async_write_at", "", "", "Argument[*2]", "remote-sink", "manual"]
-  - addsTo:
-      pack: codeql/cpp-all
-      extensible: summaryModel
-    data: # namespace, type, subtypes, name, signature, ext, input, output, kind, provenance
-      - ["boost::asio", "", False, "buffer", "", "", "Argument[*0]", "ReturnValue", "taint", "manual"]

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

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

cpp/ql/lib/qlpack.yml

@@ -1,5 +1,5 @@
 name: codeql/cpp-all
-version: 1.1.1
+version: 0.13.1-dev
 groups: cpp
 dbscheme: semmlecode.cpp.dbscheme
 extractor: cpp
@@ -14,6 +14,4 @@ dependencies:
   codeql/tutorial: ${workspace}
   codeql/util: ${workspace}
   codeql/xml: ${workspace}
-dataExtensions:
-  - ext/*.model.yml
 warnOnImplicitThis: true

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

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

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

@@ -375,33 +375,6 @@ cached
 class IRGuardCondition extends Instruction {
   Instruction branch;
 
-  /*
-   * An `IRGuardCondition` supports reasoning about four different kinds of
-   * relations:
-   * 1. A unary equality relation of the form `e == k`
-   * 2. A binary equality relation of the form `e1 == e2 + k`
-   * 3. A unary inequality relation of the form `e < k`
-   * 4. A binary inequality relation of the form `e1 < e2 + k`
-   *
-   * where `k` is a constant.
-   *
-   * Furthermore, the unary relations (i.e., case 1 and case 3) are also
-   * inferred from `switch` statement guards: equality relations are inferred
-   * from the unique `case` statement, if any, and inequality relations are
-   * inferred from the [case range](https://gcc.gnu.org/onlinedocs/gcc/Case-Ranges.html)
-   * gcc extension.
-   *
-   * The implementation of all four follows the same structure: Each relation
-   * has a cached user-facing predicate that. For example,
-   * `GuardCondition::comparesEq` calls `compares_eq`. This predicate has
-   * several cases that recursively decompose the relation to bring it to a
-   * canonical form (i.e., a relation of the form `e1 == e2 + k`). The base
-   * case for this relation (i.e., `simple_comparison_eq`) handles
-   * `CompareEQInstruction`s and `CompareNEInstruction`, and recursive
-   * predicates (e.g., `complex_eq`) rewrites larger expressions such as
-   * `e1 + k1 == e2 + k2` into canonical the form `e1 == e2 + (k2 - k1)`.
-   */
-
   cached
   IRGuardCondition() { branch = getBranchForCondition(this) }
 
@@ -592,7 +565,7 @@ class IRGuardCondition extends Instruction {
   /** Holds if (determined by this guard) `op == k` evaluates to `areEqual` if this expression evaluates to `value`. */
   cached
   predicate comparesEq(Operand op, int k, boolean areEqual, AbstractValue value) {
-    unary_compares_eq(this, op, k, areEqual, false, value)
+    compares_eq(this, op, k, areEqual, value)
   }
 
   /**
@@ -613,7 +586,7 @@ class IRGuardCondition extends Instruction {
   cached
   predicate ensuresEq(Operand op, int k, IRBlock block, boolean areEqual) {
     exists(AbstractValue value |
-      unary_compares_eq(this, op, k, areEqual, false, value) and this.valueControls(block, value)
+      compares_eq(this, op, k, areEqual, value) and this.valueControls(block, value)
     )
   }
 
@@ -638,7 +611,7 @@ class IRGuardCondition extends Instruction {
   cached
   predicate ensuresEqEdge(Operand op, int k, IRBlock pred, IRBlock succ, boolean areEqual) {
     exists(AbstractValue value |
-      unary_compares_eq(this, op, k, areEqual, false, value) and
+      compares_eq(this, op, k, areEqual, value) and
       this.valueControlsEdge(pred, succ, value)
     )
   }
@@ -764,68 +737,26 @@ private predicate compares_eq(
   )
 }
 
-/**
- * Holds if `op == k` is `areEqual` given that `test` is equal to `value`.
- *
- * Many internal predicates in this file have a `inNonZeroCase` column.
- * Ideally, the `k` column would be a type such as `Option<int>::Option`, to
- * represent whether we have a concrete value `k` such that `op == k`, or whether
- * we only know that `op != 0`.
- * However, cannot instantiate `Option` with an infinite type. Thus the boolean
- * `inNonZeroCase` is used to distinquish the `Some` (where we have a concrete
- * value `k`) and `None` cases (where we only know that `op != 0`).
- *
- * Thus, if `inNonZeroCase = true` then `op != 0` and the value of `k` is
- * meaningless.
- *
- * To see why `inNonZeroCase` is needed consider the following C program:
- * ```c
- * char* p = ...;
- * if(p) {
- *   use(p);
- * }
- * ```
- * in C++ there would be an int-to-bool conversion on `p`. However, since C
- * does not have booleans there is no conversion. We want to be able to
- * conclude that `p` is non-zero in the true branch, so we need to give `k`
- * some value. However, simply setting `k = 1` would make the rest of the
- * analysis think that `k == 1` holds inside the branch. So we distinquish
- * between the above case and
- * ```c
- * if(p == 1) {
- *   use(p)
- * }
- * ```
- * by setting `inNonZeroCase` to `true` in the former case, but not in the
- * latter.
- */
-private predicate unary_compares_eq(
-  Instruction test, Operand op, int k, boolean areEqual, boolean inNonZeroCase, AbstractValue value
+/** Holds if `op == k` is `areEqual` given that `test` is equal to `value`. */
+private predicate compares_eq(
+  Instruction test, Operand op, int k, boolean areEqual, AbstractValue value
 ) {
   /* The simple case where the test *is* the comparison so areEqual = testIsTrue xor eq. */
-  exists(AbstractValue v |
-    unary_simple_comparison_eq(test, k, inNonZeroCase, v) and op.getDef() = test
-  |
+  exists(AbstractValue v | simple_comparison_eq(test, op, k, v) |
     areEqual = true and value = v
     or
     areEqual = false and value = v.getDualValue()
   )
   or
-  unary_complex_eq(test, op, k, areEqual, inNonZeroCase, value)
+  complex_eq(test, op, k, areEqual, value)
   or
   /* (x is true => (op == k)) => (!x is false => (op == k)) */
-  exists(AbstractValue dual, boolean inNonZeroCase0 |
-    value = dual.getDualValue() and
-    unary_compares_eq(test.(LogicalNotInstruction).getUnary(), op, k, inNonZeroCase0, areEqual, dual)
-  |
-    k = 0 and inNonZeroCase = inNonZeroCase0
-    or
-    k != 0 and inNonZeroCase = true
+  exists(AbstractValue dual | value = dual.getDualValue() |
+    compares_eq(test.(LogicalNotInstruction).getUnary(), op, k, areEqual, dual)
   )
   or
   // ((test is `areEqual` => op == const + k2) and const == `k1`) =>
   // test is `areEqual` => op == k1 + k2
-  inNonZeroCase = false and
   exists(int k1, int k2, ConstantInstruction const |
     compares_eq(test, op, const.getAUse(), k2, areEqual, value) and
     int_value(const) = k1 and
@@ -850,63 +781,35 @@ private predicate simple_comparison_eq(
   value.(BooleanValue).getValue() = false
 }
 
-/**
- * Rearrange various simple comparisons into `op == k` form.
- */
-private predicate unary_simple_comparison_eq(
-  Instruction test, int k, boolean inNonZeroCase, AbstractValue value
-) {
+/** Rearrange various simple comparisons into `op == k` form. */
+private predicate simple_comparison_eq(Instruction test, Operand op, int k, AbstractValue value) {
   exists(SwitchInstruction switch, CaseEdge case |
     test = switch.getExpression() and
+    op.getDef() = test and
     case = value.(MatchValue).getCase() and
     exists(switch.getSuccessor(case)) and
-    case.getValue().toInt() = k and
-    inNonZeroCase = false
+    case.getValue().toInt() = k
   )
   or
-  // Any instruction with an integral type could potentially be part of a
-  // check for nullness when used in a guard. So we include all integral
-  // typed instructions here. However, since some of these instructions are
-  // already included as guards in other cases, we exclude those here.
-  // These are instructions that compute a binary equality or inequality
-  // relation. For example, the following:
-  // ```cpp
-  // if(a == b + 42) { ... }
-  // ```
-  // generates the following IR:
-  // ```
-  // r1(glval<int>) = VariableAddress[a]     :
-  // r2(int)        = Load[a]                : &:r1, m1
-  // r3(glval<int>) = VariableAddress[b]     :
-  // r4(int)        = Load[b]                : &:r3, m2
-  // r5(int)        = Constant[42]           :
-  // r6(int)        = Add                    : r4, r5
-  // r7(bool)       = CompareEQ              : r2, r6
-  // v1(void)       = ConditionalBranch      : r7
-  // ```
-  // and since `r7` is an integral typed instruction this predicate could
-  // include a case for when `r7` evaluates to true (in which case we would
-  // infer that `r6` was non-zero, and a case for when `r7` evaluates to false
-  // (in which case we would infer that `r6` was zero).
-  // However, since `a == b + 42` is already supported when reasoning about
-  // binary equalities we exclude those cases here.
-  not test.isGLValue() and
-  not simple_comparison_eq(test, _, _, _, _) and
-  not simple_comparison_lt(test, _, _, _) and
-  not test = any(SwitchInstruction switch).getExpression() and
-  (
-    test.getResultIRType() instanceof IRAddressType or
-    test.getResultIRType() instanceof IRIntegerType or
-    test.getResultIRType() instanceof IRBooleanType
-  ) and
-  (
-    k = 1 and
-    value.(BooleanValue).getValue() = true and
-    inNonZeroCase = true
-    or
+  // There's no implicit CompareInstruction in files compiled as C since C
+  // doesn't have implicit boolean conversions. So instead we check whether
+  // there's a branch on a value of pointer or integer type.
+  exists(ConditionalBranchInstruction branch, IRType type |
+    not test instanceof CompareInstruction and
+    type = test.getResultIRType() and
+    (type instanceof IRAddressType or type instanceof IRIntegerType) and
+    test = branch.getCondition() and
+    op.getDef() = test
+  |
+    // We'd like to also include a case such as:
+    // ```
+    // k = 1 and
+    // value.(BooleanValue).getValue() = true
+    // ```
+    // but all we know is that the value is non-zero in the true branch.
+    // So we can only conclude something in the false branch.
     k = 0 and
-    value.(BooleanValue).getValue() = false and
-    inNonZeroCase = false
+    value.(BooleanValue).getValue() = false
   )
 }
 
@@ -918,12 +821,12 @@ private predicate complex_eq(
   add_eq(cmp, left, right, k, areEqual, value)
 }
 
-private predicate unary_complex_eq(
-  Instruction test, Operand op, int k, boolean areEqual, boolean inNonZeroCase, AbstractValue value
+private predicate complex_eq(
+  Instruction test, Operand op, int k, boolean areEqual, AbstractValue value
 ) {
-  unary_sub_eq(test, op, k, areEqual, inNonZeroCase, value)
+  sub_eq(test, op, k, areEqual, value)
   or
-  unary_add_eq(test, op, k, areEqual, inNonZeroCase, value)
+  add_eq(test, op, k, areEqual, value)
 }
 
 /*
@@ -952,8 +855,7 @@ private predicate compares_lt(
 
 /** Holds if `op < k` evaluates to `isLt` given that `test` evaluates to `value`. */
 private predicate compares_lt(Instruction test, Operand op, int k, boolean isLt, AbstractValue value) {
-  unary_simple_comparison_lt(test, k, isLt, value) and
-  op.getDef() = test
+  simple_comparison_lt(test, op, k, isLt, value)
   or
   complex_lt(test, op, k, isLt, value)
   or
@@ -1000,11 +902,12 @@ private predicate simple_comparison_lt(CompareInstruction cmp, Operand left, Ope
 }
 
 /** Rearrange various simple comparisons into `op < k` form. */
-private predicate unary_simple_comparison_lt(
-  Instruction test, int k, boolean isLt, AbstractValue value
+private predicate simple_comparison_lt(
+  Instruction test, Operand op, int k, boolean isLt, AbstractValue value
 ) {
   exists(SwitchInstruction switch, CaseEdge case |
     test = switch.getExpression() and
+    op.getDef() = test and
     case = value.(MatchValue).getCase() and
     exists(switch.getSuccessor(case)) and
     case.getMaxValue() > case.getMinValue()
@@ -1187,20 +1090,16 @@ private predicate sub_eq(
 }
 
 // op - x == c => op == (c+x)
-private predicate unary_sub_eq(
-  Instruction test, Operand op, int k, boolean areEqual, boolean inNonZeroCase, AbstractValue value
-) {
-  inNonZeroCase = false and
+private predicate sub_eq(Instruction test, Operand op, int k, boolean areEqual, AbstractValue value) {
   exists(SubInstruction sub, int c, int x |
-    unary_compares_eq(test, sub.getAUse(), c, areEqual, inNonZeroCase, value) and
+    compares_eq(test, sub.getAUse(), c, areEqual, value) and
     op = sub.getLeftOperand() and
     x = int_value(sub.getRight()) and
     k = c + x
   )
   or
-  inNonZeroCase = false and
   exists(PointerSubInstruction sub, int c, int x |
-    unary_compares_eq(test, sub.getAUse(), c, areEqual, inNonZeroCase, value) and
+    compares_eq(test, sub.getAUse(), c, areEqual, value) and
     op = sub.getLeftOperand() and
     x = int_value(sub.getRight()) and
     k = c + x
@@ -1254,13 +1153,11 @@ private predicate add_eq(
 }
 
