Impl of Scan with filters (unfinished)

- Enable `mg-expr` in mg-storage-v3
-  Adapt id_mapper
- Change conversion function from PropertyValue to TypedValue
- Remove memgraph functions
- Enable expression tests on for storage

.clang-tidy

@@ -6,6 +6,7 @@ Checks: '*,
          -altera-unroll-loops,
          -android-*,
          -cert-err58-cpp,
+         -cert-str34-c,
          -cppcoreguidelines-avoid-c-arrays,
          -cppcoreguidelines-avoid-goto,
          -cppcoreguidelines-avoid-magic-numbers,
@@ -49,6 +50,7 @@ Checks: '*,
          -misc-non-private-member-variables-in-classes,
          -modernize-avoid-c-arrays,
          -modernize-concat-nested-namespaces,
+         -modernize-loop-convert,
          -modernize-pass-by-value,
          -modernize-use-equals-default,
          -modernize-use-nodiscard,
@@ -61,7 +63,9 @@ Checks: '*,
          -readability-magic-numbers,
          -readability-named-parameter,
          -misc-no-recursion,
-         -concurrency-mt-unsafe'
+         -concurrency-mt-unsafe,
+         -bugprone-easily-swappable-parameters'
+
 WarningsAsErrors: ''
 HeaderFilterRegex: 'src/.*'
 AnalyzeTemporaryDtors: false
@@ -88,4 +92,3 @@ CheckOptions:
   - key:             modernize-use-nullptr.NullMacros
     value:           'NULL'
 ...
-

.githooks/pre-commit

@@ -24,14 +24,6 @@ for file in $modified_files; do
 
   git checkout-index --prefix="$tmpdir/" -- $file
 
-  echo "Running clang-format..."
-  $project_folder/tools/git-clang-format $tmpdir/$file
-  CODE=$?
-
-  if [ $CODE -ne 0 ]; then
-    break
-  fi
-
   # Do not break header checker
   echo "Running header checker..."
   $project_folder/tools/header-checker.py $tmpdir/$file $file --amend-year
@@ -39,7 +31,6 @@ for file in $modified_files; do
   if [ $CODE -ne 0 ]; then
     FAIL=1
   fi
-
 done;
 
 return ${FAIL}

.github/pull_request_template.md

@@ -3,7 +3,9 @@
 - [ ] Update [changelog](https://docs.memgraph.com/memgraph/changelog)
 - [ ] Write E2E tests
 - [ ] Compare the [benchmarking results](https://bench-graph.memgraph.com/) between the master branch and the Epic branch
+- [ ] Provide the full content or a guide for the final git message
 
 [master < Task] PR
 - [ ] Check, and update documentation if necessary
 - [ ] Update [changelog](https://docs.memgraph.com/memgraph/changelog)
+- [ ] Provide the full content or a guide for the final git message

.github/workflows/diff.yaml

@@ -1,4 +1,7 @@
 name: Diff
+concurrency:
+  group: ${{ github.head_ref || github.sha }}
+  cancel-in-progress: true
 
 on:
   push:
@@ -67,6 +70,11 @@ jobs:
           # branches and tags. (default: 1)
           fetch-depth: 0
 
+        # This is also needed if we want do to comparison against other branches
+        # See https://github.community/t/checkout-code-fails-when-it-runs-lerna-run-test-since-master/17920
+      - name: Fetch all history for all tags and branches
+        run: git fetch
+
       - name: Build combined ASAN, UBSAN and coverage binaries
         run: |
           # Activate toolchain.
@@ -107,12 +115,22 @@ jobs:
           name: "Code coverage"
           path: tools/github/generated/code_coverage.tar.gz
 
+      - name: Set base branch
+        if: ${{ github.event_name == 'pull_request' }}
+        run: |
+          echo "BASE_BRANCH=origin/${{ github.base_ref }}" >> $GITHUB_ENV
+
+      - name: Set base branch # if we manually dispatch or push to master
+        if: ${{ github.event_name != 'pull_request' }}
+        run: |
+          echo "BASE_BRANCH=origin/master" >> $GITHUB_ENV
+
       - name: Run clang-tidy
         run: |
           source /opt/toolchain-v4/activate
 
           # Restrict clang-tidy results only to the modified parts
-          git diff -U0 master... -- src ':!*.hpp' | ./tools/github/clang-tidy/clang-tidy-diff.py -p 1 -j $THREADS -path build | tee ./build/clang_tidy_output.txt
+          git diff -U0 ${{ env.BASE_BRANCH }}... -- src | ./tools/github/clang-tidy/clang-tidy-diff.py -p 1 -j $THREADS -extra-arg="-DMG_CLANG_TIDY_CHECK" -path build | tee ./build/clang_tidy_output.txt
 
           # Fail if any warning is reported
           ! cat ./build/clang_tidy_output.txt | ./tools/github/clang-tidy/grep_error_lines.sh > /dev/null
@@ -153,7 +171,7 @@ jobs:
 
           # Run leftover CTest tests (all except unit and benchmark tests).
           cd build
-          ctest -E "(memgraph__unit|memgraph__benchmark)" --output-on-failure
+          ctest -E "(memgraph__unit|memgraph__benchmark|memgraph__simulation)" --output-on-failure
 
       - name: Run drivers tests
         run: |
@@ -244,6 +262,15 @@ jobs:
           cd build
           ctest -R memgraph__unit --output-on-failure -j$THREADS
 
+      - name: Run simulation tests
+        run: |
+          # Activate toolchain.
+          source /opt/toolchain-v4/activate
+
+          # Run unit tests.
+          cd build
+          ctest -R memgraph__simulation --output-on-failure -j$THREADS
+
       - name: Run e2e tests
         run: |
           # TODO(gitbuda): Setup mgclient and pymgclient properly.

.github/workflows/full_clang_tidy.yaml

@@ -39,7 +39,7 @@ jobs:
           source /opt/toolchain-v4/activate
 
           # The results are also written to standard output in order to retain them in the logs
-          ./tools/github/clang-tidy/run-clang-tidy.py -p build -j $THREADS -clang-tidy-binary=/opt/toolchain-v4/bin/clang-tidy "$PWD/src/*" |
+          ./tools/github/clang-tidy/run-clang-tidy.py -p build -j $THREADS -extra-arg="-DMG_CLANG_TIDY_CHECK" -clang-tidy-binary=/opt/toolchain-v4/bin/clang-tidy "$PWD/src/*" |
             tee ./build/full_clang_tidy_output.txt
 
       - name: Summarize clang-tidy results

.github/workflows/package_all.yaml

@@ -6,11 +6,11 @@ on: workflow_dispatch
 
 jobs:
   centos-7:
-    runs-on: [self-hosted, DockerMgBuild]
+    runs-on: [self-hosted, DockerMgBuild, X64]
     timeout-minutes: 60
     steps:
       - name: "Set up repository"
-        uses: actions/checkout@v2
+        uses: actions/checkout@v3
         with:
           fetch-depth: 0 # Required because of release/get_version.py
       - name: "Build package"
@@ -22,29 +22,29 @@ jobs:
           name: centos-7
           path: build/output/centos-7/memgraph*.rpm
 
-  centos-8:
-    runs-on: [self-hosted, DockerMgBuild]
+  centos-9:
+    runs-on: [self-hosted, DockerMgBuild, X64]
     timeout-minutes: 60
     steps:
       - name: "Set up repository"
-        uses: actions/checkout@v2
+        uses: actions/checkout@v3
         with:
           fetch-depth: 0 # Required because of release/get_version.py
       - name: "Build package"
         run: |
-          ./release/package/run.sh package centos-8
+          ./release/package/run.sh package centos-9
       - name: "Upload package"
         uses: actions/upload-artifact@v2
         with:
-          name: centos-8
-          path: build/output/centos-8/memgraph*.rpm
+          name: centos-9
+          path: build/output/centos-9/memgraph*.rpm
 
   debian-10:
-    runs-on: [self-hosted, DockerMgBuild]
+    runs-on: [self-hosted, DockerMgBuild, X64]
     timeout-minutes: 60
     steps:
       - name: "Set up repository"
-        uses: actions/checkout@v2
+        uses: actions/checkout@v3
         with:
           fetch-depth: 0 # Required because of release/get_version.py
       - name: "Build package"
@@ -57,11 +57,11 @@ jobs:
           path: build/output/debian-10/memgraph*.deb
 
   debian-11:
-    runs-on: [self-hosted, DockerMgBuild]
+    runs-on: [self-hosted, DockerMgBuild, X64]
     timeout-minutes: 60
     steps:
       - name: "Set up repository"
-        uses: actions/checkout@v2
+        uses: actions/checkout@v3
         with:
           fetch-depth: 0 # Required because of release/get_version.py
       - name: "Build package"
@@ -74,11 +74,11 @@ jobs:
           path: build/output/debian-11/memgraph*.deb
 
   docker:
-    runs-on: [self-hosted, DockerMgBuild]
+    runs-on: [self-hosted, DockerMgBuild, X64]
     timeout-minutes: 60
     steps:
       - name: "Set up repository"
-        uses: actions/checkout@v2
+        uses: actions/checkout@v3
         with:
           fetch-depth: 0 # Required because of release/get_version.py
       - name: "Build package"
@@ -93,11 +93,11 @@ jobs:
           path: build/output/docker/memgraph*.tar.gz
 
   ubuntu-1804:
-    runs-on: [self-hosted, DockerMgBuild]
+    runs-on: [self-hosted, DockerMgBuild, X64]
     timeout-minutes: 60
     steps:
       - name: "Set up repository"
-        uses: actions/checkout@v2
+        uses: actions/checkout@v3
         with:
           fetch-depth: 0 # Required because of release/get_version.py
       - name: "Build package"
@@ -110,11 +110,11 @@ jobs:
           path: build/output/ubuntu-18.04/memgraph*.deb
 
   ubuntu-2004:
-    runs-on: [self-hosted, DockerMgBuild]
+    runs-on: [self-hosted, DockerMgBuild, X64]
     timeout-minutes: 60
     steps:
       - name: "Set up repository"
-        uses: actions/checkout@v2
+        uses: actions/checkout@v3
         with:
           fetch-depth: 0 # Required because of release/get_version.py
       - name: "Build package"
@@ -126,12 +126,29 @@ jobs:
           name: ubuntu-2004
           path: build/output/ubuntu-20.04/memgraph*.deb
 
-  debian-11-platform:
-    runs-on: [self-hosted, DockerMgBuild]
+  ubuntu-2204:
+    runs-on: [self-hosted, DockerMgBuild, X64]
     timeout-minutes: 60
     steps:
       - name: "Set up repository"
-        uses: actions/checkout@v2
+        uses: actions/checkout@v3
+        with:
+          fetch-depth: 0 # Required because of release/get_version.py
+      - name: "Build package"
+        run: |
+          ./release/package/run.sh package ubuntu-22.04
+      - name: "Upload package"
+        uses: actions/upload-artifact@v2
+        with:
+          name: ubuntu-2204
+          path: build/output/ubuntu-22.04/memgraph*.deb
+
+  debian-11-platform:
+    runs-on: [self-hosted, DockerMgBuild, X64]
+    timeout-minutes: 60
+    steps:
+      - name: "Set up repository"
+        uses: actions/checkout@v3
         with:
           fetch-depth: 0 # Required because of release/get_version.py
       - name: "Build package"
@@ -142,3 +159,20 @@ jobs:
         with:
           name: debian-11-platform
           path: build/output/debian-11/memgraph*.deb
+
+  debian-11-arm:
+    runs-on: [self-hosted, DockerMgBuild, ARM64, strange]
+    timeout-minutes: 60
+    steps:
+      - name: "Set up repository"
+        uses: actions/checkout@v3
+        with:
+          fetch-depth: 0 # Required because of release/get_version.py
+      - name: "Build package"
+        run: |
+          ./release/package/run.sh package debian-11-arm
+      - name: "Upload package"
+        uses: actions/upload-artifact@v2
+        with:
+          name: debian-11-arm
+          path: build/output/debian-11-arm/memgraph*.deb

.github/workflows/release_docker.yaml

@@ -4,8 +4,12 @@ on:
   workflow_dispatch:
     inputs:
       version:
-        description: "Memgraph binary version to publish on Dockerhub."
+        description: "Memgraph binary version to publish on DockerHub."
         required: true
+      force_release:
+        type: boolean
+        required: false
+        default: false
 
 jobs:
   docker_publish:
@@ -36,6 +40,22 @@ jobs:
           curl -L https://download.memgraph.com/memgraph/v${{ github.event.inputs.version }}/debian-11/memgraph_${{ github.event.inputs.version }}-1_amd64.deb > memgraph-amd64.deb
           curl -L https://download.memgraph.com/memgraph/v${{ github.event.inputs.version }}/debian-11-aarch64/memgraph_${{ github.event.inputs.version }}-1_arm64.deb > memgraph-arm64.deb
 
+      - name: Check if specified version is already pushed
+        run: |
+          EXISTS=$(docker manifest inspect $DOCKER_ORGANIZATION_NAME/$DOCKER_REPOSITORY_NAME:${{ github.event.inputs.version }} > /dev/null; echo $?)
+          echo $EXISTS
+          if [[ ${EXISTS} -eq 0 ]]; then
+              echo 'The specified version has been already released to DockerHub.'
+              if [[ ${{ github.event.inputs.force_release }} = true ]]; then
+                echo 'Forcing the release!'
+              else
+                echo 'Stopping the release!'
+                exit 1
+              fi
+          else
+              echo 'All good the specified version has not been release to DockerHub.'
+          fi
+
       - name: Build & push docker images
         run: |
           cd release/docker

.gitignore

@@ -23,6 +23,8 @@ cmake-build-*
 cmake/DownloadProject/
 dist/
 src/query/frontend/opencypher/generated/
+src/query/v2/frontend/opencypher/generated/
+src/parser/opencypher/generated
 tags
 ve/
 ve3/
@@ -50,15 +52,25 @@ src/distributed/pull_produce_rpc_messages.hpp
 src/distributed/storage_gc_rpc_messages.hpp
 src/distributed/token_sharing_rpc_messages.hpp
 src/distributed/updates_rpc_messages.hpp
+src/query/v2/frontend/ast/ast.hpp
 src/query/frontend/ast/ast.hpp
+src/storage/v3/bindings/ast/ast.hpp
 src/query/distributed/frontend/ast/ast_serialization.hpp
+src/query/v2/distributed/frontend/ast/ast_serialization.hpp
 src/durability/distributed/state_delta.hpp
 src/durability/single_node/state_delta.hpp
 src/durability/single_node_ha/state_delta.hpp
 src/query/frontend/semantic/symbol.hpp
+src/query/v2/frontend/semantic/symbol.hpp
+src/expr/semantic/symbol.hpp
 src/query/distributed/frontend/semantic/symbol_serialization.hpp
+src/query/v2/distributed/frontend/semantic/symbol_serialization.hpp
 src/query/distributed/plan/ops.hpp
+src/query/v2/distributed/plan/ops.hpp
 src/query/plan/operator.hpp
+src/query/v2/plan/operator.hpp
+src/parser/opencypher/generated
+src/expr/semantic/symbol.hpp
 src/raft/log_entry.hpp
 src/raft/raft_rpc_messages.hpp
 src/raft/snapshot_metadata.hpp

.pre-commit-config.yaml

@@ -0,0 +1,24 @@
+repos:
+  - repo: https://github.com/pre-commit/pre-commit-hooks
+    rev: v2.3.0
+    hooks:
+      - id: check-yaml
+      - id: end-of-file-fixer
+      - id: trailing-whitespace
+  - repo: https://github.com/psf/black
+    rev: 22.3.0
+    hooks:
+      - id: black
+        args: # arguments to configure black
+          - --line-length=120
+          - --include='\.pyi?$'
+          # these folders wont be formatted by black
+          - --exclude="""\.git |
+            \.__pycache__|
+            build|
+            libs|
+            .cache"""
+  - repo: https://github.com/pre-commit/mirrors-clang-format
+    rev: v13.0.0
+    hooks:
+      - id: clang-format

CODEOWNERS

@@ -1 +0,0 @@
-* @antaljanosbenjamin @kostasrim

config/flags.yaml

@@ -47,6 +47,14 @@ modifications:
     value: ""
     override: false
 
+  - name: "bolt_cert_file"
+    value: "/etc/memgraph/ssl/cert.pem"
+    override: false
+
+  - name: "bolt_key_file"
+    value: "/etc/memgraph/ssl/key.pem"
+    override: false
+
   - name: "storage_properties_on_edges"
     value: "true"
     override: true

environment/os/centos-7.sh

@@ -5,6 +5,9 @@ set -Eeuo pipefail
 DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
 source "$DIR/../util.sh"
 
+check_operating_system "centos-7"
+check_architecture "x86_64"
+
 TOOLCHAIN_BUILD_DEPS=(
     coreutils gcc gcc-c++ make # generic build tools
     wget # used for archive download

environment/os/centos-8.sh

@@ -1,161 +0,0 @@
-#!/bin/bash
-
-set -Eeuo pipefail
-
-DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
-source "$DIR/../util.sh"
-
-TOOLCHAIN_BUILD_DEPS=(
-    coreutils-common gcc gcc-c++ make # generic build tools
-    wget # used for archive download
-    gnupg2 # used for archive signature verification
-    tar gzip bzip2 xz unzip # used for archive unpacking
-    zlib-devel # zlib library used for all builds
-    expat-devel libipt libipt-devel libbabeltrace-devel xz-devel python36-devel texinfo # for gdb
-    libcurl-devel # for cmake
-    curl # snappy
-    readline-devel # for cmake and llvm
-    libffi-devel libxml2-devel # for llvm
-    libedit-devel pcre-devel automake bison # for swig
-    file
-    openssl-devel
-    gmp-devel
-    gperf
-    patch
-)
-
-TOOLCHAIN_RUN_DEPS=(
-    make # generic build tools
-    tar gzip bzip2 xz # used for archive unpacking
-    zlib # zlib library used for all builds
-    expat libipt libbabeltrace xz-libs python36 # for gdb
-    readline # for cmake and llvm
-    libffi libxml2 # for llvm
-    openssl-devel
-)
-
-MEMGRAPH_BUILD_DEPS=(
-    git # source code control
-    make pkgconf-pkg-config # build system
-    curl wget # for downloading libs
-    libuuid-devel java-11-openjdk # required by antlr
-    readline-devel # for memgraph console
-    python36-devel # for query modules
-    openssl-devel
-    libseccomp-devel
-    python36 python3-virtualenv python3-pip nmap-ncat # for qa, macro_benchmark and stress tests
-    #
-    # IMPORTANT: python3-yaml does NOT exist on CentOS
-    # Install it manually using `pip3 install PyYAML`
-    #
-    PyYAML # Package name here does not correspond to the yum package!
-    libcurl-devel # mg-requests
-    rpm-build rpmlint # for RPM package building
-    doxygen graphviz # source documentation generators
-    which mono-complete dotnet-sdk-3.1 nodejs golang zip unzip java-11-openjdk-devel # for driver tests
-    sbcl # for custom Lisp C++ preprocessing
-    autoconf # for jemalloc code generation
-    libtool  # for protobuf code generation
-)
-
-list() {
-    echo "$1"
-}
-
-check() {
-    local missing=""
-    for pkg in $1; do
-        if [ "$pkg" == "PyYAML" ]; then
-            if ! python3 -c "import yaml" >/dev/null 2>/dev/null; then
-                missing="$pkg $missing"
-            fi
-            continue
-        fi
-        if ! yum list installed "$pkg" >/dev/null 2>/dev/null; then
-            missing="$pkg $missing"
-        fi
-    done
-    if [ "$missing" != "" ]; then
-        echo "MISSING PACKAGES: $missing"
-        exit 1
-    fi
-}
-
-install() {
-    cd "$DIR"
-    if [ "$EUID" -ne 0 ]; then
-        echo "Please run as root."
-        exit 1
-    fi
-    # If GitHub Actions runner is installed, append LANG to the environment.
-    # Python related tests doesn't work the LANG export.
-    if [ -d "/home/gh/actions-runner" ]; then
-        echo "LANG=en_US.utf8" >> /home/gh/actions-runner/.env
-    else
-        echo "NOTE: export LANG=en_US.utf8"
-    fi
-    dnf install -y epel-release
-    dnf install -y 'dnf-command(config-manager)'
-    dnf config-manager --set-enabled powertools # Required to install texinfo.
-    dnf update -y
-    dnf install -y wget git python36 python3-pip
-    for pkg in $1; do
-        if [ "$pkg" == libipt ]; then
-            if ! dnf list installed libipt >/dev/null 2>/dev/null; then
-                dnf install -y http://repo.okay.com.mx/centos/8/x86_64/release/libipt-1.6.1-8.el8.x86_64.rpm
-            fi
-            continue
-        fi
-        if [ "$pkg" == libipt-devel ]; then
-            if ! yum list installed libipt-devel >/dev/null 2>/dev/null; then
-                dnf install -y http://repo.okay.com.mx/centos/8/x86_64/release/libipt-devel-1.6.1-8.el8.x86_64.rpm
-            fi
-            continue
-        fi
-        # Install GDB dependencies not present in the standard repos.
-        # https://bugs.centos.org/view.php?id=17068
-        # https://centos.pkgs.org
-	    # Since 2020, there is Babeltrace2 (https://babeltrace.org). Not used
-        # within GDB yet (an assumption).
-        # http://mirror.centos.org/centos/8/PowerTools/x86_64/os/Packages/libbabeltrace-devel-1.5.4-3.el8.x86_64.rpm not working
-        if [ "$pkg" == libbabeltrace-devel ]; then
-            if ! dnf list installed libbabeltrace-devel >/dev/null 2>/dev/null; then
-                dnf install -y https://rpmfind.net/linux/centos/8-stream/PowerTools/x86_64/os/Packages/libbabeltrace-devel-1.5.4-3.el8.x86_64.rpm
-            fi
-            continue
-        fi
-        if [ "$pkg" == sbcl ]; then
-            if ! dnf list installed cl-asdf >/dev/null 2>/dev/null; then
-                dnf install -y 	https://pkgs.dyn.su/el8/base/x86_64/cl-asdf-20101028-18.el8.noarch.rpm
-            fi
-            if ! dnf list installed common-lisp-controller >/dev/null 2>/dev/null; then
-                dnf install -y https://pkgs.dyn.su/el8/base/x86_64/common-lisp-controller-7.4-20.el8.noarch.rpm
-            fi
-            if ! dnf list installed sbcl >/dev/null 2>/dev/null; then
-                dnf install -y https://pkgs.dyn.su/el8/base/x86_64/sbcl-2.0.1-4.el8.x86_64.rpm
-            fi
-            continue
-        fi
-        if [ "$pkg" == dotnet-sdk-3.1 ]; then
-            if ! dnf list installed dotnet-sdk-3.1 >/dev/null 2>/dev/null; then
-                wget -nv https://packages.microsoft.com/config/centos/8/packages-microsoft-prod.rpm -O packages-microsoft-prod.rpm
-                rpm -Uvh https://packages.microsoft.com/config/centos/8/packages-microsoft-prod.rpm
-                dnf update -y
-                dnf install -y dotnet-sdk-3.1
-            fi
-            continue
-        fi
-        if [ "$pkg" == PyYAML ]; then
-            if [ -z ${SUDO_USER+x} ]; then # Running as root (e.g. Docker).
-                pip3 install --user PyYAML
-            else # Running using sudo.
-                sudo -H -u "$SUDO_USER" bash -c "pip3 install --user PyYAML"
-            fi
-            continue
-        fi
-        dnf install -y "$pkg"
-    done
-}
-
-deps=$2"[*]"
-"$1" "${!deps}"

environment/os/centos-9.sh

@@ -5,15 +5,16 @@ set -Eeuo pipefail
 DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
 source "$DIR/../util.sh"
 
+check_operating_system "centos-9"
+check_architecture "x86_64"
+
 TOOLCHAIN_BUILD_DEPS=(
-    coreutils gcc gcc-c++ make # generic build tools
+    coreutils-common gcc gcc-c++ make # generic build tools
     wget # used for archive download
     gnupg2 # used for archive signature verification
     tar gzip bzip2 xz unzip # used for archive unpacking
     zlib-devel # zlib library used for all builds
-    expat-devel xz-devel python3-devel texinfo # for gdb
-    libcurl-devel # for cmake
-    curl # snappy
+    expat-devel xz-devel python3-devel texinfo libbabeltrace-devel # for gdb
     readline-devel # for cmake and llvm
     libffi-devel libxml2-devel # for llvm
     libedit-devel pcre-devel automake bison # for swig
@@ -21,6 +22,9 @@ TOOLCHAIN_BUILD_DEPS=(
     openssl-devel
     gmp-devel
     gperf
+    diffutils
+    libipt libipt-devel # intel
+    patch
 )
 
 TOOLCHAIN_RUN_DEPS=(
@@ -31,18 +35,19 @@ TOOLCHAIN_RUN_DEPS=(
     readline # for cmake and llvm
     libffi libxml2 # for llvm
     openssl-devel
+    perl # for openssl
 )
 
 MEMGRAPH_BUILD_DEPS=(
     git # source code control
     make pkgconf-pkg-config # build system
-    curl wget # for downloading libs
+    wget # for downloading libs
     libuuid-devel java-11-openjdk # required by antlr
     readline-devel # for memgraph console
     python3-devel # for query modules
     openssl-devel
     libseccomp-devel
-    python3 python3-virtualenv python3-pip nmap-ncat # for qa, macro_benchmark and stress tests
+    python3 python3-pip python3-virtualenv nmap-ncat # for qa, macro_benchmark and stress tests
     #
     # IMPORTANT: python3-yaml does NOT exist on CentOS
     # Install it manually using `pip3 install PyYAML`
@@ -73,12 +78,6 @@ check() {
         if [ "$pkg" == "python3-virtualenv" ]; then
             continue
         fi
-        if [ "$pkg" == sbcl ]; then
-            if ! sbcl --version &> /dev/null; then
-	        missing="$pkg $missing"
-            fi
-            continue
-        fi
         if ! yum list installed "$pkg" >/dev/null 2>/dev/null; then
             missing="$pkg $missing"
         fi
@@ -105,13 +104,37 @@ install() {
     yum update -y
     yum install -y wget git python3 python3-pip
     for pkg in $1; do
+        # Since there is no support for libipt-devel for CentOS 9 we install
+        # Fedoras version of same libs, they are the same version but released
+        # for different OS
+        # TODO Update when libipt-devel releases for CentOS 9
+        if [ "$pkg" == libipt ]; then
+            if ! dnf list installed libipt >/dev/null 2>/dev/null; then
+                dnf install -y http://repo.okay.com.mx/centos/8/x86_64/release/libipt-1.6.1-8.el8.x86_64.rpm
+            fi
+            continue
+        fi
+        if [ "$pkg" == libipt-devel ]; then
+            if ! dnf list installed libipt-devel >/dev/null 2>/dev/null; then
+                dnf install -y http://repo.okay.com.mx/centos/8/x86_64/release/libipt-devel-1.6.1-8.el8.x86_64.rpm
+            fi
+            continue
+        fi
+        if [ "$pkg" == libbabeltrace-devel ]; then
+            if ! dnf list installed libbabeltrace-devel >/dev/null 2>/dev/null; then
+                dnf install -y http://mirror.stream.centos.org/9-stream/CRB/x86_64/os/Packages/libbabeltrace-devel-1.5.8-10.el9.x86_64.rpm
+            fi
+            continue
+        fi
         if [ "$pkg" == sbcl ]; then
-            if ! sbcl --version &> /dev/null; then
-	        curl -s https://altushost-swe.dl.sourceforge.net/project/sbcl/sbcl/1.4.2/sbcl-1.4.2-arm64-linux-binary.tar.bz2 -o /tmp/sbcl-arm64.tar.bz2
-		tar xvjf /tmp/sbcl-arm64.tar.bz2 -C /tmp
-		pushd /tmp/sbcl-1.4.2-arm64-linux
-		INSTALL_ROOT=/usr/local sh install.sh
-		popd
+            if ! dnf list installed cl-asdf >/dev/null 2>/dev/null; then
+                dnf install -y 	https://pkgs.dyn.su/el8/base/x86_64/cl-asdf-20101028-18.el8.noarch.rpm
+            fi
+            if ! dnf list installed common-lisp-controller >/dev/null 2>/dev/null; then
+                dnf install -y https://pkgs.dyn.su/el8/base/x86_64/common-lisp-controller-7.4-20.el8.noarch.rpm
+            fi
+            if ! dnf list installed sbcl >/dev/null 2>/dev/null; then
+                dnf install -y https://pkgs.dyn.su/el8/base/x86_64/sbcl-2.0.1-4.el8.x86_64.rpm
             fi
             continue
         fi
@@ -125,9 +148,11 @@ install() {
         fi
         if [ "$pkg" == python3-virtualenv ]; then
             if [ -z ${SUDO_USER+x} ]; then # Running as root (e.g. Docker).
-                pip3 install --user virtualenv
+                pip3 install virtualenv
+                pip3 install virtualenvwrapper
             else # Running using sudo.
-                sudo -H -u "$SUDO_USER" bash -c "pip3 install --user virtualenv"
+                sudo -H -u "$SUDO_USER" bash -c "pip3 install virtualenv"
+                sudo -H -u "$SUDO_USER" bash -c "pip3 install virtualenvwrapper"
             fi
             continue
         fi

environment/os/debian-10.sh

@@ -5,6 +5,9 @@ set -Eeuo pipefail
 DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
 source "$DIR/../util.sh"
 
+check_operating_system "debian-10"
+check_architecture "x86_64"
+
 TOOLCHAIN_BUILD_DEPS=(
     coreutils gcc g++ build-essential make # generic build tools
     wget # used for archive download

environment/os/debian-11-arm.sh

@@ -5,6 +5,9 @@ set -Eeuo pipefail
 DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
 source "$DIR/../util.sh"
 
+check_operating_system "debian-11"
+check_architecture "arm64"
+
 TOOLCHAIN_BUILD_DEPS=(
     coreutils gcc g++ build-essential make # generic build tools
     wget # used for archive download
@@ -87,15 +90,6 @@ EOF
     fi
     apt install -y wget
     for pkg in $1; do
-        if [ "$pkg" == dotnet-sdk-3.1 ]; then
-            if ! dpkg -s "$pkg" 2>/dev/null >/dev/null; then
-                wget -nv https://packages.microsoft.com/config/debian/10/packages-microsoft-prod.deb -O packages-microsoft-prod.deb
-                dpkg -i packages-microsoft-prod.deb
-                apt-get update
-                apt-get install -y apt-transport-https dotnet-sdk-3.1
-            fi
-            continue
-        fi
         apt install -y "$pkg"
     done
 }

environment/os/debian-11.sh

@@ -5,6 +5,9 @@ set -Eeuo pipefail
 DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
 source "$DIR/../util.sh"
 
+check_operating_system "debian-11"
+check_architecture "x86_64"
+
 TOOLCHAIN_BUILD_DEPS=(
     coreutils gcc g++ build-essential make # generic build tools
     wget # used for archive download
@@ -87,7 +90,7 @@ EOF
     fi
     apt install -y wget
     for pkg in $1; do
-        if [ "$pkg" == dotnet-sdk-3.1 ]; then
+        if [ "$pkg" == dotnet-sdk-3.1  ]; then
             if ! dpkg -s "$pkg" 2>/dev/null >/dev/null; then
                 wget -nv https://packages.microsoft.com/config/debian/10/packages-microsoft-prod.deb -O packages-microsoft-prod.deb
                 dpkg -i packages-microsoft-prod.deb

environment/os/template.sh

@@ -5,6 +5,8 @@ set -Eeuo pipefail
 DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
 source "$DIR/../util.sh"
 
+check_operating_system "todo-os-name"
+
 TOOLCHAIN_BUILD_DEPS=(
     pkg
 )

environment/os/ubuntu-18.04.sh

@@ -5,6 +5,9 @@ set -Eeuo pipefail
 DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
 source "$DIR/../util.sh"
 
+check_operating_system "ubuntu-18.04"
+check_architecture "x86_64"
+
 TOOLCHAIN_BUILD_DEPS=(
     coreutils gcc g++ build-essential make # generic build tools
     wget # archive download

environment/os/ubuntu-20.04.sh

@@ -5,6 +5,9 @@ set -Eeuo pipefail
 DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
 source "$DIR/../util.sh"
 
+check_operating_system "ubuntu-20.04"
+check_architecture "x86_64"
+
 TOOLCHAIN_BUILD_DEPS=(
     coreutils gcc g++ build-essential make # generic build tools
     wget # used for archive download

environment/os/ubuntu-22.04.sh

@@ -0,0 +1,96 @@
+#!/bin/bash
+
+set -Eeuo pipefail
+
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
+source "$DIR/../util.sh"
+
+check_operating_system "ubuntu-22.04"
+check_architecture "x86_64"
+
+TOOLCHAIN_BUILD_DEPS=(
+    coreutils gcc g++ build-essential make # generic build tools
+    wget # used for archive download
+    gnupg # used for archive signature verification
+    tar gzip bzip2 xz-utils unzip # used for archive unpacking
+    zlib1g-dev # zlib library used for all builds
+    libexpat1-dev libipt-dev libbabeltrace-dev liblzma-dev python3-dev texinfo # for gdb
+    libcurl4-openssl-dev # for cmake
+    libreadline-dev # for cmake and llvm
+    libffi-dev libxml2-dev # for llvm
+    curl # snappy
+    file
+    git # for thrift
+    libgmp-dev # for gdb
+    gperf # for proxygen
+    libssl-dev
+    libedit-dev libpcre3-dev automake bison # for swig
+)
+
+TOOLCHAIN_RUN_DEPS=(
+    make # generic build tools
+    tar gzip bzip2 xz-utils # used for archive unpacking
+    zlib1g # zlib library used for all builds
+    libexpat1 libipt2 libbabeltrace1 liblzma5 python3 # for gdb
+    libcurl4 # for cmake
+    libreadline8 # for cmake and llvm
+    libffi7 libxml2 # for llvm
+    libssl-dev # for libevent
+)
+
+MEMGRAPH_BUILD_DEPS=(
+    git # source code control
+    make pkg-config # build system
+    curl wget # for downloading libs
+    uuid-dev default-jre-headless # required by antlr
+    libreadline-dev # for memgraph console
+    libpython3-dev python3-dev # for query modules
+    libssl-dev
+    libseccomp-dev
+    netcat # tests are using nc to wait for memgraph
+    python3 python3-virtualenv python3-pip # for qa, macro_benchmark and stress tests
+    python3-yaml # for the configuration generator
+    libcurl4-openssl-dev # mg-requests
+    sbcl # for custom Lisp C++ preprocessing
+    doxygen graphviz # source documentation generators
+    mono-runtime mono-mcs zip unzip default-jdk-headless # for driver tests
+    dotnet-sdk-6.0 golang nodejs npm
+    autoconf # for jemalloc code generation
+    libtool  # for protobuf code generation
+)
+
+list() {
+    echo "$1"
+}
+
+check() {
+    check_all_dpkg "$1"
+}
+
+install() {
+    cd "$DIR"
+    apt update
+    # If GitHub Actions runner is installed, append LANG to the environment.
+    # Python related tests doesn't work the LANG export.
+    if [ -d "/home/gh/actions-runner" ]; then
+        echo "LANG=en_US.utf8" >> /home/gh/actions-runner/.env
+    else
+        echo "NOTE: export LANG=en_US.utf8"
+    fi
+    apt install -y wget
+    for pkg in $1; do
+        if [ "$pkg" == dotnet-sdk-6.0 ]; then
+            if ! dpkg -s dotnet-sdk-6.0 2>/dev/null >/dev/null; then
+                wget -nv https://packages.microsoft.com/config/ubuntu/22.04/packages-microsoft-prod.deb -O packages-microsoft-prod.deb
+                dpkg -i packages-microsoft-prod.deb
+                apt-get update
+                apt-get install -y apt-transport-https dotnet-sdk-6.0
+            fi
+            continue
+        fi
+        apt install -y "$pkg"
+    done
+}
+
+deps=$2"[*]"
+"$1" "${!deps}"

environment/toolchain/v4.sh

@@ -675,7 +675,7 @@ PROXYGEN_SHA256=5360a8ccdfb2f5a6c7b3eed331ec7ab0e2c792d579c6fff499c85c516c11fe14
 SNAPPY_SHA256=75c1fbb3d618dd3a0483bff0e26d0a92b495bbe5059c8b4f1c962b478b6e06e7
 SNAPPY_VERSION=1.1.9
 XZ_VERSION=5.2.5 # for LZMA
-ZLIB_VERSION=1.2.11
+ZLIB_VERSION=1.2.12
 ZSTD_VERSION=1.5.0
 WANGLE_SHA256=1002e9c32b6f4837f6a760016e3b3e22f3509880ef3eaad191c80dc92655f23f
 
@@ -1178,12 +1178,21 @@ popd
 
 # create toolchain archive
 if [ ! -f $NAME-binaries-$DISTRO.tar.gz ]; then
-    DISTRO_FULL_NAME=$DISTRO
-    if [ "$for_arm" = true ]; then
-        DISTRO_FULL_NAME="$DISTRO_FULL_NAME-aarch64"
+    DISTRO_FULL_NAME=${DISTRO}
+    if [[ "${DISTRO}" == centos* ]]; then
+        if [[ "$for_arm" = "true" ]]; then
+            DISTRO_FULL_NAME="$DISTRO_FULL_NAME-aarch64"
+        else
+            DISTRO_FULL_NAME="$DISTRO_FULL_NAME-x86_64"
+        fi
     else
-        DISTRO_FULL_NAME="$DISTRO_FULL_NAME-x86_64"
+        if [[ "$for_arm" = "true" ]]; then
+            DISTRO_FULL_NAME="$DISTRO_FULL_NAME-arm64"
+        else
+            DISTRO_FULL_NAME="$DISTRO_FULL_NAME-amd64"
+        fi
     fi
+
     tar --owner=root --group=root -cpvzf $NAME-binaries-$DISTRO_FULL_NAME.tar.gz -C /opt $NAME
 fi
 

environment/util.sh

@@ -5,10 +5,28 @@ operating_system() {
     sort | cut -d '=' -f 2- | sed 's/"//g' | paste -s -d '-'
 }
 
+check_operating_system() {
+    if [ "$(operating_system)" != "$1" ]; then
+        echo "Not the right operating system!"
+        exit 1
+    else
+        echo "The right operating system."
+    fi
+}
+
 architecture() {
     uname -m
 }
 