 // left + x == right + c => left == right + (c-x)
-private predicate unary_add_eq(
-  Instruction test, Operand left, int k, boolean areEqual, boolean inNonZeroCase,
-  AbstractValue value
+private predicate add_eq(
+  Instruction test, Operand left, int k, boolean areEqual, AbstractValue value
 ) {
-  inNonZeroCase = false and
   exists(AddInstruction lhs, int c, int x |
-    unary_compares_eq(test, lhs.getAUse(), c, areEqual, inNonZeroCase, value) and
+    compares_eq(test, lhs.getAUse(), c, areEqual, value) and
     (
       left = lhs.getLeftOperand() and x = int_value(lhs.getRight())
       or
@@ -1269,9 +1166,8 @@ private predicate unary_add_eq(
     k = c - x
   )
   or
-  inNonZeroCase = false and
   exists(PointerAddInstruction lhs, int c, int x |
-    unary_compares_eq(test, lhs.getAUse(), c, areEqual, inNonZeroCase, value) and
+    compares_eq(test, lhs.getAUse(), c, areEqual, value) and
     (
       left = lhs.getLeftOperand() and x = int_value(lhs.getRight())
       or

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

@@ -78,7 +78,6 @@ private import internal.FlowSummaryImpl
 private import internal.FlowSummaryImpl::Public
 private import internal.FlowSummaryImpl::Private
 private import internal.FlowSummaryImpl::Private::External
-private import internal.ExternalFlowExtensions as Extensions
 private import codeql.mad.ModelValidation as SharedModelVal
 private import codeql.util.Unit
 
@@ -139,9 +138,6 @@ predicate sourceModel(
     row.splitAt(";", 7) = kind
   ) and
   provenance = "manual"
-  or
-  Extensions::sourceModel(namespace, type, subtypes, name, signature, ext, output, kind, provenance,
-    _)
 }
 
 /** Holds if a sink model exists for the given parameters. */
@@ -162,8 +158,6 @@ predicate sinkModel(
     row.splitAt(";", 7) = kind
   ) and
   provenance = "manual"
-  or
-  Extensions::sinkModel(namespace, type, subtypes, name, signature, ext, input, kind, provenance, _)
 }
 
 /** Holds if a summary model exists for the given parameters. */
@@ -185,9 +179,6 @@ predicate summaryModel(
     row.splitAt(";", 8) = kind
   ) and
   provenance = "manual"
-  or
-  Extensions::summaryModel(namespace, type, subtypes, name, signature, ext, input, output, kind,
-    provenance, _)
 }
 
 private predicate relevantNamespace(string namespace) {
@@ -212,10 +203,8 @@ private predicate canonicalNamespaceLink(string namespace, string subns) {
 }
 
 /**
- * Holds if MaD framework coverage of `namespace` is `n` api endpoints of the
- * kind `(kind, part)`, and `namespaces` is the number of subnamespaces of
- * `namespace` which have MaD framework coverage (including `namespace`
- * itself).
+ * Holds if CSV framework coverage of `namespace` is `n` api endpoints of the
+ * kind `(kind, part)`.
  */
 predicate modelCoverage(string namespace, int namespaces, string kind, string part, int n) {
   namespaces = strictcount(string subns | canonicalNamespaceLink(namespace, subns)) and
@@ -332,10 +321,10 @@ module CsvValidation {
       or
       summaryModel(namespace, type, _, name, signature, ext, _, _, _, _) and pred = "summary"
     |
-      not namespace.regexpMatch("[a-zA-Z0-9_\\.:]*") and
+      not namespace.regexpMatch("[a-zA-Z0-9_\\.]+") and
       result = "Dubious namespace \"" + namespace + "\" in " + pred + " model."
       or
-      not type.regexpMatch("[a-zA-Z0-9_<>,\\+]*") and
+      not type.regexpMatch("[a-zA-Z0-9_<>,\\+]+") and
       result = "Dubious type \"" + type + "\" in " + pred + " model."
       or
       not name.regexpMatch("[a-zA-Z0-9_<>,]*") and

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

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

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

@@ -242,17 +242,7 @@ class CastNode extends Node {
   CastNode() { none() } // stub implementation
 }
 
-class DataFlowCallable extends Function {
-  /** Gets a best-effort total ordering. */
-  int totalorder() {
-    this =
-      rank[result](DataFlowCallable c, string file, int startline, int startcolumn |
-        c.getLocation().hasLocationInfo(file, startline, startcolumn, _, _)
-      |
-        c order by file, startline, startcolumn
-      )
-  }
-}
+class DataFlowCallable = Function;
 
 class DataFlowExpr = Expr;
 
@@ -271,28 +261,10 @@ class DataFlowCall extends Expr instanceof Call {
   ExprNode getNode() { result.getExpr() = this }
 
   /** Gets the enclosing callable of this call. */
-  DataFlowCallable getEnclosingCallable() { result = this.getEnclosingFunction() }
-
-  /** Gets a best-effort total ordering. */
-  int totalorder() {
-    this =
-      rank[result](DataFlowCall c, int startline, int startcolumn |
-        c.getLocation().hasLocationInfo(_, startline, startcolumn, _, _)
-      |
-        c order by startline, startcolumn
-      )
-  }
+  Function getEnclosingCallable() { result = this.getEnclosingFunction() }
 }
 
-class NodeRegion instanceof Unit {
-  string toString() { result = "NodeRegion" }
-
-  predicate contains(Node n) { none() }
-
-  int totalOrder() { result = 1 }
-}
-
-predicate isUnreachableInCall(NodeRegion nr, DataFlowCall call) { none() } // stub implementation
+predicate isUnreachableInCall(Node n, DataFlowCall call) { none() } // stub implementation
 
 /**
  * Holds if access paths with `c` at their head always should be tracked at high

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

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

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

@@ -1062,16 +1062,6 @@ class DataFlowCallable extends TDataFlowCallable {
     result = this.asSummarizedCallable() or // SummarizedCallable = Function (in CPP)
     result = this.asSourceCallable()
   }
-
-  /** Gets a best-effort total ordering. */
-  int totalorder() {
-    this =
-      rank[result](DataFlowCallable c, string file, int startline, int startcolumn |
-        c.getLocation().hasLocationInfo(file, startline, startcolumn, _, _)
-      |
-        c order by file, startline, startcolumn
-      )
-  }
 }
 
 /**
@@ -1169,16 +1159,6 @@ class DataFlowCall extends TDataFlowCall {
    * Gets the location of this call.
    */
   Location getLocation() { none() }
-
-  /** Gets a best-effort total ordering. */
-  int totalorder() {
-    this =
-      rank[result](DataFlowCall c, int startline, int startcolumn |
-        c.getLocation().hasLocationInfo(_, startline, startcolumn, _, _)
-      |
-        c order by startline, startcolumn
-      )
-  }
 }
 
 /**
@@ -1267,53 +1247,43 @@ module IsUnreachableInCall {
     any(G::IRGuardCondition guard).ensuresLt(left, right, k, block, areEqual)
   }
 
-  class NodeRegion instanceof IRBlock {
-    string toString() { result = "NodeRegion" }
-
-    predicate contains(Node n) { this = n.getBasicBlock() }
-
-    int totalOrder() {
-      this =
-        rank[result](IRBlock b, int startline, int startcolumn |
-          b.getLocation().hasLocationInfo(_, startline, startcolumn, _, _)
-        |
-          b order by startline, startcolumn
-        )
-    }
-  }
-
-  predicate isUnreachableInCall(NodeRegion block, DataFlowCall call) {
+  predicate isUnreachableInCall(Node n, DataFlowCall call) {
     exists(
       InstructionDirectParameterNode paramNode, ConstantIntegralTypeArgumentNode arg,
-      IntegerConstantInstruction constant, int k, Operand left, Operand right, int argval
+      IntegerConstantInstruction constant, int k, Operand left, Operand right, IRBlock block
     |
       // arg flows into `paramNode`
-      DataFlowImplCommon::viableParamArg(call, pragma[only_bind_into](paramNode),
-        pragma[only_bind_into](arg)) and
+      DataFlowImplCommon::viableParamArg(call, paramNode, arg) and
       left = constant.getAUse() and
       right = valueNumber(paramNode.getInstruction()).getAUse() and
-      argval = arg.getValue()
+      block = n.getBasicBlock()
     |
       // and there's a guard condition which ensures that the result of `left == right + k` is `areEqual`
-      exists(boolean areEqual | ensuresEq(left, right, k, block, areEqual) |
+      exists(boolean areEqual |
+        ensuresEq(pragma[only_bind_into](left), pragma[only_bind_into](right),
+          pragma[only_bind_into](k), pragma[only_bind_into](block), areEqual)
+      |
         // this block ensures that left = right + k, but it holds that `left != right + k`
         areEqual = true and
-        constant.getValue().toInt() != argval + k
+        constant.getValue().toInt() != arg.getValue() + k
         or
         // this block ensures that or `left != right + k`, but it holds that `left = right + k`
         areEqual = false and
-        constant.getValue().toInt() = argval + k
+        constant.getValue().toInt() = arg.getValue() + k
       )
       or
       // or there's a guard condition which ensures that the result of `left < right + k` is `isLessThan`
-      exists(boolean isLessThan | ensuresLt(left, right, k, block, isLessThan) |
+      exists(boolean isLessThan |
+        ensuresLt(pragma[only_bind_into](left), pragma[only_bind_into](right),
+          pragma[only_bind_into](k), pragma[only_bind_into](block), isLessThan)
+      |
         isLessThan = true and
         // this block ensures that `left < right + k`, but it holds that `left >= right + k`
-        constant.getValue().toInt() >= argval + k
+        constant.getValue().toInt() >= arg.getValue() + k
         or
         // this block ensures that `left >= right + k`, but it holds that `left < right + k`
         isLessThan = false and
-        constant.getValue().toInt() < argval + k
+        constant.getValue().toInt() < arg.getValue() + k
       )
     )
   }
@@ -1336,8 +1306,6 @@ predicate nodeIsHidden(Node n) {
   n instanceof FinalGlobalValue
   or
   n instanceof InitialGlobalValue
-  or
-  n instanceof SsaPhiInputNode
 }
 
 predicate neverSkipInPathGraph(Node n) {
@@ -1636,8 +1604,6 @@ private Instruction getAnInstruction(Node n) {
   or
   result = n.(SsaPhiNode).getPhiNode().getBasicBlock().getFirstInstruction()
   or
-  result = n.(SsaPhiInputNode).getBasicBlock().getFirstInstruction()
-  or
   n.(IndirectInstruction).hasInstructionAndIndirectionIndex(result, _)
   or
   not n instanceof IndirectInstruction and
@@ -1767,7 +1733,7 @@ module IteratorFlow {
         crementCall = def.getValue().asInstruction().(StoreInstruction).getSourceValue() and
         sv = def.getSourceVariable() and
         bb.getInstruction(i) = crementCall and
-        Ssa::ssaDefReachesReadExt(sv, result.asDef(), bb, i)
+        Ssa::ssaDefReachesRead(sv, result.asDef(), bb, i)
       )
     }
 
@@ -1801,7 +1767,7 @@ module IteratorFlow {
         isIteratorWrite(writeToDeref, address) and
         operandForFullyConvertedCall(address, starCall) and
         bbStar.getInstruction(iStar) = starCall and
-        Ssa::ssaDefReachesReadExt(_, def.asDef(), bbStar, iStar) and
+        Ssa::ssaDefReachesRead(_, def.asDef(), bbStar, iStar) and
         ultimate = getAnUltimateDefinition*(def) and
         beginStore = ultimate.getValue().asInstruction() and
         operandForFullyConvertedCall(beginStore.getSourceValueOperand(), beginCall)

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

@@ -17,7 +17,6 @@ private import SsaInternals as Ssa
 private import DataFlowImplCommon as DataFlowImplCommon
 private import codeql.util.Unit
 private import Node0ToString
-import ExprNodes
 
 /**
  * The IR dataflow graph consists of the following nodes:
@@ -46,7 +45,6 @@ private newtype TIRDataFlowNode =
     or
     Ssa::isModifiableByCall(operand, indirectionIndex)
   } or
-  TSsaPhiInputNode(Ssa::PhiNode phi, IRBlock input) { phi.hasInputFromBlock(_, _, _, _, input) } or
   TSsaPhiNode(Ssa::PhiNode phi) or
   TSsaIteratorNode(IteratorFlow::IteratorFlowNode n) or
   TRawIndirectOperand0(Node0Impl node, int indirectionIndex) {
@@ -116,13 +114,6 @@ predicate conversionFlow(
       instrTo.(CheckedConvertOrNullInstruction).getUnaryOperand() = opFrom
       or
       instrTo.(InheritanceConversionInstruction).getUnaryOperand() = opFrom
-      or
-      exists(BuiltInInstruction builtIn |
-        builtIn = instrTo and
-        // __builtin_bit_cast
-        builtIn.getBuiltInOperation() instanceof BuiltInBitCast and
-        opFrom = builtIn.getAnOperand()
-      )
     )
     or
     additional = true and
@@ -167,12 +158,6 @@ class Node extends TIRDataFlowNode {
   /** Gets the operands corresponding to this node, if any. */
   Operand asOperand() { result = this.(OperandNode).getOperand() }
 