+check_architecture() {
+    if [ "$(architecture)" != "$1" ]; then
+        echo "Not the right architecture!"
+        exit 1
+    else
+        echo "The right architecture."
+    fi
+}
+
 check_all_yum() {
     local missing=""
     for pkg in $1; do

init

@@ -24,7 +24,7 @@ function setup_virtualenv () {
     fi
 
     # create new virtualenv
-    virtualenv -p python3 ve3 || exit 1
+    python3 -m virtualenv -p python3 ve3 || exit 1
     source ve3/bin/activate
     pip --timeout 1000 install -r requirements.txt || exit 1
     deactivate
@@ -66,7 +66,7 @@ fi
 
 DISTRO=$(operating_system)
 ARCHITECTURE=$(architecture)
-if [ "${ARCHITECTURE}" = "arm64" ]; then
+if [ "${ARCHITECTURE}" = "arm64" ] || [ "${ARCHITECTURE}" = "aarch64" ]; then
   OS_SCRIPT=$DIR/environment/os/$DISTRO-arm.sh
 else
   OS_SCRIPT=$DIR/environment/os/$DISTRO.sh
@@ -135,3 +135,10 @@ for hook in $(find $DIR/.githooks -type f -printf "%f\n"); do
   ln -s -f "$DIR/.githooks/$hook" "$DIR/.git/hooks/$hook"
   echo "Added $hook hook"
 done;
+
+# Install precommit hook
+python3 -m pip install pre-commit
+python3 -m pre_commit install
+
+# Link `include/mgp.py` with `release/mgp/mgp.py`
+ln -v -f include/mgp.py release/mgp/mgp.py

libs/.gitignore

@@ -4,5 +4,5 @@
 !cleanup.sh
 !CMakeLists.txt
 !__main.cpp
-!jemalloc.cmake
 !pulsar.patch
+!antlr4.10.1.patch

libs/CMakeLists.txt

@@ -106,6 +106,7 @@ import_external_library(antlr4 STATIC
              -DWITH_LIBCXX=OFF # because of debian bug
              -DCMAKE_SKIP_INSTALL_ALL_DEPENDENCY=true
              -DCMAKE_CXX_STANDARD=20
+             -DANTLR_BUILD_CPP_TESTS=OFF
   BUILD_COMMAND $(MAKE) antlr4_static
   INSTALL_COMMAND $(MAKE) install)
 

libs/antlr4.10.1.patch

@@ -0,0 +1,13 @@
+diff --git a/runtime/Cpp/runtime/CMakeLists.txt b/runtime/Cpp/runtime/CMakeLists.txt
+index baf46cac9..2e7756de8 100644
+--- a/runtime/Cpp/runtime/CMakeLists.txt
++++ b/runtime/Cpp/runtime/CMakeLists.txt
+@@ -134,7 +134,7 @@ set_target_properties(antlr4_static
+                                  ARCHIVE_OUTPUT_DIRECTORY ${LIB_OUTPUT_DIR}
+                                  COMPILE_FLAGS "${disabled_compile_warnings} ${extra_static_compile_flags}")
+
+-install(TARGETS antlr4_shared
++install(TARGETS antlr4_shared OPTIONAL
+         EXPORT antlr4-targets
+         ARCHIVE DESTINATION ${CMAKE_INSTALL_LIBDIR}
+         LIBRARY DESTINATION ${CMAKE_INSTALL_LIBDIR}

libs/antlr4.patch

@@ -1,43 +0,0 @@
-diff --git a/runtime/Cpp/runtime/CMakeLists.txt b/runtime/Cpp/runtime/CMakeLists.txt
-index a8503bb..11362cf 100644
---- a/runtime/Cpp/runtime/CMakeLists.txt
-+++ b/runtime/Cpp/runtime/CMakeLists.txt
-@@ -5,8 +5,8 @@ set(THIRDPARTY_DIR ${CMAKE_BINARY_DIR}/runtime/thirdparty)
- set(UTFCPP_DIR ${THIRDPARTY_DIR}/utfcpp)
- ExternalProject_Add(
-   utfcpp
--  GIT_REPOSITORY        "git://github.com/nemtrif/utfcpp"
--  GIT_TAG               "v3.1.1"
-+  GIT_REPOSITORY        "https://github.com/nemtrif/utfcpp"
-+  GIT_TAG               "v3.2.1"
-   SOURCE_DIR            ${UTFCPP_DIR}
-   UPDATE_DISCONNECTED   1
-   CMAKE_ARGS            -DCMAKE_INSTALL_PREFIX=${UTFCPP_DIR}/install -Dgtest_force_shared_crt=ON
-@@ -118,7 +118,7 @@ set_target_properties(antlr4_static
-                                  ARCHIVE_OUTPUT_DIRECTORY ${LIB_OUTPUT_DIR}
-                                  COMPILE_FLAGS "${disabled_compile_warnings} ${extra_static_compile_flags}")
- 
--install(TARGETS antlr4_shared
-+install(TARGETS antlr4_shared OPTIONAL
-         DESTINATION lib
-         EXPORT antlr4-targets)
- install(TARGETS antlr4_static
-diff --git a/runtime/Cpp/runtime/src/support/Any.h b/runtime/Cpp/runtime/src/support/Any.h
-index 468db98..65a473b 100644
---- a/runtime/Cpp/runtime/src/support/Any.h
-+++ b/runtime/Cpp/runtime/src/support/Any.h
-@@ -122,12 +122,12 @@ private:
-     }
- 
-   private:
--    template<int N = 0, typename std::enable_if<N == N && std::is_nothrow_copy_constructible<T>::value, int>::type = 0>
-+    template<int N = 0, typename std::enable_if<N == N && std::is_copy_constructible<T>::value, int>::type = 0>
-     Base* clone() const {
-       return new Derived<T>(value);
-     }
- 
--    template<int N = 0, typename std::enable_if<N == N && !std::is_nothrow_copy_constructible<T>::value, int>::type = 0>
-+    template<int N = 0, typename std::enable_if<N == N && !std::is_copy_constructible<T>::value, int>::type = 0>
-     Base* clone() const {
-       return nullptr;
-     }

libs/jemalloc.cmake

@@ -1,55 +0,0 @@
-set(JEMALLOC_DIR "${LIB_DIR}/jemalloc")
-
-set(JEMALLOC_SRCS
-    ${JEMALLOC_DIR}/src/arena.c
-    ${JEMALLOC_DIR}/src/background_thread.c
-    ${JEMALLOC_DIR}/src/base.c
-    ${JEMALLOC_DIR}/src/bin.c
-    ${JEMALLOC_DIR}/src/bitmap.c
-    ${JEMALLOC_DIR}/src/ckh.c
-    ${JEMALLOC_DIR}/src/ctl.c
-    ${JEMALLOC_DIR}/src/div.c
-    ${JEMALLOC_DIR}/src/extent.c
-    ${JEMALLOC_DIR}/src/extent_dss.c
-    ${JEMALLOC_DIR}/src/extent_mmap.c
-    ${JEMALLOC_DIR}/src/hash.c
-    ${JEMALLOC_DIR}/src/hook.c
-    ${JEMALLOC_DIR}/src/jemalloc.c
-    ${JEMALLOC_DIR}/src/large.c
-    ${JEMALLOC_DIR}/src/log.c
-    ${JEMALLOC_DIR}/src/malloc_io.c
-    ${JEMALLOC_DIR}/src/mutex.c
-    ${JEMALLOC_DIR}/src/mutex_pool.c
-    ${JEMALLOC_DIR}/src/nstime.c
-    ${JEMALLOC_DIR}/src/pages.c
-    ${JEMALLOC_DIR}/src/prng.c
-    ${JEMALLOC_DIR}/src/prof.c
-    ${JEMALLOC_DIR}/src/rtree.c
-    ${JEMALLOC_DIR}/src/sc.c
-    ${JEMALLOC_DIR}/src/stats.c
-    ${JEMALLOC_DIR}/src/sz.c
-    ${JEMALLOC_DIR}/src/tcache.c
-    ${JEMALLOC_DIR}/src/test_hooks.c
-    ${JEMALLOC_DIR}/src/ticker.c
-    ${JEMALLOC_DIR}/src/tsd.c
-    ${JEMALLOC_DIR}/src/witness.c
-    ${JEMALLOC_DIR}/src/safety_check.c
-)
-
-add_library(jemalloc ${JEMALLOC_SRCS})
-target_include_directories(jemalloc PUBLIC "${JEMALLOC_DIR}/include")
-
-find_package(Threads REQUIRED)
-target_link_libraries(jemalloc PUBLIC Threads::Threads)
-
-target_compile_definitions(jemalloc PRIVATE -DJEMALLOC_NO_PRIVATE_NAMESPACE)
-
-if (CMAKE_BUILD_TYPE STREQUAL "DEBUG")
-    target_compile_definitions(jemalloc PRIVATE -DJEMALLOC_DEBUG=1 -DJEMALLOC_PROF=1)
-endif()
-
-target_compile_options(jemalloc PRIVATE -Wno-redundant-decls)
-# for RTLD_NEXT
-target_compile_definitions(jemalloc PRIVATE _GNU_SOURCE)
-
-set_property(TARGET jemalloc APPEND PROPERTY INTERFACE_COMPILE_DEFINITIONS USE_JEMALLOC=1)

libs/setup.sh

@@ -105,7 +105,7 @@ repo_clone_try_double () {
 # Download from primary_urls might fail because the cache is not installed.
 declare -A primary_urls=(
   ["antlr4-code"]="http://$local_cache_host/git/antlr4.git"
-  ["antlr4-generator"]="http://$local_cache_host/file/antlr-4.9.2-complete.jar"
+  ["antlr4-generator"]="http://$local_cache_host/file/antlr-4.10.1-complete.jar"
   ["cppitertools"]="http://$local_cache_host/git/cppitertools.git"
   ["rapidcheck"]="http://$local_cache_host/git/rapidcheck.git"
   ["gbenchmark"]="http://$local_cache_host/git/benchmark.git"
@@ -130,7 +130,7 @@ declare -A primary_urls=(
 # should fail.
 declare -A secondary_urls=(
   ["antlr4-code"]="https://github.com/antlr/antlr4.git"
-  ["antlr4-generator"]="http://www.antlr.org/download/antlr-4.9.2-complete.jar"
+  ["antlr4-generator"]="https://www.antlr.org/download/antlr-4.10.1-complete.jar"
   ["cppitertools"]="https://github.com/ryanhaining/cppitertools.git"
   ["rapidcheck"]="https://github.com/emil-e/rapidcheck.git"
   ["gbenchmark"]="https://github.com/google/benchmark.git"
@@ -152,10 +152,10 @@ declare -A secondary_urls=(
 # antlr
 file_get_try_double "${primary_urls[antlr4-generator]}" "${secondary_urls[antlr4-generator]}"
 
-antlr4_tag="4.9.2" # v4.9.2
+antlr4_tag="4.10.1" # v4.10.1
 repo_clone_try_double "${primary_urls[antlr4-code]}" "${secondary_urls[antlr4-code]}" "antlr4" "$antlr4_tag" true
 pushd antlr4
-git apply ../antlr4.patch
+git apply ../antlr4.10.1.patch
 popd
 
 # cppitertools v2.0 2019-12-23
@@ -199,7 +199,7 @@ git apply ../rocksdb.patch
 popd
 
 # mgclient
-mgclient_tag="96e95c6845463cbe88948392be58d26da0d5ffd3" # (2022-02-08)
+mgclient_tag="v1.4.0" # (2022-06-14)
 repo_clone_try_double "${primary_urls[mgclient]}" "${secondary_urls[mgclient]}" "mgclient" "$mgclient_tag"
 sed -i 's/\${CMAKE_INSTALL_LIBDIR}/lib/' mgclient/src/CMakeLists.txt
 

release/CMakeLists.txt

@@ -41,7 +41,7 @@ set(CPACK_DEBIAN_PACKAGE_DESCRIPTION "${CPACK_PACKAGE_DESCRIPTION_SUMMARY}
  applications driver by real-time connected data.")
 # Add `openssl` package to dependencies list. Used to generate SSL certificates.
 # We also depend on `python3` because we embed it in Memgraph.
-set(CPACK_DEBIAN_PACKAGE_DEPENDS "openssl (>= 1.1.0), python3 (>= 3.5.0)")
+set(CPACK_DEBIAN_PACKAGE_DEPENDS "openssl (>= 1.1.0), python3 (>= 3.5.0), libstdc++6")
 
 # Setting arhitecture extension for rpm packages
 set(MG_ARCH_EXTENSION_RPM "noarch")
@@ -67,7 +67,7 @@ It aims to deliver developers the speed, simplicity and scale required to build
 the next generation of applications driver by real-time connected data.")
 # Add `openssl` package to dependencies list. Used to generate SSL certificates.
 # We also depend on `python3` because we embed it in Memgraph.
-set(CPACK_RPM_PACKAGE_REQUIRES "openssl >= 1.0.0, curl >= 7.29.0, python3 >= 3.5.0")
+set(CPACK_RPM_PACKAGE_REQUIRES "openssl >= 1.0.0, curl >= 7.29.0, python3 >= 3.5.0, libstdc >= 6, logrotate")
 
 # All variables must be set before including.
 include(CPack)

release/docker/memgraph_rpm.dockerfile

@@ -1,31 +0,0 @@
-FROM dokken/centos-stream-9
-# NOTE: If you change the base distro update release/package as well.
-
-ARG release
-
-RUN yum update && yum install -y \
-    openssl libcurl libseccomp python3 python3-pip \
-    --nobest --allowerasing \
-  && rm -rf /tmp/* \
-  && yum clean all
-
-RUN pip3 install networkx==2.4 numpy==1.21.4 scipy==1.7.3
-
-COPY ${release} /
-
-# Install memgraph package
-RUN rpm -i ${release}
-
-# Memgraph listens for Bolt Protocol on this port by default.
-EXPOSE 7687
-# Snapshots and logging volumes
-VOLUME /var/log/memgraph
-VOLUME /var/lib/memgraph
-# Configuration volume
-VOLUME /etc/memgraph
-
-USER memgraph
-WORKDIR /usr/lib/memgraph
-
-ENTRYPOINT ["/usr/lib/memgraph/memgraph"]
-CMD [""]

release/mgp/.gitignore

@@ -0,0 +1,3 @@
+.venv
+dist
+mgp.py

release/mgp/LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

release/mgp/README.md

@@ -0,0 +1,4 @@
+# mgp
+
+PyPi package used for type hinting when creating MAGE modules. The get started
+using MAGE repository checkout the repository here: https://github.com/memgraph/mage.

release/mgp/_mgp.py

@@ -0,0 +1,255 @@
+from typing import Any
+
+
+class MgpIterable:
+    def get() -> Any:
+        pass
+
+    def next() -> Any:
+        pass
+
+
+class Vertex:
+    def is_valid() -> bool:  # type: ignore
+        pass
+
+    def underlying_graph_is_mutable() -> bool:  # type: ignore
+        pass
+
+    def iter_properties() -> MgpIterable:  # type: ignore
+        pass
+
+    def get_property(self, property_name: str) -> "Property":  # type: ignore
+        pass
+
+    def set_property(self, property_name: str, value: Any) -> "Property":  # type: ignore
+        pass
+
+    def get_id() -> "VertexId":  # type: ignore
+        pass
+
+    def label_at(self, index: int) -> "Label":  # type: ignore
+        pass
+
+    def labels_count() -> int:  # type: ignore
+        pass
+
+    def add_label(self, label: Any):
+        pass
+
+    def remove_label(self, label: Any):
+        pass
+
+    def iter_in_edges() -> MgpIterable:  # type: ignore
+        pass
+
+    def iter_out_edges() -> MgpIterable:  # type: ignore
+        pass
+
+
+class Edge:
+    def is_valid() -> bool:  # type: ignore
+        pass
+
+    def underlying_graph_is_mutable() -> bool:  # type: ignore
+        pass
+
+    def iter_properties() -> MgpIterable:  # type: ignore
+        pass
+
+    def get_property(self, property_name: str) -> "Property":  # type: ignore
+        pass
+
+    def set_property(self, property_name: str, valuse: Any) -> "Property":  # type: ignore
+        pass
+
+    def get_type_name() -> str:  # type: ignore
+        pass
+
+    def get_id() -> "EdgeId":  # type: ignore
+        pass
+
+    def from_vertex() -> Vertex:  # type: ignore
+        pass
+
+    def to_vertex() -> Vertex:  # type: ignore
+        pass
+
+
+class Path:
+    def is_valid() -> bool:  # type: ignore
+        pass
+
+    @staticmethod
+    def make_with_start(vertex: Vertex) -> "Path":  # type: ignore
+        pass
+
+
+class Graph:
+    def is_valid() -> bool:  # type: ignore
+        pass
+
+
+class CypherType:
+    pass
+
+
+class Message:
+    def is_valid() -> bool:  # type: ignore
+        pass
+
+    def source_type() -> str:  # type: ignore
+        pass
+
+    def topic_name() -> str:  # type: ignore
+        pass
+
+    def key() -> bytes:  # type: ignore
+        pass
+
+    def timestamp() -> int:  # type: ignore
+        pass
+
+    def offset() -> int:  # type: ignore
+        pass
+
+    def payload() -> bytes:  # type: ignore
+        pass
+
+
+class Messages:
+    def is_valid() -> bool:  # type: ignore
+        pass
+
+    def message_at(self, id: int) -> Message:  # type: ignore
+        pass
+
+    def total_messages() -> int:  # type: ignore
+        pass
+
+
+class UnknownError(Exception):
+    pass
+
+
+class UnableToAllocateError(Exception):
+    pass
+
+
+class InsufficientBufferError(Exception):
+    pass
+
+
+class OutOfRangeError(Exception):
+    pass
+
+
+class LogicErrorError(Exception):
+
+    pass
+
+
+class DeletedObjectError(Exception):
+    pass
+
+
+class InvalidArgumentError(Exception):
+    pass
+
+
+class KeyAlreadyExistsError(Exception):
+    pass
+
+
+class ImmutableObjectError(Exception):
+    pass
+
+
+class ValueConversionError(Exception):
+    pass
+
+
+class SerializationError(Exception):
+    pass
+
+
+def type_nullable(elem: Any):
+    pass
+
+
+def type_list(elem: Any):
+    pass
+
+
+def type_bool():
+    pass
+
+
+def type_string():
+    pass
+
+
+def type_int():
+    pass
+
+
+def type_float():
+    pass
+
+
+def type_number():
+    pass
+
+
+def type_map():
+    pass
+
+
+def type_node():
+    pass
+
+
+def type_relationship():
+    pass
+
+
+def type_path():
+    pass
+
+
+def type_date():
+    pass
+
+
+def type_local_time():
+    pass
+
+
+def type_local_date_time():
+    pass
+
+
+def type_duration():
+    pass
+
+
+def type_any():
+    pass
+
+
+class _MODULE:
+    @staticmethod
+    def add_read_procedure(wrapper):
+        pass
+
+    @staticmethod
+    def add_write_procedure(wrapper):
+        pass
+
+    @staticmethod
+    def add_transformation(wrapper):
+        pass
+
+    @staticmethod
+    def add_function(wrapper):
+        pass

release/mgp/instructions.md

@@ -0,0 +1,22 @@
+# How to publish new versions
+## Prerequisites
+1. Installed poetry
+```
+pip install poetry
+```
+2. Set up [API tokens](https://pypi.org/help/#apitoken)
+3. Be a collaborator on [pypi](https://pypi.org/project/mgp/)
+
+## Making changes
+1. Make changes to the package
+2. Bump version in `pyproject.tml`
+3. `poetry build`
+4. `poetry publish`
+
+## Why is this not automatized?
+
+Because someone always has to manually bump up the version in `pyproject.toml`
+
+## Why does `_mgp.py` exists?
+Because we are mocking here all the types that are created by Memgraph
+in order to fix typing errors in `mgp.py`.

release/mgp/pyproject.toml

@@ -0,0 +1,23 @@
+[tool.poetry]
+name = "mgp"
+version = "1.0.0"
+description = "Memgraph's module for developing MAGE modules. Used only for type hinting!"
+authors = [
+    "MasterMedo <mislav.vuletic@gmail.com>",
+    "jbajic <jure.bajic@memgraph.io>",
+    "katarinasupe <katarina.supe@memgraph.io>",
+    "antejavor <ante.javor@memgraph.io>",
+    "antaljanosbenjamin <benjamin.antal@memgraph.io>",
+]
+license = "Apache-2.0"
+readme = "README.md"
+include = ["mgp.py", "_mgp.py"]
+
+[tool.poetry.dependencies]
+python = "^3.7"
+
+[tool.poetry.dev-dependencies]
+
+[build-system]
+requires = ["poetry-core>=1.0.0"]
+build-backend = "poetry.core.masonry.api"

release/package/centos-7/Dockerfile

@@ -7,8 +7,8 @@ RUN yum -y update \
 # Do NOT be smart here and clean the cache because the container is used in the
 # stateful context.
 
-RUN wget -q https://s3-eu-west-1.amazonaws.com/deps.memgraph.io/${TOOLCHAIN_VERSION}/${TOOLCHAIN_VERSION}-binaries-centos-7.tar.gz \
-    -O ${TOOLCHAIN_VERSION}-binaries-centos-7.tar.gz \
-    && tar xzvf ${TOOLCHAIN_VERSION}-binaries-centos-7.tar.gz -C /opt
+RUN wget -q https://s3-eu-west-1.amazonaws.com/deps.memgraph.io/${TOOLCHAIN_VERSION}/${TOOLCHAIN_VERSION}-binaries-centos-7-x86_64.tar.gz \
+    -O ${TOOLCHAIN_VERSION}-binaries-centos-7-x86_64.tar.gz \
+    && tar xzvf ${TOOLCHAIN_VERSION}-binaries-centos-7-x86_64.tar.gz -C /opt
 
 ENTRYPOINT ["sleep", "infinity"]

release/package/centos-8/Dockerfile

@@ -1,14 +0,0 @@
-FROM centos:8
-
-ARG TOOLCHAIN_VERSION
-
-RUN dnf -y update \
-    && dnf install -y wget git
-# Do NOT be smart here and clean the cache because the container is used in the
-# stateful context.
-
-RUN wget -q https://s3-eu-west-1.amazonaws.com/deps.memgraph.io/${TOOLCHAIN_VERSION}/${TOOLCHAIN_VERSION}-binaries-centos-8.tar.gz \
-    -O ${TOOLCHAIN_VERSION}-binaries-centos-8.tar.gz \
-    && tar xzvf ${TOOLCHAIN_VERSION}-binaries-centos-8.tar.gz -C /opt
-
-ENTRYPOINT ["sleep", "infinity"]

release/package/centos-9/Dockerfile

@@ -0,0 +1,14 @@
+FROM quay.io/centos/centos:stream9
+
+ARG TOOLCHAIN_VERSION
+
+RUN yum -y update \
+    && yum install -y wget git
+# Do NOT be smart here and clean the cache because the container is used in the
+# stateful context.
+
+RUN wget -q https://s3-eu-west-1.amazonaws.com/deps.memgraph.io/${TOOLCHAIN_VERSION}/${TOOLCHAIN_VERSION}-binaries-centos-9-x86_64.tar.gz \
+    -O ${TOOLCHAIN_VERSION}-binaries-centos-9-x86_64.tar.gz \
+    && tar xzvf ${TOOLCHAIN_VERSION}-binaries-centos-9-x86_64.tar.gz -C /opt
+
+ENTRYPOINT ["sleep", "infinity"]

release/package/debian-10/Dockerfile

@@ -10,8 +10,8 @@ RUN apt update && apt install -y \
 # Do NOT be smart here and clean the cache because the container is used in the
 # stateful context.
 
-RUN wget -q https://s3-eu-west-1.amazonaws.com/deps.memgraph.io/${TOOLCHAIN_VERSION}/${TOOLCHAIN_VERSION}-binaries-debian-10.tar.gz \
-    -O ${TOOLCHAIN_VERSION}-binaries-debian-10.tar.gz \
-    && tar xzvf ${TOOLCHAIN_VERSION}-binaries-debian-10.tar.gz -C /opt
+RUN wget -q https://s3-eu-west-1.amazonaws.com/deps.memgraph.io/${TOOLCHAIN_VERSION}/${TOOLCHAIN_VERSION}-binaries-debian-10-amd64.tar.gz \
+    -O ${TOOLCHAIN_VERSION}-binaries-debian-10-amd64.tar.gz \
+    && tar xzvf ${TOOLCHAIN_VERSION}-binaries-debian-10-amd64.tar.gz -C /opt
 
 ENTRYPOINT ["sleep", "infinity"]

release/package/debian-11-arm/Dockerfile

@@ -0,0 +1,17 @@
+FROM debian:11
+
+ARG TOOLCHAIN_VERSION
+
+# Stops tzdata interactive configuration.
+ENV DEBIAN_FRONTEND=noninteractive
+
+RUN apt update && apt install -y \
+    ca-certificates wget git
+# Do NOT be smart here and clean the cache because the container is used in the
+# stateful context.
+
+RUN wget -q https://s3-eu-west-1.amazonaws.com/deps.memgraph.io/${TOOLCHAIN_VERSION}/${TOOLCHAIN_VERSION}-binaries-debian-11-arm64.tar.gz \
+    -O ${TOOLCHAIN_VERSION}-binaries-debian-11-arm64.tar.gz \
+    && tar xzvf ${TOOLCHAIN_VERSION}-binaries-debian-11-arm64.tar.gz -C /opt
+
+ENTRYPOINT ["sleep", "infinity"]

release/package/debian-11/Dockerfile

@@ -10,8 +10,8 @@ RUN apt update && apt install -y \
 # Do NOT be smart here and clean the cache because the container is used in the
 # stateful context.
 
-RUN wget -q https://s3-eu-west-1.amazonaws.com/deps.memgraph.io/${TOOLCHAIN_VERSION}/${TOOLCHAIN_VERSION}-binaries-debian-11.tar.gz \
-    -O ${TOOLCHAIN_VERSION}-binaries-debian-11.tar.gz \
-    && tar xzvf ${TOOLCHAIN_VERSION}-binaries-debian-11.tar.gz -C /opt
+RUN wget -q https://s3-eu-west-1.amazonaws.com/deps.memgraph.io/${TOOLCHAIN_VERSION}/${TOOLCHAIN_VERSION}-binaries-debian-11-amd64.tar.gz \
+    -O ${TOOLCHAIN_VERSION}-binaries-debian-11-amd64.tar.gz \
+    && tar xzvf ${TOOLCHAIN_VERSION}-binaries-debian-11-amd64.tar.gz -C /opt
 
 ENTRYPOINT ["sleep", "infinity"]

release/package/docker-compose.yml

@@ -4,10 +4,10 @@ services:
     build:
       context: centos-7
     container_name: "mgbuild_centos-7"
-  mgbuild_centos-8:
+  mgbuild_centos-9:
     build:
-      context: centos-8
-    container_name: "mgbuild_centos-8"
+      context: centos-9
+    container_name: "mgbuild_centos-9"
   mgbuild_debian-10:
     build:
       context: debian-10
@@ -24,3 +24,7 @@ services:
     build:
       context: ubuntu-20.04
     container_name: "mgbuild_ubuntu-20.04"
+  mgbuild_ubuntu-22.04:
+    build:
+      context: ubuntu-22.04
+    container_name: "mgbuild_ubuntu-22.04"

release/package/run.sh

@@ -3,7 +3,7 @@
 set -Eeuo pipefail
 
 SCRIPT_DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"
-SUPPORTED_OS=(centos-7 centos-8 debian-10 debian-11 ubuntu-18.04 ubuntu-20.04)
+SUPPORTED_OS=(centos-7 centos-9 debian-10 debian-11 ubuntu-18.04 ubuntu-20.04 ubuntu-22.04 debian-11-arm)
 PROJECT_ROOT="$SCRIPT_DIR/../.."
 TOOLCHAIN_VERSION="toolchain-v4"
 ACTIVATE_TOOLCHAIN="source /opt/${TOOLCHAIN_VERSION}/activate"
@@ -67,14 +67,18 @@ make_package () {
     # environment/os/{os}.sh does not come within the toolchain package. When
     # migrating to the next version of toolchain do that, and remove the
     # TOOLCHAIN_RUN_DEPS installation from here.
-    echo "Installing dependencies..."
+    echo "Installing dependencies using '/memgraph/environment/os/$os.sh' script..."
     docker exec "$build_container" bash -c "/memgraph/environment/os/$os.sh install TOOLCHAIN_RUN_DEPS"
     docker exec "$build_container" bash -c "/memgraph/environment/os/$os.sh install MEMGRAPH_BUILD_DEPS"
 
     echo "Building targeted package..."
     docker exec "$build_container" bash -c "cd /memgraph && $ACTIVATE_TOOLCHAIN && ./init"
     docker exec "$build_container" bash -c "cd $container_build_dir && rm -rf ./*"
-    docker exec "$build_container" bash -c "cd $container_build_dir && $ACTIVATE_TOOLCHAIN && cmake -DCMAKE_BUILD_TYPE=release $telemetry_id_override_flag .."
+    if [[ "$os" == "debian-11-arm" ]]; then
+        docker exec "$build_container" bash -c "cd $container_build_dir && $ACTIVATE_TOOLCHAIN && cmake -DCMAKE_BUILD_TYPE=release -DMG_ARCH="ARM64" $telemetry_id_override_flag .."
+    else
+        docker exec "$build_container" bash -c "cd $container_build_dir && $ACTIVATE_TOOLCHAIN && cmake -DCMAKE_BUILD_TYPE=release $telemetry_id_override_flag .."
+    fi
     # ' is used instead of " because we need to run make within the allowed
     # container resources.
     # shellcheck disable=SC2016

release/package/ubuntu-18.04/Dockerfile

@@ -10,8 +10,8 @@ RUN apt update && apt install -y \
 # Do NOT be smart here and clean the cache because the container is used in the
 # stateful context.
 
-RUN wget -q https://s3-eu-west-1.amazonaws.com/deps.memgraph.io/${TOOLCHAIN_VERSION}/${TOOLCHAIN_VERSION}-binaries-ubuntu-18.04.tar.gz \
-    -O ${TOOLCHAIN_VERSION}-binaries-ubuntu-18.04.tar.gz \
-    && tar xzvf ${TOOLCHAIN_VERSION}-binaries-ubuntu-18.04.tar.gz -C /opt
+RUN wget -q https://s3-eu-west-1.amazonaws.com/deps.memgraph.io/${TOOLCHAIN_VERSION}/${TOOLCHAIN_VERSION}-binaries-ubuntu-18.04-amd64.tar.gz \
+    -O ${TOOLCHAIN_VERSION}-binaries-ubuntu-18.04-amd64.tar.gz \
+    && tar xzvf ${TOOLCHAIN_VERSION}-binaries-ubuntu-18.04-amd64.tar.gz -C /opt
 
 ENTRYPOINT ["sleep", "infinity"]

release/package/ubuntu-20.04/Dockerfile

@@ -10,8 +10,8 @@ RUN apt update && apt install -y \
 # Do NOT be smart here and clean the cache because the container is used in the
 # stateful context.
 
-RUN wget -q https://s3-eu-west-1.amazonaws.com/deps.memgraph.io/${TOOLCHAIN_VERSION}/${TOOLCHAIN_VERSION}-binaries-ubuntu-20.04.tar.gz \
-    -O ${TOOLCHAIN_VERSION}-binaries-ubuntu-20.04.tar.gz \
-    && tar xzvf ${TOOLCHAIN_VERSION}-binaries-ubuntu-20.04.tar.gz -C /opt
+RUN wget -q https://s3-eu-west-1.amazonaws.com/deps.memgraph.io/${TOOLCHAIN_VERSION}/${TOOLCHAIN_VERSION}-binaries-ubuntu-20.04-amd64.tar.gz \
+    -O ${TOOLCHAIN_VERSION}-binaries-ubuntu-20.04-amd64.tar.gz \
+    && tar xzvf ${TOOLCHAIN_VERSION}-binaries-ubuntu-20.04-amd64.tar.gz -C /opt
 
 ENTRYPOINT ["sleep", "infinity"]

release/package/ubuntu-22.04/Dockerfile

@@ -0,0 +1,17 @@
+FROM ubuntu:22.04
+
+ARG TOOLCHAIN_VERSION
+
+# Stops tzdata interactive configuration.
+ENV DEBIAN_FRONTEND=noninteractive
+
+RUN apt update && apt install -y \
+    ca-certificates wget git
+# Do NOT be smart here and clean the cache because the container is used in the
+# stateful context.
+
+RUN wget -q https://s3-eu-west-1.amazonaws.com/deps.memgraph.io/${TOOLCHAIN_VERSION}/${TOOLCHAIN_VERSION}-binaries-ubuntu-22.04-amd64.tar.gz \
+    -O ${TOOLCHAIN_VERSION}-binaries-ubuntu-22.04-amd64.tar.gz \
+    && tar xzvf ${TOOLCHAIN_VERSION}-binaries-ubuntu-22.04-amd64.tar.gz -C /opt
+
+ENTRYPOINT ["sleep", "infinity"]

src/CMakeLists.txt

@@ -5,16 +5,22 @@ add_subdirectory(lisp)
 add_subdirectory(utils)
 add_subdirectory(requests)
 add_subdirectory(io)
+add_subdirectory(io/simulator)
 add_subdirectory(kvstore)
 add_subdirectory(telemetry)
 add_subdirectory(communication)
 add_subdirectory(memory)
 add_subdirectory(storage/v2)
+add_subdirectory(storage/v3)
 add_subdirectory(integrations)
 add_subdirectory(query)
+add_subdirectory(query/v2)
 add_subdirectory(slk)
 add_subdirectory(rpc)
 add_subdirectory(auth)
+add_subdirectory(parser)
+add_subdirectory(expr)
+add_subdirectory(coordinator)
 
 if (MG_ENTERPRISE)
   add_subdirectory(audit)

src/auth/models.cpp

@@ -37,7 +37,7 @@ const std::vector<Permission> kPermissionsAll = {
     Permission::CONSTRAINT, Permission::DUMP,      Permission::AUTH,        Permission::REPLICATION,
     Permission::DURABILITY, Permission::READ_FILE, Permission::FREE_MEMORY, Permission::TRIGGER,
     Permission::CONFIG,     Permission::STREAM,    Permission::MODULE_READ, Permission::MODULE_WRITE,
-    Permission::WEBSOCKET};
+    Permission::WEBSOCKET,  Permission::SCHEMA};
 }  // namespace
 
 std::string PermissionToString(Permission permission) {
@@ -84,6 +84,8 @@ std::string PermissionToString(Permission permission) {
       return "MODULE_WRITE";
     case Permission::WEBSOCKET:
       return "WEBSOCKET";
+    case Permission::SCHEMA:
+      return "SCHEMA";
   }
 }
 

src/auth/models.hpp

@@ -38,7 +38,8 @@ enum class Permission : uint64_t {
   STREAM       = 1U << 17U,
   MODULE_READ  = 1U << 18U,
   MODULE_WRITE = 1U << 19U,
-  WEBSOCKET    = 1U << 20U
+  WEBSOCKET    = 1U << 20U,
+  SCHEMA       = 1U << 21U
 };
 // clang-format on
 

src/common/types.hpp

@@ -0,0 +1,19 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include <cstdint>
+
+namespace memgraph::common {
+enum class SchemaType : uint8_t { BOOL, INT, STRING, DATE, LOCALTIME, LOCALDATETIME, DURATION };
+
+}  // namespace memgraph::common

src/communication/v2/session.hpp

@@ -105,9 +105,16 @@ class WebsocketSession : public std::enable_shared_from_this<WebsocketSession<TS
     boost::asio::socket_base::keep_alive option(true);
 
     // Set a decorator to change the Server of the handshake
-    ws_.set_option(boost::beast::websocket::stream_base::decorator([](boost::beast::websocket::response_type &res) {
+    ws_.set_option(boost::beast::websocket::stream_base::decorator([&req](boost::beast::websocket::response_type &res) {
       res.set(boost::beast::http::field::server, std::string("Memgraph Bolt WS"));
-      res.set(boost::beast::http::field::sec_websocket_protocol, "binary");
+
+      // We need to do this to support WASM clients, which explicitly send this flag
+      // in their upgrade request
+      // Neo4j client breaks when this flag is sent
+      if (const auto secondary_protocol = req.base().find(boost::beast::http::field::sec_websocket_protocol);
+          secondary_protocol != res.base().end() && secondary_protocol->value() == "binary") {
+        res.set(boost::beast::http::field::sec_websocket_protocol, "binary");
+      }
     }));
     ws_.binary(true);
 
@@ -162,7 +169,7 @@ class WebsocketSession : public std::enable_shared_from_this<WebsocketSession<TS
         boost::asio::bind_executor(strand_, std::bind_front(&WebsocketSession::OnRead, shared_from_this())));
   }
 
-  void OnRead(const boost::system::error_code &ec, [[maybe_unused]] const size_t bytes_transferred) {
+  void OnRead(const boost::system::error_code &ec, const size_t bytes_transferred) {
     // This indicates that the WebsocketSession was closed
     if (ec == boost::beast::websocket::error::closed) {
       return;
@@ -246,11 +253,7 @@ class Session final : public std::enable_shared_from_this<Session<TSession, TSes
   Session(Session &&) = delete;
   Session &operator=(const Session &) = delete;
   Session &operator=(Session &&) = delete;
-  ~Session() {
-    if (IsConnected()) {
-      spdlog::error("Session: Destructor called while execution is active");
-    }
-  }
+  ~Session() = default;
 
   bool Start() {
     if (execution_active_) {
@@ -326,7 +329,8 @@ class Session final : public std::enable_shared_from_this<Session<TSession, TSes
       socket.lowest_layer().non_blocking(false);
     });
     timeout_timer_.expires_at(boost::asio::steady_timer::time_point::max());
-    spdlog::info("Accepted a connection from {}:", service_name_, remote_endpoint_.address(), remote_endpoint_.port());
+    spdlog::info("Accepted a connection from {}: {}:{}", service_name_, remote_endpoint_.address(),
+                 remote_endpoint_.port());
   }
 
   void DoRead() {
@@ -400,7 +404,6 @@ class Session final : public std::enable_shared_from_this<Session<TSession, TSes
     if (ec == boost::asio::error::operation_aborted) {
       return;
     }
-    execution_active_ = false;
 
     if (ec == boost::asio::error::eof) {
       spdlog::info("Session closed by peer");

src/coordinator/CMakeLists.txt

@@ -0,0 +1,9 @@
+set(coordinator_src_files
+    coordinator.cpp
+    shard_map.cpp)
+
+find_package(fmt REQUIRED)
+find_package(Threads REQUIRED)
+
+add_library(mg-coordinator STATIC ${coordinator_src_files})
+target_link_libraries(mg-coordinator stdc++fs Threads::Threads fmt::fmt mg-utils mg-storage-v3)

src/coordinator/coordinator.cpp

@@ -0,0 +1,129 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#include <coordinator/coordinator.hpp>
+
+namespace memgraph::coordinator {
+
+CoordinatorWriteResponses Coordinator::ApplyWrite(HeartbeatRequest &&heartbeat_request) {
+  spdlog::info("Coordinator handling HeartbeatRequest");
+
+  // add this storage engine to any under-replicated shards that it is not already a part of
+
+  auto initializing_rsms_for_shard_manager =
+      shard_map_.AssignShards(heartbeat_request.from_storage_manager, heartbeat_request.initialized_rsms);
+
+  return HeartbeatResponse{
+      .shards_to_initialize = initializing_rsms_for_shard_manager,
+  };
+}
+
+CoordinatorWriteResponses Coordinator::ApplyWrite(HlcRequest &&hlc_request) {
+  HlcResponse res{};
+
+  auto hlc_shard_map = shard_map_.GetHlc();
+
+  MG_ASSERT(!(hlc_request.last_shard_map_version.logical_id > hlc_shard_map.logical_id));
+
+  res.new_hlc = Hlc{
+      .logical_id = ++highest_allocated_timestamp_,
+      // TODO(tyler) probably pass some more context to the Coordinator here
+      // so that we can use our wall clock and enforce monotonicity.
+      // .coordinator_wall_clock = io_.Now(),
+  };
+
+  // Allways return fresher shard_map for now.
+  res.fresher_shard_map = std::make_optional(shard_map_);
+
+  return res;
+}
+
+CoordinatorWriteResponses Coordinator::ApplyWrite(AllocateEdgeIdBatchRequest &&ahr) {
+  AllocateEdgeIdBatchResponse res{};
+
+  uint64_t low = highest_allocated_edge_id_;
+
+  highest_allocated_edge_id_ += ahr.batch_size;
+
+  uint64_t high = highest_allocated_edge_id_;
+
+  res.low = low;
+  res.high = high;
+
+  return res;
+}
+
+/// This splits the shard immediately beneath the provided
+/// split key, keeping the assigned peers identical for now,
+/// but letting them be gradually migrated over time.
+CoordinatorWriteResponses Coordinator::ApplyWrite(SplitShardRequest &&split_shard_request) {
+  SplitShardResponse res{};
+
+  if (split_shard_request.previous_shard_map_version != shard_map_.shard_map_version) {
+    res.success = false;
+  } else {
+    res.success = shard_map_.SplitShard(split_shard_request.previous_shard_map_version, split_shard_request.label_id,
+                                        split_shard_request.split_key);
+  }
+
+  return res;
+}
+
+/// This adds the provided storage engine to the standby storage engine pool,
+/// which can be used to rebalance storage over time.
+CoordinatorWriteResponses Coordinator::ApplyWrite(
+    RegisterStorageEngineRequest && /* register_storage_engine_request */) {
+  RegisterStorageEngineResponse res{};
+  // TODO
+
+  return res;
+}
+
+/// This begins the process of draining the provided storage engine from all raft
+/// clusters that it might be participating in.
+CoordinatorWriteResponses Coordinator::ApplyWrite(
+    DeregisterStorageEngineRequest && /* register_storage_engine_request */) {
+  DeregisterStorageEngineResponse res{};
+  // TODO
+
+  return res;
+}
+
+CoordinatorWriteResponses Coordinator::ApplyWrite(InitializeLabelRequest &&initialize_label_request) {
+  InitializeLabelResponse res{};
+
+  std::optional<LabelId> new_label_id = shard_map_.InitializeNewLabel(
+      initialize_label_request.label_name, initialize_label_request.schema, initialize_label_request.replication_factor,
+      initialize_label_request.last_shard_map_version);
+
+  if (new_label_id) {
+    res.new_label_id = new_label_id.value();
+    res.fresher_shard_map = std::nullopt;
+    res.success = true;
+  } else {
+    res.fresher_shard_map = shard_map_;
+    res.success = false;
+  }
+
+  return res;
+}
+
+CoordinatorWriteResponses Coordinator::ApplyWrite(AllocatePropertyIdsRequest &&allocate_property_ids_request) {
+  AllocatePropertyIdsResponse res{};
+
+  auto property_ids = shard_map_.AllocatePropertyIds(allocate_property_ids_request.property_names);
+
+  res.property_ids = property_ids;
+
+  return res;
+}
+
+}  // namespace memgraph::coordinator

src/coordinator/coordinator.hpp

@@ -0,0 +1,196 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include <optional>
+#include <set>
+#include <string>
+#include <unordered_set>
+#include <variant>
+#include <vector>
+
+#include <boost/uuid/uuid.hpp>
+
+#include <coordinator/hybrid_logical_clock.hpp>
+#include <coordinator/shard_map.hpp>
+#include <io/simulator/simulator.hpp>
+#include <io/time.hpp>
+#include <io/transport.hpp>
+#include <storage/v3/id_types.hpp>
+#include <storage/v3/schemas.hpp>
+
+namespace memgraph::coordinator {
+
+using memgraph::io::Address;
+using memgraph::storage::v3::LabelId;
+using memgraph::storage::v3::PropertyId;
+using memgraph::storage::v3::SchemaProperty;
+using SimT = memgraph::io::simulator::SimulatorTransport;
+using PrimaryKey = std::vector<PropertyValue>;
+
+struct HlcRequest {
+  Hlc last_shard_map_version;
+};
+
+struct HlcResponse {
+  Hlc new_hlc;
+  std::optional<ShardMap> fresher_shard_map;
+};
+
+struct GetShardMapRequest {
+  // No state
+};
+
+struct GetShardMapResponse {
+  ShardMap shard_map;
+};
+
+struct AllocateHlcBatchRequest {
+  Hlc low;
+  Hlc high;
+};
+
+struct AllocateHlcBatchResponse {
+  bool success;
+  Hlc low;
+  Hlc high;
+};
+
+struct AllocateEdgeIdBatchRequest {
+  size_t batch_size;
+};
+
+struct AllocateEdgeIdBatchResponse {
+  uint64_t low;
+  uint64_t high;
+};
+
+struct AllocatePropertyIdsRequest {
+  std::vector<std::string> property_names;
+};
+
+struct AllocatePropertyIdsResponse {
+  std::map<std::string, PropertyId> property_ids;
+};
+
+struct SplitShardRequest {
+  Hlc previous_shard_map_version;
+  LabelId label_id;
+  PrimaryKey split_key;
+};
+
+struct SplitShardResponse {
+  bool success;
+};
+
+struct RegisterStorageEngineRequest {
+  Address address;
+};
+
+struct RegisterStorageEngineResponse {
+  bool success;
+};
+
+struct DeregisterStorageEngineRequest {
+  Address address;
+};
+
+struct DeregisterStorageEngineResponse {
+  bool success;
+};
+
+struct InitializeLabelRequest {
+  std::string label_name;
+  std::vector<SchemaProperty> schema;
+  size_t replication_factor;
+  Hlc last_shard_map_version;
+};
+
+struct InitializeLabelResponse {
+  bool success;
+  LabelId new_label_id;
+  std::optional<ShardMap> fresher_shard_map;
+};
+
+struct HeartbeatRequest {
+  Address from_storage_manager;
+  std::set<boost::uuids::uuid> initialized_rsms;
+};
+
+struct HeartbeatResponse {
+  std::vector<ShardToInitialize> shards_to_initialize;
+};
+
+using CoordinatorWriteRequests =
+    std::variant<HlcRequest, AllocateEdgeIdBatchRequest, SplitShardRequest, RegisterStorageEngineRequest,
+                 DeregisterStorageEngineRequest, InitializeLabelRequest, AllocatePropertyIdsRequest, HeartbeatRequest>;
+using CoordinatorWriteResponses = std::variant<HlcResponse, AllocateEdgeIdBatchResponse, SplitShardResponse,
+                                               RegisterStorageEngineResponse, DeregisterStorageEngineResponse,
+                                               InitializeLabelResponse, AllocatePropertyIdsResponse, HeartbeatResponse>;
+
+using CoordinatorReadRequests = std::variant<GetShardMapRequest>;
+using CoordinatorReadResponses = std::variant<GetShardMapResponse>;
+
+class Coordinator {
+ public:
+  explicit Coordinator(ShardMap sm) : shard_map_{std::move(sm)} {}
+
+  // NOLINTNEXTLINE(readability-convert-member-functions-to-static
+  CoordinatorReadResponses Read(CoordinatorReadRequests requests) {
+    return std::visit([&](auto &&request) { return HandleRead(std::forward<decltype(request)>(request)); },
+                      std::move(requests));  // NOLINT(hicpp-move-const-arg,performance-move-const-arg)
+  }
+
+  // NOLINTNEXTLINE(readability-convert-member-functions-to-static
+  CoordinatorWriteResponses Apply(CoordinatorWriteRequests requests) {
+    return std::visit([&](auto &&request) mutable { return ApplyWrite(std::forward<decltype(request)>(request)); },
+                      std::move(requests));
+  }
+
+ private:
+  ShardMap shard_map_;
+  uint64_t highest_allocated_timestamp_;
+
+  /// Query engines need to periodically request batches of unique edge IDs.
+  uint64_t highest_allocated_edge_id_;
+
+  CoordinatorReadResponses HandleRead(GetShardMapRequest && /* get_shard_map_request */) {
+    GetShardMapResponse res;
+    res.shard_map = shard_map_;
+    return res;
+  }
+
+  CoordinatorWriteResponses ApplyWrite(HeartbeatRequest &&heartbeat_request);
+
+  CoordinatorWriteResponses ApplyWrite(HlcRequest &&hlc_request);
+
+  CoordinatorWriteResponses ApplyWrite(AllocateEdgeIdBatchRequest &&ahr);
+
+  /// This splits the shard immediately beneath the provided
+  /// split key, keeping the assigned peers identical for now,
+  /// but letting them be gradually migrated over time.
+  CoordinatorWriteResponses ApplyWrite(SplitShardRequest &&split_shard_request);
+
+  /// This adds the provided storage engine to the standby storage engine pool,
+  /// which can be used to rebalance storage over time.
+  static CoordinatorWriteResponses ApplyWrite(RegisterStorageEngineRequest && /* register_storage_engine_request */);
+
+  /// This begins the process of draining the provided storage engine from all raft
+  /// clusters that it might be participating in.
+  static CoordinatorWriteResponses ApplyWrite(DeregisterStorageEngineRequest && /* register_storage_engine_request */);
+
+  CoordinatorWriteResponses ApplyWrite(InitializeLabelRequest &&initialize_label_request);
+
+  CoordinatorWriteResponses ApplyWrite(AllocatePropertyIdsRequest &&allocate_property_ids_request);
+};
+
+}  // namespace memgraph::coordinator

src/coordinator/coordinator_client.hpp

@@ -0,0 +1,24 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include "coordinator/coordinator.hpp"
+#include "io/rsm/rsm_client.hpp"
+
+namespace memgraph::coordinator {
+
+using memgraph::io::rsm::RsmClient;
+
+template <typename IoImpl>
+using CoordinatorClient = RsmClient<IoImpl, CoordinatorWriteRequests, CoordinatorWriteResponses,
+                                    CoordinatorReadRequests, CoordinatorReadResponses>;
+}  // namespace memgraph::coordinator

src/coordinator/coordinator_rsm.hpp

@@ -0,0 +1,23 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include "coordinator/coordinator.hpp"
+#include "io/rsm/raft.hpp"
+
+namespace memgraph::coordinator {
+
+template <typename IoImpl>
+using CoordinatorRsm = memgraph::io::rsm::Raft<IoImpl, Coordinator, CoordinatorWriteRequests, CoordinatorWriteResponses,
+                                               CoordinatorReadRequests, CoordinatorReadResponses>;
+
+}  // namespace memgraph::coordinator

src/coordinator/hybrid_logical_clock.hpp

@@ -0,0 +1,36 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include <compare>
+
+#include "io/time.hpp"
+
+namespace memgraph::coordinator {
+
+using Time = memgraph::io::Time;
+
+/// Hybrid-logical clock
+struct Hlc {
+  uint64_t logical_id;
+  Time coordinator_wall_clock;
+
+  auto operator<=>(const Hlc &other) const { return logical_id <=> other.logical_id; }
+
+  bool operator==(const Hlc &other) const = default;
+  bool operator<(const Hlc &other) const = default;
+  bool operator==(const uint64_t other) const { return logical_id == other; }
+  bool operator<(const uint64_t other) const { return logical_id < other; }
+  bool operator>=(const uint64_t other) const { return logical_id >= other; }
+};
+
+}  // namespace memgraph::coordinator

src/coordinator/shard_map.cpp

@@ -0,0 +1,90 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#include "coordinator/shard_map.hpp"
+#include "storage/v3/temporal.hpp"
+
+namespace memgraph::coordinator {
+
+using memgraph::common::SchemaType;
+using memgraph::storage::v3::TemporalData;
+using memgraph::storage::v3::TemporalType;
+
+PrimaryKey SchemaToMinKey(const std::vector<SchemaProperty> &schema) {
+  PrimaryKey ret{};
+
+  const int64_t min_int = std::numeric_limits<int64_t>::min();
+
+  const TemporalData date{TemporalType::Date, min_int};
+  const TemporalData local_time{TemporalType::LocalTime, min_int};
+  const TemporalData local_date_time{TemporalType::LocalDateTime, min_int};
+  const TemporalData duration{TemporalType::Duration, min_int};
+
+  for (const auto &schema_property : schema) {
+    switch (schema_property.type) {
+      case SchemaType::BOOL:
+        ret.emplace_back(PropertyValue(false));
+        break;
+      case SchemaType::INT:
+        ret.emplace_back(PropertyValue(min_int));
+        break;
+      case SchemaType::STRING:
+        ret.emplace_back(PropertyValue(""));
+        break;
+      case SchemaType::DATE:
+        ret.emplace_back(PropertyValue(date));
+        break;
+      case SchemaType::LOCALTIME:
+        ret.emplace_back(PropertyValue(local_time));
+        break;
+      case SchemaType::LOCALDATETIME:
+        ret.emplace_back(PropertyValue(local_date_time));
+        break;
+      case SchemaType::DURATION:
+        ret.emplace_back(PropertyValue(duration));
+        break;
+    }
+  }
+
+  return ret;
+}
+
+std::optional<LabelId> ShardMap::InitializeNewLabel(std::string label_name, std::vector<SchemaProperty> schema,
+                                                    size_t replication_factor, Hlc last_shard_map_version) {
+  if (shard_map_version != last_shard_map_version || labels.contains(label_name)) {
+    return std::nullopt;
+  }
+
+  const LabelId label_id = LabelId::FromUint(++max_label_id);
+
+  labels.emplace(std::move(label_name), label_id);
+
+  PrimaryKey initial_key = SchemaToMinKey(schema);
+  Shard empty_shard = {};
+
+  Shards shards = {
+      {initial_key, empty_shard},
+  };
+
+  LabelSpace label_space{
+      .schema = std::move(schema),
+      .shards = shards,
+      .replication_factor = replication_factor,
+  };
+
+  label_spaces.emplace(label_id, label_space);
+
+  IncrementShardMapVersion();
+
+  return label_id;
+}
+
+}  // namespace memgraph::coordinator

src/coordinator/shard_map.hpp

@@ -0,0 +1,276 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include <limits>
+#include <map>
+#include <set>
+#include <vector>
+
+#include <boost/uuid/uuid.hpp>
+#include <boost/uuid/uuid_generators.hpp>
+
+#include "common/types.hpp"
+#include "coordinator/hybrid_logical_clock.hpp"
+#include "io/address.hpp"
+#include "storage/v3/config.hpp"
+#include "storage/v3/id_types.hpp"
+#include "storage/v3/property_value.hpp"
+#include "storage/v3/schemas.hpp"
+#include "storage/v3/temporal.hpp"
+
+namespace memgraph::coordinator {
+
+using memgraph::io::Address;
+using memgraph::storage::v3::Config;
+using memgraph::storage::v3::LabelId;
+using memgraph::storage::v3::PropertyId;
+using memgraph::storage::v3::PropertyValue;
+using memgraph::storage::v3::SchemaProperty;
+
+enum class Status : uint8_t {
+  CONSENSUS_PARTICIPANT,
+  INITIALIZING,
+  // TODO(tyler) this will possibly have more states,
+  // depending on the reconfiguration protocol that we
+  // implement.
+};
+
+struct AddressAndStatus {
+  memgraph::io::Address address;
+  Status status;
+  friend bool operator<(const AddressAndStatus &lhs, const AddressAndStatus &rhs) { return lhs.address < rhs.address; }
+};
+
+using PrimaryKey = std::vector<PropertyValue>;
+using Shard = std::vector<AddressAndStatus>;
+using Shards = std::map<PrimaryKey, Shard>;
+using LabelName = std::string;
+using PropertyName = std::string;
+using PropertyMap = std::map<PropertyName, PropertyId>;
+
+struct ShardToInitialize {
+  boost::uuids::uuid uuid;
+  LabelId label_id;
+  PrimaryKey min_key;
+  std::optional<PrimaryKey> max_key;
+  Config config;
+};
+
+PrimaryKey SchemaToMinKey(const std::vector<SchemaProperty> &schema);
+
+struct LabelSpace {
+  std::vector<SchemaProperty> schema;
+  std::map<PrimaryKey, Shard> shards;
+  size_t replication_factor;
+};
+
+struct ShardMap {
+  Hlc shard_map_version;
+  uint64_t max_property_id;
+  std::map<PropertyName, PropertyId> properties;
+  uint64_t max_label_id;
+  std::map<LabelName, LabelId> labels;
+  std::map<LabelId, LabelSpace> label_spaces;
+  std::map<LabelId, std::vector<SchemaProperty>> schemas;
+
+  Shards GetShards(const LabelName &label) {
+    const auto id = labels.at(label);
+    auto &shards = label_spaces.at(id).shards;
+    return shards;
+  }
+
+  // TODO(gabor) later we will want to update the wallclock time with
+  // the given Io<impl>'s time as well
+  Hlc IncrementShardMapVersion() noexcept {
+    ++shard_map_version.logical_id;
+    return shard_map_version;
+  }
+
+  Hlc GetHlc() const noexcept { return shard_map_version; }
+
+  // Returns the shard UUIDs that have been assigned but not yet acknowledged for this storage manager
+  std::vector<ShardToInitialize> AssignShards(Address storage_manager, std::set<boost::uuids::uuid> initialized) {
+    std::vector<ShardToInitialize> ret{};
+
+    bool mutated = false;
+
+    for (auto &[label_id, label_space] : label_spaces) {
+      for (auto &[low_key, shard] : label_space.shards) {
+        // TODO(tyler) avoid these triple-nested loops by having the heartbeat include better info
+        bool machine_contains_shard = false;
+
+        for (auto &aas : shard) {
+          if (initialized.contains(aas.address.unique_id)) {
+            spdlog::info("marking shard as full consensus participant: {}", aas.address.unique_id);
+            aas.status = Status::CONSENSUS_PARTICIPANT;
+            machine_contains_shard = true;
+          } else {
+            const bool same_machine = aas.address.last_known_ip == storage_manager.last_known_ip &&
+                                      aas.address.last_known_port == storage_manager.last_known_port;
+            if (same_machine) {
+              machine_contains_shard = true;
+              ret.push_back(ShardToInitialize{
+                  .uuid = aas.address.unique_id,
+                  .label_id = label_id,
+                  .min_key = low_key,
+                  .max_key = std::nullopt,
+                  .config = Config{},
+              });
+            }
+          }
+        }
+
+        if (!machine_contains_shard && shard.size() < label_space.replication_factor) {
+          Address address = storage_manager;
+
+          // TODO(tyler) use deterministic UUID so that coordinators don't diverge here
+          address.unique_id = boost::uuids::uuid{boost::uuids::random_generator()()},
+
+          ret.push_back(ShardToInitialize{
+              .uuid = address.unique_id,
+              .label_id = label_id,
+              .min_key = low_key,
+              .max_key = std::nullopt,
+              .config = Config{},
+          });
+
+          AddressAndStatus aas = {
+              .address = address,
+              .status = Status::INITIALIZING,
+          };
+
+          shard.emplace_back(aas);
+        }
+      }
+    }
+
+    if (mutated) {
+      IncrementShardMapVersion();
+    }
+
+    return ret;
+  }
+
+  bool SplitShard(Hlc previous_shard_map_version, LabelId label_id, const PrimaryKey &key) {
+    if (previous_shard_map_version != shard_map_version) {
+      return false;
+    }
+
+    auto &label_space = label_spaces.at(label_id);
+    auto &shards_in_map = label_space.shards;
+
+    MG_ASSERT(!shards_in_map.empty());
+    MG_ASSERT(!shards_in_map.contains(key));
+    MG_ASSERT(label_spaces.contains(label_id));
+
+    // Finding the Shard that the new PrimaryKey should map to.
+    auto prev = std::prev(shards_in_map.upper_bound(key));
+    Shard duplicated_shard = prev->second;
+
+    // Apply the split
+    shards_in_map[key] = duplicated_shard;
+
+    return true;
+  }
+
+  std::optional<LabelId> InitializeNewLabel(std::string label_name, std::vector<SchemaProperty> schema,
+                                            size_t replication_factor, Hlc last_shard_map_version);
+
+  void AddServer(Address server_address) {
+    // Find a random place for the server to plug in
+  }
+
+  LabelId GetLabelId(const std::string &label) const { return labels.at(label); }
+
+  Shards GetShardsForRange(const LabelName &label_name, const PrimaryKey &start_key, const PrimaryKey &end_key) const {
+    MG_ASSERT(start_key <= end_key);
+    MG_ASSERT(labels.contains(label_name));
+
+    LabelId label_id = labels.at(label_name);
+
+    const auto &label_space = label_spaces.at(label_id);
+
+    const auto &shards_for_label = label_space.shards;
+
+    MG_ASSERT(shards_for_label.begin()->first <= start_key,
+              "the ShardMap must always contain a minimal key that is less than or equal to any requested key");
+
+    auto it = std::prev(shards_for_label.upper_bound(start_key));
+    const auto end_it = shards_for_label.upper_bound(end_key);
+
+    Shards shards{};
+
+    std::copy(it, end_it, std::inserter(shards, shards.end()));
+
+    return shards;
+  }
+
+  Shard GetShardForKey(const LabelName &label_name, const PrimaryKey &key) const {
+    MG_ASSERT(labels.contains(label_name));
+
+    LabelId label_id = labels.at(label_name);
+
+    const auto &label_space = label_spaces.at(label_id);
+
+    MG_ASSERT(label_space.shards.begin()->first <= key,
+              "the ShardMap must always contain a minimal key that is less than or equal to any requested key");
+
+    return std::prev(label_space.shards.upper_bound(key))->second;
+  }
+
+  Shard GetShardForKey(const LabelId &label_id, const PrimaryKey &key) const {
+    MG_ASSERT(label_spaces.contains(label_id));
+
+    const auto &label_space = label_spaces.at(label_id);
+
+    MG_ASSERT(label_space.shards.begin()->first <= key,
+              "the ShardMap must always contain a minimal key that is less than or equal to any requested key");
+
+    return std::prev(label_space.shards.upper_bound(key))->second;
+  }
+
+  PropertyMap AllocatePropertyIds(const std::vector<PropertyName> &new_properties) {
+    PropertyMap ret{};
+
+    bool mutated = false;
+
+    for (const auto &property_name : new_properties) {
+      if (properties.contains(property_name)) {
+        auto property_id = properties.at(property_name);
+        ret.emplace(property_name, property_id);
+      } else {
+        mutated = true;
+
+        const PropertyId property_id = PropertyId::FromUint(++max_property_id);
+        ret.emplace(property_name, property_id);
+        properties.emplace(property_name, property_id);
+      }
+    }
+
+    if (mutated) {
+      IncrementShardMapVersion();
+    }
+
+    return ret;
+  }
+
+  std::optional<PropertyId> GetPropertyId(const std::string &property_name) const {
+    if (properties.contains(property_name)) {
+      return properties.at(property_name);
+    }
+
+    return std::nullopt;
+  }
+};
+
+}  // namespace memgraph::coordinator

src/expr/CMakeLists.txt

@@ -0,0 +1,20 @@
+define_add_lcp(add_lcp_expr lcp_expr_cpp_files generated_lcp_expr_files)
+
+add_lcp_expr(semantic/symbol.lcp)
+
+add_custom_target(generate_lcp_expr DEPENDS ${generated_lcp_expr_files})
+
+set(mg_expr_sources
+    ${lcp_expr_cpp_files}
+    parsing.cpp)
+
+find_package(Boost REQUIRED)
+
+add_library(mg-expr STATIC ${mg_expr_sources})
+add_dependencies(mg-expr generate_lcp_expr)
+target_include_directories(mg-expr PUBLIC ${CMAKE_SOURCE_DIR}/include)
+target_include_directories(mg-expr PUBLIC ${CMAKE_CURRENT_SOURCE_DIR})
+target_include_directories(mg-expr PUBLIC ${CMAKE_CURRENT_SOURCE_DIR}/ast)
+target_include_directories(mg-expr PUBLIC ${CMAKE_CURRENT_SOURCE_DIR}/interpret)
+target_include_directories(mg-expr PUBLIC ${CMAKE_CURRENT_SOURCE_DIR}/semantic)
+target_link_libraries(mg-expr cppitertools Boost::headers mg-utils mg-parser)

src/expr/ast.hpp

@@ -0,0 +1,35 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+#pragma once
+
+#ifndef MG_AST_INCLUDE_PATH
+#ifdef MG_CLANG_TIDY_CHECK
+// NOLINTNEXTLINE(cppcoreguidelines-macro-usage)
+#define MG_AST_INCLUDE_PATH "query/v2/frontend/ast/ast.hpp"
+#else
+#error Missing AST include path
+#endif
+#endif
+
+#ifndef MG_INJECTED_NAMESPACE_NAME
+#ifdef MG_CLANG_TIDY_CHECK
+// NOLINTNEXTLINE(cppcoreguidelines-macro-usage)
+#define MG_INJECTED_NAMESPACE_NAME memgraph::query::v2
+#else
+#error Missing AST namespace
+#endif
+#endif
+
+#include MG_AST_INCLUDE_PATH
+
+namespace memgraph::expr {
+using namespace MG_INJECTED_NAMESPACE_NAME;  // NOLINT(google-build-using-namespace)
+}  // namespace memgraph::expr

src/expr/ast/ast_visitor.hpp

@@ -0,0 +1,135 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include "utils/visitor.hpp"
+
+namespace MG_INJECTED_NAMESPACE_NAME {
+
+// Forward declares for Tree visitors.
+class CypherQuery;
+class SingleQuery;
+class CypherUnion;
+class NamedExpression;
+class Identifier;
+class PropertyLookup;
+class LabelsTest;
+class Aggregation;
+class Function;
+class Reduce;
+class Coalesce;
+class Extract;
+class All;
+class Single;
+class Any;
+class None;
+class ParameterLookup;
+class CallProcedure;
+class Create;
+class Match;
+class Return;
+class With;
+class Pattern;
+class NodeAtom;
+class EdgeAtom;
+class PrimitiveLiteral;
+class ListLiteral;
+class MapLiteral;
+class OrOperator;
+class XorOperator;
+class AndOperator;
+class NotOperator;
+class AdditionOperator;
+class SubtractionOperator;
+class MultiplicationOperator;
+class DivisionOperator;
+class ModOperator;
+class UnaryPlusOperator;
+class UnaryMinusOperator;
+class IsNullOperator;
+class NotEqualOperator;
+class EqualOperator;
+class LessOperator;
+class GreaterOperator;
+class LessEqualOperator;
+class GreaterEqualOperator;
+class InListOperator;
+class SubscriptOperator;
+class ListSlicingOperator;
+class IfOperator;
+class Delete;
+class Where;
+class SetProperty;
+class SetProperties;
+class SetLabels;
+class RemoveProperty;
+class RemoveLabels;
+class Merge;
+class Unwind;
+class AuthQuery;
+class ExplainQuery;
+class ProfileQuery;
+class IndexQuery;
+class InfoQuery;
+class ConstraintQuery;
+class RegexMatch;
+class DumpQuery;
+class ReplicationQuery;
+class LockPathQuery;
+class LoadCsv;
+class FreeMemoryQuery;
+class TriggerQuery;
+class IsolationLevelQuery;
+class CreateSnapshotQuery;
+class StreamQuery;
+class SettingQuery;
+class VersionQuery;
+class Foreach;
+class SchemaQuery;
+
+using TreeCompositeVisitor = memgraph::utils::CompositeVisitor<
+    SingleQuery, CypherUnion, NamedExpression, OrOperator, XorOperator, AndOperator, NotOperator, AdditionOperator,
+    SubtractionOperator, MultiplicationOperator, DivisionOperator, ModOperator, NotEqualOperator, EqualOperator,
+    LessOperator, GreaterOperator, LessEqualOperator, GreaterEqualOperator, InListOperator, SubscriptOperator,
+    ListSlicingOperator, IfOperator, UnaryPlusOperator, UnaryMinusOperator, IsNullOperator, ListLiteral, MapLiteral,
+    PropertyLookup, LabelsTest, Aggregation, Function, Reduce, Coalesce, Extract, All, Single, Any, None, CallProcedure,
+    Create, Match, Return, With, Pattern, NodeAtom, EdgeAtom, Delete, Where, SetProperty, SetProperties, SetLabels,
+    RemoveProperty, RemoveLabels, Merge, Unwind, RegexMatch, LoadCsv, Foreach>;
+
+using TreeLeafVisitor = memgraph::utils::LeafVisitor<Identifier, PrimitiveLiteral, ParameterLookup>;
+
+class HierarchicalTreeVisitor : public TreeCompositeVisitor, public TreeLeafVisitor {
+ public:
+  using TreeCompositeVisitor::PostVisit;
+  using TreeCompositeVisitor::PreVisit;
+  using TreeLeafVisitor::Visit;
+  using typename TreeLeafVisitor::ReturnType;
+};
+
+template <class TResult>
+class ExpressionVisitor
+    : public memgraph::utils::Visitor<
+          TResult, NamedExpression, OrOperator, XorOperator, AndOperator, NotOperator, AdditionOperator,
+          SubtractionOperator, MultiplicationOperator, DivisionOperator, ModOperator, NotEqualOperator, EqualOperator,
+          LessOperator, GreaterOperator, LessEqualOperator, GreaterEqualOperator, InListOperator, SubscriptOperator,
+          ListSlicingOperator, IfOperator, UnaryPlusOperator, UnaryMinusOperator, IsNullOperator, ListLiteral,
+          MapLiteral, PropertyLookup, LabelsTest, Aggregation, Function, Reduce, Coalesce, Extract, All, Single, Any,
+          None, ParameterLookup, Identifier, PrimitiveLiteral, RegexMatch> {};
+
+template <class TResult>
+class QueryVisitor
+    : public memgraph::utils::Visitor<TResult, CypherQuery, ExplainQuery, ProfileQuery, IndexQuery, AuthQuery,
+                                      InfoQuery, ConstraintQuery, DumpQuery, ReplicationQuery, LockPathQuery,
+                                      FreeMemoryQuery, TriggerQuery, IsolationLevelQuery, CreateSnapshotQuery,
+                                      StreamQuery, SettingQuery, VersionQuery, SchemaQuery> {};
+
+}  // namespace MG_INJECTED_NAMESPACE_NAME

src/expr/ast/pretty_print.hpp

@@ -0,0 +1,271 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include <iostream>
+#include <type_traits>
+
+#include "expr/ast.hpp"
+#include "expr/typed_value.hpp"
+#include "utils/algorithm.hpp"
+#include "utils/logging.hpp"
+#include "utils/string.hpp"
+
+namespace memgraph::expr {
+namespace detail {
+template <typename T>
+void PrintObject(std::ostream *out, const T &arg) {
+  static_assert(!std::is_convertible<T, Expression *>::value,
+                "This overload shouldn't be called with pointers convertible "
+                "to Expression *. This means your other PrintObject overloads aren't "
+                "being called for certain AST nodes when they should (or perhaps such "
+                "overloads don't exist yet).");
+  *out << arg;
+}
+
+inline void PrintObject(std::ostream *out, const std::string &str) { *out << utils::Escape(str); }
+
+inline void PrintObject(std::ostream *out, Aggregation::Op op) { *out << Aggregation::OpToString(op); }
+
+inline void PrintObject(std::ostream *out, Expression *expr);
+
+inline void PrintObject(std::ostream *out, Identifier *expr) { PrintObject(out, static_cast<Expression *>(expr)); }
+
+template <typename T>
+void PrintObject(std::ostream *out, const std::vector<T> &vec) {
+  *out << "[";
+  utils::PrintIterable(*out, vec, ", ", [](auto &stream, const auto &item) { PrintObject(&stream, item); });
+  *out << "]";
+}
+
+template <typename T>
+void PrintObject(std::ostream *out, const std::vector<T, utils::Allocator<T>> &vec) {
+  *out << "[";
+  utils::PrintIterable(*out, vec, ", ", [](auto &stream, const auto &item) { PrintObject(&stream, item); });
+  *out << "]";
+}
+
+template <typename K, typename V>
+void PrintObject(std::ostream *out, const std::map<K, V> &map) {
+  *out << "{";
+  utils::PrintIterable(*out, map, ", ", [](auto &stream, const auto &item) {
+    PrintObject(&stream, item.first);
+    stream << ": ";
+    PrintObject(&stream, item.second);
+  });
+  *out << "}";
+}
+
+template <typename T>
+void PrintObject(std::ostream *out, const utils::pmr::map<utils::pmr::string, T> &map) {
+  *out << "{";
+  utils::PrintIterable(*out, map, ", ", [](auto &stream, const auto &item) {
+    PrintObject(&stream, item.first);
+    stream << ": ";
+    PrintObject(&stream, item.second);
+  });
+  *out << "}";
+}
+
+template <typename T1, typename T2, typename T3>
+inline void PrintObject(std::ostream *out, const TypedValueT<T1, T2, T3> &value) {
+  using TypedValue = TypedValueT<T1, T2, T3>;
+  switch (value.type()) {
+    case TypedValue::Type::Null:
+      *out << "null";
+      break;
+    case TypedValue::Type::String:
+      PrintObject(out, value.ValueString());
+      break;
+    case TypedValue::Type::Bool:
+      *out << (value.ValueBool() ? "true" : "false");
+      break;
+    case TypedValue::Type::Int:
+      PrintObject(out, value.ValueInt());
+      break;
+    case TypedValue::Type::Double:
+      PrintObject(out, value.ValueDouble());
+      break;
+    case TypedValue::Type::List:
+      PrintObject(out, value.ValueList());
+      break;
+    case TypedValue::Type::Map:
+      PrintObject(out, value.ValueMap());
+      break;
+    case TypedValue::Type::Date:
+      PrintObject(out, value.ValueDate());
+      break;
+    case TypedValue::Type::Duration:
+      PrintObject(out, value.ValueDuration());
+      break;
+    case TypedValue::Type::LocalTime:
+      PrintObject(out, value.ValueLocalTime());
+      break;
+    case TypedValue::Type::LocalDateTime:
+      PrintObject(out, value.ValueLocalDateTime());
+      break;
+    default:
+      MG_ASSERT(false, "PrintObject(std::ostream *out, const TypedValue &value) should not reach here");
+  }
+}
+
+template <typename T>
+void PrintOperatorArgs(std::ostream *out, const T &arg) {
+  *out << " ";
+  PrintObject(out, arg);
+  *out << ")";
+}
+
+template <typename T, typename... Ts>
+void PrintOperatorArgs(std::ostream *out, const T &arg, const Ts &...args) {
+  *out << " ";
+  PrintObject(out, arg);
+  PrintOperatorArgs(out, args...);
+}
+
+template <typename... Ts>
+void PrintOperator(std::ostream *out, const std::string &name, const Ts &...args) {
+  *out << "(" << name;
+  PrintOperatorArgs(out, args...);
+}
+}  // namespace detail
+
+class ExpressionPrettyPrinter : public ExpressionVisitor<void> {
+ public:
+  explicit ExpressionPrettyPrinter(std::ostream *out) : out_(out) {}
+
+  // Unary operators
+  // NOLINTNEXTLINE(cppcoreguidelines-macro-usage)
+#define UNARY_OPERATOR_VISIT(OP_NODE, OP_STR)      \
+  /* NOLINTNEXTLINE(bugprone-macro-parentheses) */ \
+  void Visit(OP_NODE &op) override { detail::PrintOperator(out_, OP_STR, op.expression_); }
+
+  UNARY_OPERATOR_VISIT(NotOperator, "Not");
+  UNARY_OPERATOR_VISIT(UnaryPlusOperator, "+");
+  UNARY_OPERATOR_VISIT(UnaryMinusOperator, "-");
+  UNARY_OPERATOR_VISIT(IsNullOperator, "IsNull");
+
+#undef UNARY_OPERATOR_VISIT
+
+  // Binary operators
+// NOLINTNEXTLINE(cppcoreguidelines-macro-usage)
+#define BINARY_OPERATOR_VISIT(OP_NODE, OP_STR)     \
+  /* NOLINTNEXTLINE(bugprone-macro-parentheses) */ \
+  void Visit(OP_NODE &op) override { detail::PrintOperator(out_, OP_STR, op.expression1_, op.expression2_); }
+
+  BINARY_OPERATOR_VISIT(OrOperator, "Or");
+  BINARY_OPERATOR_VISIT(XorOperator, "Xor");
+  BINARY_OPERATOR_VISIT(AndOperator, "And");
+  BINARY_OPERATOR_VISIT(AdditionOperator, "+");
+  BINARY_OPERATOR_VISIT(SubtractionOperator, "-");
+  BINARY_OPERATOR_VISIT(MultiplicationOperator, "*");
+  BINARY_OPERATOR_VISIT(DivisionOperator, "/");
+  BINARY_OPERATOR_VISIT(ModOperator, "%");
+  BINARY_OPERATOR_VISIT(NotEqualOperator, "!=");
+  BINARY_OPERATOR_VISIT(EqualOperator, "==");
+  BINARY_OPERATOR_VISIT(LessOperator, "<");
+  BINARY_OPERATOR_VISIT(GreaterOperator, ">");
+  BINARY_OPERATOR_VISIT(LessEqualOperator, "<=");
+  BINARY_OPERATOR_VISIT(GreaterEqualOperator, ">=");
+  BINARY_OPERATOR_VISIT(InListOperator, "In");
+  BINARY_OPERATOR_VISIT(SubscriptOperator, "Subscript");
+
+#undef BINARY_OPERATOR_VISIT
+
+  // Other
+  void Visit(ListSlicingOperator &op) override {
+    detail::PrintOperator(out_, "ListSlicing", op.list_, op.lower_bound_, op.upper_bound_);
+  }
+
+  void Visit(IfOperator &op) override {
+    detail::PrintOperator(out_, "If", op.condition_, op.then_expression_, op.else_expression_);
+  }
+
+  void Visit(ListLiteral &op) override { detail::PrintOperator(out_, "ListLiteral", op.elements_); }
+
+  void Visit(MapLiteral &op) override {
+    std::map<std::string, Expression *> map;
+    for (const auto &kv : op.elements_) {
+      map[kv.first.name] = kv.second;
+    }
+    detail::PrintObject(out_, map);
+  }
+
+  void Visit(LabelsTest &op) override { detail::PrintOperator(out_, "LabelsTest", op.expression_); }
+
+  void Visit(Aggregation &op) override { detail::PrintOperator(out_, "Aggregation", op.op_); }
+
+  void Visit(Function &op) override { detail::PrintOperator(out_, "Function", op.function_name_, op.arguments_); }
+
+  void Visit(Reduce &op) override {
+    detail::PrintOperator(out_, "Reduce", op.accumulator_, op.initializer_, op.identifier_, op.list_, op.expression_);
+  }
+
+  void Visit(Coalesce &op) override { detail::PrintOperator(out_, "Coalesce", op.expressions_); }
+
+  void Visit(Extract &op) override { detail::PrintOperator(out_, "Extract", op.identifier_, op.list_, op.expression_); }
+
+  void Visit(All &op) override {
+    detail::PrintOperator(out_, "All", op.identifier_, op.list_expression_, op.where_->expression_);
+  }
+
+  void Visit(Single &op) override {
+    detail::PrintOperator(out_, "Single", op.identifier_, op.list_expression_, op.where_->expression_);
+  }
+
+  void Visit(Any &op) override {
+    detail::PrintOperator(out_, "Any", op.identifier_, op.list_expression_, op.where_->expression_);
+  }
+
+  void Visit(None &op) override {
+    detail::PrintOperator(out_, "None", op.identifier_, op.list_expression_, op.where_->expression_);
+  }
+
+  void Visit(Identifier &op) override { detail::PrintOperator(out_, "Identifier", op.name_); }
+
+  void Visit(PrimitiveLiteral &op) override { detail::PrintObject(out_, op.value_); }
+
+  void Visit(PropertyLookup &op) override {
+    detail::PrintOperator(out_, "PropertyLookup", op.expression_, op.property_.name);
+  }
+
+  void Visit(ParameterLookup &op) override { detail::PrintOperator(out_, "ParameterLookup", op.token_position_); }
+
+  void Visit(NamedExpression &op) override { detail::PrintOperator(out_, "NamedExpression", op.name_, op.expression_); }
+
+  void Visit(RegexMatch &op) override { detail::PrintOperator(out_, "=~", op.string_expr_, op.regex_); }
+
+ private:
+  std::ostream *out_;
+};
+
+namespace detail {
+inline void PrintObject(std::ostream *out, Expression *expr) {
+  if (expr) {
+    ExpressionPrettyPrinter printer{out};
+    expr->Accept(printer);
+  } else {
+    *out << "<null>";
+  }
+}
+}  // namespace detail
+
+inline void PrintExpression(Expression *expr, std::ostream *out) {
+  ExpressionPrettyPrinter printer{out};
+  expr->Accept(printer);
+}
+
+inline void PrintExpression(NamedExpression *expr, std::ostream *out) {
+  ExpressionPrettyPrinter printer{out};
+  expr->Accept(printer);
+}
+}  // namespace memgraph::expr