-  /**
-   * Gets the operand that is indirectly tracked by this node behind `index`
-   * number of indirections.
-   */
-  Operand asIndirectOperand(int index) { hasOperandAndIndex(this, result, index) }
-
   /**
    * Holds if this node is at index `i` in basic block `block`.
    *
@@ -185,9 +170,6 @@ class Node extends TIRDataFlowNode {
     or
     this.(SsaPhiNode).getPhiNode().getBasicBlock() = block and i = -1
     or
-    this.(SsaPhiInputNode).getBlock() = block and
-    i = block.getInstructionCount()
-    or
     this.(RawIndirectOperand).getOperand().getUse() = block.getInstruction(i)
     or
     this.(RawIndirectInstruction).getInstruction() = block.getInstruction(i)
@@ -640,7 +622,7 @@ class SsaPhiNode extends Node, TSsaPhiNode {
 
   final override Location getLocationImpl() { result = phi.getBasicBlock().getLocation() }
 
-  override string toStringImpl() { result = phi.toString() }
+  override string toStringImpl() { result = "Phi" }
 
   /**
    * Gets a node that is used as input to this phi node.
@@ -649,7 +631,7 @@ class SsaPhiNode extends Node, TSsaPhiNode {
    */
   cached
   final Node getAnInput(boolean fromBackEdge) {
-    result.(SsaPhiInputNode).getPhiNode() = phi and
+    localFlowStep(result, this) and
     exists(IRBlock bPhi, IRBlock bResult |
       bPhi = phi.getBasicBlock() and bResult = result.getBasicBlock()
     |
@@ -672,58 +654,6 @@ class SsaPhiNode extends Node, TSsaPhiNode {
   predicate isPhiRead() { phi.isPhiRead() }
 }
 
-/**
- * INTERNAL: Do not use.
- *
- * A node that is used as an input to a phi node.
- *
- * This class exists to allow more powerful barrier guards. Consider this
- * example:
- *
- * ```cpp
- * int x = source();
- * if(!safe(x)) {
- *   x = clear();
- * }
- * // phi node for x here
- * sink(x);
- * ```
- *
- * At the phi node for `x` it is neither the case that `x` is dominated by
- * `safe(x)`, or is the case that the phi is dominated by a clearing of `x`.
- *
- * By inserting a "phi input" node as the last entry in the basic block that
- * defines the inputs to the phi we can conclude that each of those inputs are
- * safe to pass to `sink`.
- */
-class SsaPhiInputNode extends Node, TSsaPhiInputNode {
-  Ssa::PhiNode phi;
-  IRBlock block;
-
-  SsaPhiInputNode() { this = TSsaPhiInputNode(phi, block) }
-
-  /** Gets the phi node associated with this node. */
-  Ssa::PhiNode getPhiNode() { result = phi }
-
-  /** Gets the basic block in which this input originates. */
-  IRBlock getBlock() { result = block }
-
-  override Declaration getEnclosingCallable() { result = this.getFunction() }
-
-  override Declaration getFunction() { result = phi.getBasicBlock().getEnclosingFunction() }
-
-  override DataFlowType getType() { result = this.getSourceVariable().getType() }
-
-  override predicate isGLValue() { phi.getSourceVariable().isGLValue() }
-
-  final override Location getLocationImpl() { result = block.getLastInstruction().getLocation() }
-
-  override string toStringImpl() { result = "Phi input" }
-
-  /** Gets the source variable underlying this phi node. */
-  Ssa::SourceVariable getSourceVariable() { result = phi.getSourceVariable() }
-}
-
 /**
  * INTERNAL: do not use.
  *
@@ -1297,6 +1227,466 @@ class UninitializedNode extends Node {
   LocalVariable getLocalVariable() { result = v }
 }
 
+private module GetConvertedResultExpression {
+  private import semmle.code.cpp.ir.implementation.raw.internal.TranslatedExpr
+  private import semmle.code.cpp.ir.implementation.raw.internal.InstructionTag
+
+  private Operand getAnInitializeDynamicAllocationInstructionAddress() {
+    result = any(InitializeDynamicAllocationInstruction init).getAllocationAddressOperand()
+  }
+
+  /**
+   * Gets the expression that should be returned as the result expression from `instr`.
+   *
+   * Note that this predicate may return multiple results in cases where a conversion belongs to a
+   * different AST element than its operand.
+   */
+  Expr getConvertedResultExpression(Instruction instr, int n) {
+    // Only fully converted instructions have a result for `asConvertedExpr`
+    not conversionFlow(unique(Operand op |
+        // The address operand of a `InitializeDynamicAllocationInstruction` is
+        // special: we need to handle it during dataflow (since it's
+        // effectively a store to an indirection), but it doesn't appear in
+        // source syntax, so dataflow node <-> expression conversion shouldn't
+        // care about it.
+        op = getAUse(instr) and not op = getAnInitializeDynamicAllocationInstructionAddress()
+      |
+        op
+      ), _, false, false) and
+    result = getConvertedResultExpressionImpl(instr) and
+    n = 0
+    or
+    // If the conversion also has a result then we return multiple results
+    exists(Operand operand | conversionFlow(operand, instr, false, false) |
+      n = 1 and
+      result = getConvertedResultExpressionImpl(operand.getDef())
+      or
+      result = getConvertedResultExpression(operand.getDef(), n - 1)
+    )
+  }
+
+  private Expr getConvertedResultExpressionImpl0(Instruction instr) {
+    // IR construction inserts an additional cast to a `size_t` on the extent
+    // of a `new[]` expression. The resulting `ConvertInstruction` doesn't have
+    // a result for `getConvertedResultExpression`. We remap this here so that
+    // this `ConvertInstruction` maps to the result of the expression that
+    // represents the extent.
+    exists(TranslatedNonConstantAllocationSize tas |
+      result = tas.getExtent().getExpr() and
+      instr = tas.getInstruction(AllocationExtentConvertTag())
+    )
+    or
+    // There's no instruction that returns `ParenthesisExpr`, but some queries
+    // expect this
+    exists(TranslatedTransparentConversion ttc |
+      result = ttc.getExpr().(ParenthesisExpr) and
+      instr = ttc.getResult()
+    )
+    or
+    // Certain expressions generate `CopyValueInstruction`s only when they
+    // are needed. Examples of this include crement operations and compound
+    // assignment operations. For example:
+    // ```cpp
+    // int x = ...
+    // int y = x++;
+    // ```
+    // this generate IR like:
+    // ```
+    // r1(glval<int>) = VariableAddress[x] :
+    // r2(int)        = Constant[0]        :
+    // m3(int)        = Store[x]           : &:r1, r2
+    // r4(glval<int>) = VariableAddress[y] :
+    // r5(glval<int>) = VariableAddress[x] :
+    // r6(int)        = Load[x]            : &:r5, m3
+    // r7(int)        = Constant[1]        :
+    // r8(int)        = Add                : r6, r7
+    // m9(int)        = Store[x]           : &:r5, r8
+    // r11(int)       = CopyValue         : r6
+    // m12(int)       = Store[y]          : &:r4, r11
+    // ```
+    // When the `CopyValueInstruction` is not generated there is no instruction
+    // whose `getConvertedResultExpression` maps back to the expression. When
+    // such an instruction doesn't exist it means that the old value is not
+    // needed, and in that case the only value that will propagate forward in
+    // the program is the value that's been updated. So in those cases we just
+    // use the result of `node.asDefinition()` as the result of `node.asExpr()`.
+    exists(TranslatedCoreExpr tco |
+      tco.getInstruction(_) = instr and
+      tco.producesExprResult() and
+      result = asDefinitionImpl0(instr)
+    )
+  }
+
+  private Expr getConvertedResultExpressionImpl(Instruction instr) {
+    result = getConvertedResultExpressionImpl0(instr)
+    or
+    not exists(getConvertedResultExpressionImpl0(instr)) and
+    result = instr.getConvertedResultExpression()
+  }
+
+  /**
+   * Gets the result for `node.asDefinition()` (when `node` is the instruction
+   * node that wraps `store`) in the cases where `store.getAst()` should not be
+   * used to define the result of `node.asDefinition()`.
+   */
+  private Expr asDefinitionImpl0(StoreInstruction store) {
+    // For an expression such as `i += 2` we pretend that the generated
+    // `StoreInstruction` contains the result of the expression even though
+    // this isn't totally aligned with the C/C++ standard.
+    exists(TranslatedAssignOperation tao |
+      store = tao.getInstruction(AssignmentStoreTag()) and
+      result = tao.getExpr()
+    )
+    or
+    // Similarly for `i++` and `++i` we pretend that the generated
+    // `StoreInstruction` is contains the result of the expression even though
+    // this isn't totally aligned with the C/C++ standard.
+    exists(TranslatedCrementOperation tco |
+      store = tco.getInstruction(CrementStoreTag()) and
+      result = tco.getExpr()
+    )
+  }
+
+  /**
+   * Holds if the expression returned by `store.getAst()` should not be
+   * returned as the result of `node.asDefinition()` when `node` is the
+   * instruction node that wraps `store`.
+   */
+  private predicate excludeAsDefinitionResult(StoreInstruction store) {
+    // Exclude the store to the temporary generated by a ternary expression.
+    exists(TranslatedConditionalExpr tce |
+      store = tce.getInstruction(ConditionValueFalseStoreTag())
+      or
+      store = tce.getInstruction(ConditionValueTrueStoreTag())
+    )
+  }
+
+  /**
+   * Gets the expression that represents the result of `StoreInstruction` for
+   * dataflow purposes.
+   *
+   * For example, consider the following example
+   * ```cpp
+   * int x = 42;     // 1
+   * x = 34;         // 2
+   * ++x;            // 3
+   * x++;            // 4
+   * x += 1;         // 5
+   * int y = x += 2; // 6
+   * ```
+   * For (1) the result is `42`.
+   * For (2) the result is `x = 34`.
+   * For (3) the result is `++x`.
+   * For (4) the result is `x++`.
+   * For (5) the result is `x += 1`.
+   * For (6) there are two results:
+   *   - For the `StoreInstruction` generated by `x += 2` the result
+   *     is `x += 2`
+   *   - For the `StoreInstruction` generated by `int y = ...` the result
+   *     is also `x += 2`
+   */
+  Expr asDefinitionImpl(StoreInstruction store) {
+    not exists(asDefinitionImpl0(store)) and
+    not excludeAsDefinitionResult(store) and
+    result = store.getAst().(Expr).getUnconverted()
+    or
+    result = asDefinitionImpl0(store)
+  }
+}
+
+private import GetConvertedResultExpression
+
+/** Holds if `node` is an `OperandNode` that should map `node.asExpr()` to `e`. */
+predicate exprNodeShouldBeOperand(OperandNode node, Expr e, int n) {
+  not exprNodeShouldBeIndirectOperand(_, e, n) and
+  exists(Instruction def |
+    unique( | | getAUse(def)) = node.getOperand() and
+    e = getConvertedResultExpression(def, n)
+  )
+}
+
+/** Holds if `node` should be an `IndirectOperand` that maps `node.asIndirectExpr()` to `e`. */
+private predicate indirectExprNodeShouldBeIndirectOperand(
+  IndirectOperand node, Expr e, int n, int indirectionIndex
+) {
+  exists(Instruction def |
+    node.hasOperandAndIndirectionIndex(unique( | | getAUse(def)), indirectionIndex) and
+    e = getConvertedResultExpression(def, n)
+  )
+}
+
+/** Holds if `node` should be an `IndirectOperand` that maps `node.asExpr()` to `e`. */
+private predicate exprNodeShouldBeIndirectOperand(IndirectOperand node, Expr e, int n) {
+  exists(ArgumentOperand operand |
+    // When an argument (qualifier or positional) is a prvalue and the
+    // parameter (qualifier or positional) is a (const) reference, IR
+    // construction introduces a temporary `IRVariable`. The `VariableAddress`
+    // instruction has the argument as its `getConvertedResultExpression`
+    // result. However, the instruction actually represents the _address_ of
+    // the argument. So to fix this mismatch, we have the indirection of the
+    // `VariableAddressInstruction` map to the expression.
+    node.hasOperandAndIndirectionIndex(operand, 1) and
+    e = getConvertedResultExpression(operand.getDef(), n) and
+    operand.getDef().(VariableAddressInstruction).getIRVariable() instanceof IRTempVariable
+  )
+}
+
+private predicate exprNodeShouldBeIndirectOutNode(IndirectArgumentOutNode node, Expr e, int n) {
+  exists(CallInstruction call |
+    call.getStaticCallTarget() instanceof Constructor and
+    e = getConvertedResultExpression(call, n) and
+    call.getThisArgumentOperand() = node.getAddressOperand()
+  )
+}
+
+/** Holds if `node` should be an instruction node that maps `node.asExpr()` to `e`. */
+predicate exprNodeShouldBeInstruction(Node node, Expr e, int n) {
+  not exprNodeShouldBeOperand(_, e, n) and
+  not exprNodeShouldBeIndirectOutNode(_, e, n) and
+  not exprNodeShouldBeIndirectOperand(_, e, n) and
+  e = getConvertedResultExpression(node.asInstruction(), n)
+}
+
+/** Holds if `node` should be an `IndirectInstruction` that maps `node.asIndirectExpr()` to `e`. */
+predicate indirectExprNodeShouldBeIndirectInstruction(
+  IndirectInstruction node, Expr e, int n, int indirectionIndex
+) {
+  not indirectExprNodeShouldBeIndirectOperand(_, e, n, indirectionIndex) and
+  exists(Instruction instr |
+    node.hasInstructionAndIndirectionIndex(instr, indirectionIndex) and
+    e = getConvertedResultExpression(instr, n)
+  )
+}
+
+abstract private class ExprNodeBase extends Node {
+  /**
+   * Gets the expression corresponding to this node, if any. The returned
+   * expression may be a `Conversion`.
+   */
+  abstract Expr getConvertedExpr(int n);
+
+  /** Gets the non-conversion expression corresponding to this node, if any. */
+  final Expr getExpr(int n) { result = this.getConvertedExpr(n).getUnconverted() }
+}
+
+/**
+ * Holds if there exists a dataflow node whose `asExpr(n)` should evaluate
+ * to `e`.
+ */
+private predicate exprNodeShouldBe(Expr e, int n) {
+  exprNodeShouldBeInstruction(_, e, n) or
+  exprNodeShouldBeOperand(_, e, n) or
+  exprNodeShouldBeIndirectOutNode(_, e, n) or
+  exprNodeShouldBeIndirectOperand(_, e, n)
+}
+
+private class InstructionExprNode extends ExprNodeBase, InstructionNode {
+  InstructionExprNode() {
+    exists(Expr e, int n |
+      exprNodeShouldBeInstruction(this, e, n) and
+      not exists(Expr conv |
+        exprNodeShouldBe(conv, n + 1) and
+        conv.getUnconverted() = e.getUnconverted()
+      )
+    )
+  }
+
+  final override Expr getConvertedExpr(int n) { exprNodeShouldBeInstruction(this, result, n) }
+}
+
+private class OperandExprNode extends ExprNodeBase, OperandNode {
+  OperandExprNode() {
+    exists(Expr e, int n |
+      exprNodeShouldBeOperand(this, e, n) and
+      not exists(Expr conv |
+        exprNodeShouldBe(conv, n + 1) and
+        conv.getUnconverted() = e.getUnconverted()
+      )
+    )
+  }
+
+  final override Expr getConvertedExpr(int n) { exprNodeShouldBeOperand(this, result, n) }
+}
+
+abstract private class IndirectExprNodeBase extends Node {
+  /**
+   * Gets the expression corresponding to this node, if any. The returned
+   * expression may be a `Conversion`.
+   */
+  abstract Expr getConvertedExpr(int n, int indirectionIndex);
+
+  /** Gets the non-conversion expression corresponding to this node, if any. */
+  final Expr getExpr(int n, int indirectionIndex) {
+    result = this.getConvertedExpr(n, indirectionIndex).getUnconverted()
+  }
+}
+
+/** A signature for converting an indirect node to an expression. */
+private signature module IndirectNodeToIndirectExprSig {
+  /** The indirect node class to be converted to an expression */
+  class IndirectNode;
+
+  /**
+   * Holds if the indirect expression at indirection index `indirectionIndex`
+   * of `node` is `e`. The integer `n` specifies how many conversions has been
+   * applied to `node`.
+   */
+  predicate indirectNodeHasIndirectExpr(IndirectNode node, Expr e, int n, int indirectionIndex);
+}
+
+/**
+ * A module that implements the logic for deciding whether an indirect node
+ * should be an `IndirectExprNode`.
+ */
+private module IndirectNodeToIndirectExpr<IndirectNodeToIndirectExprSig Sig> {
+  import Sig
+
+  /**
+   * This predicate shifts the indirection index by one when `conv` is a
+   * `ReferenceDereferenceExpr`.
+   *
+   * This is necessary because `ReferenceDereferenceExpr` is a conversion
+   * in the AST, but appears as a `LoadInstruction` in the IR.
+   */
+  bindingset[e, indirectionIndex]
+  private predicate adjustForReference(
+    Expr e, int indirectionIndex, Expr conv, int adjustedIndirectionIndex
+  ) {
+    conv.(ReferenceDereferenceExpr).getExpr() = e and
+    adjustedIndirectionIndex = indirectionIndex - 1
+    or
+    not conv instanceof ReferenceDereferenceExpr and
+    conv = e and
+    adjustedIndirectionIndex = indirectionIndex
+  }
+
+  /** Holds if `node` should be an `IndirectExprNode`. */
+  predicate charpred(IndirectNode node) {
+    exists(Expr e, int n, int indirectionIndex |
+      indirectNodeHasIndirectExpr(node, e, n, indirectionIndex) and
+      not exists(Expr conv, int adjustedIndirectionIndex |
+        adjustForReference(e, indirectionIndex, conv, adjustedIndirectionIndex) and
+        indirectExprNodeShouldBe(conv, n + 1, adjustedIndirectionIndex)
+      )
+    )
+  }
+}
+
+private predicate indirectExprNodeShouldBe(Expr e, int n, int indirectionIndex) {
+  indirectExprNodeShouldBeIndirectOperand(_, e, n, indirectionIndex) or
+  indirectExprNodeShouldBeIndirectInstruction(_, e, n, indirectionIndex)
+}
+
+private module IndirectOperandIndirectExprNodeImpl implements IndirectNodeToIndirectExprSig {
+  class IndirectNode = IndirectOperand;
+
+  predicate indirectNodeHasIndirectExpr = indirectExprNodeShouldBeIndirectOperand/4;
+}
+
+module IndirectOperandToIndirectExpr =
+  IndirectNodeToIndirectExpr<IndirectOperandIndirectExprNodeImpl>;
+
+private class IndirectOperandIndirectExprNode extends IndirectExprNodeBase instanceof IndirectOperand
+{
+  IndirectOperandIndirectExprNode() { IndirectOperandToIndirectExpr::charpred(this) }
+
+  final override Expr getConvertedExpr(int n, int index) {
+    IndirectOperandToIndirectExpr::indirectNodeHasIndirectExpr(this, result, n, index)
+  }
+}
+
+private module IndirectInstructionIndirectExprNodeImpl implements IndirectNodeToIndirectExprSig {
+  class IndirectNode = IndirectInstruction;
+
+  predicate indirectNodeHasIndirectExpr = indirectExprNodeShouldBeIndirectInstruction/4;
+}
+
+module IndirectInstructionToIndirectExpr =
+  IndirectNodeToIndirectExpr<IndirectInstructionIndirectExprNodeImpl>;
+
+private class IndirectInstructionIndirectExprNode extends IndirectExprNodeBase instanceof IndirectInstruction
+{
+  IndirectInstructionIndirectExprNode() { IndirectInstructionToIndirectExpr::charpred(this) }
+
+  final override Expr getConvertedExpr(int n, int index) {
+    IndirectInstructionToIndirectExpr::indirectNodeHasIndirectExpr(this, result, n, index)
+  }
+}
+
+private class IndirectArgumentOutExprNode extends ExprNodeBase, IndirectArgumentOutNode {
+  IndirectArgumentOutExprNode() { exprNodeShouldBeIndirectOutNode(this, _, _) }
+
+  final override Expr getConvertedExpr(int n) { exprNodeShouldBeIndirectOutNode(this, result, n) }
+}
+
+private class IndirectOperandExprNode extends ExprNodeBase instanceof IndirectOperand {
+  IndirectOperandExprNode() { exprNodeShouldBeIndirectOperand(this, _, _) }
+
+  final override Expr getConvertedExpr(int n) { exprNodeShouldBeIndirectOperand(this, result, n) }
+}
+
+/**
+ * An expression, viewed as a node in a data flow graph.
+ */
+class ExprNode extends Node instanceof ExprNodeBase {
+  /**
+   * INTERNAL: Do not use.
+   */
+  Expr getExpr(int n) { result = super.getExpr(n) }
+
+  /**
+   * Gets the non-conversion expression corresponding to this node, if any. If
+   * this node strictly (in the sense of `getConvertedExpr`) corresponds to a
+   * `Conversion`, then the result is that `Conversion`'s non-`Conversion` base
+   * expression.
+   */
+  final Expr getExpr() { result = this.getExpr(_) }
+
+  /**
+   * INTERNAL: Do not use.
+   */
+  Expr getConvertedExpr(int n) { result = super.getConvertedExpr(n) }
+
+  /**
+   * Gets the expression corresponding to this node, if any. The returned
+   * expression may be a `Conversion`.
+   */
+  final Expr getConvertedExpr() { result = this.getConvertedExpr(_) }
+}
+
+/**
+ * An indirect expression, viewed as a node in a data flow graph.
+ */
+class IndirectExprNode extends Node instanceof IndirectExprNodeBase {
+  /**
+   * Gets the non-conversion expression corresponding to this node, if any. If
+   * this node strictly (in the sense of `getConvertedExpr`) corresponds to a
+   * `Conversion`, then the result is that `Conversion`'s non-`Conversion` base
+   * expression.
+   */
+  final Expr getExpr(int indirectionIndex) { result = this.getExpr(_, indirectionIndex) }
+
+  /**
+   * INTERNAL: Do not use.
+   */
+  Expr getExpr(int n, int indirectionIndex) { result = super.getExpr(n, indirectionIndex) }
+
+  /**
+   * INTERNAL: Do not use.
+   */
+  Expr getConvertedExpr(int n, int indirectionIndex) {
+    result = super.getConvertedExpr(n, indirectionIndex)
+  }
+
+  /**
+   * Gets the expression corresponding to this node, if any. The returned
+   * expression may be a `Conversion`.
+   */
+  Expr getConvertedExpr(int indirectionIndex) {
+    result = this.getConvertedExpr(_, indirectionIndex)
+  }
+}
+
 abstract private class AbstractParameterNode extends Node {
   /**
    * Holds if this node is the parameter of `f` at the specified position. The
@@ -1786,9 +2176,6 @@ private module Cached {
       // Def-use/Use-use flow
       Ssa::ssaFlow(nodeFrom, nodeTo)
       or
-      // Phi input -> Phi
-      nodeFrom.(SsaPhiInputNode).getPhiNode() = nodeTo.(SsaPhiNode).getPhiNode()
-      or
       IteratorFlow::localFlowStep(nodeFrom, nodeTo)
       or
       // Operand -> Instruction flow
@@ -2227,22 +2614,6 @@ class ContentSet instanceof Content {
   }
 }
 
-pragma[nomagic]
-private predicate guardControlsPhiInput(
-  IRGuardCondition g, boolean branch, Ssa::Definition def, IRBlock input, Ssa::PhiNode phi
-) {
-  phi.hasInputFromBlock(def, _, _, _, input) and
-  (
-    g.controls(input, branch)
-    or
-    exists(EdgeKind kind |
-      g.getBlock() = input and
-      kind = getConditionalEdge(branch) and
-      input.getSuccessor(kind) = phi.getBasicBlock()
-    )
-  )
-}
-
 /**
  * Holds if the guard `g` validates the expression `e` upon evaluating to `branch`.
  *
@@ -2291,21 +2662,13 @@ module BarrierGuard<guardChecksSig/3 guardChecks> {
    *
    * NOTE: If an indirect expression is tracked, use `getAnIndirectBarrierNode` instead.
    */
-  Node getABarrierNode() {
+  ExprNode getABarrierNode() {
     exists(IRGuardCondition g, Expr e, ValueNumber value, boolean edge |
       e = value.getAnInstruction().getConvertedResultExpression() and
-      result.asConvertedExpr() = e and
+      result.getConvertedExpr() = e and
       guardChecks(g, value.getAnInstruction().getConvertedResultExpression(), edge) and
       g.controls(result.getBasicBlock(), edge)
     )
-    or
-    exists(
-      IRGuardCondition g, boolean branch, Ssa::DefinitionExt def, IRBlock input, Ssa::PhiNode phi
-    |
-      guardChecks(g, def.getARead().asOperand().getDef().getConvertedResultExpression(), branch) and
-      guardControlsPhiInput(g, branch, def, input, phi) and
-      result = TSsaPhiInputNode(phi, input)
-    )
   }
 
   /**
@@ -2341,7 +2704,7 @@ module BarrierGuard<guardChecksSig/3 guardChecks> {
    *
    * NOTE: If a non-indirect expression is tracked, use `getABarrierNode` instead.
    */
-  Node getAnIndirectBarrierNode() { result = getAnIndirectBarrierNode(_) }
+  IndirectExprNode getAnIndirectBarrierNode() { result = getAnIndirectBarrierNode(_) }
 
   /**
    * Gets an indirect expression node with indirection index `indirectionIndex` that is
@@ -2377,23 +2740,13 @@ module BarrierGuard<guardChecksSig/3 guardChecks> {
    *
    * NOTE: If a non-indirect expression is tracked, use `getABarrierNode` instead.
    */
-  Node getAnIndirectBarrierNode(int indirectionIndex) {
+  IndirectExprNode getAnIndirectBarrierNode(int indirectionIndex) {
     exists(IRGuardCondition g, Expr e, ValueNumber value, boolean edge |
       e = value.getAnInstruction().getConvertedResultExpression() and
-      result.asIndirectConvertedExpr(indirectionIndex) = e and
+      result.getConvertedExpr(indirectionIndex) = e and
       guardChecks(g, value.getAnInstruction().getConvertedResultExpression(), edge) and
       g.controls(result.getBasicBlock(), edge)
     )
-    or
-    exists(
-      IRGuardCondition g, boolean branch, Ssa::DefinitionExt def, IRBlock input, Ssa::PhiNode phi
-    |
-      guardChecks(g,
-        def.getARead().asIndirectOperand(indirectionIndex).getDef().getConvertedResultExpression(),
-        branch) and
-      guardControlsPhiInput(g, branch, def, input, phi) and
-      result = TSsaPhiInputNode(phi, input)
-    )
   }
 }
 