src/expr/exceptions.hpp

@@ -0,0 +1,52 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include "utils/exceptions.hpp"
+
+namespace memgraph::expr {
+
+class SyntaxException : public utils::BasicException {
+ public:
+  using utils::BasicException::BasicException;
+  SyntaxException() : SyntaxException("") {}
+};
+
+class SemanticException : public utils::BasicException {
+ public:
+  using utils::BasicException::BasicException;
+  SemanticException() : BasicException("") {}
+};
+
+class ExpressionRuntimeException : public utils::BasicException {
+ public:
+  using utils::BasicException::BasicException;
+};
+
+class RedeclareVariableError : public SemanticException {
+ public:
+  explicit RedeclareVariableError(const std::string &name) : SemanticException("Redeclaring variable: " + name + ".") {}
+};
+
+class UnboundVariableError : public SemanticException {
+ public:
+  explicit UnboundVariableError(const std::string &name) : SemanticException("Unbound variable: " + name + ".") {}
+};
+
+class TypeMismatchError : public SemanticException {
+ public:
+  TypeMismatchError(const std::string &name, const std::string &datum, const std::string &expected)
+      : SemanticException(fmt::format("Type mismatch: {} already defined as {}, expected {}.", name, datum, expected)) {
+  }
+};
+
+}  // namespace memgraph::expr

src/expr/interpret/eval.hpp

@@ -0,0 +1,826 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+/// @file
+#pragma once
+
+#include <algorithm>
+#include <limits>
+#include <map>
+#include <optional>
+#include <regex>
+#include <type_traits>
+#include <vector>
+
+#include "expr/ast.hpp"
+#include "expr/exceptions.hpp"
+#include "expr/interpret/frame.hpp"
+#include "expr/semantic/symbol_table.hpp"
+#include "utils/exceptions.hpp"
+
+namespace memgraph::expr {
+
+struct StorageTag {};
+struct QueryEngineTag {};
+
+template <typename TypedValue, typename EvaluationContext, typename DbAccessor, typename StorageView, typename LabelId,
+          typename PropertyValue, typename ConvFunctor, typename Error, typename Tag = StorageTag>
+class ExpressionEvaluator : public ExpressionVisitor<TypedValue> {
+ public:
+  ExpressionEvaluator(Frame<TypedValue> *frame, const SymbolTable &symbol_table, const EvaluationContext &ctx,
+                      DbAccessor *dba, StorageView view)
+      : frame_(frame), symbol_table_(&symbol_table), ctx_(&ctx), dba_(dba), view_(view) {}
+
+  using ExpressionVisitor<TypedValue>::Visit;
+
+  utils::MemoryResource *GetMemoryResource() const { return ctx_->memory; }
+
+  TypedValue Visit(NamedExpression &named_expression) override {
+    const auto &symbol = symbol_table_->at(named_expression);
+    auto value = named_expression.expression_->Accept(*this);
+    frame_->at(symbol) = value;
+    return value;
+  }
+
+  TypedValue Visit(Identifier &ident) override {
+    return TypedValue(frame_->at(symbol_table_->at(ident)), ctx_->memory);
+  }
+
+  // NOLINTNEXTLINE(cppcoreguidelines-macro-usage)
+#define BINARY_OPERATOR_VISITOR(OP_NODE, CPP_OP, CYPHER_OP)                                                         \
+  /* NOLINTNEXTLINE(bugprone-macro-parentheses) */                                                                  \
+  TypedValue Visit(OP_NODE &op) override {                                                                          \
+    auto val1 = op.expression1_->Accept(*this);                                                                     \
+    auto val2 = op.expression2_->Accept(*this);                                                                     \
+    try {                                                                                                           \
+      return val1 CPP_OP val2;                                                                                      \
+    } catch (const TypedValueException &) {                                                                         \
+      throw ExpressionRuntimeException("Invalid types: {} and {} for '{}'.", val1.type(), val2.type(), #CYPHER_OP); \
+    }                                                                                                               \
+  }
+
+  // NOLINTNEXTLINE(cppcoreguidelines-macro-usage)
+#define UNARY_OPERATOR_VISITOR(OP_NODE, CPP_OP, CYPHER_OP)                                   \
+  /* NOLINTNEXTLINE(bugprone-macro-parentheses) */                                           \
+  TypedValue Visit(OP_NODE &op) override {                                                   \
+    auto val = op.expression_->Accept(*this);                                                \
+    try {                                                                                    \
+      return CPP_OP val;                                                                     \
+    } catch (const TypedValueException &) {                                                  \
+      throw ExpressionRuntimeException("Invalid type {} for '{}'.", val.type(), #CYPHER_OP); \
+    }                                                                                        \
+  }
+
+  BINARY_OPERATOR_VISITOR(OrOperator, ||, OR);
+  BINARY_OPERATOR_VISITOR(XorOperator, ^, XOR);
+  BINARY_OPERATOR_VISITOR(AdditionOperator, +, +);
+  BINARY_OPERATOR_VISITOR(SubtractionOperator, -, -);
+  BINARY_OPERATOR_VISITOR(MultiplicationOperator, *, *);
+  BINARY_OPERATOR_VISITOR(DivisionOperator, /, /);
+  BINARY_OPERATOR_VISITOR(ModOperator, %, %);
+  BINARY_OPERATOR_VISITOR(NotEqualOperator, !=, <>);
+  BINARY_OPERATOR_VISITOR(EqualOperator, ==, =);
+  BINARY_OPERATOR_VISITOR(LessOperator, <, <);
+  BINARY_OPERATOR_VISITOR(GreaterOperator, >, >);
+  BINARY_OPERATOR_VISITOR(LessEqualOperator, <=, <=);
+  BINARY_OPERATOR_VISITOR(GreaterEqualOperator, >=, >=);
+
+  UNARY_OPERATOR_VISITOR(NotOperator, !, NOT);
+  UNARY_OPERATOR_VISITOR(UnaryPlusOperator, +, +);
+  UNARY_OPERATOR_VISITOR(UnaryMinusOperator, -, -);
+
+#undef BINARY_OPERATOR_VISITOR
+#undef UNARY_OPERATOR_VISITOR
+
+  TypedValue Visit(AndOperator &op) override {
+    auto value1 = op.expression1_->Accept(*this);
+    if (value1.IsBool() && !value1.ValueBool()) {
+      // If first expression is false, don't evaluate the second one.
+      return value1;
+    }
+    auto value2 = op.expression2_->Accept(*this);
+    try {
+      return value1 && value2;
+    } catch (const TypedValueException &) {
+      throw ExpressionRuntimeException("Invalid types: {} and {} for AND.", value1.type(), value2.type());
+    }
+  }
+
+  TypedValue Visit(IfOperator &if_operator) override {
+    auto condition = if_operator.condition_->Accept(*this);
+    if (condition.IsNull()) {
+      return if_operator.else_expression_->Accept(*this);
+    }
+    if (condition.type() != TypedValue::Type::Bool) {
+      // At the moment IfOperator is used only in CASE construct.
+      throw ExpressionRuntimeException("CASE expected boolean expression, got {}.", condition.type());
+    }
+    if (condition.ValueBool()) {
+      return if_operator.then_expression_->Accept(*this);
+    }
+    return if_operator.else_expression_->Accept(*this);
+  }
+
+  TypedValue Visit(InListOperator &in_list) override {
+    auto literal = in_list.expression1_->Accept(*this);
+    auto _list = in_list.expression2_->Accept(*this);
+    if (_list.IsNull()) {
+      return TypedValue(ctx_->memory);
+    }
+    // Exceptions have higher priority than returning nulls when list expression
+    // is not null.
+    if (_list.type() != TypedValue::Type::List) {
+      throw ExpressionRuntimeException("IN expected a list, got {}.", _list.type());
+    }
+    const auto &list = _list.ValueList();
+
+    // If literal is NULL there is no need to try to compare it with every
+    // element in the list since result of every comparison will be NULL. There
+    // is one special case that we must test explicitly: if list is empty then
+    // result is false since no comparison will be performed.
+    if (list.empty()) return TypedValue(false, ctx_->memory);
+    if (literal.IsNull()) return TypedValue(ctx_->memory);
+
+    auto has_null = false;
+    for (const auto &element : list) {
+      auto result = literal == element;
+      if (result.IsNull()) {
+        has_null = true;
+      } else if (result.ValueBool()) {
+        return TypedValue(true, ctx_->memory);
+      }
+    }
+    if (has_null) {
+      return TypedValue(ctx_->memory);
+    }
+    return TypedValue(false, ctx_->memory);
+  }
+
+  TypedValue Visit(SubscriptOperator &list_indexing) override {
+    auto lhs = list_indexing.expression1_->Accept(*this);
+    auto index = list_indexing.expression2_->Accept(*this);
+    if (!lhs.IsList() && !lhs.IsMap() && !lhs.IsVertex() && !lhs.IsEdge() && !lhs.IsNull())
+      throw ExpressionRuntimeException(
+          "Expected a list, a map, a node or an edge to index with '[]', got "
+          "{}.",
+          lhs.type());
+    if (lhs.IsNull() || index.IsNull()) return TypedValue(ctx_->memory);
+    if (lhs.IsList()) {
+      if (!index.IsInt())
+        throw ExpressionRuntimeException("Expected an integer as a list index, got {}.", index.type());
+      auto index_int = index.ValueInt();
+      // NOTE: Take non-const reference to list, so that we can move out the
+      // indexed element as the result.
+      auto &list = lhs.ValueList();
+      if (index_int < 0) {
+        index_int += static_cast<int64_t>(list.size());
+      }
+      if (index_int >= static_cast<int64_t>(list.size()) || index_int < 0) return TypedValue(ctx_->memory);
+      // NOTE: Explicit move is needed, so that we return the move constructed
+      // value and preserve the correct MemoryResource.
+      return std::move(list[index_int]);
+    }
+
+    if (lhs.IsMap()) {
+      if (!index.IsString())
+        throw ExpressionRuntimeException("Expected a string as a map index, got {}.", index.type());
+      // NOTE: Take non-const reference to map, so that we can move out the
+      // looked-up element as the result.
+      auto &map = lhs.ValueMap();
+      auto found = map.find(index.ValueString());
+      if (found == map.end()) return TypedValue(ctx_->memory);
+      // NOTE: Explicit move is needed, so that we return the move constructed
+      // value and preserve the correct MemoryResource.
+      return std::move(found->second);
+    }
+
+    if (lhs.IsVertex()) {
+      if (!index.IsString())
+        throw ExpressionRuntimeException("Expected a string as a property name, got {}.", index.type());
+      return TypedValue(GetProperty(lhs.ValueVertex(), index.ValueString()), ctx_->memory);
+    }
+
+    if (lhs.IsEdge()) {
+      if (!index.IsString())
+        throw ExpressionRuntimeException("Expected a string as a property name, got {}.", index.type());
+      return TypedValue(GetProperty(lhs.ValueEdge(), index.ValueString()), ctx_->memory);
+    }
+
+    // lhs is Null
+    return TypedValue(ctx_->memory);
+  }
+
+  TypedValue Visit(ListSlicingOperator &op) override {
+    // If some type is null we can't return null, because throwing exception
+    // on illegal type has higher priority.
+    auto is_null = false;
+    auto get_bound = [&](Expression *bound_expr, int64_t default_value) {
+      if (bound_expr) {
+        auto bound = bound_expr->Accept(*this);
+        if (bound.type() == TypedValue::Type::Null) {
+          is_null = true;
+        } else if (bound.type() != TypedValue::Type::Int) {
+          throw ExpressionRuntimeException("Expected an integer for a bound in list slicing, got {}.", bound.type());
+        }
+        return bound;
+      }
+      return TypedValue(default_value, ctx_->memory);
+    };
+    auto _upper_bound = get_bound(op.upper_bound_, std::numeric_limits<int64_t>::max());
+    auto _lower_bound = get_bound(op.lower_bound_, 0);
+
+    auto _list = op.list_->Accept(*this);
+    if (_list.type() == TypedValue::Type::Null) {
+      is_null = true;
+    } else if (_list.type() != TypedValue::Type::List) {
+      throw ExpressionRuntimeException("Expected a list to slice, got {}.", _list.type());
+    }
+
+    if (is_null) {
+      return TypedValue(ctx_->memory);
+    }
+    const auto &list = _list.ValueList();
+    auto normalise_bound = [&](int64_t bound) {
+      if (bound < 0) {
+        bound = static_cast<int64_t>(list.size()) + bound;
+      }
+      return std::max(static_cast<int64_t>(0), std::min(bound, static_cast<int64_t>(list.size())));
+    };
+    auto lower_bound = normalise_bound(_lower_bound.ValueInt());
+    auto upper_bound = normalise_bound(_upper_bound.ValueInt());
+    if (upper_bound <= lower_bound) {
+      return TypedValue(typename TypedValue::TVector(ctx_->memory), ctx_->memory);
+    }
+    return TypedValue(
+        typename TypedValue::TVector(list.begin() + lower_bound, list.begin() + upper_bound, ctx_->memory));
+  }
+
+  TypedValue Visit(IsNullOperator &is_null) override {
+    auto value = is_null.expression_->Accept(*this);
+    return TypedValue(value.IsNull(), ctx_->memory);
+  }
+
+  TypedValue Visit(PropertyLookup &property_lookup) override {
+    auto expression_result = property_lookup.expression_->Accept(*this);
+    auto maybe_date = [this](const auto &date, const auto &prop_name) -> std::optional<TypedValue> {
+      if (prop_name == "year") {
+        return TypedValue(date.year, ctx_->memory);
+      }
+      if (prop_name == "month") {
+        return TypedValue(date.month, ctx_->memory);
+      }
+      if (prop_name == "day") {
+        return TypedValue(date.day, ctx_->memory);
+      }
+      return std::nullopt;
+    };
+    auto maybe_local_time = [this](const auto &lt, const auto &prop_name) -> std::optional<TypedValue> {
+      if (prop_name == "hour") {
+        return TypedValue(lt.hour, ctx_->memory);
+      }
+      if (prop_name == "minute") {
+        return TypedValue(lt.minute, ctx_->memory);
+      }
+      if (prop_name == "second") {
+        return TypedValue(lt.second, ctx_->memory);
+      }
+      if (prop_name == "millisecond") {
+        return TypedValue(lt.millisecond, ctx_->memory);
+      }
+      if (prop_name == "microsecond") {
+        return TypedValue(lt.microsecond, ctx_->memory);
+      }
+      return std::nullopt;
+    };
+    auto maybe_duration = [this](const auto &dur, const auto &prop_name) -> std::optional<TypedValue> {
+      if (prop_name == "day") {
+        return TypedValue(dur.Days(), ctx_->memory);
+      }
+      if (prop_name == "hour") {
+        return TypedValue(dur.SubDaysAsHours(), ctx_->memory);
+      }
+      if (prop_name == "minute") {
+        return TypedValue(dur.SubDaysAsMinutes(), ctx_->memory);
+      }
+      if (prop_name == "second") {
+        return TypedValue(dur.SubDaysAsSeconds(), ctx_->memory);
+      }
+      if (prop_name == "millisecond") {
+        return TypedValue(dur.SubDaysAsMilliseconds(), ctx_->memory);
+      }
+      if (prop_name == "microsecond") {
+        return TypedValue(dur.SubDaysAsMicroseconds(), ctx_->memory);
+      }
+      if (prop_name == "nanosecond") {
+        return TypedValue(dur.SubDaysAsNanoseconds(), ctx_->memory);
+      }
+      return std::nullopt;
+    };
+    switch (expression_result.type()) {
+      case TypedValue::Type::Null:
+        return TypedValue(ctx_->memory);
+      case TypedValue::Type::Vertex:
+        return GetProperty(expression_result.ValueVertex(), property_lookup.property_);
+      case TypedValue::Type::Edge:
+        return GetProperty(expression_result.ValueEdge(), property_lookup.property_);
+      case TypedValue::Type::Map: {
+        // NOTE: Take non-const reference to map, so that we can move out the
+        // looked-up element as the result.
+        auto &map = expression_result.ValueMap();
+        auto found = map.find(property_lookup.property_.name.c_str());
+        if (found == map.end()) return TypedValue(ctx_->memory);
+        // NOTE: Explicit move is needed, so that we return the move constructed
+        // value and preserve the correct MemoryResource.
+        return std::move(found->second);
+      }
+      case TypedValue::Type::Duration: {
+        const auto &prop_name = property_lookup.property_.name;
+        const auto &dur = expression_result.ValueDuration();
+        if (auto dur_field = maybe_duration(dur, prop_name); dur_field) {
+          return std::move(*dur_field);
+        }
+        throw ExpressionRuntimeException("Invalid property name {} for Duration", prop_name);
+      }
+      case TypedValue::Type::Date: {
+        const auto &prop_name = property_lookup.property_.name;
+        const auto &date = expression_result.ValueDate();
+        if (auto date_field = maybe_date(date, prop_name); date_field) {
+          return std::move(*date_field);
+        }
+        throw ExpressionRuntimeException("Invalid property name {} for Date", prop_name);
+      }
+      case TypedValue::Type::LocalTime: {
+        const auto &prop_name = property_lookup.property_.name;
+        const auto &lt = expression_result.ValueLocalTime();
+        if (auto lt_field = maybe_local_time(lt, prop_name); lt_field) {
+          return std::move(*lt_field);
+        }
+        throw ExpressionRuntimeException("Invalid property name {} for LocalTime", prop_name);
+      }
+      case TypedValue::Type::LocalDateTime: {
+        const auto &prop_name = property_lookup.property_.name;
+        const auto &ldt = expression_result.ValueLocalDateTime();
+        if (auto date_field = maybe_date(ldt.date, prop_name); date_field) {
+          return std::move(*date_field);
+        }
+        if (auto lt_field = maybe_local_time(ldt.local_time, prop_name); lt_field) {
+          return std::move(*lt_field);
+        }
+        throw ExpressionRuntimeException("Invalid property name {} for LocalDateTime", prop_name);
+      }
+      default:
+        throw ExpressionRuntimeException("Only nodes, edges, maps and temporal types have properties to be looked-up.");
+    }
+  }
+
+  template <typename VertexAccessor, typename TTag = Tag,
+            typename TReturnType = std::enable_if_t<std::is_same_v<TTag, StorageTag>, bool>>
+  TReturnType HasLabelImpl(const VertexAccessor &vertex, const LabelIx &label, StorageTag /*tag*/) {
+    auto has_label = vertex.HasLabel(view_, GetLabel(label));
+    if (has_label.HasError() && has_label.GetError() == Error::NONEXISTENT_OBJECT) {
+      // This is a very nasty and temporary hack in order to make MERGE
+      // work. The old storage had the following logic when returning an
+      // `OLD` view: `return old ? old : new`. That means that if the
+      // `OLD` view didn't exist, it returned the NEW view. With this hack
+      // we simulate that behavior.
+      // TODO (mferencevic, teon.banek): Remove once MERGE is
+      // reimplemented.
+      has_label = vertex.HasLabel(StorageView::NEW, GetLabel(label));
+    }
+    if (has_label.HasError()) {
+      switch (has_label.GetError()) {
+        case Error::DELETED_OBJECT:
+          throw ExpressionRuntimeException("Trying to access labels on a deleted node.");
+        case Error::NONEXISTENT_OBJECT:
+          throw ExpressionRuntimeException("Trying to access labels from a node that doesn't exist.");
+        case Error::SERIALIZATION_ERROR:
+        case Error::VERTEX_HAS_EDGES:
+        case Error::PROPERTIES_DISABLED:
+          throw ExpressionRuntimeException("Unexpected error when accessing labels.");
+      }
+    }
+    return *has_label;
+  }
+
+  template <typename VertexAccessor, typename TTag = Tag,
+            typename TReturnType = std::enable_if_t<std::is_same_v<TTag, QueryEngineTag>, bool>>
+  TReturnType HasLabelImpl(const VertexAccessor &vertex, const LabelIx &label_ix, QueryEngineTag /*tag*/) {
+    auto label = typename VertexAccessor::Label{LabelId::FromUint(label_ix.ix)};
+    auto has_label = vertex.HasLabel(label);
+    return !has_label;
+  }
+
+  TypedValue Visit(LabelsTest &labels_test) override {
+    auto expression_result = labels_test.expression_->Accept(*this);
+    switch (expression_result.type()) {
+      case TypedValue::Type::Null:
+        return TypedValue(ctx_->memory);
+      case TypedValue::Type::Vertex: {
+        const auto &vertex = expression_result.ValueVertex();
+        if (std::ranges::all_of(labels_test.labels_, [&vertex, this](const auto label_test) {
+              return this->HasLabelImpl(vertex, label_test, Tag{});
+            })) {
+          return TypedValue(true, ctx_->memory);
+        }
+        return TypedValue(false, ctx_->memory);
+      }
+      default:
+        throw ExpressionRuntimeException("Only nodes have labels.");
+    }
+  }
+
+  TypedValue Visit(PrimitiveLiteral &literal) override {
+    // TODO: no need to evaluate constants, we can write it to frame in one
+    // of the previous phases.
+    return TypedValue(literal.value_, ctx_->memory);
+  }
+
+  TypedValue Visit(ListLiteral &literal) override {
+    typename TypedValue::TVector result(ctx_->memory);
+    result.reserve(literal.elements_.size());
+    for (const auto &expression : literal.elements_) result.emplace_back(expression->Accept(*this));
+    return TypedValue(result, ctx_->memory);
+  }
+
+  TypedValue Visit(MapLiteral &literal) override {
+    typename TypedValue::TMap result(ctx_->memory);
+    for (const auto &pair : literal.elements_) result.emplace(pair.first.name, pair.second->Accept(*this));
+    return TypedValue(result, ctx_->memory);
+  }
+
+  TypedValue Visit(Aggregation &aggregation) override {
+    return TypedValue(frame_->at(symbol_table_->at(aggregation)), ctx_->memory);
+  }
+
+  TypedValue Visit(Coalesce &coalesce) override {
+    auto &exprs = coalesce.expressions_;
+
+    if (exprs.size() == 0) {
+      throw ExpressionRuntimeException("'coalesce' requires at least one argument.");
+    }
+
+    for (int64_t i = 0; i < exprs.size(); ++i) {
+      TypedValue val(exprs[i]->Accept(*this), ctx_->memory);
+      if (!val.IsNull()) {
+        return val;
+      }
+    }
+
+    return TypedValue(ctx_->memory);
+  }
+
+  TypedValue Visit(Function &function) override {
+    FunctionContext function_ctx{dba_, ctx_->memory, ctx_->timestamp, &ctx_->counters, view_};
+    // Stack allocate evaluated arguments when there's a small number of them.
+    if (function.arguments_.size() <= 8) {
+      TypedValue arguments[8] = {TypedValue(ctx_->memory), TypedValue(ctx_->memory), TypedValue(ctx_->memory),
+                                 TypedValue(ctx_->memory), TypedValue(ctx_->memory), TypedValue(ctx_->memory),
+                                 TypedValue(ctx_->memory), TypedValue(ctx_->memory)};
+      for (size_t i = 0; i < function.arguments_.size(); ++i) {
+        arguments[i] = function.arguments_[i]->Accept(*this);
+      }
+      auto res = function.function_(arguments, function.arguments_.size(), function_ctx);
+      MG_ASSERT(res.GetMemoryResource() == ctx_->memory);
+      return res;
+    } else {
+      typename TypedValue::TVector arguments(ctx_->memory);
+      arguments.reserve(function.arguments_.size());
+      for (const auto &argument : function.arguments_) {
+        arguments.emplace_back(argument->Accept(*this));
+      }
+      auto res = function.function_(arguments.data(), arguments.size(), function_ctx);
+      MG_ASSERT(res.GetMemoryResource() == ctx_->memory);
+      return res;
+    }
+  }
+
+  TypedValue Visit(Reduce &reduce) override {
+    auto list_value = reduce.list_->Accept(*this);
+    if (list_value.IsNull()) {
+      return TypedValue(ctx_->memory);
+    }
+    if (list_value.type() != TypedValue::Type::List) {
+      throw ExpressionRuntimeException("REDUCE expected a list, got {}.", list_value.type());
+    }
+    const auto &list = list_value.ValueList();
+    const auto &element_symbol = symbol_table_->at(*reduce.identifier_);
+    const auto &accumulator_symbol = symbol_table_->at(*reduce.accumulator_);
+    auto accumulator = reduce.initializer_->Accept(*this);
+    for (const auto &element : list) {
+      frame_->at(accumulator_symbol) = accumulator;
+      frame_->at(element_symbol) = element;
+      accumulator = reduce.expression_->Accept(*this);
+    }
+    return accumulator;
+  }
+
+  TypedValue Visit(Extract &extract) override {
+    auto list_value = extract.list_->Accept(*this);
+    if (list_value.IsNull()) {
+      return TypedValue(ctx_->memory);
+    }
+    if (list_value.type() != TypedValue::Type::List) {
+      throw ExpressionRuntimeException("EXTRACT expected a list, got {}.", list_value.type());
+    }
+    const auto &list = list_value.ValueList();
+    const auto &element_symbol = symbol_table_->at(*extract.identifier_);
+    typename TypedValue::TVector result(ctx_->memory);
+    result.reserve(list.size());
+    for (const auto &element : list) {
+      if (element.IsNull()) {
+        result.emplace_back();
+      } else {
+        frame_->at(element_symbol) = element;
+        result.emplace_back(extract.expression_->Accept(*this));
+      }
+    }
+    return TypedValue(result, ctx_->memory);
+  }
+
+  TypedValue Visit(All &all) override {
+    auto list_value = all.list_expression_->Accept(*this);
+    if (list_value.IsNull()) {
+      return TypedValue(ctx_->memory);
+    }
+    if (list_value.type() != TypedValue::Type::List) {
+      throw ExpressionRuntimeException("ALL expected a list, got {}.", list_value.type());
+    }
+    const auto &list = list_value.ValueList();
+    const auto &symbol = symbol_table_->at(*all.identifier_);
+    bool has_null_elements = false;
+    bool has_value = false;
+    for (const auto &element : list) {
+      frame_->at(symbol) = element;
+      auto result = all.where_->expression_->Accept(*this);
+      if (!result.IsNull() && result.type() != TypedValue::Type::Bool) {
+        throw ExpressionRuntimeException("Predicate of ALL must evaluate to boolean, got {}.", result.type());
+      }
+      if (!result.IsNull()) {
+        has_value = true;
+        if (!result.ValueBool()) {
+          return TypedValue(false, ctx_->memory);
+        }
+      } else {
+        has_null_elements = true;
+      }
+    }
+    if (!has_value) {
+      return TypedValue(ctx_->memory);
+    }
+    if (has_null_elements) {
+      return TypedValue(false, ctx_->memory);
+    } else {
+      return TypedValue(true, ctx_->memory);
+    }
+  }
+
+  TypedValue Visit(Single &single) override {
+    auto list_value = single.list_expression_->Accept(*this);
+    if (list_value.IsNull()) {
+      return TypedValue(ctx_->memory);
+    }
+    if (list_value.type() != TypedValue::Type::List) {
+      throw ExpressionRuntimeException("SINGLE expected a list, got {}.", list_value.type());
+    }
+    const auto &list = list_value.ValueList();
+    const auto &symbol = symbol_table_->at(*single.identifier_);
+    bool has_value = false;
+    bool predicate_satisfied = false;
+    for (const auto &element : list) {
+      frame_->at(symbol) = element;
+      auto result = single.where_->expression_->Accept(*this);
+      if (!result.IsNull() && result.type() != TypedValue::Type::Bool) {
+        throw ExpressionRuntimeException("Predicate of SINGLE must evaluate to boolean, got {}.", result.type());
+      }
+      if (result.type() == TypedValue::Type::Bool) {
+        has_value = true;
+      }
+      if (result.IsNull() || !result.ValueBool()) {
+        continue;
+      }
+      // Return false if more than one element satisfies the predicate.
+      if (predicate_satisfied) {
+        return TypedValue(false, ctx_->memory);
+      } else {
+        predicate_satisfied = true;
+      }
+    }
+    if (!has_value) {
+      return TypedValue(ctx_->memory);
+    } else {
+      return TypedValue(predicate_satisfied, ctx_->memory);
+    }
+  }
+
+  TypedValue Visit(Any &any) override {
+    auto list_value = any.list_expression_->Accept(*this);
+    if (list_value.IsNull()) {
+      return TypedValue(ctx_->memory);
+    }
+    if (list_value.type() != TypedValue::Type::List) {
+      throw ExpressionRuntimeException("ANY expected a list, got {}.", list_value.type());
+    }
+    const auto &list = list_value.ValueList();
+    const auto &symbol = symbol_table_->at(*any.identifier_);
+    bool has_value = false;
+    for (const auto &element : list) {
+      frame_->at(symbol) = element;
+      auto result = any.where_->expression_->Accept(*this);
+      if (!result.IsNull() && result.type() != TypedValue::Type::Bool) {
+        throw ExpressionRuntimeException("Predicate of ANY must evaluate to boolean, got {}.", result.type());
+      }
+      if (!result.IsNull()) {
+        has_value = true;
+        if (result.ValueBool()) {
+          return TypedValue(true, ctx_->memory);
+        }
+      }
+    }
+    // Return Null if all elements are Null
+    if (!has_value) {
+      return TypedValue(ctx_->memory);
+    } else {
+      return TypedValue(false, ctx_->memory);
+    }
+  }
+
+  TypedValue Visit(None &none) override {
+    auto list_value = none.list_expression_->Accept(*this);
+    if (list_value.IsNull()) {
+      return TypedValue(ctx_->memory);
+    }
+    if (list_value.type() != TypedValue::Type::List) {
+      throw ExpressionRuntimeException("NONE expected a list, got {}.", list_value.type());
+    }
+    const auto &list = list_value.ValueList();
+    const auto &symbol = symbol_table_->at(*none.identifier_);
+    bool has_value = false;
+    for (const auto &element : list) {
+      frame_->at(symbol) = element;
+      auto result = none.where_->expression_->Accept(*this);
+      if (!result.IsNull() && result.type() != TypedValue::Type::Bool) {
+        throw ExpressionRuntimeException("Predicate of NONE must evaluate to boolean, got {}.", result.type());
+      }
+      if (!result.IsNull()) {
+        has_value = true;
+        if (result.ValueBool()) {
+          return TypedValue(false, ctx_->memory);
+        }
+      }
+    }
+    // Return Null if all elements are Null
+    if (!has_value) {
+      return TypedValue(ctx_->memory);
+    } else {
+      return TypedValue(true, ctx_->memory);
+    }
+  }
+
+  TypedValue Visit(ParameterLookup &param_lookup) override {
+    return TypedValue(conv_(ctx_->parameters.AtTokenPosition(param_lookup.token_position_)), ctx_->memory);
+  }
+
+  TypedValue Visit(RegexMatch &regex_match) override {
+    auto target_string_value = regex_match.string_expr_->Accept(*this);
+    auto regex_value = regex_match.regex_->Accept(*this);
+    if (target_string_value.IsNull() || regex_value.IsNull()) {
+      return TypedValue(ctx_->memory);
+    }
+    if (regex_value.type() != TypedValue::Type::String) {
+      throw ExpressionRuntimeException("Regular expression must evaluate to a string, got {}.", regex_value.type());
+    }
+    if (target_string_value.type() != TypedValue::Type::String) {
+      // Instead of error, we return Null which makes it compatible in case we
+      // use indexed lookup which filters out any non-string properties.
+      // Assuming a property lookup is the target_string_value.
+      return TypedValue(ctx_->memory);
+    }
+    const auto &target_string = target_string_value.ValueString();
+    try {
+      std::regex regex(regex_value.ValueString());
+      return TypedValue(std::regex_match(target_string, regex), ctx_->memory);
+    } catch (const std::regex_error &e) {
+      throw ExpressionRuntimeException("Regex error in '{}': {}", regex_value.ValueString(), e.what());
+    }
+  }
+
+ private:
+  template <class TRecordAccessor, class TTag = Tag,
+            class TReturnType = std::enable_if_t<std::is_same_v<TTag, QueryEngineTag>, TypedValue>>
+  TReturnType GetProperty(const TRecordAccessor &record_accessor, PropertyIx prop) {
+    auto maybe_prop = record_accessor.GetProperty(prop.name);
+    // Handler non existent property
+    return conv_(maybe_prop);
+  }
+
+  template <class TRecordAccessor, class TTag = Tag,
+            class TReturnType = std::enable_if_t<std::is_same_v<TTag, QueryEngineTag>, TypedValue>>
+  TReturnType GetProperty(const TRecordAccessor &record_accessor, const std::string_view name) {
+    auto maybe_prop = record_accessor.GetProperty(std::string(name));
+    // Handler non existent property
+    return conv_(maybe_prop);
+  }
+
+  template <class TRecordAccessor, class TTag = Tag,
+            class TReturnType = std::enable_if_t<std::is_same_v<TTag, StorageTag>, TypedValue>>
+  TypedValue GetProperty(const TRecordAccessor &record_accessor, PropertyIx prop) {
+    auto maybe_prop = record_accessor.GetProperty(view_, ctx_->properties[prop.ix]);
+    if (maybe_prop.HasError() && maybe_prop.GetError() == Error::NONEXISTENT_OBJECT) {
+      // This is a very nasty and temporary hack in order to make MERGE work.
+      // The old storage had the following logic when returning an `OLD` view:
+      // `return old ? old : new`. That means that if the `OLD` view didn't
+      // exist, it returned the NEW view. With this hack we simulate that
+      // behavior.
+      // TODO (mferencevic, teon.banek): Remove once MERGE is reimplemented.
+      maybe_prop = record_accessor.GetProperty(StorageView::NEW, ctx_->properties[prop.ix]);
+    }
+    if (maybe_prop.HasError()) {
+      switch (maybe_prop.GetError()) {
+        case Error::DELETED_OBJECT:
+          throw ExpressionRuntimeException("Trying to get a property from a deleted object.");
+        case Error::NONEXISTENT_OBJECT:
+          throw ExpressionRuntimeException("Trying to get a property from an object that doesn't exist.");
+        case Error::SERIALIZATION_ERROR:
+        case Error::VERTEX_HAS_EDGES:
+        case Error::PROPERTIES_DISABLED:
+          throw ExpressionRuntimeException("Unexpected error when getting a property.");
+      }
+    }
+    return conv_(*maybe_prop, ctx_->memory);
+  }
+
+  template <class TRecordAccessor, class TTag = Tag,
+            class TReturnType = std::enable_if_t<std::is_same_v<TTag, StorageTag>, TypedValue>>
+  TypedValue GetProperty(const TRecordAccessor &record_accessor, const std::string_view name) {
+    auto maybe_prop = record_accessor.GetProperty(view_, dba_->NameToProperty(name));
+    if (maybe_prop.HasError() && maybe_prop.GetError() == Error::NONEXISTENT_OBJECT) {
+      // This is a very nasty and temporary hack in order to make MERGE work.
+      // The old storage had the following logic when returning an `OLD` view:
+      // `return old ? old : new`. That means that if the `OLD` view didn't
+      // exist, it returned the NEW view. With this hack we simulate that
+      // behavior.
+      // TODO (mferencevic, teon.banek): Remove once MERGE is reimplemented.
+      maybe_prop = record_accessor.GetProperty(view_, dba_->NameToProperty(name));
+    }
+    if (maybe_prop.HasError()) {
+      switch (maybe_prop.GetError()) {
+        case Error::DELETED_OBJECT:
+          throw ExpressionRuntimeException("Trying to get a property from a deleted object.");
+        case Error::NONEXISTENT_OBJECT:
+          throw ExpressionRuntimeException("Trying to get a property from an object that doesn't exist.");
+        case Error::SERIALIZATION_ERROR:
+        case Error::VERTEX_HAS_EDGES:
+        case Error::PROPERTIES_DISABLED:
+          throw ExpressionRuntimeException("Unexpected error when getting a property.");
+      }
+    }
+    return conv_(*maybe_prop, ctx_->memory);
+  }
+
+  LabelId GetLabel(LabelIx label) { return ctx_->labels[label.ix]; }
+
+  Frame<TypedValue> *frame_;
+  const SymbolTable *symbol_table_;
+  const EvaluationContext *ctx_;
+  DbAccessor *dba_;
+  // which switching approach should be used when evaluating
+  StorageView view_;
+  ConvFunctor conv_;
+};
+
+/// A helper function for evaluating an expression that's an int.
+///
+/// @param what - Name of what's getting evaluated. Used for user feedback (via
+///               exception) when the evaluated value is not an int.
+/// @throw ExpressionRuntimeException if expression doesn't evaluate to an int.
+template <typename ExpressionEvaluator>
+int64_t EvaluateInt(ExpressionEvaluator *evaluator, Expression *expr, const std::string &what) {
+  TypedValue value = expr->Accept(*evaluator);
+  try {
+    return value.ValueInt();
+  } catch (TypedValueException &e) {
+    throw ExpressionRuntimeException(what + " must be an int");
+  }
+}
+
+template <typename ExpressionEvaluator>
+std::optional<size_t> EvaluateMemoryLimit(ExpressionEvaluator *eval, Expression *memory_limit, size_t memory_scale) {
+  if (!memory_limit) return std::nullopt;
+  auto limit_value = memory_limit->Accept(*eval);
+  if (!limit_value.IsInt() || limit_value.ValueInt() <= 0)
+    throw ExpressionRuntimeException("Memory limit must be a non-negative integer.");
+  size_t limit = limit_value.ValueInt();
+  if (std::numeric_limits<size_t>::max() / memory_scale < limit)
+    throw ExpressionRuntimeException("Memory limit overflow.");
+  return limit * memory_scale;
+}
+
+}  // namespace memgraph::expr