@@ -2402,14 +2755,6 @@ module BarrierGuard<guardChecksSig/3 guardChecks> {
  */
 signature predicate instructionGuardChecksSig(IRGuardCondition g, Instruction instr, boolean branch);
 
-private EdgeKind getConditionalEdge(boolean branch) {
-  branch = true and
-  result instanceof TrueEdge
-  or
-  branch = false and
-  result instanceof FalseEdge
-}
-
 /**
  * Provides a set of barrier nodes for a guard that validates an instruction.
  *
@@ -2418,20 +2763,12 @@ private EdgeKind getConditionalEdge(boolean branch) {
  */
 module InstructionBarrierGuard<instructionGuardChecksSig/3 instructionGuardChecks> {
   /** Gets a node that is safely guarded by the given guard check. */
-  Node getABarrierNode() {
+  ExprNode getABarrierNode() {
     exists(IRGuardCondition g, ValueNumber value, boolean edge, Operand use |
       instructionGuardChecks(g, value.getAnInstruction(), edge) and
       use = value.getAnInstruction().getAUse() and
       result.asOperand() = use and
-      g.controls(result.getBasicBlock(), edge)
-    )
-    or
-    exists(
-      IRGuardCondition g, boolean branch, Ssa::DefinitionExt def, IRBlock input, Ssa::PhiNode phi
-    |
-      instructionGuardChecks(g, def.getARead().asOperand().getDef(), branch) and
-      guardControlsPhiInput(g, branch, def, input, phi) and
-      result = TSsaPhiInputNode(phi, input)
+      g.controls(use.getDef().getBlock(), edge)
     )
   }
 }

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

@@ -1,479 +0,0 @@
-/**
- * Provides the classes `ExprNode` and `IndirectExprNode` for converting between `Expr` and `Node`.
- */
-
-private import cpp
-private import semmle.code.cpp.ir.IR
-private import DataFlowUtil
-private import DataFlowPrivate
-private import semmle.code.cpp.ir.implementation.raw.internal.TranslatedExpr
-private import semmle.code.cpp.ir.implementation.raw.internal.InstructionTag
-
-cached
-private module Cached {
-  private Operand getAnInitializeDynamicAllocationInstructionAddress() {
-    result = any(InitializeDynamicAllocationInstruction init).getAllocationAddressOperand()
-  }
-
-  /**
-   * Gets the expression that should be returned as the result expression from `instr`.
-   *
-   * Note that this predicate may return multiple results in cases where a conversion belongs to a
-   * different AST element than its operand.
-   */
-  private Expr getConvertedResultExpression(Instruction instr, int n) {
-    // Only fully converted instructions have a result for `asConvertedExpr`
-    not conversionFlow(unique(Operand op |
-        // The address operand of a `InitializeDynamicAllocationInstruction` is
-        // special: we need to handle it during dataflow (since it's
-        // effectively a store to an indirection), but it doesn't appear in
-        // source syntax, so dataflow node <-> expression conversion shouldn't
-        // care about it.
-        op = getAUse(instr) and not op = getAnInitializeDynamicAllocationInstructionAddress()
-      |
-        op
-      ), _, false, false) and
-    result = getConvertedResultExpressionImpl(instr) and
-    n = 0
-    or
-    // If the conversion also has a result then we return multiple results
-    exists(Operand operand | conversionFlow(operand, instr, false, false) |
-      n = 1 and
-      result = getConvertedResultExpressionImpl(operand.getDef())
-      or
-      result = getConvertedResultExpression(operand.getDef(), n - 1)
-    )
-  }
-
-  private Expr getConvertedResultExpressionImpl0(Instruction instr) {
-    // IR construction inserts an additional cast to a `size_t` on the extent
-    // of a `new[]` expression. The resulting `ConvertInstruction` doesn't have
-    // a result for `getConvertedResultExpression`. We remap this here so that
-    // this `ConvertInstruction` maps to the result of the expression that
-    // represents the extent.
-    exists(TranslatedNonConstantAllocationSize tas |
-      result = tas.getExtent().getExpr() and
-      instr = tas.getInstruction(AllocationExtentConvertTag())
-    )
-    or
-    // There's no instruction that returns `ParenthesisExpr`, but some queries
-    // expect this
-    exists(TranslatedTransparentConversion ttc |
-      result = ttc.getExpr().(ParenthesisExpr) and
-      instr = ttc.getResult()
-    )
-    or
-    // Certain expressions generate `CopyValueInstruction`s only when they
-    // are needed. Examples of this include crement operations and compound
-    // assignment operations. For example:
-    // ```cpp
-    // int x = ...
-    // int y = x++;
-    // ```
-    // this generate IR like:
-    // ```
-    // r1(glval<int>) = VariableAddress[x] :
-    // r2(int)        = Constant[0]        :
-    // m3(int)        = Store[x]           : &:r1, r2
-    // r4(glval<int>) = VariableAddress[y] :
-    // r5(glval<int>) = VariableAddress[x] :
-    // r6(int)        = Load[x]            : &:r5, m3
-    // r7(int)        = Constant[1]        :
-    // r8(int)        = Add                : r6, r7
-    // m9(int)        = Store[x]           : &:r5, r8
-    // r11(int)       = CopyValue         : r6
-    // m12(int)       = Store[y]          : &:r4, r11
-    // ```
-    // When the `CopyValueInstruction` is not generated there is no instruction
-    // whose `getConvertedResultExpression` maps back to the expression. When
-    // such an instruction doesn't exist it means that the old value is not
-    // needed, and in that case the only value that will propagate forward in
-    // the program is the value that's been updated. So in those cases we just
-    // use the result of `node.asDefinition()` as the result of `node.asExpr()`.
-    exists(TranslatedCoreExpr tco |
-      tco.getInstruction(_) = instr and
-      tco.producesExprResult() and
-      result = asDefinitionImpl0(instr)
-    )
-  }
-
-  private Expr getConvertedResultExpressionImpl(Instruction instr) {
-    result = getConvertedResultExpressionImpl0(instr)
-    or
-    not exists(getConvertedResultExpressionImpl0(instr)) and
-    result = instr.getConvertedResultExpression()
-  }
-
-  /**
-   * Gets the result for `node.asDefinition()` (when `node` is the instruction
-   * node that wraps `store`) in the cases where `store.getAst()` should not be
-   * used to define the result of `node.asDefinition()`.
-   */
-  private Expr asDefinitionImpl0(StoreInstruction store) {
-    // For an expression such as `i += 2` we pretend that the generated
-    // `StoreInstruction` contains the result of the expression even though
-    // this isn't totally aligned with the C/C++ standard.
-    exists(TranslatedAssignOperation tao |
-      store = tao.getInstruction(AssignmentStoreTag()) and
-      result = tao.getExpr()
-    )
-    or
-    // Similarly for `i++` and `++i` we pretend that the generated
-    // `StoreInstruction` is contains the result of the expression even though
-    // this isn't totally aligned with the C/C++ standard.
-    exists(TranslatedCrementOperation tco |
-      store = tco.getInstruction(CrementStoreTag()) and
-      result = tco.getExpr()
-    )
-  }
-
-  /**
-   * Holds if the expression returned by `store.getAst()` should not be
-   * returned as the result of `node.asDefinition()` when `node` is the
-   * instruction node that wraps `store`.
-   */
-  private predicate excludeAsDefinitionResult(StoreInstruction store) {
-    // Exclude the store to the temporary generated by a ternary expression.
-    exists(TranslatedConditionalExpr tce |
-      store = tce.getInstruction(ConditionValueFalseStoreTag())
-      or
-      store = tce.getInstruction(ConditionValueTrueStoreTag())
-    )
-  }
-
-  /**
-   * Gets the expression that represents the result of `StoreInstruction` for
-   * dataflow purposes.
-   *
-   * For example, consider the following example
-   * ```cpp
-   * int x = 42;     // 1
-   * x = 34;         // 2
-   * ++x;            // 3
-   * x++;            // 4
-   * x += 1;         // 5
-   * int y = x += 2; // 6
-   * ```
-   * For (1) the result is `42`.
-   * For (2) the result is `x = 34`.
-   * For (3) the result is `++x`.
-   * For (4) the result is `x++`.
-   * For (5) the result is `x += 1`.
-   * For (6) there are two results:
-   *   - For the `StoreInstruction` generated by `x += 2` the result
-   *     is `x += 2`
-   *   - For the `StoreInstruction` generated by `int y = ...` the result
-   *     is also `x += 2`
-   */
-  cached
-  Expr asDefinitionImpl(StoreInstruction store) {
-    not exists(asDefinitionImpl0(store)) and
-    not excludeAsDefinitionResult(store) and
-    result = store.getAst().(Expr).getUnconverted()
-    or
-    result = asDefinitionImpl0(store)
-  }
-
-  /** Holds if `node` is an `OperandNode` that should map `node.asExpr()` to `e`. */
-  private predicate exprNodeShouldBeOperand(OperandNode node, Expr e, int n) {
-    not exprNodeShouldBeIndirectOperand(_, e, n) and
-    exists(Instruction def |
-      unique( | | getAUse(def)) = node.getOperand() and
-      e = getConvertedResultExpression(def, n)
-    )
-  }
-
-  /** Holds if `node` should be an `IndirectOperand` that maps `node.asIndirectExpr()` to `e`. */
-  private predicate indirectExprNodeShouldBeIndirectOperand(
-    IndirectOperand node, Expr e, int n, int indirectionIndex
-  ) {
-    exists(Instruction def |
-      node.hasOperandAndIndirectionIndex(unique( | | getAUse(def)), indirectionIndex) and
-      e = getConvertedResultExpression(def, n)
-    )
-  }
-
-  /** Holds if `node` should be an `IndirectOperand` that maps `node.asExpr()` to `e`. */
-  private predicate exprNodeShouldBeIndirectOperand(IndirectOperand node, Expr e, int n) {
-    exists(ArgumentOperand operand |
-      // When an argument (qualifier or positional) is a prvalue and the
-      // parameter (qualifier or positional) is a (const) reference, IR
-      // construction introduces a temporary `IRVariable`. The `VariableAddress`
-      // instruction has the argument as its `getConvertedResultExpression`
-      // result. However, the instruction actually represents the _address_ of
-      // the argument. So to fix this mismatch, we have the indirection of the
-      // `VariableAddressInstruction` map to the expression.
-      node.hasOperandAndIndirectionIndex(operand, 1) and
-      e = getConvertedResultExpression(operand.getDef(), n) and
-      operand.getDef().(VariableAddressInstruction).getIRVariable() instanceof IRTempVariable
-    )
-  }
-
-  private predicate exprNodeShouldBeIndirectOutNode(IndirectArgumentOutNode node, Expr e, int n) {
-    exists(CallInstruction call |
-      call.getStaticCallTarget() instanceof Constructor and
-      e = getConvertedResultExpression(call, n) and
-      call.getThisArgumentOperand() = node.getAddressOperand()
-    )
-  }
-
-  /** Holds if `node` should be an instruction node that maps `node.asExpr()` to `e`. */
-  private predicate exprNodeShouldBeInstruction(Node node, Expr e, int n) {
-    not exprNodeShouldBeOperand(_, e, n) and
-    not exprNodeShouldBeIndirectOutNode(_, e, n) and
-    not exprNodeShouldBeIndirectOperand(_, e, n) and
-    e = getConvertedResultExpression(node.asInstruction(), n)
-  }
-
-  /** Holds if `node` should be an `IndirectInstruction` that maps `node.asIndirectExpr()` to `e`. */
-  private predicate indirectExprNodeShouldBeIndirectInstruction(
-    IndirectInstruction node, Expr e, int n, int indirectionIndex
-  ) {
-    not indirectExprNodeShouldBeIndirectOperand(_, e, n, indirectionIndex) and
-    exists(Instruction instr |
-      node.hasInstructionAndIndirectionIndex(instr, indirectionIndex) and
-      e = getConvertedResultExpression(instr, n)
-    )
-  }
-
-  abstract private class ExprNodeBase extends Node {
-    /**
-     * Gets the expression corresponding to this node, if any. The returned
-     * expression may be a `Conversion`.
-     */
-    abstract Expr getConvertedExpr(int n);
-
-    /** Gets the non-conversion expression corresponding to this node, if any. */
-    final Expr getExpr(int n) { result = this.getConvertedExpr(n).getUnconverted() }
-  }
-
-  /**
-   * Holds if there exists a dataflow node whose `asExpr(n)` should evaluate
-   * to `e`.
-   */
-  private predicate exprNodeShouldBe(Expr e, int n) {
-    exprNodeShouldBeInstruction(_, e, n) or
-    exprNodeShouldBeOperand(_, e, n) or
-    exprNodeShouldBeIndirectOutNode(_, e, n) or
-    exprNodeShouldBeIndirectOperand(_, e, n)
-  }
-
-  private class InstructionExprNode extends ExprNodeBase, InstructionNode {
-    InstructionExprNode() {
-      exists(Expr e, int n |
-        exprNodeShouldBeInstruction(this, e, n) and
-        not exists(Expr conv |
-          exprNodeShouldBe(conv, n + 1) and
-          conv.getUnconverted() = e.getUnconverted()
-        )
-      )
-    }
-
-    final override Expr getConvertedExpr(int n) { exprNodeShouldBeInstruction(this, result, n) }
-  }
-
-  private class OperandExprNode extends ExprNodeBase, OperandNode {
-    OperandExprNode() {
-      exists(Expr e, int n |
-        exprNodeShouldBeOperand(this, e, n) and
-        not exists(Expr conv |
-          exprNodeShouldBe(conv, n + 1) and
-          conv.getUnconverted() = e.getUnconverted()
-        )
-      )
-    }
-
-    final override Expr getConvertedExpr(int n) { exprNodeShouldBeOperand(this, result, n) }
-  }
-
-  abstract private class IndirectExprNodeBase extends Node {
-    /**
-     * Gets the expression corresponding to this node, if any. The returned
-     * expression may be a `Conversion`.
-     */
-    abstract Expr getConvertedExpr(int n, int indirectionIndex);
-
-    /** Gets the non-conversion expression corresponding to this node, if any. */
-    final Expr getExpr(int n, int indirectionIndex) {
-      result = this.getConvertedExpr(n, indirectionIndex).getUnconverted()
-    }
-  }
-
-  /** A signature for converting an indirect node to an expression. */
-  private signature module IndirectNodeToIndirectExprSig {
-    /** The indirect node class to be converted to an expression */
-    class IndirectNode;
-
-    /**
-     * Holds if the indirect expression at indirection index `indirectionIndex`
-     * of `node` is `e`. The integer `n` specifies how many conversions has been
-     * applied to `node`.
-     */
-    predicate indirectNodeHasIndirectExpr(IndirectNode node, Expr e, int n, int indirectionIndex);
-  }
-
-  /**
-   * A module that implements the logic for deciding whether an indirect node
-   * should be an `IndirectExprNode`.
-   */
-  private module IndirectNodeToIndirectExpr<IndirectNodeToIndirectExprSig Sig> {
-    import Sig
-
-    /**
-     * This predicate shifts the indirection index by one when `conv` is a
-     * `ReferenceDereferenceExpr`.
-     *
-     * This is necessary because `ReferenceDereferenceExpr` is a conversion
-     * in the AST, but appears as a `LoadInstruction` in the IR.
-     */
-    bindingset[e, indirectionIndex]
-    private predicate adjustForReference(
-      Expr e, int indirectionIndex, Expr conv, int adjustedIndirectionIndex
-    ) {
-      conv.(ReferenceDereferenceExpr).getExpr() = e and
-      adjustedIndirectionIndex = indirectionIndex - 1
-      or
-      not conv instanceof ReferenceDereferenceExpr and
-      conv = e and
-      adjustedIndirectionIndex = indirectionIndex
-    }
-
-    /** Holds if `node` should be an `IndirectExprNode`. */
-    predicate charpred(IndirectNode node) {
-      exists(Expr e, int n, int indirectionIndex |
-        indirectNodeHasIndirectExpr(node, e, n, indirectionIndex) and
-        not exists(Expr conv, int adjustedIndirectionIndex |
-          adjustForReference(e, indirectionIndex, conv, adjustedIndirectionIndex) and
-          indirectExprNodeShouldBe(conv, n + 1, adjustedIndirectionIndex)
-        )
-      )
-    }
-  }
-
-  private predicate indirectExprNodeShouldBe(Expr e, int n, int indirectionIndex) {
-    indirectExprNodeShouldBeIndirectOperand(_, e, n, indirectionIndex) or
-    indirectExprNodeShouldBeIndirectInstruction(_, e, n, indirectionIndex)
-  }
-
-  private module IndirectOperandIndirectExprNodeImpl implements IndirectNodeToIndirectExprSig {
-    class IndirectNode = IndirectOperand;
-
-    predicate indirectNodeHasIndirectExpr = indirectExprNodeShouldBeIndirectOperand/4;
-  }
-
-  module IndirectOperandToIndirectExpr =
-    IndirectNodeToIndirectExpr<IndirectOperandIndirectExprNodeImpl>;
-
-  private class IndirectOperandIndirectExprNode extends IndirectExprNodeBase instanceof IndirectOperand
-  {
-    IndirectOperandIndirectExprNode() { IndirectOperandToIndirectExpr::charpred(this) }
-
-    final override Expr getConvertedExpr(int n, int index) {
-      IndirectOperandToIndirectExpr::indirectNodeHasIndirectExpr(this, result, n, index)
-    }
-  }
-
-  private module IndirectInstructionIndirectExprNodeImpl implements IndirectNodeToIndirectExprSig {
-    class IndirectNode = IndirectInstruction;
-
-    predicate indirectNodeHasIndirectExpr = indirectExprNodeShouldBeIndirectInstruction/4;
-  }
-
-  module IndirectInstructionToIndirectExpr =
-    IndirectNodeToIndirectExpr<IndirectInstructionIndirectExprNodeImpl>;
-
-  private class IndirectInstructionIndirectExprNode extends IndirectExprNodeBase instanceof IndirectInstruction
-  {
-    IndirectInstructionIndirectExprNode() { IndirectInstructionToIndirectExpr::charpred(this) }
-
-    final override Expr getConvertedExpr(int n, int index) {
-      IndirectInstructionToIndirectExpr::indirectNodeHasIndirectExpr(this, result, n, index)
-    }
-  }
-
-  private class IndirectArgumentOutExprNode extends ExprNodeBase, IndirectArgumentOutNode {
-    IndirectArgumentOutExprNode() { exprNodeShouldBeIndirectOutNode(this, _, _) }
-
-    final override Expr getConvertedExpr(int n) { exprNodeShouldBeIndirectOutNode(this, result, n) }
-  }
-
-  private class IndirectOperandExprNode extends ExprNodeBase instanceof IndirectOperand {
-    IndirectOperandExprNode() { exprNodeShouldBeIndirectOperand(this, _, _) }
-
-    final override Expr getConvertedExpr(int n) { exprNodeShouldBeIndirectOperand(this, result, n) }
-  }
-
-  /**
-   * An expression, viewed as a node in a data flow graph.
-   */
-  cached
-  class ExprNode extends Node instanceof ExprNodeBase {
-    /**
-     * INTERNAL: Do not use.
-     */
-    cached
-    Expr getExpr(int n) { result = super.getExpr(n) }
-
-    /**
-     * Gets the non-conversion expression corresponding to this node, if any. If
-     * this node strictly (in the sense of `getConvertedExpr`) corresponds to a
-     * `Conversion`, then the result is that `Conversion`'s non-`Conversion` base
-     * expression.
-     */
-    cached
-    final Expr getExpr() { result = this.getExpr(_) }
-
-    /**
-     * INTERNAL: Do not use.
-     */
-    cached
-    Expr getConvertedExpr(int n) { result = super.getConvertedExpr(n) }
-
-    /**
-     * Gets the expression corresponding to this node, if any. The returned
-     * expression may be a `Conversion`.
-     */
-    cached
-    final Expr getConvertedExpr() { result = this.getConvertedExpr(_) }
-  }
-
-  /**
-   * An indirect expression, viewed as a node in a data flow graph.
-   */
-  cached
-  class IndirectExprNode extends Node instanceof IndirectExprNodeBase {
-    /**
-     * Gets the non-conversion expression corresponding to this node, if any. If
-     * this node strictly (in the sense of `getConvertedExpr`) corresponds to a
-     * `Conversion`, then the result is that `Conversion`'s non-`Conversion` base
-     * expression.
-     */
-    cached
-    final Expr getExpr(int indirectionIndex) { result = this.getExpr(_, indirectionIndex) }
-
-    /**
-     * INTERNAL: Do not use.
-     */
-    cached
-    Expr getExpr(int n, int indirectionIndex) { result = super.getExpr(n, indirectionIndex) }
-
-    /**
-     * INTERNAL: Do not use.
-     */
-    cached
-    Expr getConvertedExpr(int n, int indirectionIndex) {
-      result = super.getConvertedExpr(n, indirectionIndex)
-    }
-
-    /**
-     * Gets the expression corresponding to this node, if any. The returned
-     * expression may be a `Conversion`.
-     */
-    cached
-    Expr getConvertedExpr(int indirectionIndex) {
-      result = this.getConvertedExpr(_, indirectionIndex)
-    }
-  }
-}
-
-import Cached