src/expr/interpret/frame.hpp

@@ -0,0 +1,45 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include <vector>
+
+#include "expr/semantic/symbol_table.hpp"
+#include "utils/logging.hpp"
+#include "utils/memory.hpp"
+#include "utils/pmr/vector.hpp"
+
+namespace memgraph::expr {
+
+template <typename TypedValue>
+class Frame {
+ public:
+  /// Create a Frame of given size backed by a utils::NewDeleteResource()
+  explicit Frame(int64_t size) : elems_(size, utils::NewDeleteResource()) { MG_ASSERT(size >= 0); }
+
+  Frame(int64_t size, utils::MemoryResource *memory) : elems_(size, memory) { MG_ASSERT(size >= 0); }
+
+  TypedValue &operator[](const Symbol &symbol) { return elems_[symbol.position()]; }
+  const TypedValue &operator[](const Symbol &symbol) const { return elems_[symbol.position()]; }
+
+  TypedValue &at(const Symbol &symbol) { return elems_.at(symbol.position()); }
+  const TypedValue &at(const Symbol &symbol) const { return elems_.at(symbol.position()); }
+
+  auto &elems() { return elems_; }
+
+  utils::MemoryResource *GetMemoryResource() const { return elems_.get_allocator().GetMemoryResource(); }
+
+ private:
+  utils::pmr::vector<TypedValue> elems_;
+};
+
+}  // namespace memgraph::expr

src/expr/parsing.cpp

@@ -0,0 +1,184 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#include "expr/parsing.hpp"
+
+#include <cctype>
+#include <codecvt>
+#include <locale>
+#include <stdexcept>
+
+#include "expr/exceptions.hpp"
+#include "utils/logging.hpp"
+#include "utils/string.hpp"
+
+namespace memgraph::expr {
+
+int64_t ParseIntegerLiteral(const std::string &s) {
+  try {
+    // Not really correct since long long can have a bigger range than int64_t.
+    return static_cast<int64_t>(std::stoll(s, 0, 0));
+  } catch (const std::out_of_range &) {
+    throw SemanticException("Integer literal exceeds 64 bits.");
+  }
+}
+
+std::string ParseStringLiteral(const std::string &s) {
+  // These functions is declared as lambda since its semantics is highly
+  // specific for this conxtext and shouldn't be used elsewhere.
+  auto EncodeEscapedUnicodeCodepointUtf32 = [](const std::string &s, int &i) {
+    const int kLongUnicodeLength = 8;
+    int j = i + 1;
+    while (j < static_cast<int>(s.size()) - 1 && j < i + kLongUnicodeLength + 1 && isxdigit(s[j])) {
+      ++j;
+    }
+    if (j - i == kLongUnicodeLength + 1) {
+      char32_t t = stoi(s.substr(i + 1, kLongUnicodeLength), 0, 16);
+      i += kLongUnicodeLength;
+      std::wstring_convert<std::codecvt_utf8<char32_t>, char32_t> converter;
+      return converter.to_bytes(t);
+    }
+    throw SyntaxException(
+        "Expected 8 hex digits as unicode codepoint started with \\U. "
+        "Use \\u for 4 hex digits format.");
+  };
+  auto EncodeEscapedUnicodeCodepointUtf16 = [](const std::string &s, int &i) {
+    const int kShortUnicodeLength = 4;
+    int j = i + 1;
+    while (j < static_cast<int>(s.size()) - 1 && j < i + kShortUnicodeLength + 1 && isxdigit(s[j])) {
+      ++j;
+    }
+    if (j - i >= kShortUnicodeLength + 1) {
+      char16_t t = stoi(s.substr(i + 1, kShortUnicodeLength), 0, 16);
+      if (t >= 0xD800 && t <= 0xDBFF) {
+        // t is high surrogate pair. Expect one more utf16 codepoint.
+        j = i + kShortUnicodeLength + 1;
+        if (j >= static_cast<int>(s.size()) - 1 || s[j] != '\\') {
+          throw SemanticException("Invalid UTF codepoint.");
+        }
+        ++j;
+        if (j >= static_cast<int>(s.size()) - 1 || (s[j] != 'u' && s[j] != 'U')) {
+          throw SemanticException("Invalid UTF codepoint.");
+        }
+        ++j;
+        int k = j;
+        while (k < static_cast<int>(s.size()) - 1 && k < j + kShortUnicodeLength && isxdigit(s[k])) {
+          ++k;
+        }
+        if (k != j + kShortUnicodeLength) {
+          throw SemanticException("Invalid UTF codepoint.");
+        }
+        char16_t surrogates[3] = {t, static_cast<char16_t>(stoi(s.substr(j, kShortUnicodeLength), 0, 16)), 0};
+        i += kShortUnicodeLength + 2 + kShortUnicodeLength;
+        std::wstring_convert<std::codecvt_utf8_utf16<char16_t>, char16_t> converter;
+        return converter.to_bytes(surrogates);
+      } else {
+        i += kShortUnicodeLength;
+        std::wstring_convert<std::codecvt_utf8_utf16<char16_t>, char16_t> converter;
+        return converter.to_bytes(t);
+      }
+    }
+    throw SyntaxException(
+        "Expected 4 hex digits as unicode codepoint started with \\u. "
+        "Use \\U for 8 hex digits format.");
+  };
+
+  std::string unescaped;
+  bool escape = false;
+
+  // First and last char is quote, we don't need to look at them.
+  for (int i = 1; i < static_cast<int>(s.size()) - 1; ++i) {
+    if (escape) {
+      switch (s[i]) {
+        case '\\':
+          unescaped += '\\';
+          break;
+        case '\'':
+          unescaped += '\'';
+          break;
+        case '"':
+          unescaped += '"';
+          break;
+        case 'B':
+        case 'b':
+          unescaped += '\b';
+          break;
+        case 'F':
+        case 'f':
+          unescaped += '\f';
+          break;
+        case 'N':
+        case 'n':
+          unescaped += '\n';
+          break;
+        case 'R':
+        case 'r':
+          unescaped += '\r';
+          break;
+        case 'T':
+        case 't':
+          unescaped += '\t';
+          break;
+        case 'U':
+          try {
+            unescaped += EncodeEscapedUnicodeCodepointUtf32(s, i);
+          } catch (const std::range_error &) {
+            throw SemanticException("Invalid UTF codepoint.");
+          }
+          break;
+        case 'u':
+          try {
+            unescaped += EncodeEscapedUnicodeCodepointUtf16(s, i);
+          } catch (const std::range_error &) {
+            throw SemanticException("Invalid UTF codepoint.");
+          }
+          break;
+        default:
+          // This should never happen, except grammar changes and we don't
+          // notice change in this production.
+          DLOG_FATAL("can't happen");
+          throw std::exception();
+      }
+      escape = false;
+    } else if (s[i] == '\\') {
+      escape = true;
+    } else {
+      unescaped += s[i];
+    }
+  }
+  return unescaped;
+}
+
+double ParseDoubleLiteral(const std::string &s) {
+  try {
+    return utils::ParseDouble(s);
+  } catch (const utils::BasicException &) {
+    throw SemanticException("Couldn't parse string to double.");
+  }
+}
+
+std::string ParseParameter(const std::string &s) {
+  DMG_ASSERT(s[0] == '$', "Invalid string passed as parameter name");
+  if (s[1] != '`') return s.substr(1);
+  // If parameter name is escaped symbolic name then symbolic name should be
+  // unescaped and leading and trailing backquote should be removed.
+  DMG_ASSERT(s.size() > 3U && s.back() == '`', "Invalid string passed as parameter name");
+  std::string out;
+  for (int i = 2; i < static_cast<int>(s.size()) - 1; ++i) {
+    if (s[i] == '`') {
+      ++i;
+    }
+    out.push_back(s[i]);
+  }
+  return out;
+}
+
+}  // namespace memgraph::expr

src/expr/parsing.hpp

@@ -0,0 +1,27 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+/// @file
+#pragma once
+
+#include <cstdint>
+#include <string>
+
+namespace memgraph::expr {
+
+// These are the functions for parsing literals and parameter names from
+// opencypher query.
+int64_t ParseIntegerLiteral(const std::string &s);
+std::string ParseStringLiteral(const std::string &s);
+double ParseDoubleLiteral(const std::string &s);
+std::string ParseParameter(const std::string &s);
+
+}  // namespace memgraph::expr

src/expr/semantic/symbol.lcp

@@ -0,0 +1,87 @@
+;; Copyright 2022 Memgraph Ltd.
+;;
+;; Use of this software is governed by the Business Source License
+;; included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+;; License, and you may not use this file except in compliance with the Business Source License.
+;;
+;; As of the Change Date specified in that file, in accordance with
+;; the Business Source License, use of this software will be governed
+;; by the Apache License, Version 2.0, included in the file
+;; licenses/APL.txt.
+
+#>cpp
+#pragma once
+
+#include <string>
+
+#include "utils/typeinfo.hpp"
+cpp<#
+
+(lcp:namespace memgraph)
+(lcp:namespace expr)
+
+(lcp:define-class symbol ()
+  ((name "std::string" :scope :public)
+   (position :int64_t :scope :public)
+   (user-declared :bool :initval "true" :scope :public)
+   (type "Type" :initval "Type::ANY" :scope :public)
+   (token-position :int64_t :initval "-1" :scope :public))
+  (:public
+    ;; This is similar to TypedValue::Type, but this has `Any` type.
+    ;; TODO: Make a better Type structure which can store a generic List.
+    (lcp:define-enum type (any vertex edge path number edge-list)
+      (:serialize))
+    #>cpp
+    // TODO: Generate enum to string conversion from LCP. Note, that this is
+    // displayed to the end user, so we may want to have a pretty name of each
+    // value.
+    static std::string TypeToString(Type type) {
+      const char *enum_string[] = {"Any",  "Vertex", "Edge",
+                                   "Path", "Number", "EdgeList"};
+      return enum_string[static_cast<int>(type)];
+    }
+
+    Symbol() {}
+    Symbol(const std::string &name, int position, bool user_declared,
+           Type type = Type::ANY, int token_position = -1)
+        : name_(name),
+          position_(position),
+          user_declared_(user_declared),
+          type_(type),
+          token_position_(token_position) {}
+
+    bool operator==(const Symbol &other) const {
+      return position_ == other.position_ && name_ == other.name_ &&
+             type_ == other.type_;
+    }
+    bool operator!=(const Symbol &other) const { return !operator==(other); }
+
+    // TODO: Remove these since members are public
+    const auto &name() const { return name_; }
+    int position() const { return position_; }
+    Type type() const { return type_; }
+    bool user_declared() const { return user_declared_; }
+    int token_position() const { return token_position_; }
+    cpp<#)
+  (:serialize (:slk)))
+
+(lcp:pop-namespace) ;; expr
+(lcp:pop-namespace) ;; memgraph
+
+#>cpp
+namespace std {
+
+template <>
+struct hash<memgraph::expr::Symbol> {
+  size_t operator()(const memgraph::expr::Symbol &symbol) const {
+    size_t prime = 265443599u;
+    size_t hash = std::hash<int>{}(symbol.position());
+    hash ^= prime * std::hash<std::string>{}(symbol.name());
+    hash ^= prime * std::hash<int>{}(static_cast<int>(symbol.type()));
+    return hash;
+  }
+};
+
+}  // namespace std
+
+cpp<#

src/expr/semantic/symbol_generator.hpp

@@ -0,0 +1,712 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+// Copyright 2017 Memgraph
+//
+// Created by Teon Banek on 11-03-2017
+
+#pragma once
+
+#include <algorithm>
+#include <optional>
+#include <ranges>
+#include <unordered_map>
+#include <unordered_set>
+#include <variant>
+#include <vector>
+
+#include "expr/ast.hpp"
+#include "expr/ast/ast_visitor.hpp"
+#include "expr/exceptions.hpp"
+#include "expr/semantic/symbol_table.hpp"
+
+namespace memgraph::expr {
+namespace detail {
+inline std::unordered_map<std::string, Identifier *> GeneratePredefinedIdentifierMap(
+    const std::vector<Identifier *> &predefined_identifiers) {
+  std::unordered_map<std::string, Identifier *> identifier_map;
+  for (const auto &identifier : predefined_identifiers) {
+    identifier_map.emplace(identifier->name_, identifier);
+  }
+
+  return identifier_map;
+}
+}  // namespace detail
+
+/// Visits the AST and generates symbols for variables.
+///
+/// During the process of symbol generation, simple semantic checks are
+/// performed. Such as, redeclaring a variable or conflicting expectations of
+/// variable types.
+class SymbolGenerator : public HierarchicalTreeVisitor {
+ public:
+  explicit SymbolGenerator(SymbolTable *symbol_table, const std::vector<Identifier *> &predefined_identifiers)
+      : symbol_table_(symbol_table),
+        predefined_identifiers_{detail::GeneratePredefinedIdentifierMap(predefined_identifiers)},
+        scopes_(1, Scope()) {}
+
+  using HierarchicalTreeVisitor::PostVisit;
+  using HierarchicalTreeVisitor::PreVisit;
+  using HierarchicalTreeVisitor::Visit;
+  using typename HierarchicalTreeVisitor::ReturnType;
+
+  // Query
+  bool PreVisit(SingleQuery & /*unused*/) override {
+    prev_return_names_ = curr_return_names_;
+    curr_return_names_.clear();
+    return true;
+  }
+
+  // Union
+  bool PreVisit(CypherUnion & /*unused*/) override {
+    scopes_.back() = Scope();
+    return true;
+  }
+
+  bool PostVisit(CypherUnion &cypher_union) override {
+    if (prev_return_names_ != curr_return_names_) {
+      throw SemanticException("All subqueries in an UNION must have the same column names.");
+    }
+
+    // create new symbols for the result of the union
+    for (const auto &name : curr_return_names_) {
+      auto symbol = CreateSymbol(name, false);
+      cypher_union.union_symbols_.push_back(symbol);
+    }
+
+    return true;
+  }
+
+  // Clauses
+  bool PreVisit(Create & /*unused*/) override {
+    scopes_.back().in_create = true;
+    return true;
+  }
+
+  bool PostVisit(Create & /*unused*/) override {
+    scopes_.back().in_create = false;
+    return true;
+  }
+
+  bool PreVisit(CallProcedure &call_proc) override {
+    for (auto *expr : call_proc.arguments_) {
+      expr->Accept(*this);
+    }
+    return false;
+  }
+
+  bool PostVisit(CallProcedure &call_proc) override {
+    for (auto *ident : call_proc.result_identifiers_) {
+      if (HasSymbolLocalScope(ident->name_)) {
+        throw RedeclareVariableError(ident->name_);
+      }
+      ident->MapTo(CreateSymbol(ident->name_, true));
+    }
+    return true;
+  }
+
+  bool PreVisit(LoadCsv & /*unused*/) override { return false; }
+
+  bool PostVisit(LoadCsv &load_csv) override {
+    if (HasSymbolLocalScope(load_csv.row_var_->name_)) {
+      throw RedeclareVariableError(load_csv.row_var_->name_);
+    }
+    load_csv.row_var_->MapTo(CreateSymbol(load_csv.row_var_->name_, true));
+    return true;
+  }
+
+  bool PreVisit(Return &ret) override {
+    auto &scope = scopes_.back();
+    scope.in_return = true;
+    VisitReturnBody(ret.body_);
+    scope.in_return = false;
+    return false;  // We handled the traversal ourselves.
+  }
+
+  bool PostVisit(Return & /*unused*/) override {
+    for (const auto &name_symbol : scopes_.back().symbols) curr_return_names_.insert(name_symbol.first);
+    return true;
+  }
+
+  bool PreVisit(With &with) override {
+    auto &scope = scopes_.back();
+    scope.in_with = true;
+    VisitReturnBody(with.body_, with.where_);
+    scope.in_with = false;
+    return false;  // We handled the traversal ourselves.
+  }
+
+  bool PreVisit(Where & /*unused*/) override {
+    scopes_.back().in_where = true;
+    return true;
+  }
+
+  bool PostVisit(Where & /*unused*/) override {
+    scopes_.back().in_where = false;
+    return true;
+  }
+
+  bool PreVisit(Merge & /*unused*/) override {
+    scopes_.back().in_merge = true;
+    return true;
+  }
+
+  bool PostVisit(Merge & /*unused*/) override {
+    scopes_.back().in_merge = false;
+    return true;
+  }
+
+  bool PostVisit(Unwind &unwind) override {
+    const auto &name = unwind.named_expression_->name_;
+    if (HasSymbolLocalScope(name)) {
+      throw RedeclareVariableError(name);
+    }
+    unwind.named_expression_->MapTo(CreateSymbol(name, true));
+    return true;
+  }
+
+  bool PreVisit(Match & /*unused*/) override {
+    scopes_.back().in_match = true;
+    return true;
+  }
+
+  bool PostVisit(Match & /*unused*/) override {
+    auto &scope = scopes_.back();
+    scope.in_match = false;
+    // Check variables in property maps after visiting Match, so that they can
+    // reference symbols out of bind order.
+    for (auto &ident : scope.identifiers_in_match) {
+      if (!HasSymbolLocalScope(ident->name_) && !ConsumePredefinedIdentifier(ident->name_))
+        throw UnboundVariableError(ident->name_);
+      ident->MapTo(scope.symbols[ident->name_]);
+    }
+    scope.identifiers_in_match.clear();
+    return true;
+  }
+
+  bool PreVisit(Foreach &for_each) override {
+    const auto &name = for_each.named_expression_->name_;
+    scopes_.emplace_back(Scope());
+    scopes_.back().in_foreach = true;
+    for_each.named_expression_->MapTo(
+        CreateSymbol(name, true, Symbol::Type::ANY, for_each.named_expression_->token_position_));
+    return true;
+  }
+
+  bool PostVisit(Foreach & /*unused*/) override {
+    scopes_.pop_back();
+    return true;
+  }
+
+  // Expressions
+  ReturnType Visit(Identifier &ident) override {
+    auto &scope = scopes_.back();
+    if (scope.in_skip || scope.in_limit) {
+      throw SemanticException("Variables are not allowed in {}.", scope.in_skip ? "SKIP" : "LIMIT");
+    }
+    Symbol symbol;
+    if (scope.in_pattern && !(scope.in_node_atom || scope.visiting_edge)) {
+      // If we are in the pattern, and outside of a node or an edge, the
+      // identifier is the pattern name.
+      symbol = GetOrCreateSymbolLocalScope(ident.name_, ident.user_declared_, Symbol::Type::PATH);
+    } else if (scope.in_pattern && scope.in_pattern_atom_identifier) {
+      //  Patterns used to create nodes and edges cannot redeclare already
+      //  established bindings. Declaration only happens in single node
+      //  patterns and in edge patterns. OpenCypher example,
+      //  `MATCH (n) CREATE (n)` should throw an error that `n` is already
+      //  declared. While `MATCH (n) CREATE (n) -[:R]-> (n)` is allowed,
+      //  since `n` now references the bound node instead of declaring it.
+      if ((scope.in_create_node || scope.in_create_edge) && HasSymbolLocalScope(ident.name_)) {
+        throw RedeclareVariableError(ident.name_);
+      }
+      auto type = Symbol::Type::VERTEX;
+      if (scope.visiting_edge) {
+        // Edge referencing is not allowed (like in Neo4j):
+        // `MATCH (n) - [r] -> (n) - [r] -> (n) RETURN r` is not allowed.
+        if (HasSymbolLocalScope(ident.name_)) {
+          throw RedeclareVariableError(ident.name_);
+        }
+        type = scope.visiting_edge->IsVariable() ? Symbol::Type::EDGE_LIST : Symbol::Type::EDGE;
+      }
+      symbol = GetOrCreateSymbolLocalScope(ident.name_, ident.user_declared_, type);
+    } else if (scope.in_pattern && !scope.in_pattern_atom_identifier && scope.in_match) {
+      if (scope.in_edge_range && scope.visiting_edge->identifier_->name_ == ident.name_) {
+        // Prevent variable path bounds to reference the identifier which is bound
+        // by the variable path itself.
+        throw UnboundVariableError(ident.name_);
+      }
+      // Variables in property maps or bounds of variable length path during MATCH
+      // can reference symbols bound later in the same MATCH. We collect them
+      // here, so that they can be checked after visiting Match.
+      scope.identifiers_in_match.emplace_back(&ident);
+    } else {
+      // Everything else references a bound symbol.
+      if (!HasSymbol(ident.name_) && !ConsumePredefinedIdentifier(ident.name_)) throw UnboundVariableError(ident.name_);
+      symbol = GetOrCreateSymbol(ident.name_, ident.user_declared_, Symbol::Type::ANY);
+    }
+    ident.MapTo(symbol);
+    return true;
+  }
+
+  ReturnType Visit(PrimitiveLiteral & /*unused*/) override { return true; }
+
+  ReturnType Visit(ParameterLookup & /*unused*/) override { return true; }
+
+  bool PreVisit(Aggregation &aggr) override {
+    auto &scope = scopes_.back();
+    // Check if the aggregation can be used in this context. This check should
+    // probably move to a separate phase, which checks if the query is well
+    // formed.
+    if ((!scope.in_return && !scope.in_with) || scope.in_order_by || scope.in_skip || scope.in_limit ||
+        scope.in_where) {
+      throw SemanticException("Aggregation functions are only allowed in WITH and RETURN.");
+    }
+    if (scope.in_aggregation) {
+      throw SemanticException(
+          "Using aggregation functions inside aggregation functions is not "
+          "allowed.");
+    }
+    if (scope.num_if_operators) {
+      // Neo allows aggregations here and produces very interesting behaviors.
+      // To simplify implementation at this moment we decided to completely
+      // disallow aggregations inside of the CASE.
+      // However, in some cases aggregation makes perfect sense, for example:
+      //    CASE count(n) WHEN 10 THEN "YES" ELSE "NO" END.
+      // TODO: Rethink of allowing aggregations in some parts of the CASE
+      // construct.
+      throw SemanticException("Using aggregation functions inside of CASE is not allowed.");
+    }
+    // Create a virtual symbol for aggregation result.
+    // Currently, we only have aggregation operators which return numbers.
+    auto aggr_name = Aggregation::OpToString(aggr.op_) + std::to_string(aggr.symbol_pos_);
+    aggr.MapTo(CreateSymbol(aggr_name, false, Symbol::Type::NUMBER));
+    scope.in_aggregation = true;
+    scope.has_aggregation = true;
+    return true;
+  }
+
+  bool PostVisit(Aggregation & /*unused*/) override {
+    scopes_.back().in_aggregation = false;
+    return true;
+  }
+
+  bool PreVisit(IfOperator & /*unused*/) override {
+    ++scopes_.back().num_if_operators;
+    return true;
+  }
+
+  bool PostVisit(IfOperator & /*unused*/) override {
+    --scopes_.back().num_if_operators;
+    return true;
+  }
+
+  bool PreVisit(All &all) override {
+    all.list_expression_->Accept(*this);
+    VisitWithIdentifiers(all.where_->expression_, {all.identifier_});
+    return false;
+  }
+
+  bool PreVisit(Single &single) override {
+    single.list_expression_->Accept(*this);
+    VisitWithIdentifiers(single.where_->expression_, {single.identifier_});
+    return false;
+  }
+
+  bool PreVisit(Any &any) override {
+    any.list_expression_->Accept(*this);
+    VisitWithIdentifiers(any.where_->expression_, {any.identifier_});
+    return false;
+  }
+
+  bool PreVisit(None &none) override {
+    none.list_expression_->Accept(*this);
+    VisitWithIdentifiers(none.where_->expression_, {none.identifier_});
+    return false;
+  }
+
+  bool PreVisit(Reduce &reduce) override {
+    reduce.initializer_->Accept(*this);
+    reduce.list_->Accept(*this);
+    VisitWithIdentifiers(reduce.expression_, {reduce.accumulator_, reduce.identifier_});
+    return false;
+  }
+
+  bool PreVisit(Extract &extract) override {
+    extract.list_->Accept(*this);
+    VisitWithIdentifiers(extract.expression_, {extract.identifier_});
+    return false;
+  }
+
+  // Pattern and its subparts.
+  bool PreVisit(Pattern &pattern) override {
+    auto &scope = scopes_.back();
+    scope.in_pattern = true;
+    if ((scope.in_create || scope.in_merge) && pattern.atoms_.size() == 1U) {
+      MG_ASSERT(utils::IsSubtype(*pattern.atoms_[0], NodeAtom::kType), "Expected a single NodeAtom in Pattern");
+      scope.in_create_node = true;
+    }
+    return true;
+  }
+
+  bool PostVisit(Pattern & /*unused*/) override {
+    auto &scope = scopes_.back();
+    scope.in_pattern = false;
+    scope.in_create_node = false;
+    return true;
+  }
+
+  bool PreVisit(NodeAtom &node_atom) override {
+    auto &scope = scopes_.back();
+    auto check_node_semantic = [&node_atom, &scope, this](const bool props_or_labels) {
+      const auto &node_name = node_atom.identifier_->name_;
+      if ((scope.in_create || scope.in_merge) && props_or_labels && HasSymbolLocalScope(node_name)) {
+        throw SemanticException("Cannot create node '" + node_name +
+                                "' with labels or properties, because it is already declared.");
+      }
+      scope.in_pattern_atom_identifier = true;
+      node_atom.identifier_->Accept(*this);
+      scope.in_pattern_atom_identifier = false;
+    };
+
+    scope.in_node_atom = true;
+    if (auto *properties = std::get_if<std::unordered_map<PropertyIx, Expression *>>(&node_atom.properties_)) {
+      bool props_or_labels = !properties->empty() || !node_atom.labels_.empty();
+
+      check_node_semantic(props_or_labels);
+      for (auto kv : *properties) {
+        kv.second->Accept(*this);
+      }
+
+      return false;
+    }
+    auto &properties_parameter = std::get<ParameterLookup *>(node_atom.properties_);
+    bool props_or_labels = !properties_parameter || !node_atom.labels_.empty();
+
+    check_node_semantic(props_or_labels);
+    properties_parameter->Accept(*this);
+    return false;
+  }
+
+  bool PostVisit(NodeAtom & /*unused*/) override {
+    scopes_.back().in_node_atom = false;
+    return true;
+  }
+
+  bool PreVisit(EdgeAtom &edge_atom) override {
+    auto &scope = scopes_.back();
+    scope.visiting_edge = &edge_atom;
+    if (scope.in_create || scope.in_merge) {
+      scope.in_create_edge = true;
+      if (edge_atom.edge_types_.size() != 1U) {
+        throw SemanticException(
+            "A single relationship type must be specified "
+            "when creating an edge.");
+      }
+      if (scope.in_create &&  // Merge allows bidirectionality
+          edge_atom.direction_ == EdgeAtom::Direction::BOTH) {
+        throw SemanticException(
+            "Bidirectional relationship are not supported "
+            "when creating an edge");
+      }
+      if (edge_atom.IsVariable()) {
+        throw SemanticException(
+            "Variable length relationships are not supported when creating an "
+            "edge.");
+      }
+    }
+    if (auto *properties = std::get_if<std::unordered_map<PropertyIx, Expression *>>(&edge_atom.properties_)) {
+      for (auto kv : *properties) {
+        kv.second->Accept(*this);
+      }
+    } else {
+      std::get<ParameterLookup *>(edge_atom.properties_)->Accept(*this);
+    }
+    if (edge_atom.IsVariable()) {
+      scope.in_edge_range = true;
+      if (edge_atom.lower_bound_) {
+        edge_atom.lower_bound_->Accept(*this);
+      }
+      if (edge_atom.upper_bound_) {
+        edge_atom.upper_bound_->Accept(*this);
+      }
+      scope.in_edge_range = false;
+      scope.in_pattern = false;
+      if (edge_atom.filter_lambda_.expression) {
+        VisitWithIdentifiers(edge_atom.filter_lambda_.expression,
+                             {edge_atom.filter_lambda_.inner_edge, edge_atom.filter_lambda_.inner_node});
+      } else {
+        // Create inner symbols, but don't bind them in scope, since they are to
+        // be used in the missing filter expression.
+        auto *inner_edge = edge_atom.filter_lambda_.inner_edge;
+        inner_edge->MapTo(
+            symbol_table_->CreateSymbol(inner_edge->name_, inner_edge->user_declared_, Symbol::Type::EDGE));
+        auto *inner_node = edge_atom.filter_lambda_.inner_node;
+        inner_node->MapTo(
+            symbol_table_->CreateSymbol(inner_node->name_, inner_node->user_declared_, Symbol::Type::VERTEX));
+      }
+      if (edge_atom.weight_lambda_.expression) {
+        VisitWithIdentifiers(edge_atom.weight_lambda_.expression,
+                             {edge_atom.weight_lambda_.inner_edge, edge_atom.weight_lambda_.inner_node});
+      }
+      scope.in_pattern = true;
+    }
+    scope.in_pattern_atom_identifier = true;
+    edge_atom.identifier_->Accept(*this);
+    scope.in_pattern_atom_identifier = false;
+    if (edge_atom.total_weight_) {
+      if (HasSymbolLocalScope(edge_atom.total_weight_->name_)) {
+        throw RedeclareVariableError(edge_atom.total_weight_->name_);
+      }
+      edge_atom.total_weight_->MapTo(GetOrCreateSymbolLocalScope(
+          edge_atom.total_weight_->name_, edge_atom.total_weight_->user_declared_, Symbol::Type::NUMBER));
+    }
+    return false;
+  }
+
+  bool PostVisit(EdgeAtom & /*unused*/) override {
+    auto &scope = scopes_.back();
+    scope.visiting_edge = nullptr;
+    scope.in_create_edge = false;
+    return true;
+  }
+
+ private:
+  // Scope stores the state of where we are when visiting the AST and a map of
+  // names to symbols.
+  struct Scope {
+    bool in_pattern{false};
+    bool in_merge{false};
+    bool in_create{false};
+    // in_create_node is true if we are creating or merging *only* a node.
+    // Therefore, it is *not* equivalent to (in_create || in_merge) &&
+    // in_node_atom.
+    bool in_create_node{false};
+    // True if creating an edge;
+    // shortcut for (in_create || in_merge) && visiting_edge.
+    bool in_create_edge{false};
+    bool in_node_atom{false};
+    EdgeAtom *visiting_edge{nullptr};
+    bool in_aggregation{false};
+    bool in_return{false};
+    bool in_with{false};
+    bool in_skip{false};
+    bool in_limit{false};
+    bool in_order_by{false};
+    bool in_where{false};
+    bool in_match{false};
+    bool in_foreach{false};
+    // True when visiting a pattern atom (node or edge) identifier, which can be
+    // reused or created in the pattern itself.
+    bool in_pattern_atom_identifier{false};
+    // True when visiting range bounds of a variable path.
+    bool in_edge_range{false};
+    // True if the return/with contains an aggregation in any named expression.
+    bool has_aggregation{false};
+    // Map from variable names to symbols.
+    std::map<std::string, Symbol> symbols;
+    // Identifiers found in property maps of patterns or as variable length path
+    // bounds in a single Match clause. They need to be checked after visiting
+    // Match. Identifiers created by naming vertices, edges and paths are *not*
+    // stored in here.
+    std::vector<Identifier *> identifiers_in_match;
+    // Number of nested IfOperators.
+    int num_if_operators{0};
+  };
+
+  inline static std::optional<Symbol> FindSymbolInScope(const std::string &name, const Scope &scope,
+                                                        Symbol::Type type) {
+    if (auto it = scope.symbols.find(name); it != scope.symbols.end()) {
+      const auto &symbol = it->second;
+      // Unless we have `ANY` type, check that types match.
+      if (type != Symbol::Type::ANY && symbol.type() != Symbol::Type::ANY && type != symbol.type()) {
+        throw TypeMismatchError(name, Symbol::TypeToString(symbol.type()), Symbol::TypeToString(type));
+      }
+      return symbol;
+    }
+    return std::nullopt;
+  }
+
+  bool HasSymbol(const std::string &name) const {
+    return std::ranges::any_of(scopes_, [&name](const auto &scope) { return scope.symbols.contains(name); });
+  }
+
+  bool HasSymbolLocalScope(const std::string &name) const { return scopes_.back().symbols.contains(name); }
+
+  // @return true if it added a predefined identifier with that name
+  bool ConsumePredefinedIdentifier(const std::string &name) {
+    auto it = predefined_identifiers_.find(name);
+
+    if (it == predefined_identifiers_.end()) {
+      return false;
+    }
+
+    // we can only use the predefined identifier in a single scope so we remove it after creating
+    // a symbol for it
+    auto &identifier = it->second;
+    MG_ASSERT(!identifier->user_declared_, "Predefined symbols cannot be user declared!");
+    identifier->MapTo(CreateSymbol(identifier->name_, identifier->user_declared_));
+    predefined_identifiers_.erase(it);
+    return true;
+  }
+
+  // Returns a freshly generated symbol. Previous mapping of the same name to a
+  // different symbol is replaced with the new one.
+  Symbol CreateSymbol(const std::string &name, bool user_declared, Symbol::Type type = Symbol::Type::ANY,
+                      int token_position = -1) {
+    auto symbol = symbol_table_->CreateSymbol(name, user_declared, type, token_position);
+    scopes_.back().symbols[name] = symbol;
+    return symbol;
+  }
+
+  Symbol GetOrCreateSymbol(const std::string &name, bool user_declared, Symbol::Type type = Symbol::Type::ANY) {
+    // NOLINTNEXTLINE
+    for (auto scope = scopes_.rbegin(); scope != scopes_.rend(); ++scope) {
+      if (auto maybe_symbol = FindSymbolInScope(name, *scope, type); maybe_symbol) {
+        return *maybe_symbol;
+      }
+    }
+    return CreateSymbol(name, user_declared, type);
+  }
+
+  // Returns the symbol by name. If the mapping already exists, checks if the
+  // types match. Otherwise, returns a new symbol.
+  Symbol GetOrCreateSymbolLocalScope(const std::string &name, bool user_declared,
+                                     Symbol::Type type = Symbol::Type::ANY) {
+    auto &scope = scopes_.back();
+    if (auto maybe_symbol = FindSymbolInScope(name, scope, type); maybe_symbol) {
+      return *maybe_symbol;
+    }
+    return CreateSymbol(name, user_declared, type);
+  }
+
+  void VisitReturnBody(ReturnBody &body, Where *where = nullptr) {
+    auto &scope = scopes_.back();
+    for (auto &expr : body.named_expressions) {
+      expr->Accept(*this);
+    }
+    std::vector<Symbol> user_symbols;
+    if (body.all_identifiers) {
+      // Carry over user symbols because '*' appeared.
+      for (const auto &sym_pair : scope.symbols) {
+        if (!sym_pair.second.user_declared()) {
+          continue;
+        }
+        user_symbols.emplace_back(sym_pair.second);
+      }
+      if (user_symbols.empty()) {
+        throw SemanticException("There are no variables in scope to use for '*'.");
+      }
+    }
+    // WITH/RETURN clause removes declarations of all the previous variables and
+    // declares only those established through named expressions. New declarations
+    // must not be visible inside named expressions themselves.
+    bool removed_old_names = false;
+    if ((!where && body.order_by.empty()) || scope.has_aggregation) {
+      // WHERE and ORDER BY need to see both the old and new symbols, unless we
+      // have an aggregation. Therefore, we can clear the symbols immediately if
+      // there is neither ORDER BY nor WHERE, or we have an aggregation.
+      scope.symbols.clear();
+      removed_old_names = true;
+    }
+    // Create symbols for named expressions.
+    std::unordered_set<std::string> new_names;
+    for (const auto &user_sym : user_symbols) {
+      new_names.insert(user_sym.name());
+      scope.symbols[user_sym.name()] = user_sym;
+    }
+    for (auto &named_expr : body.named_expressions) {
+      const auto &name = named_expr->name_;
+      if (!new_names.insert(name).second) {
+        throw SemanticException("Multiple results with the same name '{}' are not allowed.", name);
+      }
+      // An improvement would be to infer the type of the expression, so that the
+      // new symbol would have a more specific type.
+      named_expr->MapTo(CreateSymbol(name, true, Symbol::Type::ANY, named_expr->token_position_));
+    }
+    scope.in_order_by = true;
+    for (const auto &order_pair : body.order_by) {
+      order_pair.expression->Accept(*this);
+    }
+    scope.in_order_by = false;
+    if (body.skip) {
+      scope.in_skip = true;
+      body.skip->Accept(*this);
+      scope.in_skip = false;
+    }
+    if (body.limit) {
+      scope.in_limit = true;
+      body.limit->Accept(*this);
+      scope.in_limit = false;
+    }
+    if (where) where->Accept(*this);
+    if (!removed_old_names) {
+      // We have an ORDER BY or WHERE, but no aggregation, which means we didn't
+      // clear the old symbols, so do it now. We cannot just call clear, because
+      // we've added new symbols.
+      for (auto sym_it = scope.symbols.begin(); sym_it != scope.symbols.end();) {
+        if (new_names.find(sym_it->first) == new_names.end()) {
+          sym_it = scope.symbols.erase(sym_it);
+        } else {
+          sym_it++;
+        }
+      }
+    }
+    scopes_.back().has_aggregation = false;
+  }
+
+  void VisitWithIdentifiers(Expression *expr, const std::vector<Identifier *> &identifiers) {
+    auto &scope = scopes_.back();
+    std::vector<std::pair<std::optional<Symbol>, Identifier *>> prev_symbols;
+    // Collect previous symbols if they exist.
+    for (const auto &identifier : identifiers) {
+      std::optional<Symbol> prev_symbol;
+      auto prev_symbol_it = scope.symbols.find(identifier->name_);
+      if (prev_symbol_it != scope.symbols.end()) {
+        prev_symbol = prev_symbol_it->second;
+      }
+      identifier->MapTo(CreateSymbol(identifier->name_, identifier->user_declared_));
+      prev_symbols.emplace_back(prev_symbol, identifier);
+    }
+    // Visit the expression with the new symbols bound.
+    expr->Accept(*this);
+    // Restore back to previous symbols.
+    for (const auto &prev : prev_symbols) {
+      const auto &prev_symbol = prev.first;
+      const auto &identifier = prev.second;
+      if (prev_symbol) {
+        scope.symbols[identifier->name_] = *prev_symbol;
+      } else {
+        scope.symbols.erase(identifier->name_);
+      }
+    }
+  }
+
+  SymbolTable *symbol_table_;
+
+  // Identifiers which are injected from outside the query. Each identifier
+  // is mapped by its name.
+  std::unordered_map<std::string, Identifier *> predefined_identifiers_;
+  std::vector<Scope> scopes_;
+  std::unordered_set<std::string> prev_return_names_;
+  std::unordered_set<std::string> curr_return_names_;
+};
+
+inline SymbolTable MakeSymbolTable(CypherQuery *query, const std::vector<Identifier *> &predefined_identifiers = {}) {
+  SymbolTable symbol_table;
+  SymbolGenerator symbol_generator(&symbol_table, predefined_identifiers);
+  query->single_query_->Accept(symbol_generator);
+  for (auto *cypher_union : query->cypher_unions_) {
+    cypher_union->Accept(symbol_generator);
+  }
+  return symbol_table;
+}
+
+}  // namespace memgraph::expr