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

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

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

@@ -9,7 +9,6 @@ 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 DataFlowPrivate
 import SsaInternalsCommon
 
@@ -105,8 +104,8 @@ predicate hasRawIndirectInstruction(Instruction instr, int indirectionIndex) {
 cached
 private newtype TDefImpl =
   TDefAddressImpl(BaseIRVariable v) or
-  TDirectDefImpl(Operand address, int indirectionIndex) {
-    isDef(_, _, address, _, _, indirectionIndex)
+  TDirectDefImpl(BaseSourceVariableInstruction base, Operand address, int indirectionIndex) {
+    isDef(_, _, address, base, _, indirectionIndex)
   } or
   TGlobalDefImpl(GlobalLikeVariable v, IRFunction f, int indirectionIndex) {
     // Represents the initial "definition" of a global variable when entering
@@ -116,8 +115,8 @@ private newtype TDefImpl =
 
 cached
 private newtype TUseImpl =
-  TDirectUseImpl(Operand operand, int indirectionIndex) {
-    isUse(_, operand, _, _, indirectionIndex) and
+  TDirectUseImpl(BaseSourceVariableInstruction base, Operand operand, int indirectionIndex) {
+    isUse(_, operand, base, _, indirectionIndex) and
     not isDef(true, _, operand, _, _, _)
   } or
   TGlobalUse(GlobalLikeVariable v, IRFunction f, int indirectionIndex) {
@@ -211,11 +210,19 @@ abstract class DefImpl extends TDefImpl {
    */
   abstract int getIndirection();
 
+  /**
+   * Gets the instruction that computes the base of this definition or use.
+   * This is always a `VariableAddressInstruction` or an `CallInstruction`.
+   */
+  abstract BaseSourceVariableInstruction getBase();
+
   /**
    * Gets the base source variable (i.e., the variable without
    * any indirection) of this definition or use.
    */
-  abstract BaseSourceVariable getBaseSourceVariable();
+  final BaseSourceVariable getBaseSourceVariable() {
+    this.getBase().getBaseSourceVariable() = result
+  }
 
   /** Gets the variable that is defined or used. */
   SourceVariable getSourceVariable() {
@@ -275,11 +282,19 @@ abstract class UseImpl extends TUseImpl {
   /** Gets the indirection index of this use. */
   final int getIndirectionIndex() { result = indirectionIndex }
 
+  /**
+   * Gets the instruction that computes the base of this definition or use.
+   * This is always a `VariableAddressInstruction` or an `CallInstruction`.
+   */
+  abstract BaseSourceVariableInstruction getBase();
+
   /**
    * Gets the base source variable (i.e., the variable without
    * any indirection) of this definition or use.
    */
-  abstract BaseSourceVariable getBaseSourceVariable();
+  final BaseSourceVariable getBaseSourceVariable() {
+    this.getBase().getBaseSourceVariable() = result
+  }
 
   /** Gets the variable that is defined or used. */
   SourceVariable getSourceVariable() {
@@ -314,17 +329,6 @@ private predicate sourceVariableHasBaseAndIndex(SourceVariable v, BaseSourceVari
   v.getIndirection() = ind
 }
 
-/**
- * Gets the instruction that computes the address that's used to
- * initialize `v`.
- */
-private Instruction getInitializationTargetAddress(IRVariable v) {
-  exists(TranslatedVariableInitialization init |
-    init.getIRVariable() = v and
-    result = init.getTargetAddress()
-  )
-}
-
 /** An initial definition of an `IRVariable`'s address. */
 private class DefAddressImpl extends DefImpl, TDefAddressImpl {
   BaseIRVariable v;
@@ -343,15 +347,8 @@ private class DefAddressImpl extends DefImpl, TDefAddressImpl {
   final override Node0Impl getValue() { none() }
 
   final override predicate hasIndexInBlock(IRBlock block, int index) {
-    exists(IRVariable var | var = v.getIRVariable() |
-      block.getInstruction(index) = getInitializationTargetAddress(var)
-      or
-      // If there is no translatated element that does initialization of the
-      // variable we place the SSA definition at the entry block of the function.
-      not exists(getInitializationTargetAddress(var)) and
-      block = var.getEnclosingIRFunction().getEntryBlock() and
-      index = 0
-    )
+    block = v.getIRVariable().getEnclosingIRFunction().getEntryBlock() and
+    index = 0
   }
 
   override Cpp::Location getLocation() { result = v.getIRVariable().getLocation() }
@@ -361,13 +358,14 @@ private class DefAddressImpl extends DefImpl, TDefAddressImpl {
     result.getIndirection() = 0
   }
 
-  final override BaseSourceVariable getBaseSourceVariable() { result = v }
+  final override BaseSourceVariableInstruction getBase() { none() }
 }
 
 private class DirectDef extends DefImpl, TDirectDefImpl {
   Operand address;
+  BaseSourceVariableInstruction base;
 
-  DirectDef() { this = TDirectDefImpl(address, indirectionIndex) }
+  DirectDef() { this = TDirectDefImpl(base, address, indirectionIndex) }
 
   override Cpp::Location getLocation() { result = this.getAddressOperand().getUse().getLocation() }
 
@@ -379,36 +377,30 @@ private class DirectDef extends DefImpl, TDirectDefImpl {
 
   override Operand getAddressOperand() { result = address }
 
-  private BaseSourceVariableInstruction getBase() {
-    isDef(_, _, address, result, _, indirectionIndex)
-  }
+  override BaseSourceVariableInstruction getBase() { result = base }
 
-  override BaseSourceVariable getBaseSourceVariable() {
-    result = this.getBase().getBaseSourceVariable()
-  }
+  override int getIndirection() { isDef(_, _, address, base, result, indirectionIndex) }
 
-  override int getIndirection() { isDef(_, _, address, _, result, indirectionIndex) }
+  override Node0Impl getValue() { isDef(_, result, address, base, _, _) }
 
-  override Node0Impl getValue() { isDef(_, result, address, _, _, _) }
-
-  override predicate isCertain() { isDef(true, _, address, _, _, indirectionIndex) }
+  override predicate isCertain() { isDef(true, _, address, base, _, indirectionIndex) }
 }
 
 private class DirectUseImpl extends UseImpl, TDirectUseImpl {
   Operand operand;
+  BaseSourceVariableInstruction base;
 
-  DirectUseImpl() { this = TDirectUseImpl(operand, indirectionIndex) }
+  DirectUseImpl() { this = TDirectUseImpl(base, operand, indirectionIndex) }
 
   override string toString() { result = "Use of " + this.getSourceVariable() }
 
   final override predicate hasIndexInBlock(IRBlock block, int index) {
     // See the comment in `ssa0`'s `OperandBasedUse` for an explanation of this
     // predicate's implementation.
-    if this.getBase().getAst() = any(Cpp::PostfixCrementOperation c).getOperand()
+    if base.getAst() = any(Cpp::PostfixCrementOperation c).getOperand()
     then
-      exists(Operand op, int indirection, Instruction base |
+      exists(Operand op, int indirection |
         indirection = this.getIndirection() and
-        base = this.getBase() and
         op =
           min(Operand cand, int i |
             isUse(_, cand, base, indirection, indirectionIndex) and
@@ -421,19 +413,15 @@ private class DirectUseImpl extends UseImpl, TDirectUseImpl {
     else operand.getUse() = block.getInstruction(index)
   }
 
-  private BaseSourceVariableInstruction getBase() { isUse(_, operand, result, _, indirectionIndex) }
-
-  override BaseSourceVariable getBaseSourceVariable() {
-    result = this.getBase().getBaseSourceVariable()
-  }
+  final override BaseSourceVariableInstruction getBase() { result = base }
 
   final Operand getOperand() { result = operand }
 
   final override Cpp::Location getLocation() { result = operand.getLocation() }
 
-  override int getIndirection() { isUse(_, operand, _, result, indirectionIndex) }
+  override int getIndirection() { isUse(_, operand, base, result, indirectionIndex) }
 
-  override predicate isCertain() { isUse(true, operand, _, _, indirectionIndex) }
+  override predicate isCertain() { isUse(true, operand, base, _, indirectionIndex) }
 
   override Node getNode() { nodeHasOperand(result, operand, indirectionIndex) }
 }
@@ -492,7 +480,13 @@ class FinalParameterUse extends UseImpl, TFinalParameterUse {
     result instanceof UnknownDefaultLocation
   }
 
-  override BaseIRVariable getBaseSourceVariable() { result.getIRVariable().getAst() = p }
+  override BaseSourceVariableInstruction getBase() {
+    exists(InitializeParameterInstruction init |
+      init.getParameter() = p and
+      // This is always a `VariableAddressInstruction`
+      result = init.getAnOperand().getDef()
+    )
+  }
 }
 
 /**
@@ -578,8 +572,8 @@ class GlobalUse extends UseImpl, TGlobalUse {
     )
   }
 
-  override BaseSourceVariable getBaseSourceVariable() {
-    baseSourceVariableIsGlobal(result, global, f)
+  override SourceVariable getSourceVariable() {
+    sourceVariableIsGlobal(result, global, f, this.getIndirection())
   }
 
   final override Cpp::Location getLocation() { result = f.getLocation() }
@@ -596,6 +590,8 @@ class GlobalUse extends UseImpl, TGlobalUse {
   Type getUnderlyingType() { result = global.getUnderlyingType() }
 
   override predicate isCertain() { any() }
+
+  override BaseSourceVariableInstruction getBase() { none() }
 }
 
 /**
@@ -625,8 +621,8 @@ class GlobalDefImpl extends DefImpl, TGlobalDefImpl {
   }
 
   /** Gets the global variable associated with this definition. */
-  override BaseSourceVariable getBaseSourceVariable() {
-    baseSourceVariableIsGlobal(result, global, f)
+  override SourceVariable getSourceVariable() {
+    sourceVariableIsGlobal(result, global, f, this.getIndirection())
   }
 
   override int getIndirection() { result = indirectionIndex }
@@ -649,6 +645,8 @@ class GlobalDefImpl extends DefImpl, TGlobalDefImpl {
   override string toString() { result = "Def of " + this.getSourceVariable() }
 
   override Location getLocation() { result = f.getLocation() }
+
+  override BaseSourceVariableInstruction getBase() { none() }
 }
 
 /**
@@ -657,9 +655,19 @@ class GlobalDefImpl extends DefImpl, TGlobalDefImpl {
  */
 predicate adjacentDefRead(IRBlock bb1, int i1, SourceVariable sv, IRBlock bb2, int i2) {
   adjacentDefReadExt(_, sv, bb1, i1, bb2, i2)
+  or
+  exists(PhiNode phi |
+    lastRefRedefExt(_, sv, bb1, i1, phi) and
+    phi.definesAt(sv, bb2, i2, _)
+  )
 }
 
 predicate useToNode(IRBlock bb, int i, SourceVariable sv, Node nodeTo) {
+  exists(Phi phi |
+    phi.asPhi().definesAt(sv, bb, i, _) and
+    nodeTo = phi.getNode()
+  )
+  or
   exists(UseImpl use |
     use.hasIndexInBlock(bb, i, sv) and
     nodeTo = use.getNode()
@@ -713,26 +721,46 @@ predicate nodeToDefOrUse(Node node, SourceVariable sv, IRBlock bb, int i, boolea
  */
 private predicate indirectConversionFlowStep(Node nFrom, Node nTo) {
   not exists(SourceVariable sv, IRBlock bb2, int i2 |
-    useToNode(bb2, i2, sv, nTo) and
+    nodeToDefOrUse(nTo, sv, bb2, i2, _) and
     adjacentDefRead(bb2, i2, sv, _, _)
   ) and
-  exists(Operand op1, Operand op2, int indirectionIndex, Instruction instr |
-    hasOperandAndIndex(nFrom, op1, pragma[only_bind_into](indirectionIndex)) and
-    hasOperandAndIndex(nTo, op2, pragma[only_bind_into](indirectionIndex)) and
-    instr = op2.getDef() and
-    conversionFlow(op1, instr, _, _)
+  (
+    exists(Operand op1, Operand op2, int indirectionIndex, Instruction instr |
+      hasOperandAndIndex(nFrom, op1, pragma[only_bind_into](indirectionIndex)) and
+      hasOperandAndIndex(nTo, op2, pragma[only_bind_into](indirectionIndex)) and
+      instr = op2.getDef() and
+      conversionFlow(op1, instr, _, _)
+    )
+    or
+    exists(Operand op1, Operand op2, int indirectionIndex, Instruction instr |
+      hasOperandAndIndex(nFrom, op1, pragma[only_bind_into](indirectionIndex)) and
+      hasOperandAndIndex(nTo, op2, indirectionIndex - 1) and
+      instr = op2.getDef() and
+      isDereference(instr, op1, _)
+    )
   )
 }
 
 /**
- * Holds if `node` is a phi input node that should receive flow from the
- * definition to (or use of) `sv` at `(bb1, i1)`.
+ * The reason for this predicate is a bit annoying:
+ * We cannot mark a `PointerArithmeticInstruction` that computes an offset based on some SSA
+ * variable `x` as a use of `x` since this creates taint-flow in the following example:
+ * ```c
+ * int x = array[source]
+ * sink(*array)
+ * ```
+ * This is because `source` would flow from the operand of `PointerArithmeticInstruction` to the
+ * result of the instruction, and into the `IndirectOperand` that represents the value of `*array`.
+ * Then, via use-use flow, flow will arrive at `*array` in `sink(*array)`.
+ *
+ * So this predicate recurses back along conversions and `PointerArithmeticInstruction`s to find the
+ * first use that has provides use-use flow, and uses that target as the target of the `nodeFrom`.
  */
-private predicate phiToNode(SsaPhiInputNode node, SourceVariable sv, IRBlock bb1, int i1) {
-  exists(PhiNode phi, IRBlock input |
-    phi.hasInputFromBlock(_, sv, bb1, i1, input) and
-    node.getPhiNode() = phi and
-    node.getBlock() = input
+private predicate adjustForPointerArith(PostUpdateNode pun, SourceVariable sv, IRBlock bb2, int i2) {
+  exists(IRBlock bb1, int i1, Node adjusted |
+    indirectConversionFlowStep*(adjusted, pun.getPreUpdateNode()) and
+    nodeToDefOrUse(adjusted, sv, bb1, i1, _) and
+    adjacentDefRead(bb1, i1, sv, bb2, i2)
   )
 }
 
@@ -747,14 +775,10 @@ private predicate phiToNode(SsaPhiInputNode node, SourceVariable sv, IRBlock bb1
 private predicate ssaFlowImpl(
   IRBlock bb1, int i1, SourceVariable sv, Node nodeFrom, Node nodeTo, boolean uncertain
 ) {
-  nodeToDefOrUse(nodeFrom, sv, bb1, i1, uncertain) and
-  (
-    exists(IRBlock bb2, int i2 |
-      adjacentDefRead(bb1, i1, sv, bb2, i2) and
-      useToNode(bb2, i2, sv, nodeTo)
-    )
-    or
-    phiToNode(nodeTo, sv, bb1, i1)
+  exists(IRBlock bb2, int i2 |
+    nodeToDefOrUse(nodeFrom, sv, bb1, i1, uncertain) and
+    adjacentDefRead(bb1, i1, sv, bb2, i2) and
+    useToNode(bb2, i2, sv, nodeTo)
   ) and
   nodeFrom != nodeTo
 }
@@ -763,7 +787,7 @@ private predicate ssaFlowImpl(
 private Node getAPriorDefinition(DefinitionExt next) {
   exists(IRBlock bb, int i, SourceVariable sv |
     lastRefRedefExt(_, pragma[only_bind_into](sv), pragma[only_bind_into](bb),
-      pragma[only_bind_into](i), _, next) and
+      pragma[only_bind_into](i), next) and
     nodeToDefOrUse(result, sv, bb, i, _)
   )
 }
@@ -870,31 +894,9 @@ private predicate isArgumentOfCallable(DataFlowCall call, Node n) {
  * Holds if there is use-use flow from `pun`'s pre-update node to `n`.
  */
 private predicate postUpdateNodeToFirstUse(PostUpdateNode pun, Node n) {
-  // We cannot mark a `PointerArithmeticInstruction` that computes an offset
-  // based on some SSA
-  // variable `x` as a use of `x` since this creates taint-flow in the
-  // following example:
-  // ```c
-  // int x = array[source]
-  // sink(*array)
-  // ```
-  // This is because `source` would flow from the operand of `PointerArithmetic`
-  // instruction to the result of the instruction, and into the `IndirectOperand`
-  // that represents the value of `*array`. Then, via use-use flow, flow will
-  // arrive at `*array` in `sink(*array)`.
-  // So this predicate recurses back along conversions and `PointerArithmetic`
-  // instructions to find the first use that has provides use-use flow, and
-  // uses that target as the target of the `nodeFrom`.
-  exists(Node adjusted, IRBlock bb1, int i1, SourceVariable sv |
-    indirectConversionFlowStep*(adjusted, pun.getPreUpdateNode()) and
-    useToNode(bb1, i1, sv, adjusted)
-  |
-    exists(IRBlock bb2, int i2 |
-      adjacentDefRead(bb1, i1, sv, bb2, i2) and
-      useToNode(bb2, i2, sv, n)
-    )
-    or
-    phiToNode(n, sv, bb1, i1)
+  exists(SourceVariable sv, IRBlock bb2, int i2 |
+    adjustForPointerArith(pun, sv, bb2, i2) and
+    useToNode(bb2, i2, sv, n)
   )
 }
 
@@ -949,23 +951,19 @@ predicate postUpdateFlow(PostUpdateNode pun, Node nodeTo) {
 
 /** Holds if `nodeTo` receives flow from the phi node `nodeFrom`. */
 predicate fromPhiNode(SsaPhiNode nodeFrom, Node nodeTo) {
-  exists(PhiNode phi, SourceVariable sv, IRBlock bb1, int i1 |
+  exists(PhiNode phi, SourceVariable sv, IRBlock bb1, int i1, IRBlock bb2, int i2 |
     phi = nodeFrom.getPhiNode() and
-    phi.definesAt(sv, bb1, i1, _)
-  |
-    exists(IRBlock bb2, int i2 |
-      adjacentDefRead(bb1, i1, sv, bb2, i2) and
-      useToNode(bb2, i2, sv, nodeTo)
-    )
-    or
-    phiToNode(nodeTo, sv, bb1, i1)
+    phi.definesAt(sv, bb1, i1, _) and
+    adjacentDefRead(bb1, i1, sv, bb2, i2) and
+    useToNode(bb2, i2, sv, nodeTo)
   )
 }
 