src/expr/semantic/symbol_table.hpp

@@ -0,0 +1,64 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include <map>
+#include <string>
+
+#include "expr/ast.hpp"
+#include "expr/semantic/symbol.hpp"
+#include "utils/logging.hpp"
+
+namespace memgraph::expr {
+
+class SymbolTable final {
+ public:
+  SymbolTable() {}
+  const Symbol &CreateSymbol(const std::string &name, bool user_declared, Symbol::Type type = Symbol::Type::ANY,
+                             int32_t token_position = -1) {
+    MG_ASSERT(table_.size() <= std::numeric_limits<int32_t>::max(),
+              "SymbolTable size doesn't fit into 32-bit integer!");
+    auto got = table_.emplace(position_, Symbol(name, position_, user_declared, type, token_position));
+    MG_ASSERT(got.second, "Duplicate symbol ID!");
+    position_++;
+    return got.first->second;
+  }
+
+  // TODO(buda): This is the same logic as in the cypher_main_visitor. During
+  // parsing phase symbol table doesn't exist. Figure out a better solution.
+  const Symbol &CreateAnonymousSymbol(Symbol::Type type = Symbol::Type::ANY) {
+    int id = 1;
+    while (true) {
+      static const std::string &kAnonPrefix = "anon";
+      std::string name_candidate = kAnonPrefix + std::to_string(id++);
+      if (std::find_if(std::begin(table_), std::end(table_), [&name_candidate](const auto &item) -> bool {
+            return item.second.name_ == name_candidate;
+          }) == std::end(table_)) {
+        return CreateSymbol(name_candidate, false, type);
+      }
+    }
+  }
+
+  const Symbol &at(const Identifier &ident) const { return table_.at(ident.symbol_pos_); }
+  const Symbol &at(const NamedExpression &nexpr) const { return table_.at(nexpr.symbol_pos_); }
+  const Symbol &at(const Aggregation &aggr) const { return table_.at(aggr.symbol_pos_); }
+
+  // TODO: Remove these since members are public
+  int32_t max_position() const { return static_cast<int32_t>(table_.size()); }
+
+  const auto &table() const { return table_; }
+
+  int32_t position_{0};
+  std::map<int32_t, Symbol> table_;
+};
+
+}  // namespace memgraph::expr

src/glue/v2/auth.cpp

@@ -0,0 +1,64 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#include "glue/v2/auth.hpp"
+
+namespace memgraph::glue::v2 {
+
+auth::Permission PrivilegeToPermission(query::v2::AuthQuery::Privilege privilege) {
+  switch (privilege) {
+    case query::v2::AuthQuery::Privilege::MATCH:
+      return auth::Permission::MATCH;
+    case query::v2::AuthQuery::Privilege::CREATE:
+      return auth::Permission::CREATE;
+    case query::v2::AuthQuery::Privilege::MERGE:
+      return auth::Permission::MERGE;
+    case query::v2::AuthQuery::Privilege::DELETE:
+      return auth::Permission::DELETE;
+    case query::v2::AuthQuery::Privilege::SET:
+      return auth::Permission::SET;
+    case query::v2::AuthQuery::Privilege::REMOVE:
+      return auth::Permission::REMOVE;
+    case query::v2::AuthQuery::Privilege::INDEX:
+      return auth::Permission::INDEX;
+    case query::v2::AuthQuery::Privilege::STATS:
+      return auth::Permission::STATS;
+    case query::v2::AuthQuery::Privilege::CONSTRAINT:
+      return auth::Permission::CONSTRAINT;
+    case query::v2::AuthQuery::Privilege::DUMP:
+      return auth::Permission::DUMP;
+    case query::v2::AuthQuery::Privilege::REPLICATION:
+      return auth::Permission::REPLICATION;
+    case query::v2::AuthQuery::Privilege::DURABILITY:
+      return auth::Permission::DURABILITY;
+    case query::v2::AuthQuery::Privilege::READ_FILE:
+      return auth::Permission::READ_FILE;
+    case query::v2::AuthQuery::Privilege::FREE_MEMORY:
+      return auth::Permission::FREE_MEMORY;
+    case query::v2::AuthQuery::Privilege::TRIGGER:
+      return auth::Permission::TRIGGER;
+    case query::v2::AuthQuery::Privilege::CONFIG:
+      return auth::Permission::CONFIG;
+    case query::v2::AuthQuery::Privilege::AUTH:
+      return auth::Permission::AUTH;
+    case query::v2::AuthQuery::Privilege::STREAM:
+      return auth::Permission::STREAM;
+    case query::v2::AuthQuery::Privilege::MODULE_READ:
+      return auth::Permission::MODULE_READ;
+    case query::v2::AuthQuery::Privilege::MODULE_WRITE:
+      return auth::Permission::MODULE_WRITE;
+    case query::v2::AuthQuery::Privilege::WEBSOCKET:
+      return auth::Permission::WEBSOCKET;
+    case query::v2::AuthQuery::Privilege::SCHEMA:
+      return auth::Permission::SCHEMA;
+  }
+}
+}  // namespace memgraph::glue::v2

src/glue/v2/auth.hpp

@@ -0,0 +1,23 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#include "auth/models.hpp"
+#include "query/v2/frontend/ast/ast.hpp"
+
+namespace memgraph::glue::v2 {
+
+/**
+ * This function converts query::AuthQuery::Privilege to its corresponding
+ * auth::Permission.
+ */
+auth::Permission PrivilegeToPermission(query::v2::AuthQuery::Privilege privilege);
+
+}  // namespace memgraph::glue::v2

src/glue/v2/communication.cpp

@@ -0,0 +1,277 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#include "glue/v2/communication.hpp"
+
+#include <map>
+#include <string>
+#include <vector>
+
+#include "storage/v3/edge_accessor.hpp"
+#include "storage/v3/shard.hpp"
+#include "storage/v3/vertex_accessor.hpp"
+#include "utils/temporal.hpp"
+
+using memgraph::communication::bolt::Value;
+
+namespace memgraph::glue::v2 {
+
+query::v2::TypedValue ToTypedValue(const Value &value) {
+  switch (value.type()) {
+    case Value::Type::Null:
+      return {};
+    case Value::Type::Bool:
+      return query::v2::TypedValue(value.ValueBool());
+    case Value::Type::Int:
+      return query::v2::TypedValue(value.ValueInt());
+    case Value::Type::Double:
+      return query::v2::TypedValue(value.ValueDouble());
+    case Value::Type::String:
+      return query::v2::TypedValue(value.ValueString());
+    case Value::Type::List: {
+      std::vector<query::v2::TypedValue> list;
+      list.reserve(value.ValueList().size());
+      for (const auto &v : value.ValueList()) list.push_back(ToTypedValue(v));
+      return query::v2::TypedValue(std::move(list));
+    }
+    case Value::Type::Map: {
+      std::map<std::string, query::v2::TypedValue> map;
+      for (const auto &kv : value.ValueMap()) map.emplace(kv.first, ToTypedValue(kv.second));
+      return query::v2::TypedValue(std::move(map));
+    }
+    case Value::Type::Vertex:
+    case Value::Type::Edge:
+    case Value::Type::UnboundedEdge:
+    case Value::Type::Path:
+      throw communication::bolt::ValueException("Unsupported conversion from Value to TypedValue");
+    case Value::Type::Date:
+      return query::v2::TypedValue(value.ValueDate());
+    case Value::Type::LocalTime:
+      return query::v2::TypedValue(value.ValueLocalTime());
+    case Value::Type::LocalDateTime:
+      return query::v2::TypedValue(value.ValueLocalDateTime());
+    case Value::Type::Duration:
+      return query::v2::TypedValue(value.ValueDuration());
+  }
+}
+
+storage::v3::Result<communication::bolt::Vertex> ToBoltVertex(const query::v2::VertexAccessor &vertex,
+                                                              const storage::v3::Shard &db, storage::v3::View view) {
+  return ToBoltVertex(vertex.impl_, db, view);
+}
+
+storage::v3::Result<communication::bolt::Edge> ToBoltEdge(const query::v2::EdgeAccessor &edge,
+                                                          const storage::v3::Shard &db, storage::v3::View view) {
+  return ToBoltEdge(edge.impl_, db, view);
+}
+
+storage::v3::Result<Value> ToBoltValue(const query::v2::TypedValue &value, const storage::v3::Shard &db,
+                                       storage::v3::View view) {
+  switch (value.type()) {
+    case query::v2::TypedValue::Type::Null:
+      return Value();
+    case query::v2::TypedValue::Type::Bool:
+      return Value(value.ValueBool());
+    case query::v2::TypedValue::Type::Int:
+      return Value(value.ValueInt());
+    case query::v2::TypedValue::Type::Double:
+      return Value(value.ValueDouble());
+    case query::v2::TypedValue::Type::String:
+      return Value(std::string(value.ValueString()));
+    case query::v2::TypedValue::Type::List: {
+      std::vector<Value> values;
+      values.reserve(value.ValueList().size());
+      for (const auto &v : value.ValueList()) {
+        auto maybe_value = ToBoltValue(v, db, view);
+        if (maybe_value.HasError()) return maybe_value.GetError();
+        values.emplace_back(std::move(*maybe_value));
+      }
+      return Value(std::move(values));
+    }
+    case query::v2::TypedValue::Type::Map: {
+      std::map<std::string, Value> map;
+      for (const auto &kv : value.ValueMap()) {
+        auto maybe_value = ToBoltValue(kv.second, db, view);
+        if (maybe_value.HasError()) return maybe_value.GetError();
+        map.emplace(kv.first, std::move(*maybe_value));
+      }
+      return Value(std::move(map));
+    }
+    case query::v2::TypedValue::Type::Vertex: {
+      auto maybe_vertex = ToBoltVertex(value.ValueVertex(), db, view);
+      if (maybe_vertex.HasError()) return maybe_vertex.GetError();
+      return Value(std::move(*maybe_vertex));
+    }
+    case query::v2::TypedValue::Type::Edge: {
+      auto maybe_edge = ToBoltEdge(value.ValueEdge(), db, view);
+      if (maybe_edge.HasError()) return maybe_edge.GetError();
+      return Value(std::move(*maybe_edge));
+    }
+    case query::v2::TypedValue::Type::Path: {
+      auto maybe_path = ToBoltPath(value.ValuePath(), db, view);
+      if (maybe_path.HasError()) return maybe_path.GetError();
+      return Value(std::move(*maybe_path));
+    }
+    case query::v2::TypedValue::Type::Date:
+      return Value(value.ValueDate());
+    case query::v2::TypedValue::Type::LocalTime:
+      return Value(value.ValueLocalTime());
+    case query::v2::TypedValue::Type::LocalDateTime:
+      return Value(value.ValueLocalDateTime());
+    case query::v2::TypedValue::Type::Duration:
+      return Value(value.ValueDuration());
+  }
+}
+
+storage::v3::Result<communication::bolt::Vertex> ToBoltVertex(const storage::v3::VertexAccessor &vertex,
+                                                              const storage::v3::Shard &db, storage::v3::View view) {
+  // TODO(jbajic) Fix bolt communication
+  auto id = communication::bolt::Id::FromUint(0);
+  auto maybe_labels = vertex.Labels(view);
+  if (maybe_labels.HasError()) return maybe_labels.GetError();
+  std::vector<std::string> labels;
+  labels.reserve(maybe_labels->size());
+  for (const auto &label : *maybe_labels) {
+    labels.push_back(db.LabelToName(label));
+  }
+  auto maybe_properties = vertex.Properties(view);
+  if (maybe_properties.HasError()) return maybe_properties.GetError();
+  std::map<std::string, Value> properties;
+  for (const auto &prop : *maybe_properties) {
+    properties[db.PropertyToName(prop.first)] = ToBoltValue(prop.second);
+  }
+  return communication::bolt::Vertex{id, labels, properties};
+}
+
+storage::v3::Result<communication::bolt::Edge> ToBoltEdge(const storage::v3::EdgeAccessor &edge,
+                                                          const storage::v3::Shard &db, storage::v3::View view) {
+  // TODO(jbajic) Fix bolt communication
+  auto id = communication::bolt::Id::FromUint(0);
+  auto from = communication::bolt::Id::FromUint(0);
+  auto to = communication::bolt::Id::FromUint(0);
+  const auto &type = db.EdgeTypeToName(edge.EdgeType());
+  auto maybe_properties = edge.Properties(view);
+  if (maybe_properties.HasError()) return maybe_properties.GetError();
+  std::map<std::string, Value> properties;
+  for (const auto &prop : *maybe_properties) {
+    properties[db.PropertyToName(prop.first)] = ToBoltValue(prop.second);
+  }
+  return communication::bolt::Edge{id, from, to, type, properties};
+}
+
+storage::v3::Result<communication::bolt::Path> ToBoltPath(const query::v2::Path &path, const storage::v3::Shard &db,
+                                                          storage::v3::View view) {
+  std::vector<communication::bolt::Vertex> vertices;
+  vertices.reserve(path.vertices().size());
+  for (const auto &v : path.vertices()) {
+    auto maybe_vertex = ToBoltVertex(v, db, view);
+    if (maybe_vertex.HasError()) return maybe_vertex.GetError();
+    vertices.emplace_back(std::move(*maybe_vertex));
+  }
+  std::vector<communication::bolt::Edge> edges;
+  edges.reserve(path.edges().size());
+  for (const auto &e : path.edges()) {
+    auto maybe_edge = ToBoltEdge(e, db, view);
+    if (maybe_edge.HasError()) return maybe_edge.GetError();
+    edges.emplace_back(std::move(*maybe_edge));
+  }
+  return communication::bolt::Path(vertices, edges);
+}
+
+storage::v3::PropertyValue ToPropertyValue(const Value &value) {
+  switch (value.type()) {
+    case Value::Type::Null:
+      return {};
+    case Value::Type::Bool:
+      return storage::v3::PropertyValue(value.ValueBool());
+    case Value::Type::Int:
+      return storage::v3::PropertyValue(value.ValueInt());
+    case Value::Type::Double:
+      return storage::v3::PropertyValue(value.ValueDouble());
+    case Value::Type::String:
+      return storage::v3::PropertyValue(value.ValueString());
+    case Value::Type::List: {
+      std::vector<storage::v3::PropertyValue> vec;
+      vec.reserve(value.ValueList().size());
+      for (const auto &value : value.ValueList()) vec.emplace_back(ToPropertyValue(value));
+      return storage::v3::PropertyValue(std::move(vec));
+    }
+    case Value::Type::Map: {
+      std::map<std::string, storage::v3::PropertyValue> map;
+      for (const auto &kv : value.ValueMap()) map.emplace(kv.first, ToPropertyValue(kv.second));
+      return storage::v3::PropertyValue(std::move(map));
+    }
+    case Value::Type::Vertex:
+    case Value::Type::Edge:
+    case Value::Type::UnboundedEdge:
+    case Value::Type::Path:
+      throw communication::bolt::ValueException("Unsupported conversion from Value to PropertyValue");
+    case Value::Type::Date:
+      return storage::v3::PropertyValue(
+          storage::v3::TemporalData(storage::v3::TemporalType::Date, value.ValueDate().MicrosecondsSinceEpoch()));
+    case Value::Type::LocalTime:
+      return storage::v3::PropertyValue(storage::v3::TemporalData(storage::v3::TemporalType::LocalTime,
+                                                                  value.ValueLocalTime().MicrosecondsSinceEpoch()));
+    case Value::Type::LocalDateTime:
+      return storage::v3::PropertyValue(storage::v3::TemporalData(storage::v3::TemporalType::LocalDateTime,
+                                                                  value.ValueLocalDateTime().MicrosecondsSinceEpoch()));
+    case Value::Type::Duration:
+      return storage::v3::PropertyValue(
+          storage::v3::TemporalData(storage::v3::TemporalType::Duration, value.ValueDuration().microseconds));
+  }
+}
+
+Value ToBoltValue(const storage::v3::PropertyValue &value) {
+  switch (value.type()) {
+    case storage::v3::PropertyValue::Type::Null:
+      return {};
+    case storage::v3::PropertyValue::Type::Bool:
+      return {value.ValueBool()};
+    case storage::v3::PropertyValue::Type::Int:
+      return {value.ValueInt()};
+      break;
+    case storage::v3::PropertyValue::Type::Double:
+      return {value.ValueDouble()};
+    case storage::v3::PropertyValue::Type::String:
+      return {value.ValueString()};
+    case storage::v3::PropertyValue::Type::List: {
+      const auto &values = value.ValueList();
+      std::vector<Value> vec;
+      vec.reserve(values.size());
+      for (const auto &v : values) {
+        vec.push_back(ToBoltValue(v));
+      }
+      return {std::move(vec)};
+    }
+    case storage::v3::PropertyValue::Type::Map: {
+      const auto &map = value.ValueMap();
+      std::map<std::string, Value> dv_map;
+      for (const auto &kv : map) {
+        dv_map.emplace(kv.first, ToBoltValue(kv.second));
+      }
+      return {std::move(dv_map)};
+    }
+    case storage::v3::PropertyValue::Type::TemporalData:
+      const auto &type = value.ValueTemporalData();
+      switch (type.type) {
+        case storage::v3::TemporalType::Date:
+          return {utils::Date(type.microseconds)};
+        case storage::v3::TemporalType::LocalTime:
+          return {utils::LocalTime(type.microseconds)};
+        case storage::v3::TemporalType::LocalDateTime:
+          return {utils::LocalDateTime(type.microseconds)};
+        case storage::v3::TemporalType::Duration:
+          return {utils::Duration(type.microseconds)};
+      }
+  }
+}
+
+}  // namespace memgraph::glue::v2

src/glue/v2/communication.hpp

@@ -0,0 +1,68 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+/// @file Conversion functions between Value and other memgraph types.
+#pragma once
+
+#include "communication/bolt/v1/value.hpp"
+#include "query/v2/bindings/typed_value.hpp"
+#include "storage/v3/property_value.hpp"
+#include "storage/v3/result.hpp"
+#include "storage/v3/view.hpp"
+
+namespace memgraph::storage::v3 {
+class EdgeAccessor;
+class Storage;
+class VertexAccessor;
+}  // namespace memgraph::storage::v3
+
+namespace memgraph::glue::v2 {
+
+/// @param storage::v3::VertexAccessor for converting to
+///        communication::bolt::Vertex.
+/// @param storage::v3::Shard for getting label and property names.
+/// @param storage::v3::View for deciding which vertex attributes are visible.
+///
+/// @throw std::bad_alloc
+storage::v3::Result<communication::bolt::Vertex> ToBoltVertex(const storage::v3::VertexAccessor &vertex,
+                                                              const storage::v3::Shard &db, storage::v3::View view);
+
+/// @param storage::v3::EdgeAccessor for converting to communication::bolt::Edge.
+/// @param storage::v3::Shard for getting edge type and property names.
+/// @param storage::v3::View for deciding which edge attributes are visible.
+///
+/// @throw std::bad_alloc
+storage::v3::Result<communication::bolt::Edge> ToBoltEdge(const storage::v3::EdgeAccessor &edge,
+                                                          const storage::v3::Shard &db, storage::v3::View view);
+
+/// @param query::v2::Path for converting to communication::bolt::Path.
+/// @param storage::v3::Shard for ToBoltVertex and ToBoltEdge.
+/// @param storage::v3::View for ToBoltVertex and ToBoltEdge.
+///
+/// @throw std::bad_alloc
+storage::v3::Result<communication::bolt::Path> ToBoltPath(const query::v2::Path &path, const storage::v3::Shard &db,
+                                                          storage::v3::View view);
+
+/// @param query::v2::TypedValue for converting to communication::bolt::Value.
+/// @param storage::v3::Shard for ToBoltVertex and ToBoltEdge.
+/// @param storage::v3::View for ToBoltVertex and ToBoltEdge.
+///
+/// @throw std::bad_alloc
+storage::v3::Result<communication::bolt::Value> ToBoltValue(const query::v2::TypedValue &value,
+                                                            const storage::v3::Shard &db, storage::v3::View view);
+
+query::v2::TypedValue ToTypedValue(const communication::bolt::Value &value);
+
+communication::bolt::Value ToBoltValue(const storage::v3::PropertyValue &value);
+
+storage::v3::PropertyValue ToPropertyValue(const communication::bolt::Value &value);
+
+}  // namespace memgraph::glue::v2

src/integrations/constants.hpp

@@ -17,6 +17,7 @@
 namespace memgraph::integrations {
 
 inline constexpr int64_t kDefaultCheckBatchLimit{1};
+inline constexpr int64_t kMinimumStartBatchLimit{1};
 inline constexpr std::chrono::milliseconds kDefaultCheckTimeout{30000};
 inline constexpr std::chrono::milliseconds kMinimumInterval{1};
 inline constexpr int64_t kMinimumSize{1};

src/integrations/kafka/consumer.cpp

@@ -74,6 +74,36 @@ utils::BasicResult<std::string, std::vector<Message>> GetBatch(RdKafka::KafkaCon
 
   return std::move(batch);
 }
+
+void CheckAndDestroyLastAssignmentIfNeeded(RdKafka::KafkaConsumer &consumer, const ConsumerInfo &info,
+                                           std::vector<RdKafka::TopicPartition *> &last_assignment) {
+  if (!last_assignment.empty()) {
+    if (const auto err = consumer.assign(last_assignment); err != RdKafka::ERR_NO_ERROR) {
+      throw ConsumerStartFailedException(info.consumer_name,
+                                         fmt::format("Couldn't restore commited offsets: '{}'", RdKafka::err2str(err)));
+    }
+    RdKafka::TopicPartition::destroy(last_assignment);
+  }
+}
+
+void TryToConsumeBatch(RdKafka::KafkaConsumer &consumer, const ConsumerInfo &info,
+                       const ConsumerFunction &consumer_function, const std::vector<Message> &batch) {
+  consumer_function(batch);
+  std::vector<RdKafka::TopicPartition *> partitions;
+  utils::OnScopeExit clear_partitions([&]() { RdKafka::TopicPartition::destroy(partitions); });
+
+  if (const auto err = consumer.assignment(partitions); err != RdKafka::ERR_NO_ERROR) {
+    throw ConsumerCommitFailedException(
+        info.consumer_name, fmt::format("Couldn't get assignment to commit offsets: {}", RdKafka::err2str(err)));
+  }
+  if (const auto err = consumer.position(partitions); err != RdKafka::ERR_NO_ERROR) {
+    throw ConsumerCommitFailedException(info.consumer_name,
+                                        fmt::format("Couldn't get offsets from librdkafka {}", RdKafka::err2str(err)));
+  }
+  if (const auto err = consumer.commitSync(partitions); err != RdKafka::ERR_NO_ERROR) {
+    throw ConsumerCommitFailedException(info.consumer_name, RdKafka::err2str(err));
+  }
+}
 }  // namespace
 
 Message::Message(std::unique_ptr<RdKafka::Message> &&message) : message_{std::move(message)} {
@@ -221,10 +251,21 @@ void Consumer::Start() {
   StartConsuming();
 }
 
-void Consumer::StartIfStopped() {
-  if (!is_running_) {
-    StartConsuming();
+void Consumer::StartWithLimit(const uint64_t limit_batches, std::optional<std::chrono::milliseconds> timeout) const {
+  if (is_running_) {
+    throw ConsumerRunningException(info_.consumer_name);
   }
+  if (limit_batches < kMinimumStartBatchLimit) {
+    throw ConsumerStartFailedException(
+        info_.consumer_name, fmt::format("Batch limit has to be greater than or equal to {}", kMinimumStartBatchLimit));
+  }
+  if (timeout.value_or(kMinimumInterval) < kMinimumInterval) {
+    throw ConsumerStartFailedException(
+        info_.consumer_name,
+        fmt::format("Timeout has to be greater than or equal to {} milliseconds", kMinimumInterval.count()));
+  }
+
+  StartConsumingWithLimit(limit_batches, timeout);
 }
 
 void Consumer::Stop() {
@@ -244,7 +285,7 @@ void Consumer::StopIfRunning() {
   }
 }
 
-void Consumer::Check(std::optional<std::chrono::milliseconds> timeout, std::optional<int64_t> limit_batches,
+void Consumer::Check(std::optional<std::chrono::milliseconds> timeout, std::optional<uint64_t> limit_batches,
                      const ConsumerFunction &check_consumer_function) const {
   // NOLINTNEXTLINE (modernize-use-nullptr)
   if (timeout.value_or(kMinimumInterval) < kMinimumInterval) {
@@ -344,13 +385,7 @@ void Consumer::StartConsuming() {
 
   is_running_.store(true);
 
-  if (!last_assignment_.empty()) {
-    if (const auto err = consumer_->assign(last_assignment_); err != RdKafka::ERR_NO_ERROR) {
-      throw ConsumerStartFailedException(info_.consumer_name,
-                                         fmt::format("Couldn't restore commited offsets: '{}'", RdKafka::err2str(err)));
-    }
-    RdKafka::TopicPartition::destroy(last_assignment_);
-  }
+  CheckAndDestroyLastAssignmentIfNeeded(*consumer_, info_, last_assignment_);
 
   thread_ = std::thread([this] {
     static constexpr auto kMaxThreadNameSize = utils::GetMaxThreadNameSize();
@@ -361,33 +396,18 @@ void Consumer::StartConsuming() {
     while (is_running_) {
       auto maybe_batch = GetBatch(*consumer_, info_, is_running_);
       if (maybe_batch.HasError()) {
-        spdlog::warn("Error happened in consumer {} while fetching messages: {}!", info_.consumer_name,
-                     maybe_batch.GetError());
-        break;
+        throw ConsumerReadMessagesFailedException(info_.consumer_name, maybe_batch.GetError());
       }
       const auto &batch = maybe_batch.GetValue();
 
-      if (batch.empty()) continue;
+      if (batch.empty()) {
+        continue;
+      }
 
       spdlog::info("Kafka consumer {} is processing a batch", info_.consumer_name);
 
       try {
-        consumer_function_(batch);
-        std::vector<RdKafka::TopicPartition *> partitions;
-        utils::OnScopeExit clear_partitions([&]() { RdKafka::TopicPartition::destroy(partitions); });
-
-        if (const auto err = consumer_->assignment(partitions); err != RdKafka::ERR_NO_ERROR) {
-          throw ConsumerCheckFailedException(
-              info_.consumer_name, fmt::format("Couldn't get assignment to commit offsets: {}", RdKafka::err2str(err)));
-        }
-        if (const auto err = consumer_->position(partitions); err != RdKafka::ERR_NO_ERROR) {
-          throw ConsumerCheckFailedException(
-              info_.consumer_name, fmt::format("Couldn't get offsets from librdkafka {}", RdKafka::err2str(err)));
-        }
-        if (const auto err = consumer_->commitSync(partitions); err != RdKafka::ERR_NO_ERROR) {
-          spdlog::warn("Committing offset of consumer {} failed: {}", info_.consumer_name, RdKafka::err2str(err));
-          break;
-        }
+        TryToConsumeBatch(*consumer_, info_, consumer_function_, batch);
       } catch (const std::exception &e) {
         spdlog::warn("Error happened in consumer {} while processing a batch: {}!", info_.consumer_name, e.what());
         break;
@@ -398,6 +418,44 @@ void Consumer::StartConsuming() {
   });
 }
 
+void Consumer::StartConsumingWithLimit(uint64_t limit_batches, std::optional<std::chrono::milliseconds> timeout) const {
+  MG_ASSERT(!is_running_, "Cannot start already running consumer!");
+
+  if (is_running_.exchange(true)) {
+    throw ConsumerRunningException(info_.consumer_name);
+  }
+  utils::OnScopeExit restore_is_running([this] { is_running_.store(false); });
+
+  CheckAndDestroyLastAssignmentIfNeeded(*consumer_, info_, last_assignment_);
+
+  const auto timeout_to_use = timeout.value_or(kDefaultCheckTimeout);
+  const auto start = std::chrono::steady_clock::now();
+
+  for (uint64_t batch_count = 0; batch_count < limit_batches;) {
+    const auto now = std::chrono::steady_clock::now();
+    if (now - start >= timeout_to_use) {
+      throw ConsumerStartFailedException(info_.consumer_name, "Timeout reached");
+    }
+
+    const auto maybe_batch = GetBatch(*consumer_, info_, is_running_);
+    if (maybe_batch.HasError()) {
+      throw ConsumerReadMessagesFailedException(info_.consumer_name, maybe_batch.GetError());
+    }
+    const auto &batch = maybe_batch.GetValue();
+
+    if (batch.empty()) {
+      continue;
+    }
+    ++batch_count;
+
+    spdlog::info("Kafka consumer {} is processing a batch", info_.consumer_name);
+
+    TryToConsumeBatch(*consumer_, info_, consumer_function_, batch);
+
+    spdlog::info("Kafka consumer {} finished processing", info_.consumer_name);
+  }
+}
+
 void Consumer::StopConsuming() {
   is_running_.store(false);
   if (thread_.joinable()) thread_.join();

src/integrations/kafka/consumer.hpp

@@ -113,11 +113,19 @@ class Consumer final : public RdKafka::EventCb {
   /// This method will start a new thread which will poll all the topics for messages.
   ///
   /// @throws ConsumerRunningException if the consumer is already running
+  /// @throws ConsumerStartFailedException if the commited offsets cannot be restored
   void Start();
 
-  /// Starts consuming messages if it is not started already.
+  /// Starts consuming messages.
   ///
-  void StartIfStopped();
+  /// This method will start a new thread which will poll all the topics for messages.
+  ///
+  /// @param limit_batches the consumer will only consume the given number of batches.
+  /// @param timeout the maximum duration during which the command should run.
+  ///
+  /// @throws ConsumerRunningException if the consumer is already running
+  /// @throws ConsumerStartFailedException if the commited offsets cannot be restored
+  void StartWithLimit(uint64_t limit_batches, std::optional<std::chrono::milliseconds> timeout) const;
 
   /// Stops consuming messages.
   ///
@@ -136,9 +144,9 @@ class Consumer final : public RdKafka::EventCb {
   ///                      used.
   /// @param check_consumer_function a function to feed the received messages in, only used during this dry-run.
   ///
-  /// @throws ConsumerRunningException if the consumer is alredy running.
+  /// @throws ConsumerRunningException if the consumer is already running.
   /// @throws ConsumerCheckFailedException if check isn't successful.
-  void Check(std::optional<std::chrono::milliseconds> timeout, std::optional<int64_t> limit_batches,
+  void Check(std::optional<std::chrono::milliseconds> timeout, std::optional<uint64_t> limit_batches,
              const ConsumerFunction &check_consumer_function) const;
 
   /// Returns true if the consumer is actively consuming messages.
@@ -157,6 +165,7 @@ class Consumer final : public RdKafka::EventCb {
   void event_cb(RdKafka::Event &event) override;
 
   void StartConsuming();
+  void StartConsumingWithLimit(uint64_t limit_batches, std::optional<std::chrono::milliseconds> timeout) const;
 
   void StopConsuming();
 
@@ -178,7 +187,6 @@ class Consumer final : public RdKafka::EventCb {
   ConsumerFunction consumer_function_;
   mutable std::atomic<bool> is_running_{false};
   mutable std::vector<RdKafka::TopicPartition *> last_assignment_;  // Protected by is_running_
-  std::optional<int64_t> limit_batches_{std::nullopt};
   std::unique_ptr<RdKafka::KafkaConsumer, std::function<void(RdKafka::KafkaConsumer *)>> consumer_;
   std::thread thread_;
   ConsumerRebalanceCb cb_;

src/integrations/kafka/exceptions.hpp

@@ -64,4 +64,16 @@ class TopicNotFoundException : public KafkaStreamException {
   TopicNotFoundException(const std::string_view consumer_name, const std::string_view topic_name)
       : KafkaStreamException("Kafka consumer {} cannot find topic {}", consumer_name, topic_name) {}
 };
+
+class ConsumerCommitFailedException : public KafkaStreamException {
+ public:
+  ConsumerCommitFailedException(const std::string_view consumer_name, const std::string_view error)
+      : KafkaStreamException("Committing offset of consumer {} failed: {}", consumer_name, error) {}
+};
+
+class ConsumerReadMessagesFailedException : public KafkaStreamException {
+ public:
+  ConsumerReadMessagesFailedException(const std::string_view consumer_name, const std::string_view error)
+      : KafkaStreamException("Error happened in consumer {} while fetching messages: {}", consumer_name, error) {}
+};
 }  // namespace memgraph::integrations::kafka

src/integrations/pulsar/consumer.cpp

@@ -11,13 +11,14 @@
 
 #include "integrations/pulsar/consumer.hpp"
 
+#include <algorithm>
+#include <chrono>
+#include <thread>
+
 #include <fmt/format.h>
 #include <pulsar/Client.h>
 #include <pulsar/InitialPosition.h>
 
-#include <chrono>
-#include <thread>
-
 #include "integrations/constants.hpp"
 #include "integrations/pulsar/exceptions.hpp"
 #include "utils/concepts.hpp"
@@ -33,6 +34,10 @@ namespace {
 template <typename T>
 concept PulsarConsumer = utils::SameAsAnyOf<T, pulsar_client::Consumer, pulsar_client::Reader>;
 
+template <typename TFunc>
+concept PulsarMessageGetter =
+    std::same_as<const pulsar_client::Message &, std::invoke_result_t<TFunc, const Message &>>;
+
 pulsar_client::Result ConsumeMessage(pulsar_client::Consumer &consumer, pulsar_client::Message &message,
                                      int remaining_timeout_in_ms) {
   return consumer.receive(message, remaining_timeout_in_ms);
@@ -97,6 +102,26 @@ pulsar_client::Client CreateClient(const std::string &service_url) {
   conf.setLogger(new SpdlogLoggerFactory);
   return {service_url, conf};
 }
+
+template <PulsarConsumer TConsumer, PulsarMessageGetter TPulsarMessageGetter>
+void TryToConsumeBatch(TConsumer &consumer, const ConsumerInfo &info, const ConsumerFunction &consumer_function,
+                       pulsar_client::MessageId &last_message_id, const std::vector<Message> &batch,
+                       const TPulsarMessageGetter &message_getter) {
+  consumer_function(batch);
+
+  auto has_message_failed = [&consumer, &info, &last_message_id, &message_getter](const auto &message) {
+    if (const auto result = consumer.acknowledge(message_getter(message)); result != pulsar_client::ResultOk) {
+      spdlog::warn("Acknowledging a message of consumer {} failed: {}", info.consumer_name, result);
+      return true;
+    }
+    last_message_id = message_getter(message).getMessageId();
+    return false;
+  };
+
+  if (std::ranges::any_of(batch, has_message_failed)) {
+    throw ConsumerAcknowledgeMessagesFailedException(info.consumer_name);
+  }
+}
 }  // namespace
 
 Message::Message(pulsar_client::Message &&message) : message_{std::move(message)} {}
@@ -137,6 +162,24 @@ void Consumer::Start() {
   StartConsuming();
 }
 
+void Consumer::StartWithLimit(const uint64_t limit_batches,
+                              const std::optional<std::chrono::milliseconds> timeout) const {
+  if (is_running_) {
+    throw ConsumerRunningException(info_.consumer_name);
+  }
+  if (limit_batches < kMinimumStartBatchLimit) {
+    throw ConsumerStartFailedException(
+        info_.consumer_name, fmt::format("Batch limit has to be greater than or equal to {}", kMinimumStartBatchLimit));
+  }
+  if (timeout.value_or(kMinimumInterval) < kMinimumInterval) {
+    throw ConsumerStartFailedException(
+        info_.consumer_name,
+        fmt::format("Timeout has to be greater than or equal to {} milliseconds", kMinimumInterval.count()));
+  }
+
+  StartConsumingWithLimit(limit_batches, timeout);
+}
+
 void Consumer::Stop() {
   if (!is_running_) {
     throw ConsumerStoppedException(info_.consumer_name);
@@ -154,7 +197,7 @@ void Consumer::StopIfRunning() {
   }
 }
 
-void Consumer::Check(std::optional<std::chrono::milliseconds> timeout, std::optional<int64_t> limit_batches,
+void Consumer::Check(std::optional<std::chrono::milliseconds> timeout, std::optional<uint64_t> limit_batches,
                      const ConsumerFunction &check_consumer_function) const {
   // NOLINTNEXTLINE (modernize-use-nullptr)
   if (timeout.value_or(kMinimumInterval) < kMinimumInterval) {
@@ -240,9 +283,7 @@ void Consumer::StartConsuming() {
       auto maybe_batch = GetBatch(consumer_, info_, is_running_, last_message_id_);
 
       if (maybe_batch.HasError()) {
-        spdlog::warn("Error happened in consumer {} while fetching messages: {}!", info_.consumer_name,
-                     maybe_batch.GetError());
-        break;
+        throw ConsumerReadMessagesFailedException(info_.consumer_name, maybe_batch.GetError());
       }
 
       const auto &batch = maybe_batch.GetValue();
@@ -254,18 +295,8 @@ void Consumer::StartConsuming() {
       spdlog::info("Pulsar consumer {} is processing a batch", info_.consumer_name);
 
       try {
-        consumer_function_(batch);
-
-        if (std::any_of(batch.begin(), batch.end(), [&](const auto &message) {
-              if (const auto result = consumer_.acknowledge(message.message_); result != pulsar_client::ResultOk) {
-                spdlog::warn("Acknowledging a message of consumer {} failed: {}", info_.consumer_name, result);
-                return true;
-              }
-              last_message_id_ = message.message_.getMessageId();
-              return false;
-            })) {
-          break;
-        }
+        TryToConsumeBatch(consumer_, info_, consumer_function_, last_message_id_, batch,
+                          [&](const Message &message) -> const pulsar_client::Message & { return message.message_; });
       } catch (const std::exception &e) {
         spdlog::warn("Error happened in consumer {} while processing a batch: {}!", info_.consumer_name, e.what());
         break;
@@ -277,6 +308,43 @@ void Consumer::StartConsuming() {
   });
 }
 
+void Consumer::StartConsumingWithLimit(uint64_t limit_batches, std::optional<std::chrono::milliseconds> timeout) const {
+  if (is_running_.exchange(true)) {
+    throw ConsumerRunningException(info_.consumer_name);
+  }
+  utils::OnScopeExit restore_is_running([this] { is_running_.store(false); });
+
+  const auto timeout_to_use = timeout.value_or(kDefaultCheckTimeout);
+  const auto start = std::chrono::steady_clock::now();
+
+  for (uint64_t batch_count = 0; batch_count < limit_batches;) {
+    const auto now = std::chrono::steady_clock::now();
+    if (now - start >= timeout_to_use) {
+      throw ConsumerCheckFailedException(info_.consumer_name, "Timeout reached");
+    }
+
+    const auto maybe_batch = GetBatch(consumer_, info_, is_running_, last_message_id_);
+
+    if (maybe_batch.HasError()) {
+      throw ConsumerReadMessagesFailedException(info_.consumer_name, maybe_batch.GetError());
+    }
+
+    const auto &batch = maybe_batch.GetValue();
+
+    if (batch.empty()) {
+      continue;
+    }
+    ++batch_count;
+
+    spdlog::info("Pulsar consumer {} is processing a batch", info_.consumer_name);
+
+    TryToConsumeBatch(consumer_, info_, consumer_function_, last_message_id_, batch,
+                      [](const Message &message) -> const pulsar_client::Message & { return message.message_; });
+
+    spdlog::info("Pulsar consumer {} finished processing", info_.consumer_name);
+  }
+}
+
 void Consumer::StopConsuming() {
   is_running_.store(false);
   if (thread_.joinable()) {

src/integrations/pulsar/consumer.hpp

@@ -58,25 +58,27 @@ class Consumer final {
 
   bool IsRunning() const;
   void Start();
+  void StartWithLimit(uint64_t limit_batches, std::optional<std::chrono::milliseconds> timeout) const;
   void Stop();
   void StopIfRunning();
 
-  void Check(std::optional<std::chrono::milliseconds> timeout, std::optional<int64_t> limit_batches,
+  void Check(std::optional<std::chrono::milliseconds> timeout, std::optional<uint64_t> limit_batches,
              const ConsumerFunction &check_consumer_function) const;
 
   const ConsumerInfo &Info() const;
 
  private:
   void StartConsuming();
+  void StartConsumingWithLimit(uint64_t limit_batches, std::optional<std::chrono::milliseconds> timeout) const;
   void StopConsuming();
 
   ConsumerInfo info_;
   mutable pulsar_client::Client client_;
-  pulsar_client::Consumer consumer_;
+  mutable pulsar_client::Consumer consumer_;
   ConsumerFunction consumer_function_;
 
   mutable std::atomic<bool> is_running_{false};
-  pulsar_client::MessageId last_message_id_{pulsar_client::MessageId::earliest()};
+  mutable pulsar_client::MessageId last_message_id_{pulsar_client::MessageId::earliest()};  // Protected by is_running_
   std::thread thread_;
 };
 }  // namespace memgraph::integrations::pulsar

src/integrations/pulsar/exceptions.hpp

@@ -55,4 +55,16 @@ class TopicNotFoundException : public PulsarStreamException {
   TopicNotFoundException(const std::string &consumer_name, const std::string &topic_name)
       : PulsarStreamException("Pulsar consumer {} cannot find topic {}", consumer_name, topic_name) {}
 };
+
+class ConsumerReadMessagesFailedException : public PulsarStreamException {
+ public:
+  ConsumerReadMessagesFailedException(const std::string_view consumer_name, const std::string_view error)
+      : PulsarStreamException("Error happened in consumer {} while fetching messages: {}", consumer_name, error) {}
+};
+
+class ConsumerAcknowledgeMessagesFailedException : public PulsarStreamException {
+ public:
+  explicit ConsumerAcknowledgeMessagesFailedException(const std::string_view consumer_name)
+      : PulsarStreamException("Acknowledging a message of consumer {} has failed!", consumer_name) {}
+};
 }  // namespace memgraph::integrations::pulsar

src/io/address.hpp

@@ -0,0 +1,77 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include <compare>
+
+#include <fmt/format.h>
+#include <boost/asio/ip/tcp.hpp>
+#include <boost/uuid/uuid.hpp>
+#include <boost/uuid/uuid_generators.hpp>
+#include <boost/uuid/uuid_io.hpp>
+
+namespace memgraph::io {
+struct Address {
+  // It's important for all participants to have a
+  // unique identifier - IP and port alone are not
+  // enough, and may change over the lifecycle of
+  // the nodes. Particularly storage nodes may change
+  // their IP addresses over time, and the system
+  // should gracefully update its information
+  // about them.
+  boost::uuids::uuid unique_id;
+  boost::asio::ip::address last_known_ip;
+  uint16_t last_known_port;
+
+  static Address TestAddress(uint16_t port) {
+    return Address{
+        .unique_id = boost::uuids::uuid{boost::uuids::random_generator()()},
+        .last_known_port = port,
+    };
+  }
+
+  static Address UniqueLocalAddress() {
+    return Address{
+        .unique_id = boost::uuids::uuid{boost::uuids::random_generator()()},
+    };
+  }
+
+  /// Returns a new ID with the same IP and port but a unique UUID.
+  Address ForkUniqueAddress() {
+    return Address{
+        .unique_id = boost::uuids::uuid{boost::uuids::random_generator()()},
+        .last_known_ip = last_known_ip,
+        .last_known_port = last_known_port,
+    };
+  }
+
+  friend bool operator==(const Address &lhs, const Address &rhs) = default;
+
+  /// unique_id is most dominant for ordering, then last_known_ip, then last_known_port
+  friend bool operator<(const Address &lhs, const Address &rhs) {
+    if (lhs.unique_id != rhs.unique_id) {
+      return lhs.unique_id < rhs.unique_id;
+    }
+
+    if (lhs.last_known_ip != rhs.last_known_ip) {
+      return lhs.last_known_ip < rhs.last_known_ip;
+    }
+
+    return lhs.last_known_port < rhs.last_known_port;
+  }
+
+  std::string ToString() const {
+    return fmt::format("Address {{ unique_id: {}, last_known_ip: {}, last_known_port: {} }}",
+                       boost::uuids::to_string(unique_id), last_known_ip.to_string(), last_known_port);
+  }
+};
+};  // namespace memgraph::io

src/io/errors.hpp

@@ -0,0 +1,26 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+namespace memgraph::io {
+// Signifies that a retriable operation was unable to
+// complete after a configured number of retries.
+struct RetriesExhausted {};
+
+// Signifies that a request was unable to receive a response
+// within some configured timeout duration. It is important
+// to remember that in distributed systems, a timeout does
+// not signify that a request was not received or processed.
+// It may be the case that the request was fully processed
+// but that the response was not received.
+struct TimedOut {};
+};  // namespace memgraph::io

src/io/future.hpp

@@ -0,0 +1,256 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include <condition_variable>
+#include <memory>
+#include <mutex>
+#include <optional>
+#include <thread>
+#include <utility>
+
+#include "io/errors.hpp"
+#include "utils/logging.hpp"
+
+namespace memgraph::io {
+
+// Shared is in an anonymous namespace, and the only way to
+// construct a Promise or Future is to pass a Shared in. This
+// ensures that Promises and Futures can only be constructed
+// in this translation unit.
+namespace details {
+template <typename T>
+class Shared {
+  mutable std::condition_variable cv_;
+  mutable std::mutex mu_;
+  std::optional<T> item_;
+  bool consumed_ = false;
+  bool waiting_ = false;
+  std::function<bool()> simulator_notifier_ = nullptr;
+
+ public:
+  explicit Shared(std::function<bool()> simulator_notifier) : simulator_notifier_(simulator_notifier) {}
+  Shared() = default;
+  Shared(Shared &&) = delete;
+  Shared &operator=(Shared &&) = delete;
+  Shared(const Shared &) = delete;
+  Shared &operator=(const Shared &) = delete;
+  ~Shared() = default;
+
+  /// Takes the item out of our optional item_ and returns it.
+  T Take() {
+    MG_ASSERT(item_, "Take called without item_ being present");
+    MG_ASSERT(!consumed_, "Take called on already-consumed Future");
+
+    T ret = std::move(item_).value();
+    item_.reset();
+
+    consumed_ = true;
+
+    return ret;
+  }
+
+  T Wait() {
+    std::unique_lock<std::mutex> lock(mu_);
+    waiting_ = true;
+
+    while (!item_) {
+      bool simulator_progressed = false;
+      if (simulator_notifier_) [[unlikely]] {
+        // We can't hold our own lock while notifying
+        // the simulator because notifying the simulator
+        // involves acquiring the simulator's mutex
+        // to guarantee that our notification linearizes
+        // with the simulator's condition variable.
+        // However, the simulator may acquire our
+        // mutex to check if we are being awaited,
+        // while determining system quiescence,
+        // so we have to get out of its way to avoid
+        // a cyclical deadlock.
+        lock.unlock();
+        simulator_progressed = std::invoke(simulator_notifier_);
+        lock.lock();
+        if (item_) {
+          // item may have been filled while we
+          // had dropped our mutex while notifying
+          // the simulator of our waiting_ status.
+          break;
+        }
+      }
+      if (!simulator_progressed) [[likely]] {
+        cv_.wait(lock);
+      }
+      MG_ASSERT(!consumed_, "Future consumed twice!");
+    }
+
+    waiting_ = false;
+
+    return Take();
+  }
+
+  bool IsReady() const {
+    std::unique_lock<std::mutex> lock(mu_);
+    return item_.has_value();
+  }
+
+  std::optional<T> TryGet() {
+    std::unique_lock<std::mutex> lock(mu_);
+
+    if (item_) {
+      return Take();
+    }
+
+    return std::nullopt;
+  }
+
+  void Fill(T item) {
+    {
+      std::unique_lock<std::mutex> lock(mu_);
+
+      MG_ASSERT(!consumed_, "Promise filled after it was already consumed!");
+      MG_ASSERT(!item_, "Promise filled twice!");
+
+      item_ = item;
+    }  // lock released before condition variable notification
+
+    cv_.notify_all();
+  }
+
+  bool IsAwaited() const {
+    std::unique_lock<std::mutex> lock(mu_);
+    return waiting_;
+  }
+};
+}  // namespace details
+
+template <typename T>
+class Future {
+  bool consumed_or_moved_ = false;
+  std::shared_ptr<details::Shared<T>> shared_;
+
+ public:
+  explicit Future(std::shared_ptr<details::Shared<T>> shared) : shared_(shared) {}
+
+  Future() = delete;
+  Future(Future &&old) noexcept {
+    MG_ASSERT(!old.consumed_or_moved_, "Future moved from after already being moved from or consumed.");
+    shared_ = std::move(old.shared_);
+    consumed_or_moved_ = old.consumed_or_moved_;
+    old.consumed_or_moved_ = true;
+  }
+
+  Future(const Future &) = delete;
+  Future &operator=(const Future &) = delete;
+  ~Future() = default;
+
+  /// Returns true if the Future is ready to
+  /// be consumed using TryGet or Wait (prefer Wait
+  /// if you know it's ready, because it doesn't
+  /// return an optional.
+  bool IsReady() {
+    MG_ASSERT(!consumed_or_moved_, "Called IsReady after Future already consumed!");
+    return shared_->IsReady();
+  }
+
+  /// Non-blocking method that returns the inner
+  /// item if it's already ready, or std::nullopt
+  /// if it is not ready yet.
+  std::optional<T> TryGet() {
+    MG_ASSERT(!consumed_or_moved_, "Called TryGet after Future already consumed!");
+    std::optional<T> ret = shared_->TryGet();
+    if (ret) {
+      consumed_or_moved_ = true;
+    }
+    return ret;
+  }
+
+  /// Block on the corresponding promise to be filled,
+  /// returning the inner item when ready.
+  T Wait() && {
+    MG_ASSERT(!consumed_or_moved_, "Future should only be consumed with Wait once!");
+    T ret = shared_->Wait();
+    consumed_or_moved_ = true;
+    return ret;
+  }
+
+  /// Marks this Future as canceled.
+  void Cancel() {
+    MG_ASSERT(!consumed_or_moved_, "Future::Cancel called on a future that was already moved or consumed!");
+    consumed_or_moved_ = true;
+  }
+};
+
+template <typename T>
+class Promise {
+  std::shared_ptr<details::Shared<T>> shared_;
+  bool filled_or_moved_ = false;
+
+ public:
+  explicit Promise(std::shared_ptr<details::Shared<T>> shared) : shared_(shared) {}
+
+  Promise() = delete;
+  Promise(Promise &&old) noexcept {
+    MG_ASSERT(!old.filled_or_moved_, "Promise moved from after already being moved from or filled.");
+    shared_ = std::move(old.shared_);
+    old.filled_or_moved_ = true;
+  }
+
+  Promise &operator=(Promise &&old) noexcept {
+    MG_ASSERT(!old.filled_or_moved_, "Promise moved from after already being moved from or filled.");
+    shared_ = std::move(old.shared_);
+    old.filled_or_moved_ = true;
+  }
+  Promise(const Promise &) = delete;
+  Promise &operator=(const Promise &) = delete;
+
+  ~Promise() { MG_ASSERT(filled_or_moved_, "Promise destroyed before its associated Future was filled!"); }
+
+  // Fill the expected item into the Future.
+  void Fill(T item) {
+    MG_ASSERT(!filled_or_moved_, "Promise::Fill called on a promise that is already filled or moved!");
+    shared_->Fill(item);
+    filled_or_moved_ = true;
+  }
+
+  bool IsAwaited() { return shared_->IsAwaited(); }
+
+  /// Moves this Promise into a unique_ptr.
+  std::unique_ptr<Promise<T>> ToUnique() && {
+    std::unique_ptr<Promise<T>> up = std::make_unique<Promise<T>>(std::move(shared_));
+
+    filled_or_moved_ = true;
+
+    return up;
+  }
+};
+
+template <typename T>
+std::pair<Future<T>, Promise<T>> FuturePromisePair() {
+  std::shared_ptr<details::Shared<T>> shared = std::make_shared<details::Shared<T>>();
+
+  Future<T> future = Future<T>(shared);
+  Promise<T> promise = Promise<T>(shared);
+
+  return std::make_pair(std::move(future), std::move(promise));
+}
+
+template <typename T>
+std::pair<Future<T>, Promise<T>> FuturePromisePairWithNotifier(std::function<bool()> simulator_notifier) {
+  std::shared_ptr<details::Shared<T>> shared = std::make_shared<details::Shared<T>>(simulator_notifier);
+
+  Future<T> future = Future<T>(shared);
+  Promise<T> promise = Promise<T>(shared);
+
+  return std::make_pair(std::move(future), std::move(promise));
+}
+
+};  // namespace memgraph::io

src/io/local_transport/local_system.hpp

@@ -0,0 +1,35 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include <random>
+
+#include "io/address.hpp"
+#include "io/local_transport/local_transport.hpp"
+#include "io/local_transport/local_transport_handle.hpp"
+#include "io/transport.hpp"
+
+namespace memgraph::io::local_transport {
+
+class LocalSystem {
+  std::shared_ptr<LocalTransportHandle> local_transport_handle_ = std::make_shared<LocalTransportHandle>();
+
+ public:
+  Io<LocalTransport> Register(Address address) {
+    LocalTransport local_transport(local_transport_handle_);
+    return Io{local_transport, address};
+  }
+
+  void ShutDown() { local_transport_handle_->ShutDown(); }
+};
+
+}  // namespace memgraph::io::local_transport

src/io/local_transport/local_transport.hpp

@@ -0,0 +1,64 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include <chrono>
+#include <memory>
+#include <random>
+#include <utility>
+
+#include "io/address.hpp"
+#include "io/local_transport/local_transport_handle.hpp"
+#include "io/time.hpp"
+#include "io/transport.hpp"
+
+namespace memgraph::io::local_transport {
+
+class LocalTransport {
+  std::shared_ptr<LocalTransportHandle> local_transport_handle_;
+
+ public:
+  explicit LocalTransport(std::shared_ptr<LocalTransportHandle> local_transport_handle)
+      : local_transport_handle_(std::move(local_transport_handle)) {}
+
+  template <Message RequestT, Message ResponseT>
+  ResponseFuture<ResponseT> Request(Address to_address, Address from_address, RequestId request_id, RequestT request,
+                                    Duration timeout) {
+    auto [future, promise] = memgraph::io::FuturePromisePair<ResponseResult<ResponseT>>();
+
+    local_transport_handle_->SubmitRequest(to_address, from_address, request_id, std::move(request), timeout,
+                                           std::move(promise));
+
+    return std::move(future);
+  }
+
+  template <Message... Ms>
+  requires(sizeof...(Ms) > 0) RequestResult<Ms...> Receive(Address receiver_address, Duration timeout) {
+    return local_transport_handle_->template Receive<Ms...>(receiver_address, timeout);
+  }
+
+  template <Message M>
+  void Send(Address to_address, Address from_address, RequestId request_id, M &&message) {
+    return local_transport_handle_->template Send<M>(to_address, from_address, request_id, std::forward<M>(message));
+  }
+
+  Time Now() const { return local_transport_handle_->Now(); }
+
+  bool ShouldShutDown() const { return local_transport_handle_->ShouldShutDown(); }
+
+  template <class D = std::poisson_distribution<>, class Return = uint64_t>
+  Return Rand(D distrib) {
+    std::random_device rng;
+    return distrib(rng);
+  }
+};
+};  // namespace memgraph::io::local_transport

src/io/local_transport/local_transport_handle.hpp

@@ -0,0 +1,151 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include <chrono>
+#include <condition_variable>
+#include <iostream>
+#include <map>
+#include <mutex>
+
+#include "io/errors.hpp"
+#include "io/message_conversion.hpp"
+#include "io/time.hpp"
+#include "io/transport.hpp"
+
+namespace memgraph::io::local_transport {
+
+class LocalTransportHandle {
+  mutable std::mutex mu_{};
+  mutable std::condition_variable cv_;
+  bool should_shut_down_ = false;
+
+  // the responses to requests that are being waited on
+  std::map<PromiseKey, DeadlineAndOpaquePromise> promises_;
+
+  // messages that are sent to servers that may later receive them
+  std::vector<OpaqueMessage> can_receive_;
+
+ public:
+  ~LocalTransportHandle() {
+    for (auto &&[pk, promise] : promises_) {
+      std::move(promise.promise).TimeOut();
+    }
+    promises_.clear();
+  }
+
+  void ShutDown() {
+    std::unique_lock<std::mutex> lock(mu_);
+    should_shut_down_ = true;
+    cv_.notify_all();
+  }
+
+  bool ShouldShutDown() const {
+    std::unique_lock<std::mutex> lock(mu_);
+    return should_shut_down_;
+  }
+
+  static Time Now() {
+    auto nano_time = std::chrono::system_clock::now();
+    return std::chrono::time_point_cast<std::chrono::microseconds>(nano_time);
+  }
+
+  template <Message... Ms>
+  requires(sizeof...(Ms) > 0) RequestResult<Ms...> Receive(Address /* receiver_address */, Duration timeout) {
+    std::unique_lock lock(mu_);
+
+    Time before = Now();
+
+    spdlog::info("can_receive_ size: {}", can_receive_.size());
+
+    while (can_receive_.empty()) {
+      Time now = Now();
+
+      // protection against non-monotonic timesources
+      auto maxed_now = std::max(now, before);
+      auto elapsed = maxed_now - before;
+
+      if (timeout < elapsed) {
+        return TimedOut{};
+      }
+
+      Duration relative_timeout = timeout - elapsed;
+
+      std::cv_status cv_status_value = cv_.wait_for(lock, relative_timeout);
+
+      if (cv_status_value == std::cv_status::timeout) {
+        return TimedOut{};
+      }
+    }
+
+    auto current_message = std::move(can_receive_.back());
+    can_receive_.pop_back();
+
+    auto m_opt = std::move(current_message).Take<Ms...>();
+
+    return std::move(m_opt).value();
+  }
+
+  template <Message M>
+  void Send(Address to_address, Address from_address, RequestId request_id, M &&message) {
+    std::any message_any(std::forward<M>(message));
+    OpaqueMessage opaque_message{.to_address = to_address,
+                                 .from_address = from_address,
+                                 .request_id = request_id,
+                                 .message = std::move(message_any)};
+
+    PromiseKey promise_key{
+        .requester_address = to_address, .request_id = opaque_message.request_id, .replier_address = from_address};
+
+    {
+      std::unique_lock<std::mutex> lock(mu_);
+
+      if (promises_.contains(promise_key)) {
+        spdlog::info("using message to fill promise");
+        // complete waiting promise if it's there
+        DeadlineAndOpaquePromise dop = std::move(promises_.at(promise_key));
+        promises_.erase(promise_key);
+
+        dop.promise.Fill(std::move(opaque_message));
+      } else {
+        spdlog::info("placing message in can_receive_");
+        can_receive_.emplace_back(std::move(opaque_message));
+      }
+    }  // lock dropped
+
+    cv_.notify_all();
+  }
+
+  template <Message RequestT, Message ResponseT>
+  void SubmitRequest(Address to_address, Address from_address, RequestId request_id, RequestT &&request,
+                     Duration timeout, ResponsePromise<ResponseT> promise) {
+    const bool port_matches = to_address.last_known_port == from_address.last_known_port;
+    const bool ip_matches = to_address.last_known_ip == from_address.last_known_ip;
+
+    MG_ASSERT(port_matches && ip_matches);
+    const Time deadline = Now() + timeout;
+
+    {
+      std::unique_lock<std::mutex> lock(mu_);
+
+      PromiseKey promise_key{
+          .requester_address = from_address, .request_id = request_id, .replier_address = to_address};
+      OpaquePromise opaque_promise(std::move(promise).ToUnique());
+      DeadlineAndOpaquePromise dop{.deadline = deadline, .promise = std::move(opaque_promise)};
+      promises_.emplace(std::move(promise_key), std::move(dop));
+    }  // lock dropped
+
+    Send(to_address, from_address, request_id, std::forward<RequestT>(request));
+  }
+};
+
+}  // namespace memgraph::io::local_transport

src/io/message_conversion.hpp

@@ -0,0 +1,232 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include "io/transport.hpp"
+
+namespace memgraph::io {
+
+using memgraph::io::Duration;
+using memgraph::io::Message;
+using memgraph::io::Time;
+
+struct PromiseKey {
+  Address requester_address;
+  uint64_t request_id;
+  // TODO(tyler) possibly remove replier_address from promise key
+  // once we want to support DSR.
+  Address replier_address;
+
+ public:
+  friend bool operator<(const PromiseKey &lhs, const PromiseKey &rhs) {
+    if (lhs.requester_address != rhs.requester_address) {
+      return lhs.requester_address < rhs.requester_address;
+    }
+
+    if (lhs.request_id != rhs.request_id) {
+      return lhs.request_id < rhs.request_id;
+    }
+
+    return lhs.replier_address < rhs.replier_address;
+  }
+};
+
+struct OpaqueMessage {
+  Address to_address;
+  Address from_address;
+  uint64_t request_id;
+  std::any message;
+
+  /// Recursively tries to match a specific type from the outer
+  /// variant's parameter pack against the type of the std::any,
+  /// and if it matches, make it concrete and return it. Otherwise,
+  /// move on and compare the any with the next type from the
+  /// parameter pack.
+  ///
+  /// Return is the full std::variant<Ts...> type that holds the
+  /// full parameter pack without interfering with recursive
+  /// narrowing expansion.
+  template <typename Return, Message Head, Message... Rest>
+  std::optional<Return> Unpack(std::any &&a) {
+    if (typeid(Head) == a.type()) {
+      Head concrete = std::any_cast<Head>(std::move(a));
+      return concrete;
+    }
+
+    if constexpr (sizeof...(Rest) > 0) {
+      return Unpack<Return, Rest...>(std::move(a));
+    } else {
+      return std::nullopt;
+    }
+  }
+
+  /// High level "user-facing" conversion function that lets
+  /// people interested in conversion only supply a single
+  /// parameter pack for the types that they want to compare
+  /// with the any and potentially include in the returned
+  /// variant.
+  template <Message... Ms>
+  requires(sizeof...(Ms) > 0) std::optional<std::variant<Ms...>> VariantFromAny(std::any &&a) {
+    return Unpack<std::variant<Ms...>, Ms...>(std::move(a));
+  }
+
+  template <Message... Ms>
+  requires(sizeof...(Ms) > 0) std::optional<RequestEnvelope<Ms...>> Take() && {
+    std::optional<std::variant<Ms...>> m_opt = VariantFromAny<Ms...>(std::move(message));
+
+    if (m_opt) {
+      return RequestEnvelope<Ms...>{
+          .message = std::move(*m_opt),
+          .request_id = request_id,
+          .to_address = to_address,
+          .from_address = from_address,
+      };
+    }
+
+    return std::nullopt;
+  }
+};
+
+class OpaquePromiseTraitBase {
+ public:
+  virtual const std::type_info *TypeInfo() const = 0;
+  virtual bool IsAwaited(void *ptr) const = 0;
+  virtual void Fill(void *ptr, OpaqueMessage &&) const = 0;
+  virtual void TimeOut(void *ptr) const = 0;
+
+  virtual ~OpaquePromiseTraitBase() = default;
+  OpaquePromiseTraitBase() = default;
+  OpaquePromiseTraitBase(const OpaquePromiseTraitBase &) = delete;
+  OpaquePromiseTraitBase &operator=(const OpaquePromiseTraitBase &) = delete;
+  OpaquePromiseTraitBase(OpaquePromiseTraitBase &&old) = delete;
+  OpaquePromiseTraitBase &operator=(OpaquePromiseTraitBase &&) = delete;
+};
+
+template <typename T>
+class OpaquePromiseTrait : public OpaquePromiseTraitBase {
+ public:
+  const std::type_info *TypeInfo() const override { return &typeid(T); };
+
+  bool IsAwaited(void *ptr) const override { return static_cast<ResponsePromise<T> *>(ptr)->IsAwaited(); };
+
+  void Fill(void *ptr, OpaqueMessage &&opaque_message) const override {
+    T message = std::any_cast<T>(std::move(opaque_message.message));
+    auto response_envelope = ResponseEnvelope<T>{.message = std::move(message),
+                                                 .request_id = opaque_message.request_id,
+                                                 .to_address = opaque_message.to_address,
+                                                 .from_address = opaque_message.from_address};
+    auto promise = static_cast<ResponsePromise<T> *>(ptr);
+    auto unique_promise = std::unique_ptr<ResponsePromise<T>>(promise);
+    unique_promise->Fill(std::move(response_envelope));
+  };
+
+  void TimeOut(void *ptr) const override {
+    auto promise = static_cast<ResponsePromise<T> *>(ptr);
+    auto unique_promise = std::unique_ptr<ResponsePromise<T>>(promise);
+    ResponseResult<T> result = TimedOut{};
+    unique_promise->Fill(std::move(result));
+  }
+};
+
+class OpaquePromise {
+  void *ptr_;
+  std::unique_ptr<OpaquePromiseTraitBase> trait_;
+
+ public:
+  OpaquePromise(OpaquePromise &&old) noexcept : ptr_(old.ptr_), trait_(std::move(old.trait_)) {
+    MG_ASSERT(old.ptr_ != nullptr);
+    old.ptr_ = nullptr;
+  }
+
+  OpaquePromise &operator=(OpaquePromise &&old) noexcept {
+    MG_ASSERT(ptr_ == nullptr);
+    MG_ASSERT(old.ptr_ != nullptr);
+    MG_ASSERT(this != &old);
+    ptr_ = old.ptr_;
+    trait_ = std::move(old.trait_);
+    old.ptr_ = nullptr;
+    return *this;
+  }
+
+  OpaquePromise(const OpaquePromise &) = delete;
+  OpaquePromise &operator=(const OpaquePromise &) = delete;
+
+  template <typename T>
+  std::unique_ptr<ResponsePromise<T>> Take() && {
+    MG_ASSERT(typeid(T) == *trait_->TypeInfo());
+    MG_ASSERT(ptr_ != nullptr);
+
+    auto ptr = static_cast<ResponsePromise<T> *>(ptr_);
+
+    ptr_ = nullptr;
+
+    return std::unique_ptr<T>(ptr);
+  }
+
+  template <typename T>
+  explicit OpaquePromise(std::unique_ptr<ResponsePromise<T>> promise)
+      : ptr_(static_cast<void *>(promise.release())), trait_(std::make_unique<OpaquePromiseTrait<T>>()) {}
+
+  bool IsAwaited() {
+    MG_ASSERT(ptr_ != nullptr);
+    return trait_->IsAwaited(ptr_);
+  }
+
+  void TimeOut() {
+    MG_ASSERT(ptr_ != nullptr);
+    trait_->TimeOut(ptr_);
+    ptr_ = nullptr;
+  }
+
+  void Fill(OpaqueMessage &&opaque_message) {
+    MG_ASSERT(ptr_ != nullptr);
+    trait_->Fill(ptr_, std::move(opaque_message));
+    ptr_ = nullptr;
+  }
+
+  ~OpaquePromise() {
+    MG_ASSERT(ptr_ == nullptr, "OpaquePromise destroyed without being explicitly timed out or filled");
+  }
+};
+
+struct DeadlineAndOpaquePromise {
+  Time deadline;
+  OpaquePromise promise;
+};
+
+template <class From>
+std::type_info const &type_info_for_variant(From const &from) {
+  return std::visit([](auto &&x) -> decltype(auto) { return typeid(x); }, from);
+}
+
+template <typename From, typename Return, typename Head, typename... Rest>
+std::optional<Return> ConvertVariantInner(From &&a) {
+  if (typeid(Head) == type_info_for_variant(a)) {
+    Head concrete = std::get<Head>(std::forward<From>(a));
+    return concrete;
+  }
+
+  if constexpr (sizeof...(Rest) > 0) {
+    return ConvertVariantInner<From, Return, Rest...>(std::forward<From>(a));
+  } else {
+    return std::nullopt;
+  }
+}
+
+/// This function converts a variant to another variant holding a subset OR superset of
+/// possible types.
+template <class From, class... Ms>
+requires(sizeof...(Ms) > 0) std::optional<std::variant<Ms...>> ConvertVariant(From &&from) {
+  return ConvertVariantInner<From, std::variant<Ms...>, Ms...>(std::forward<From>(from));
+}
+
+}  // namespace memgraph::io

src/io/messages.hpp

@@ -0,0 +1,46 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include <variant>
+
+#include <coordinator/coordinator.hpp>
+#include <io/rsm/raft.hpp>
+#include "query/v2/requests.hpp"
+#include "utils/concepts.hpp"
+
+namespace memgraph::io::messages {
+
+using memgraph::coordinator::CoordinatorReadRequests;
+using memgraph::coordinator::CoordinatorWriteRequests;
+using memgraph::coordinator::CoordinatorWriteResponses;
+using StorageReadRequest = msgs::ReadRequests;
+using StorageWriteRequest = msgs::WriteRequests;
+
+using memgraph::io::rsm::AppendRequest;
+using memgraph::io::rsm::AppendResponse;
+using memgraph::io::rsm::ReadRequest;
+using memgraph::io::rsm::VoteRequest;
+using memgraph::io::rsm::VoteResponse;
+using memgraph::io::rsm::WriteRequest;
+using memgraph::io::rsm::WriteResponse;
+
+using CoordinatorMessages =
+    std::variant<ReadRequest<CoordinatorReadRequests>, AppendRequest<CoordinatorWriteRequests>, AppendResponse,
+                 WriteRequest<CoordinatorWriteRequests>, VoteRequest, VoteResponse>;
+
+using ShardMessages = std::variant<ReadRequest<StorageReadRequest>, AppendRequest<StorageWriteRequest>, AppendResponse,
+                                   WriteRequest<StorageWriteRequest>, VoteRequest, VoteResponse>;
+
+using ShardManagerMessages = std::variant<WriteResponse<CoordinatorWriteResponses>>;
+
+}  // namespace memgraph::io::messages