-private predicate baseSourceVariableIsGlobal(
-  BaseIRVariable base, GlobalLikeVariable global, IRFunction func
+private predicate sourceVariableIsGlobal(
+  SourceVariable sv, GlobalLikeVariable global, IRFunction func, int indirectionIndex
 ) {
-  exists(IRVariable irVar |
+  exists(IRVariable irVar, BaseIRVariable base |
+    sourceVariableHasBaseAndIndex(sv, base, indirectionIndex) and
     irVar = base.getIRVariable() and
     irVar.getEnclosingIRFunction() = func and
     global = irVar.getAst() and
@@ -1032,26 +1030,22 @@ module SsaCached {
    * Holds if the node at index `i` in `bb` is a last reference to SSA definition
    * `def`. The reference is last because it can reach another write `next`,
    * without passing through another read or write.
-   *
-   * The path from node `i` in `bb` to `next` goes via basic block `input`,
-   * which is either a predecessor of the basic block of `next`, or `input` =
-   * `bb` in case `next` occurs in basic block `bb`.
    */
   cached
   predicate lastRefRedefExt(
-    DefinitionExt def, SourceVariable sv, IRBlock bb, int i, IRBlock input, DefinitionExt next
+    DefinitionExt def, SourceVariable sv, IRBlock bb, int i, DefinitionExt next
   ) {
-    SsaImpl::lastRefRedefExt(def, sv, bb, i, input, next)
+    SsaImpl::lastRefRedefExt(def, sv, bb, i, next)
   }
 
   cached
-  Definition phiHasInputFromBlockExt(PhiNode phi, IRBlock bb) {
-    SsaImpl::phiHasInputFromBlockExt(phi, result, bb)
+  Definition phiHasInputFromBlock(PhiNode phi, IRBlock bb) {
+    SsaImpl::phiHasInputFromBlock(phi, result, bb)
   }
 
   cached
-  predicate ssaDefReachesReadExt(SourceVariable v, DefinitionExt def, IRBlock bb, int i) {
-    SsaImpl::ssaDefReachesReadExt(v, def, bb, i)
+  predicate ssaDefReachesRead(SourceVariable v, Definition def, IRBlock bb, int i) {
+    SsaImpl::ssaDefReachesRead(v, def, bb, i)
   }
 
   predicate variableRead = SsaInput::variableRead/4;
@@ -1203,11 +1197,11 @@ class Phi extends TPhi, SsaDef {
 
   final override Location getLocation() { result = phi.getBasicBlock().getLocation() }
 
-  override string toString() { result = phi.toString() }
+  override string toString() { result = "Phi" }
 
-  SsaPhiInputNode getNode(IRBlock block) { result.getPhiNode() = phi and result.getBlock() = block }
+  SsaPhiNode getNode() { result.getPhiNode() = phi }
 
-  predicate hasInputFromBlock(Definition inp, IRBlock bb) { inp = phiHasInputFromBlockExt(phi, bb) }
+  predicate hasInputFromBlock(Definition inp, IRBlock bb) { inp = phiHasInputFromBlock(phi, bb) }
 
   final Definition getAnInput() { this.hasInputFromBlock(result, _) }
 }
@@ -1233,21 +1227,13 @@ class PhiNode extends SsaImpl::DefinitionExt {
    */
   predicate isPhiRead() { this instanceof SsaImpl::PhiReadNode }
 
-  /**
-   * Holds if the node at index `i` in `bb` is a last reference to SSA
-   * definition `def` of `sv`. The reference is last because it can reach
-   * this phi node, without passing through another read or write.
-   *
-   * The path from node `i` in `bb` to this phi node goes via basic block
-   * `input`, which is either a predecessor of the basic block of this phi
-   * node, or `input` = `bb` in case this phi node occurs in basic block `bb`.
-   */
-  predicate hasInputFromBlock(DefinitionExt def, SourceVariable sv, IRBlock bb, int i, IRBlock input) {
-    SsaCached::lastRefRedefExt(def, sv, bb, i, input, this)
+  /** Holds if `inp` is an input to this phi node along the edge originating in `bb`. */
+  predicate hasInputFromBlock(Definition inp, IRBlock bb) {
+    inp = SsaCached::phiHasInputFromBlock(this, bb)
   }
 
   /** Gets a definition that is an input to this phi node. */
-  final Definition getAnInput() { this.hasInputFromBlock(result, _, _, _, _) }
+  final Definition getAnInput() { this.hasInputFromBlock(result, _) }
 }
 
 /** An static single assignment (SSA) definition. */
@@ -1262,15 +1248,6 @@ class DefinitionExt extends SsaImpl::DefinitionExt {
     result = this.getAPhiInputOrPriorDefinition*() and
     not result instanceof PhiNode
   }
-
-  /** Gets a node that represents a read of this SSA definition. */
-  Node getARead() {
-    exists(SourceVariable sv, IRBlock bb, int i | SsaCached::ssaDefReachesReadExt(sv, this, bb, i) |
-      useToNode(bb, i, sv, result)
-      or
-      phiToNode(result, sv, bb, i)
-    )
-  }
 }
 
 class Definition = SsaImpl::Definition;

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

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

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

@@ -1844,6 +1844,9 @@ class TranslatedAssignExpr extends TranslatedNonConstantExpr {
     child = this.getRightOperand() and
     result = this.getLeftOperand().getFirstInstruction(kind)
     or
+    child = this.getRightOperand() and
+    result = this.getLeftOperand().getFirstInstruction(kind)
+    or
     kind instanceof GotoEdge and
     child = this.getLeftOperand() and
     result = this.getInstruction(AssignmentStoreTag())
@@ -3208,20 +3211,9 @@ class TranslatedBuiltInOperation extends TranslatedNonConstantExpr {
 
   final override Instruction getResult() { result = this.getInstruction(OnlyInstructionTag()) }
 
-  /**
-   * Gets the rnk'th (0-indexed) child for which a `TranslatedElement` exists.
-   *
-   * We use this predicate to filter out `TypeName` expressions that sometimes
-   * occur in builtin operations since the IR doesn't have an instruction to
-   * represent a reference to a type.
-   */
-  private TranslatedElement getRankedChild(int rnk) {
-    result = rank[rnk + 1](int id, TranslatedElement te | te = this.getChild(id) | te order by id)
-  }
-
   final override Instruction getFirstInstruction(EdgeKind kind) {
-    if exists(this.getRankedChild(0))
-    then result = this.getRankedChild(0).getFirstInstruction(kind)
+    if exists(this.getChild(0))
+    then result = this.getChild(0).getFirstInstruction(kind)
     else (
       kind instanceof GotoEdge and result = this.getInstruction(OnlyInstructionTag())
     )
@@ -3241,11 +3233,11 @@ class TranslatedBuiltInOperation extends TranslatedNonConstantExpr {
   }
 
   final override Instruction getChildSuccessorInternal(TranslatedElement child, EdgeKind kind) {
-    exists(int id | child = this.getRankedChild(id) |
-      result = this.getRankedChild(id + 1).getFirstInstruction(kind)
+    exists(int id | child = this.getChild(id) |
+      result = this.getChild(id + 1).getFirstInstruction(kind)
       or
       kind instanceof GotoEdge and
-      not exists(this.getRankedChild(id + 1)) and
+      not exists(this.getChild(id + 1)) and
       result = this.getInstruction(OnlyInstructionTag())
     )
   }
@@ -3260,7 +3252,7 @@ class TranslatedBuiltInOperation extends TranslatedNonConstantExpr {
     tag = OnlyInstructionTag() and
     exists(int index |
       operandTag = positionalArgumentOperand(index) and
-      result = this.getRankedChild(index).(TranslatedExpr).getResult()
+      result = this.getChild(index).(TranslatedExpr).getResult()
     )
   }
 

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

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

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

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

cpp/ql/src/CHANGELOG.md

@@ -1,30 +1,3 @@
-## 1.0.2
-
-No user-facing changes.
-
-## 1.0.1
-
-### Minor Analysis Improvements
-
-* The `cpp/dangerous-function-overflow` no longer produces a false positive alert when the `gets` function does not have exactly one parameter.
-
-## 1.0.0
-
-### Breaking Changes
-
-* CodeQL package management is now generally available, and all GitHub-produced CodeQL packages have had their version numbers increased to 1.0.0.
-
-### Minor Analysis Improvements
-
-* The "Use of unique pointer after lifetime ends" query (`cpp/use-of-unique-pointer-after-lifetime-ends`) no longer reports an alert when the pointer is converted to a boolean
-* The "Variable not initialized before use" query (`cpp/not-initialised`) no longer reports an alert on static variables.
-
-## 0.9.12
-
-### New Queries
-
-* Added a new query, `cpp/iterator-to-expired-container`, to detect the creation of iterators owned by a temporary objects that are about to be destroyed.
-
 ## 0.9.11
 
 ### Minor Analysis Improvements

cpp/ql/src/Critical/DoubleFree.qhelp

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

cpp/ql/src/Critical/DoubleFreeBad2.cpp

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

cpp/ql/src/Critical/DoubleFreeGood.cpp

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

cpp/ql/src/Critical/DoubleFreeGood2.cpp

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

cpp/ql/src/Critical/NotInitialised.ql

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

@@ -37,19 +37,6 @@ class AllocaCall extends FunctionCall {
   }
 }
 
-/**
- * Gets an expression associated with a dataflow node.
- */
-private Expr getExpr(DataFlow::Node node) {
-  result = node.asInstruction().getAst()
-  or
-  result = node.asOperand().getUse().getAst()
-  or
-  result = node.(DataFlow::RawIndirectInstruction).getInstruction().getAst()
-  or
-  result = node.(DataFlow::RawIndirectOperand).getOperand().getUse().getAst()
-}
-
 /**
  * A loop that contains an `alloca` call.
  */
@@ -198,6 +185,19 @@ class LoopWithAlloca extends Stmt {
     not this.conditionReachesWithoutUpdate(var, this.(Loop).getCondition())
   }
 
+  /**
+   * Gets an expression associated with a dataflow node.
+   */
+  private Expr getExpr(DataFlow::Node node) {
+    result = node.asInstruction().getAst()
+    or
+    result = node.asOperand().getUse().getAst()
+    or
+    result = node.(DataFlow::RawIndirectInstruction).getInstruction().getAst()
+    or
+    result = node.(DataFlow::RawIndirectOperand).getOperand().getUse().getAst()
+  }
+
   /**
    * Gets a definition that may be the most recent definition of the
    * controlling variable `var` before this loop.
@@ -209,9 +209,8 @@ class LoopWithAlloca extends Stmt {
       DataFlow::localFlow(result, DataFlow::exprNode(va)) and
       // Phi nodes will be preceded by nodes that represent actual definitions
       not result instanceof DataFlow::SsaPhiNode and
-      not result instanceof DataFlow::SsaPhiInputNode and
       // A source is outside the loop if it's not inside the loop
-      not exists(Expr e | e = getExpr(result) | this = getAnEnclosingLoopOfExpr(e))
+      not exists(Expr e | e = this.getExpr(result) | this = getAnEnclosingLoopOfExpr(e))
     )
   }
 
@@ -222,9 +221,9 @@ class LoopWithAlloca extends Stmt {
   private int getAControllingVarInitialValue(Variable var, DataFlow::Node source) {
     source = this.getAPrecedingDef(var) and
     (
-      result = getExpr(source).getValue().toInt()
+      result = this.getExpr(source).getValue().toInt()
       or
-      result = getExpr(source).(Assignment).getRValue().getValue().toInt()
+      result = this.getExpr(source).(Assignment).getRValue().getValue().toInt()
     )
   }
 

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

cpp/ql/src/change-notes/2024-04-29-iterator-to-expired-container.md

@@ -1,5 +1,4 @@
-## 0.9.12
-
-### New Queries
-
+---
+category: newQuery
+---
 * Added a new query, `cpp/iterator-to-expired-container`, to detect the creation of iterators owned by a temporary objects that are about to be destroyed.

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

@@ -1,10 +0,0 @@
-## 1.0.0
-
-### Breaking Changes
-
-* CodeQL package management is now generally available, and all GitHub-produced CodeQL packages have had their version numbers increased to 1.0.0.
-
-### Minor Analysis Improvements
-
-* The "Use of unique pointer after lifetime ends" query (`cpp/use-of-unique-pointer-after-lifetime-ends`) no longer reports an alert when the pointer is converted to a boolean
-* The "Variable not initialized before use" query (`cpp/not-initialised`) no longer reports an alert on static variables.

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

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

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

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

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

@@ -1,2 +1,2 @@
 ---
-lastReleaseVersion: 1.0.2
+lastReleaseVersion: 0.9.11

cpp/ql/src/experimental/Likely Bugs/DerefNullResult.cpp

@@ -1,23 +0,0 @@
-char * create (int arg) {
-    if (arg > 42) {
-        // this function may return NULL
-        return NULL;
-    }
-    char * r = malloc(arg);
-    snprintf(r, arg -1, "Hello");
-    return r;
-}
-
-void process(char *str) {
-    // str is dereferenced
-    if (str[0] == 'H') {
-        printf("Hello H\n");
-    }
-}
-
-void test(int arg) {
-    // first function returns a pointer that may be NULL
-    char *str = create(arg);
-    // str is not checked for nullness before being passed to process function
-    process(str);
-}

cpp/ql/src/experimental/Likely Bugs/DerefNullResult.qhelp

@@ -1,26 +0,0 @@
-<!DOCTYPE qhelp PUBLIC
-  "-//Semmle//qhelp//EN"
-  "qhelp.dtd">
-<qhelp>
-
-<overview>
-<p>This rule finds a dereference of a function parameter, whose value comes from another function call that may return NULL, without checks in the meantime.</p>
-</overview>
-
-<recommendation>
-<p>A check should be added between the return of the function which may return NULL, and its use by the function dereferencing ths pointer.</p>
-</recommendation>
-
-<example>
-<sample src="DerefNullResult.cpp" />
-</example>
-
-<references>
-<li>
-  <a href="https://www.owasp.org/index.php/Null_Dereference">
-    Null Dereference
-  </a> 
-</li>
-</references>
-
-</qhelp>