src/io/rsm/raft.hpp

@@ -0,0 +1,926 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+// TODO(tyler) buffer out-of-order Append buffers on the Followers to reassemble more quickly
+// TODO(tyler) handle granular batch sizes based on simple flow control
+
+#pragma once
+
+#include <deque>
+#include <iostream>
+#include <map>
+#include <set>
+#include <thread>
+#include <unordered_map>
+#include <vector>
+
+#include "io/simulator/simulator.hpp"
+#include "io/transport.hpp"
+#include "utils/concepts.hpp"
+
+namespace memgraph::io::rsm {
+
+/// Timeout and replication tunables
+using namespace std::chrono_literals;
+static constexpr auto kMinimumElectionTimeout = 100ms;
+static constexpr auto kMaximumElectionTimeout = 200ms;
+static constexpr auto kMinimumBroadcastTimeout = 40ms;
+static constexpr auto kMaximumBroadcastTimeout = 60ms;
+static constexpr auto kMinimumCronInterval = 1ms;
+static constexpr auto kMaximumCronInterval = 2ms;
+static constexpr auto kMinimumReceiveTimeout = 40ms;
+static constexpr auto kMaximumReceiveTimeout = 60ms;
+static_assert(kMinimumElectionTimeout > kMaximumBroadcastTimeout,
+              "The broadcast timeout has to be smaller than the election timeout!");
+static_assert(kMinimumElectionTimeout < kMaximumElectionTimeout,
+              "The minimum election timeout has to be smaller than the maximum election timeout!");
+static_assert(kMinimumBroadcastTimeout < kMaximumBroadcastTimeout,
+              "The minimum broadcast timeout has to be smaller than the maximum broadcast timeout!");
+static_assert(kMinimumCronInterval < kMaximumCronInterval,
+              "The minimum cron interval has to be smaller than the maximum cron interval!");
+static_assert(kMinimumReceiveTimeout < kMaximumReceiveTimeout,
+              "The minimum receive timeout has to be smaller than the maximum receive timeout!");
+static constexpr size_t kMaximumAppendBatchSize = 1024;
+
+using Term = uint64_t;
+using LogIndex = uint64_t;
+using LogSize = uint64_t;
+
+template <typename WriteOperation>
+struct WriteRequest {
+  WriteOperation operation;
+};
+
+/// WriteResponse is returned to a client after
+/// their WriteRequest was entered in to the raft
+/// log and it reached consensus.
+///
+/// WriteReturn is the result of applying the WriteRequest to
+/// ReplicatedState, and if the ReplicatedState::write
+/// method is deterministic, all replicas will
+/// have the same ReplicatedState after applying
+/// the submitted WriteRequest.
+template <typename WriteReturn>
+struct WriteResponse {
+  bool success;
+  WriteReturn write_return;
+  std::optional<Address> retry_leader;
+  LogIndex raft_index;
+};
+
+template <typename ReadOperation>
+struct ReadRequest {
+  ReadOperation operation;
+};
+
+template <typename ReadReturn>
+struct ReadResponse {
+  bool success;
+  ReadReturn read_return;
+  std::optional<Address> retry_leader;
+};
+
+/// AppendRequest is a raft-level message that the Leader
+/// periodically broadcasts to all Follower peers. This
+/// serves three main roles:
+/// 1. acts as a heartbeat from the Leader to the Follower
+/// 2. replicates new data that the Leader has received to the Follower
+/// 3. informs Follower peers when the commit index has increased,
+///    signalling that it is now safe to apply log items to the
+///    replicated state machine
+template <typename WriteRequest>
+struct AppendRequest {
+  Term term = 0;
+  LogIndex batch_start_log_index;
+  Term last_log_term;
+  std::vector<std::pair<Term, WriteRequest>> entries;
+  LogSize leader_commit;
+};
+
+struct AppendResponse {
+  bool success;
+  Term term;
+  Term last_log_term;
+  // a small optimization over the raft paper, tells
+  // the leader the offset that we are interested in
+  // to send log offsets from for us. This will only
+  // be useful at the beginning of a leader's term.
+  LogSize log_size;
+};
+
+struct VoteRequest {
+  Term term = 0;
+  LogSize log_size;
+  Term last_log_term;
+};
+
+struct VoteResponse {
+  Term term = 0;
+  LogSize committed_log_size;
+  bool vote_granted = false;
+};
+
+template <typename WriteRequest>
+struct CommonState {
+  Term term = 0;
+  std::vector<std::pair<Term, WriteRequest>> log;
+  LogSize committed_log_size = 0;
+  LogSize applied_size = 0;
+};
+
+struct FollowerTracker {
+  LogIndex next_index = 0;
+  LogSize confirmed_log_size = 0;
+};
+
+struct PendingClientRequest {
+  RequestId request_id;
+  Address address;
+  Time received_at;
+};
+
+struct Leader {
+  std::map<Address, FollowerTracker> followers;
+  std::unordered_map<LogIndex, PendingClientRequest> pending_client_requests;
+  Time last_broadcast = Time::min();
+
+  std::string static ToString() { return "\tLeader   \t"; }
+};
+
+struct Candidate {
+  std::map<Address, LogSize> successful_votes;
+  Time election_began = Time::min();
+  std::set<Address> outstanding_votes;
+
+  std::string static ToString() { return "\tCandidate\t"; }
+};
+
+struct Follower {
+  Time last_received_append_entries_timestamp;
+  Address leader_address;
+
+  std::string static ToString() { return "\tFollower \t"; }
+};
+
+using Role = std::variant<Candidate, Leader, Follower>;
+
+template <typename Role>
+concept AllRoles = memgraph::utils::SameAsAnyOf<Role, Leader, Follower, Candidate>;
+
+template <typename Role>
+concept LeaderOrFollower = memgraph::utils::SameAsAnyOf<Role, Leader, Follower>;
+
+template <typename Role>
+concept FollowerOrCandidate = memgraph::utils::SameAsAnyOf<Role, Follower, Candidate>;
+
+/*
+all ReplicatedState classes should have an Apply method
+that returns our WriteResponseValue after consensus, and
+a Read method that returns our ReadResponseValue without
+requiring consensus.
+
+ReadResponse Read(ReadOperation);
+WriteResponseValue ReplicatedState::Apply(WriteRequest);
+
+For example:
+If the state is uint64_t, and WriteRequest is `struct PlusOne {};`,
+and WriteResponseValue is also uint64_t (the new value), then
+each call to state.Apply(PlusOne{}) will return the new value
+after incrementing it. 0, 1, 2, 3... and this will be sent back
+to the client that requested the mutation.
+
+In practice, these mutations will usually be predicated on some
+previous value, so that they are idempotent, functioning similarly
+to a CAS operation.
+*/
+template <typename WriteOperation, typename ReadOperation, typename ReplicatedState, typename WriteResponseValue,
+          typename ReadResponseValue>
+concept Rsm = requires(ReplicatedState state, WriteOperation w, ReadOperation r) {
+  { state.Read(r) } -> std::same_as<ReadResponseValue>;
+  { state.Apply(w) } -> std::same_as<WriteResponseValue>;
+};
+
+/// Parameter           Purpose
+/// --------------------------
+/// IoImpl              the concrete Io provider - SimulatorTransport, ThriftTransport, etc...
+/// ReplicatedState     the high-level data structure that is managed by the raft-backed replicated state machine
+/// WriteOperation      the individual operation type that is applied to the ReplicatedState in identical order
+///                     across each replica
+/// WriteResponseValue  the return value of calling ReplicatedState::Apply(WriteOperation), which is executed in
+///                     identical order across all replicas after an WriteOperation reaches consensus.
+/// ReadOperation       the type of operations that do not require consensus before executing directly
+///                     on a const ReplicatedState &
+/// ReadResponseValue   the return value of calling ReplicatedState::Read(ReadOperation), which is executed directly
+///                     without going through consensus first
+template <typename IoImpl, typename ReplicatedState, typename WriteOperation, typename WriteResponseValue,
+          typename ReadOperation, typename ReadResponseValue>
+requires Rsm<WriteOperation, ReadOperation, ReplicatedState, WriteResponseValue, ReadResponseValue>
+class Raft {
+  CommonState<WriteOperation> state_;
+  Role role_ = Candidate{};
+  Io<IoImpl> io_;
+  std::vector<Address> peers_;
+  ReplicatedState replicated_state_;
+  Time next_cron_;
+
+ public:
+  Raft(Io<IoImpl> &&io, std::vector<Address> peers, ReplicatedState &&replicated_state)
+      : io_(std::forward<Io<IoImpl>>(io)),
+        peers_(peers),
+        replicated_state_(std::forward<ReplicatedState>(replicated_state)) {
+    if (peers.empty()) {
+      role_ = Leader{};
+    }
+  }
+
+  /// Periodic protocol maintenance. Returns the time that Cron should be called again
+  /// in the future.
+  Time Cron() {
+    // dispatch periodic logic based on our role to a specific Cron method.
+    std::optional<Role> new_role = std::visit([&](auto &role) { return Cron(role); }, role_);
+
+    if (new_role) {
+      role_ = std::move(new_role).value();
+    }
+    const Duration random_cron_interval = RandomTimeout(kMinimumCronInterval, kMaximumCronInterval);
+
+    return io_.Now() + random_cron_interval;
+  }
+
+  /// Returns the Address for our underlying Io implementation
+  Address GetAddress() { return io_.GetAddress(); }
+
+  using ReceiveVariant = std::variant<ReadRequest<ReadOperation>, AppendRequest<WriteOperation>, AppendResponse,
+                                      WriteRequest<WriteOperation>, VoteRequest, VoteResponse>;
+
+  void Handle(ReceiveVariant &&message_variant, RequestId request_id, Address from_address) {
+    // dispatch the message to a handler based on our role,
+    // which can be specified in the Handle first argument,
+    // or it can be `auto` if it's a handler for several roles
+    // or messages.
+    std::optional<Role> new_role = std::visit(
+        [&](auto &&msg, auto &role) mutable {
+          return Handle(role, std::forward<decltype(msg)>(msg), request_id, from_address);
+        },
+        std::forward<ReceiveVariant>(message_variant), role_);
+
+    // TODO(tyler) (M3) maybe replace std::visit with get_if for explicit prioritized matching, [[likely]] etc...
+    if (new_role) {
+      role_ = std::move(new_role).value();
+    }
+  }
+
+  void Run() {
+    while (!io_.ShouldShutDown()) {
+      const auto now = io_.Now();
+      if (now >= next_cron_) {
+        next_cron_ = Cron();
+      }
+
+      const Duration receive_timeout = RandomTimeout(kMinimumReceiveTimeout, kMaximumReceiveTimeout);
+
+      auto request_result =
+          io_.template ReceiveWithTimeout<ReadRequest<ReadOperation>, AppendRequest<WriteOperation>, AppendResponse,
+                                          WriteRequest<WriteOperation>, VoteRequest, VoteResponse>(receive_timeout);
+      if (request_result.HasError()) {
+        continue;
+      }
+
+      auto request = std::move(request_result.GetValue());
+
+      Handle(std::move(request.message), request.request_id, request.from_address);
+    }
+  }
+
+ private:
+  // Raft paper - 5.3
+  // When the entry has been safely replicated, the leader applies the
+  // entry to its state machine and returns the result of that
+  // execution to the client.
+  //
+  // "Safely replicated" is defined as being known to be present
+  // on at least a majority of all peers (inclusive of the Leader).
+  void BumpCommitIndexAndReplyToClients(Leader &leader) {
+    auto confirmed_log_sizes = std::vector<LogSize>{};
+
+    // We include our own log size in the calculation of the log
+    // confirmed log size that is present on at least a majority of all peers.
+    confirmed_log_sizes.push_back(state_.log.size());
+
+    for (const auto &[addr, f] : leader.followers) {
+      confirmed_log_sizes.push_back(f.confirmed_log_size);
+      Log("Follower at port ", addr.last_known_port, " has confirmed log size of: ", f.confirmed_log_size);
+    }
+
+    // reverse sort from highest to lowest (using std::ranges::greater)
+    std::ranges::sort(confirmed_log_sizes, std::ranges::greater());
+
+    // This is a particularly correctness-critical calculation because it
+    // determines the committed log size that will be broadcast in
+    // the next AppendRequest.
+    //
+    // If the following sizes are recorded for clusters of different numbers of peers,
+    // these are the expected sizes that are considered to have reached consensus:
+    //
+    // state           | expected value | (confirmed_log_sizes.size() / 2)
+    // [1]              1                 (1 / 2) => 0
+    // [2, 1]           1                 (2 / 2) => 1
+    // [3, 2, 1]        2                 (3 / 2) => 1
+    // [4, 3, 2, 1]     2                 (4 / 2) => 2
+    // [5, 4, 3, 2, 1]  3                 (5 / 2) => 2
+    const size_t majority_index = confirmed_log_sizes.size() / 2;
+    const LogSize new_committed_log_size = confirmed_log_sizes[majority_index];
+
+    // We never go backwards in history.
+    MG_ASSERT(state_.committed_log_size <= new_committed_log_size,
+              "as a Leader, we have previously set our committed_log_size to {}, but our Followers have a majority "
+              "committed_log_size of {}",
+              state_.committed_log_size, new_committed_log_size);
+
+    state_.committed_log_size = new_committed_log_size;
+
+    // For each size between the old size and the new one (inclusive),
+    // Apply that log's WriteOperation to our replicated_state_,
+    // and use the specific return value of the ReplicatedState::Apply
+    // method (WriteResponseValue) to respond to the requester.
+    for (; state_.applied_size < state_.committed_log_size; state_.applied_size++) {
+      const LogIndex apply_index = state_.applied_size;
+      const auto &write_request = state_.log[apply_index].second;
+      const WriteResponseValue write_return = replicated_state_.Apply(write_request);
+
+      if (leader.pending_client_requests.contains(apply_index)) {
+        const PendingClientRequest client_request = std::move(leader.pending_client_requests.at(apply_index));
+        leader.pending_client_requests.erase(apply_index);
+
+        const WriteResponse<WriteResponseValue> resp{
+            .success = true,
+            .write_return = std::move(write_return),
+            .raft_index = apply_index,
+        };
+
+        io_.Send(client_request.address, client_request.request_id, std::move(resp));
+      }
+    }
+
+    Log("committed_log_size is now ", state_.committed_log_size);
+  }
+
+  // Raft paper - 5.1
+  // AppendEntries RPCs are initiated by leaders to replicate log entries and to provide a form of heartbeat
+  void BroadcastAppendEntries(std::map<Address, FollowerTracker> &followers) {
+    for (auto &[address, follower] : followers) {
+      const LogIndex next_index = follower.next_index;
+
+      const auto missing = state_.log.size() - next_index;
+      const auto batch_size = std::min(missing, kMaximumAppendBatchSize);
+      const auto start_index = next_index;
+      const auto end_index = start_index + batch_size;
+
+      // advance follower's next index
+      follower.next_index += batch_size;
+
+      std::vector<std::pair<Term, WriteOperation>> entries;
+
+      entries.insert(entries.begin(), state_.log.begin() + start_index, state_.log.begin() + end_index);
+
+      const Term previous_term_from_index = PreviousTermFromIndex(start_index);
+
+      Log("sending ", entries.size(), " entries to Follower ", address.last_known_port,
+          " which are above its next_index of ", next_index);
+
+      AppendRequest<WriteOperation> ar{
+          .term = state_.term,
+          .batch_start_log_index = start_index,
+          .last_log_term = previous_term_from_index,
+          .entries = std::move(entries),
+          .leader_commit = state_.committed_log_size,
+      };
+
+      // request_id not necessary to set because it's not a Future-backed Request.
+      static constexpr RequestId request_id = 0;
+
+      io_.Send(address, request_id, std::move(ar));
+    }
+  }
+
+  // Raft paper - 5.2
+  // Raft uses randomized election timeouts to ensure that split votes are rare and that they are resolved quickly
+  Duration RandomTimeout(Duration min, Duration max) {
+    std::uniform_int_distribution time_distrib(min.count(), max.count());
+
+    const auto rand_micros = io_.Rand(time_distrib);
+
+    return Duration{rand_micros};
+  }
+
+  Duration RandomTimeout(int min_micros, int max_micros) {
+    std::uniform_int_distribution time_distrib(min_micros, max_micros);
+
+    const int rand_micros = io_.Rand(time_distrib);
+
+    return std::chrono::microseconds{rand_micros};
+  }
+
+  Term PreviousTermFromIndex(LogIndex index) const {
+    if (index == 0 || state_.log.size() + 1 <= index) {
+      return 0;
+    }
+
+    const auto &[term, data] = state_.log.at(index - 1);
+    return term;
+  }
+
+  Term CommittedLogTerm() {
+    MG_ASSERT(state_.log.size() >= state_.committed_log_size);
+    if (state_.log.empty() || state_.committed_log_size == 0) {
+      return 0;
+    }
+
+    const auto &[term, data] = state_.log.at(state_.committed_log_size - 1);
+    return term;
+  }
+
+  Term LastLogTerm() const {
+    if (state_.log.empty()) {
+      return 0;
+    }
+
+    const auto &[term, data] = state_.log.back();
+    return term;
+  }
+
+  template <typename... Ts>
+  void Log(Ts &&...args) {
+    const Time now = io_.Now();
+    const auto micros = std::chrono::duration_cast<std::chrono::milliseconds>(now.time_since_epoch()).count();
+    const Term term = state_.term;
+    const std::string role_string = std::visit([&](const auto &role) { return role.ToString(); }, role_);
+
+    std::ostringstream out;
+
+    out << '\t' << static_cast<int>(micros) << "\t" << term << "\t" << io_.GetAddress().last_known_port;
+
+    out << role_string;
+
+    (out << ... << args);
+
+    spdlog::info(out.str());
+  }
+
+  /////////////////////////////////////////////////////////////
+  /// Raft-related Cron methods
+  ///
+  /// Cron + std::visit is how events are dispatched
+  /// to certain code based on Raft role.
+  ///
+  /// Cron(role) takes as the first argument a reference to its
+  /// role, and as the second argument, the message that has
+  /// been received.
+  /////////////////////////////////////////////////////////////
+
+  // Raft paper - 5.2
+  // Candidates keep sending Vote to peers until:
+  // 1. receiving Append with a higher term (become Follower)
+  // 2. receiving Vote with a higher term (become a Follower)
+  // 3. receiving a quorum of responses to our last batch of Vote (become a Leader)
+  std::optional<Role> Cron(Candidate &candidate) {
+    const auto now = io_.Now();
+    const Duration election_timeout = RandomTimeout(kMinimumElectionTimeout, kMaximumElectionTimeout);
+    const auto election_timeout_us = std::chrono::duration_cast<std::chrono::milliseconds>(election_timeout).count();
+
+    if (now - candidate.election_began > election_timeout) {
+      state_.term++;
+      Log("becoming Candidate for term ", state_.term, " after leader timeout of ", election_timeout_us,
+          "ms elapsed since last election attempt");
+
+      const VoteRequest request{
+          .term = state_.term,
+          .log_size = state_.log.size(),
+          .last_log_term = LastLogTerm(),
+      };
+
+      auto outstanding_votes = std::set<Address>();
+
+      for (const auto &peer : peers_) {
+        // request_id not necessary to set because it's not a Future-backed Request.
+        static constexpr auto request_id = 0;
+        io_.template Send<VoteRequest>(peer, request_id, request);
+        outstanding_votes.insert(peer);
+      }
+
+      return Candidate{
+          .successful_votes = std::map<Address, LogIndex>(),
+          .election_began = now,
+          .outstanding_votes = outstanding_votes,
+      };
+    }
+    return std::nullopt;
+  }
+
+  // Raft paper - 5.2
+  // Followers become candidates if we haven't heard from the leader
+  // after a randomized timeout.
+  std::optional<Role> Cron(Follower &follower) {
+    const auto now = io_.Now();
+    const auto time_since_last_append_entries = now - follower.last_received_append_entries_timestamp;
+
+    // randomized follower timeout
+    const Duration election_timeout = RandomTimeout(kMinimumElectionTimeout, kMaximumElectionTimeout);
+
+    if (time_since_last_append_entries > election_timeout) {
+      // become a Candidate if we haven't heard from the Leader after this timeout
+      return Candidate{};
+    }
+
+    return std::nullopt;
+  }
+
+  // Leaders (re)send AppendRequest to followers.
+  std::optional<Role> Cron(Leader &leader) {
+    const Time now = io_.Now();
+    const Duration broadcast_timeout = RandomTimeout(kMinimumBroadcastTimeout, kMaximumBroadcastTimeout);
+
+    if (now - leader.last_broadcast > broadcast_timeout) {
+      BroadcastAppendEntries(leader.followers);
+      leader.last_broadcast = now;
+    }
+
+    return std::nullopt;
+  }
+
+  /////////////////////////////////////////////////////////////
+  /// Raft-related Handle methods
+  ///
+  /// Handle + std::visit is how events are dispatched
+  /// to certain code based on Raft role.
+  ///
+  /// Handle(role, message, ...)
+  /// takes as the first argument a reference
+  /// to its role, and as the second argument, the
+  /// message that has been received.
+  /////////////////////////////////////////////////////////////
+
+  // all roles can receive Vote and possibly become a follower
+  template <AllRoles ALL>
+  std::optional<Role> Handle(ALL & /* variable */, VoteRequest &&req, RequestId request_id, Address from_address) {
+    Log("received VoteRequest from ", from_address.last_known_port, " with term ", req.term);
+    const bool last_log_term_dominates = req.last_log_term >= LastLogTerm();
+    const bool term_dominates = req.term > state_.term;
+    const bool log_size_dominates = req.log_size >= state_.log.size();
+    const bool new_leader = last_log_term_dominates && term_dominates && log_size_dominates;
+
+    if (new_leader) {
+      MG_ASSERT(req.term > state_.term);
+      MG_ASSERT(std::max(req.term, state_.term) == req.term);
+    }
+
+    const VoteResponse res{
+        .term = std::max(req.term, state_.term),
+        .committed_log_size = state_.committed_log_size,
+        .vote_granted = new_leader,
+    };
+
+    io_.Send(from_address, request_id, res);
+
+    if (new_leader) {
+      // become a follower
+      state_.term = req.term;
+      return Follower{
+          .last_received_append_entries_timestamp = io_.Now(),
+          .leader_address = from_address,
+      };
+    }
+
+    if (term_dominates) {
+      Log("received a vote from an inferior candidate. Becoming Candidate");
+      state_.term = std::max(state_.term, req.term) + 1;
+      return Candidate{};
+    }
+
+    return std::nullopt;
+  }
+
+  std::optional<Role> Handle(Candidate &candidate, VoteResponse &&res, RequestId /* variable */, Address from_address) {
+    Log("received VoteResponse");
+
+    if (!res.vote_granted || res.term != state_.term) {
+      Log("received unsuccessful VoteResponse from term ", res.term, " when our candidacy term is ", state_.term);
+      // we received a delayed VoteResponse from the past, which has to do with an election that is
+      // no longer valid. We can simply drop this.
+      return std::nullopt;
+    }
+
+    MG_ASSERT(candidate.outstanding_votes.contains(from_address),
+              "Received unexpected VoteResponse from server not present in Candidate's outstanding_votes!");
+    candidate.outstanding_votes.erase(from_address);
+
+    MG_ASSERT(!candidate.successful_votes.contains(from_address),
+              "Received unexpected VoteResponse from server already in Candidate's successful_votes!");
+    candidate.successful_votes.insert({from_address, res.committed_log_size});
+
+    if (candidate.successful_votes.size() >= candidate.outstanding_votes.size()) {
+      std::map<Address, FollowerTracker> followers{};
+
+      for (const auto &[address, committed_log_size] : candidate.successful_votes) {
+        FollowerTracker follower{
+            .next_index = committed_log_size,
+            .confirmed_log_size = committed_log_size,
+        };
+        followers.insert({address, follower});
+      }
+      for (const auto &address : candidate.outstanding_votes) {
+        FollowerTracker follower{
+            .next_index = state_.log.size(),
+            .confirmed_log_size = 0,
+        };
+        followers.insert({address, follower});
+      }
+
+      Log("becoming Leader at term ", state_.term);
+
+      BroadcastAppendEntries(followers);
+
+      return Leader{
+          .followers = std::move(followers),
+          .pending_client_requests = std::unordered_map<LogIndex, PendingClientRequest>(),
+      };
+    }
+
+    return std::nullopt;
+  }
+
+  template <LeaderOrFollower LOF>
+  std::optional<Role> Handle(LOF & /* variable */, VoteResponse && /* variable */, RequestId /* variable */,
+                             Address /* variable */) {
+    Log("non-Candidate received VoteResponse");
+    return std::nullopt;
+  }
+
+  template <AllRoles ALL>
+  std::optional<Role> Handle(ALL &role, AppendRequest<WriteOperation> &&req, RequestId request_id,
+                             Address from_address) {
+    // log size starts out as state_.committed_log_size and only if everything is successful do we
+    // switch it to the log length.
+    AppendResponse res{
+        .success = false,
+        .term = state_.term,
+        .last_log_term = CommittedLogTerm(),
+        .log_size = state_.log.size(),
+    };
+
+    if constexpr (std::is_same<ALL, Leader>()) {
+      MG_ASSERT(req.term != state_.term, "Multiple leaders are acting under the term ", req.term);
+    }
+
+    const bool is_candidate = std::is_same<ALL, Candidate>();
+    const bool is_failed_competitor = is_candidate && req.term == state_.term;
+    const Time now = io_.Now();
+
+    // Raft paper - 5.2
+    // While waiting for votes, a candidate may receive an
+    // AppendEntries RPC from another server claiming to be leader. If
+    // the leader’s term (included in its RPC) is at least as large as
+    // the candidate’s current term, then the candidate recognizes the
+    // leader as legitimate and returns to follower state.
+    if (req.term > state_.term || is_failed_competitor) {
+      // become follower of this leader, reply with our log status
+      state_.term = req.term;
+
+      io_.Send(from_address, request_id, res);
+
+      Log("becoming Follower of Leader ", from_address.last_known_port, " at term ", req.term);
+      return Follower{
+          .last_received_append_entries_timestamp = now,
+          .leader_address = from_address,
+      };
+    }
+
+    if (req.term < state_.term) {
+      // nack this request from an old leader
+      io_.Send(from_address, request_id, res);
+
+      return std::nullopt;
+    }
+
+    // at this point, we're dealing with our own leader
+    if constexpr (std::is_same<ALL, Follower>()) {
+      // small specialization for when we're already a Follower
+      MG_ASSERT(role.leader_address == from_address, "Multiple Leaders are acting under the same term number!");
+      role.last_received_append_entries_timestamp = now;
+    } else {
+      Log("Somehow entered Follower-specific logic as a non-Follower");
+      MG_ASSERT(false, "Somehow entered Follower-specific logic as a non-Follower");
+    }
+
+    // Handle steady-state conditions.
+    if (req.batch_start_log_index != state_.log.size()) {
+      Log("req.batch_start_log_index of ", req.batch_start_log_index, " does not match our log size of ",
+          state_.log.size());
+    } else if (req.last_log_term != LastLogTerm()) {
+      Log("req.last_log_term differs from our leader term at that slot, expected: ", LastLogTerm(), " but got ",
+          req.last_log_term);
+    } else {
+      // happy path - Apply log
+      Log("applying batch of ", req.entries.size(), " entries to our log starting at index ",
+          req.batch_start_log_index);
+
+      const auto resize_length = req.batch_start_log_index;
+
+      MG_ASSERT(resize_length >= state_.committed_log_size,
+                "Applied history from Leader which goes back in time from our commit_index");
+
+      // possibly chop-off stuff that was replaced by
+      // things with different terms (we got data that
+      // hasn't reached consensus yet, which is normal)
+      state_.log.resize(resize_length);
+
+      if (req.entries.size() > 0) {
+        auto &[first_term, op] = req.entries.at(0);
+        MG_ASSERT(LastLogTerm() <= first_term);
+      }
+
+      state_.log.insert(state_.log.end(), std::make_move_iterator(req.entries.begin()),
+                        std::make_move_iterator(req.entries.end()));
+
+      MG_ASSERT(req.leader_commit >= state_.committed_log_size);
+      state_.committed_log_size = std::min(req.leader_commit, state_.log.size());
+
+      for (; state_.applied_size < state_.committed_log_size; state_.applied_size++) {
+        const auto &write_request = state_.log[state_.applied_size].second;
+        replicated_state_.Apply(write_request);
+      }
+
+      res.success = true;
+    }
+
+    res.last_log_term = LastLogTerm();
+    res.log_size = state_.log.size();
+
+    Log("returning log_size of ", res.log_size);
+
+    io_.Send(from_address, request_id, res);
+
+    return std::nullopt;
+  }
+
+  std::optional<Role> Handle(Leader &leader, AppendResponse &&res, RequestId /* variable */, Address from_address) {
+    if (res.term != state_.term) {
+      Log("received AppendResponse related to a previous term when we (presumably) were the leader");
+      return std::nullopt;
+    }
+
+    // TODO(tyler) when we have dynamic membership, this assert will become incorrect, but we should
+    // keep it in-place until then because it has bug finding value.
+    MG_ASSERT(leader.followers.contains(from_address), "received AppendResponse from unknown Follower");
+
+    // at this point, we know the term matches and we know this Follower
+
+    FollowerTracker &follower = leader.followers.at(from_address);
+
+    if (res.success) {
+      Log("got successful AppendResponse from ", from_address.last_known_port, " with log_size of ", res.log_size);
+      follower.next_index = std::max(follower.next_index, res.log_size);
+    } else {
+      Log("got unsuccessful AppendResponse from ", from_address.last_known_port, " with log_size of ", res.log_size);
+      follower.next_index = res.log_size;
+    }
+
+    follower.confirmed_log_size = std::max(follower.confirmed_log_size, res.log_size);
+
+    BumpCommitIndexAndReplyToClients(leader);
+
+    return std::nullopt;
+  }
+
+  template <FollowerOrCandidate FOC>
+  std::optional<Role> Handle(FOC & /* variable */, AppendResponse && /* variable */, RequestId /* variable */,
+                             Address /* variable */) {
+    // we used to be the leader, and are getting old delayed responses
+    return std::nullopt;
+  }
+
+  /////////////////////////////////////////////////////////////
+  /// RSM-related handle methods
+  /////////////////////////////////////////////////////////////
+
+  // Leaders are able to immediately respond to the requester (with a ReadResponseValue) applied to the ReplicatedState
+  std::optional<Role> Handle(Leader & /* variable */, ReadRequest<ReadOperation> &&req, RequestId request_id,
+                             Address from_address) {
+    Log("handling ReadOperation");
+    ReadOperation read_operation = req.operation;
+
+    ReadResponseValue read_return = replicated_state_.Read(read_operation);
+
+    const ReadResponse<ReadResponseValue> resp{
+        .success = true,
+        .read_return = std::move(read_return),
+        .retry_leader = std::nullopt,
+    };
+
+    io_.Send(from_address, request_id, resp);
+
+    return std::nullopt;
+  }
+
+  // Candidates should respond with a failure, similar to the Candidate + WriteRequest failure below
+  std::optional<Role> Handle(Candidate & /* variable */, ReadRequest<ReadOperation> && /* variable */,
+                             RequestId request_id, Address from_address) {
+    Log("received ReadOperation - not redirecting because no Leader is known");
+    const ReadResponse<ReadResponseValue> res{
+        .success = false,
+    };
+
+    io_.Send(from_address, request_id, res);
+
+    Cron();
+
+    return std::nullopt;
+  }
+
+  // Followers should respond with a redirection, similar to the Follower + WriteRequest response below
+  std::optional<Role> Handle(Follower &follower, ReadRequest<ReadOperation> && /* variable */, RequestId request_id,
+                             Address from_address) {
+    Log("redirecting client to known Leader with port ", follower.leader_address.last_known_port);
+
+    const ReadResponse<ReadResponseValue> res{
+        .success = false,
+        .retry_leader = follower.leader_address,
+    };
+
+    io_.Send(from_address, request_id, res);
+
+    return std::nullopt;
+  }
+
+  // Raft paper - 8
+  // When a client first starts up, it connects to a randomly chosen
+  // server. If the client’s first choice is not the leader, that
+  // server will reject the client’s request and supply information
+  // about the most recent leader it has heard from.
+  std::optional<Role> Handle(Follower &follower, WriteRequest<WriteOperation> && /* variable */, RequestId request_id,
+                             Address from_address) {
+    Log("redirecting client to known Leader with port ", follower.leader_address.last_known_port);
+
+    const WriteResponse<WriteResponseValue> res{
+        .success = false,
+        .retry_leader = follower.leader_address,
+    };
+
+    io_.Send(from_address, request_id, res);
+
+    return std::nullopt;
+  }
+
+  std::optional<Role> Handle(Candidate & /* variable */, WriteRequest<WriteOperation> && /* variable */,
+                             RequestId request_id, Address from_address) {
+    Log("received WriteRequest - not redirecting because no Leader is known");
+
+    const WriteResponse<WriteResponseValue> res{
+        .success = false,
+    };
+
+    io_.Send(from_address, request_id, res);
+
+    Cron();
+
+    return std::nullopt;
+  }
+
+  // only leaders actually handle replication requests from clients
+  std::optional<Role> Handle(Leader &leader, WriteRequest<WriteOperation> &&req, RequestId request_id,
+                             Address from_address) {
+    Log("handling WriteRequest");
+
+    // we are the leader. add item to log and send Append to peers
+    MG_ASSERT(state_.term >= LastLogTerm());
+    state_.log.emplace_back(std::pair(state_.term, std::move(req.operation)));
+
+    LogIndex log_index = state_.log.size() - 1;
+
+    PendingClientRequest pcr{
+        .request_id = request_id,
+        .address = from_address,
+        .received_at = io_.Now(),
+    };
+
+    leader.pending_client_requests.emplace(log_index, pcr);
+
+    if (peers_.empty()) {
+      BumpCommitIndexAndReplyToClients(leader);
+    } else {
+      BroadcastAppendEntries(leader.followers);
+    }
+
+    return std::nullopt;
+  }
+};
+
+};  // namespace memgraph::io::rsm

src/io/rsm/rsm_client.hpp

@@ -0,0 +1,136 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+#include <iostream>
+#include <optional>
+#include <vector>
+
+#include "io/address.hpp"
+#include "io/rsm/raft.hpp"
+#include "utils/result.hpp"
+
+namespace memgraph::io::rsm {
+
+using memgraph::io::Address;
+using memgraph::io::Duration;
+using memgraph::io::ResponseEnvelope;
+using memgraph::io::ResponseFuture;
+using memgraph::io::ResponseResult;
+using memgraph::io::Time;
+using memgraph::io::TimedOut;
+using memgraph::io::rsm::ReadRequest;
+using memgraph::io::rsm::ReadResponse;
+using memgraph::io::rsm::WriteRequest;
+using memgraph::io::rsm::WriteResponse;
+using memgraph::utils::BasicResult;
+
+template <typename IoImpl, typename WriteRequestT, typename WriteResponseT, typename ReadRequestT,
+          typename ReadResponseT>
+class RsmClient {
+  using ServerPool = std::vector<Address>;
+
+  Io<IoImpl> io_;
+  Address leader_;
+  ServerPool server_addrs_;
+
+  template <typename ResponseT>
+  void PossiblyRedirectLeader(const ResponseT &response) {
+    if (response.retry_leader) {
+      MG_ASSERT(!response.success, "retry_leader should never be set for successful responses");
+      leader_ = response.retry_leader.value();
+      spdlog::debug("client redirected to leader server {}", leader_.ToString());
+    } else if (!response.success) {
+      std::uniform_int_distribution<size_t> addr_distrib(0, (server_addrs_.size() - 1));
+      size_t addr_index = io_.Rand(addr_distrib);
+      leader_ = server_addrs_[addr_index];
+
+      spdlog::debug(
+          "client NOT redirected to leader server despite our success failing to be processed (it probably was sent to "
+          "a RSM Candidate) trying a random one at index {} with address {}",
+          addr_index, leader_.ToString());
+    }
+  }
+
+ public:
+  RsmClient(Io<IoImpl> io, Address leader, ServerPool server_addrs)
+      : io_{io}, leader_{leader}, server_addrs_{server_addrs} {}
+
+  RsmClient() = delete;
+
+  BasicResult<TimedOut, WriteResponseT> SendWriteRequest(WriteRequestT req) {
+    WriteRequest<WriteRequestT> client_req;
+    client_req.operation = req;
+
+    const Duration overall_timeout = io_.GetDefaultTimeout();
+    const Time before = io_.Now();
+
+    do {
+      spdlog::debug("client sending WriteRequest to Leader {}", leader_.ToString());
+      ResponseFuture<WriteResponse<WriteResponseT>> response_future =
+          io_.template Request<WriteRequest<WriteRequestT>, WriteResponse<WriteResponseT>>(leader_, client_req);
+      ResponseResult<WriteResponse<WriteResponseT>> response_result = std::move(response_future).Wait();
+
+      if (response_result.HasError()) {
+        spdlog::debug("client timed out while trying to communicate with leader server {}", leader_.ToString());
+        return response_result.GetError();
+      }
+
+      ResponseEnvelope<WriteResponse<WriteResponseT>> &&response_envelope = std::move(response_result.GetValue());
+      WriteResponse<WriteResponseT> &&write_response = std::move(response_envelope.message);
+
+      if (write_response.success) {
+        return std::move(write_response.write_return);
+      }
+
+      PossiblyRedirectLeader(write_response);
+    } while (io_.Now() < before + overall_timeout);
+
+    return TimedOut{};
+  }
+
+  BasicResult<TimedOut, ReadResponseT> SendReadRequest(ReadRequestT req) {
+    ReadRequest<ReadRequestT> read_req;
+    read_req.operation = req;
+
+    const Duration overall_timeout = io_.GetDefaultTimeout();
+    const Time before = io_.Now();
+
+    do {
+      spdlog::debug("client sending ReadRequest to Leader {}", leader_.ToString());
+
+      ResponseFuture<ReadResponse<ReadResponseT>> get_response_future =
+          io_.template Request<ReadRequest<ReadRequestT>, ReadResponse<ReadResponseT>>(leader_, read_req);
+
+      // receive response
+      ResponseResult<ReadResponse<ReadResponseT>> get_response_result = std::move(get_response_future).Wait();
+
+      if (get_response_result.HasError()) {
+        spdlog::debug("client timed out while trying to communicate with leader server {}", leader_.ToString());
+        return get_response_result.GetError();
+      }
+
+      ResponseEnvelope<ReadResponse<ReadResponseT>> &&get_response_envelope = std::move(get_response_result.GetValue());
+      ReadResponse<ReadResponseT> &&read_get_response = std::move(get_response_envelope.message);
+
+      if (read_get_response.success) {
+        return std::move(read_get_response.read_return);
+      }
+
+      PossiblyRedirectLeader(read_get_response);
+    } while (io_.Now() < before + overall_timeout);
+
+    return TimedOut{};
+  }
+};
+
+}  // namespace memgraph::io::rsm

src/io/rsm/shard_rsm.hpp

@@ -0,0 +1,155 @@
+// Copyright 2022 Memgraph Ltd.
+//
+// Use of this software is governed by the Business Source License
+// included in the file licenses/BSL.txt; by using this file, you agree to be bound by the terms of the Business Source
+// License, and you may not use this file except in compliance with the Business Source License.
+//
+// As of the Change Date specified in that file, in accordance with
+// the Business Source License, use of this software will be governed
+// by the Apache License, Version 2.0, included in the file
+// licenses/APL.txt.
+
+#pragma once
+
+/// The ShardRsm is a simple in-memory raft-backed kv store that can be used for simple testing
+/// and implementation of some query engine logic before storage engines are fully implemented.
+///
+/// To implement multiple read and write commands, change the StorageRead* and StorageWrite* requests
+/// and responses to a std::variant of the different options, and route them to specific handlers in
+/// the ShardRsm's Read and Apply methods. Remember that Read is called immediately when the Raft
+/// leader receives the request, and does not replicate anything over Raft. Apply is called only
+/// AFTER the StorageWriteRequest is replicated to a majority of Raft peers, and the result of calling
+/// ShardRsm::Apply(StorageWriteRequest) is returned to the client that submitted the request.
+
+#include <deque>
+#include <iostream>
+#include <map>
+#include <optional>
+#include <set>
+#include <thread>
+#include <vector>
+
+#include "coordinator/hybrid_logical_clock.hpp"
+#include "io/address.hpp"
+#include "io/rsm/raft.hpp"
+#include "storage/v3/id_types.hpp"
+#include "storage/v3/property_value.hpp"
+#include "utils/logging.hpp"
+
+namespace memgraph::io::rsm {
+
+using memgraph::coordinator::Hlc;
+using memgraph::storage::v3::LabelId;
+using memgraph::storage::v3::PropertyValue;
+
+using ShardRsmKey = std::vector<PropertyValue>;
+
+struct StorageWriteRequest {
+  LabelId label_id;
+  Hlc transaction_id;
+  ShardRsmKey key;
+  std::optional<int> value;
+};
+
+struct StorageWriteResponse {
+  bool shard_rsm_success;
+  std::optional<int> last_value;
+  // Only has a value if the given shard does not contain the requested key
+  std::optional<Hlc> latest_known_shard_map_version{std::nullopt};
+};
+
+struct StorageReadRequest {
+  LabelId label_id;
+  Hlc transaction_id;
+  ShardRsmKey key;
+};
+
+struct StorageReadResponse {
+  bool shard_rsm_success;
+  std::optional<int> value;
+  // Only has a value if the given shard does not contain the requested key
+  std::optional<Hlc> latest_known_shard_map_version{std::nullopt};
+};
+
+class ShardRsm {
+  std::map<ShardRsmKey, int> state_;
+  ShardRsmKey minimum_key_;
+  std::optional<ShardRsmKey> maximum_key_{std::nullopt};
+  Hlc shard_map_version_;
+
+  // The key is not located in this shard
+  bool IsKeyInRange(const ShardRsmKey &key) const {
+    if (maximum_key_) [[likely]] {
+      return (key >= minimum_key_ && key <= maximum_key_);
+    }
+    return key >= minimum_key_;
+  }
+
+ public:
+  StorageReadResponse Read(StorageReadRequest request) const {
+    StorageReadResponse ret;
+
+    if (!IsKeyInRange(request.key)) {
+      ret.latest_known_shard_map_version = shard_map_version_;
+      ret.shard_rsm_success = false;
+    } else if (state_.contains(request.key)) {
+      ret.value = state_.at(request.key);
+      ret.shard_rsm_success = true;
+    } else {
+      ret.shard_rsm_success = false;
+      ret.value = std::nullopt;
+    }
+    return ret;
+  }
+
+  StorageWriteResponse Apply(StorageWriteRequest request) {
+    StorageWriteResponse ret;
+
+    // Key is outside the prohibited range
+    if (!IsKeyInRange(request.key)) {
+      ret.latest_known_shard_map_version = shard_map_version_;
+      ret.shard_rsm_success = false;
+    }
+    // Key exist
+    else if (state_.contains(request.key)) {
+      auto &val = state_[request.key];
+
+      /*
+       *   Delete
+       */
+      if (!request.value) {
+        ret.shard_rsm_success = true;
+        ret.last_value = val;
+        state_.erase(state_.find(request.key));
+      }
+
+      /*
+       *   Update
+       */
+      // Does old_value match?
+      if (request.value == val) {
+        ret.last_value = val;
+        ret.shard_rsm_success = true;
+
+        val = request.value.value();
+
+      } else {
+        ret.last_value = val;
+        ret.shard_rsm_success = false;
+      }
+    }
+    /*
+     *   Create
+     */
+    else {
+      ret.last_value = std::nullopt;
+      ret.shard_rsm_success = true;
+
+      state_.emplace(request.key, request.value.value());
+    }
+
+    return ret;
+  }
+};
+
+}  // namespace memgraph::io::rsm