kuzudb · mewim · Jan 7, 2023 · Jan 7, 2023 · Jan 7, 2023
diff --git a/.github/workflows/ci-workflow.yml b/.github/workflows/ci-workflow.yml
@@ -27,13 +27,13 @@ jobs:
     runs-on: kuzu-self-hosted-testing
     steps:
       - uses: actions/checkout@v2
-      - run: pip install -r tools/python_api/requirements_dev.txt
+      - run: pip3 install --user -r tools/python_api/requirements_dev.txt
 
       - name: build
-        run: CC=clang-13 CXX=clang++-13 make release NUM_THREADS=32
+        run: CC=clang-14 CXX=clang++-14 make release NUM_THREADS=32
 
       - name: test
-        run: CC=clang-13 CXX=clang++-13 make test
+        run: CC=clang-14 CXX=clang++-14 make test
 
   clang-formatting-check:
     name: clang-formatting-check
@@ -44,8 +44,8 @@ jobs:
         with:
           repository: Sarcasm/run-clang-format
           path: run-clang-format
-      - run: python3 run-clang-format/run-clang-format.py --clang-format-executable /usr/bin/clang-format -r src/
-      - run: python3 run-clang-format/run-clang-format.py --clang-format-executable /usr/bin/clang-format -r test/
+      - run: python3 run-clang-format/run-clang-format.py --clang-format-executable /usr/bin/clang-format-11 -r src/
+      - run: python3 run-clang-format/run-clang-format.py --clang-format-executable /usr/bin/clang-format-11 -r test/
 
   benchmark:
     name: benchmark

diff --git a/.gitignore b/.gitignore
@@ -13,3 +13,4 @@ __pycache__/
 *.py[cod]
 *$py.class
 cmake-build-debug/
+test/unittest_temp/
diff --git a/CMakeLists.txt b/CMakeLists.txt
@@ -5,9 +5,6 @@ project(Kuzu VERSION 0.0.1.1 LANGUAGES CXX)
 find_package(Threads REQUIRED)
 
 set(CMAKE_FIND_PACKAGE_RESOLVE_SYMLINKS TRUE)
-find_package(Arrow REQUIRED)
-find_package(Parquet REQUIRED)
-
 set(CMAKE_CXX_STANDARD 20)
 set(CMAKE_CXX_STANDARD_REQUIRED True)
 set(CMAKE_POSITION_INDEPENDENT_CODE ON)
@@ -66,17 +63,22 @@ endfunction()
 
 add_definitions(-DKUZU_ROOT_DIRECTORY="${PROJECT_SOURCE_DIR}")
 
-set(ARROW_INSTALL ${PROJECT_SOURCE_DIR}/external/build/arrow/install)
+set(ARROW_INSTALL ${CMAKE_CURRENT_SOURCE_DIR}/external/build/arrow/install)
+add_library(arrow_deps STATIC IMPORTED)
+set_target_properties(arrow_deps PROPERTIES IMPORTED_LOCATION ${ARROW_INSTALL}/lib/libarrow_bundled_dependencies.a)
+add_library(parquet_lib STATIC IMPORTED)
+set_target_properties(parquet_lib PROPERTIES IMPORTED_LOCATION ${ARROW_INSTALL}/lib/libparquet.a)
+add_library(arrow_lib STATIC IMPORTED)
+set_target_properties(arrow_lib PROPERTIES IMPORTED_LOCATION ${ARROW_INSTALL}/lib/libarrow.a)
 include_directories(${ARROW_INSTALL}/include)
+
 include_directories(src/include)
 include_directories(third_party/antlr4_cypher/include)
 include_directories(third_party/antlr4_runtime/src)
 include_directories(third_party/spdlog)
 include_directories(third_party/nlohmann_json)
 include_directories(third_party/utf8proc/include)
 include_directories(third_party/pybind11/include)
-include_directories(${ARROW_INCLUDE_DIR})
-include_directories(${PARQUET_INCLUDE_DIR})
 
 add_subdirectory(third_party)
 add_subdirectory(src)

diff --git a/Makefile b/Makefile
@@ -1,4 +1,4 @@
-.PHONY: all release debug test clean
+.PHONY: all release debug test clean arrow clean-external clean-all
 
 all: release
 
@@ -27,22 +27,34 @@ ifeq ($(UBSAN), 1)
 	SANITIZER_FLAG=-DENABLE_ADDRESS_SANITIZER=FALSE -DENABLE_THREAD_SANITIZER=TRUE -DENABLE_UBSAN=TRUE
 endif
 
-release:
+arrow:
+	cd external && \
+	mkdir -p build && \
+	cd build && \
+	cmake $(GENERATOR) $(FORCE_COLOR) $(SANITIZER_FLAG) -DCMAKE_BUILD_TYPE=Release ../ && \
+	cmake --build . --config Release -- -j $(NUM_THREADS)
+
+release: arrow
 	mkdir -p build/release && \
 	cd build/release && \
 	cmake $(GENERATOR) $(FORCE_COLOR) $(SANITIZER_FLAG) -DCMAKE_BUILD_TYPE=Release ../.. && \
 	cmake --build . --config Release -- -j $(NUM_THREADS)
 
-debug:
+debug: arrow
 	mkdir -p build/debug && \
 	cd build/debug && \
 	cmake $(GENERATOR) $(FORCE_COLOR) $(SANITIZER_FLAG) -DCMAKE_BUILD_TYPE=Debug ../.. && \
 	cmake --build . --config Debug -- -j $(NUM_THREADS)
 
 
-test:
+test: release
 	cd $(ROOT_DIR)/build/release/test && \
 	ctest
 
+clean-external:
+	rm -rf external/build
+
 clean:
 	rm -rf build
+
+clean-all: clean-external clean
diff --git a/README.md b/README.md
@@ -17,7 +17,7 @@ Kùzu is an in-process property graph database management system (GDBMS) built f
 Kùzu is being actively developed at University of Waterloo as a feature-rich and usable GDBMS. Kùzu is available under a permissible license. So try it out and help us make it better! We welcome your feedback and feature requests.
 
 ## Build
-To build from source code, Kùzu requires Cmake(>=3.11), Python 3, [Apache Arrow](https://arrow.apache.org/), [xsimd](https://xsimd.readthedocs.io/en/latest/), and a compiler that supports `C++20`.
+To build from source code, Kùzu requires Cmake(>=3.11), Python 3, and a compiler that supports `C++20`.
 - Perform a full clean build
   - `make clean && make`
 - Run tests (optional)

diff --git a/benchmark/Dockerfile b/benchmark/Dockerfile
@@ -4,11 +4,7 @@ ENV CSV_DIR /csv
 ENV SERIALIZED_DIR /serialized
 ENV DEBIAN_FRONTEND=noninteractive
 RUN apt-get update && apt-get install -y --no-install-recommends apt-utils
-RUN apt-get update && apt-get install -y lsb-release wget
-RUN wget https://apache.jfrog.io/artifactory/arrow/$(lsb_release --id --short | tr 'A-Z' 'a-z')/apache-arrow-apt-source-latest-$(lsb_release --codename --short).deb
-RUN apt-get install -y ./apache-arrow-apt-source-latest-$(lsb_release --codename --short).deb
-RUN rm ./apache-arrow*
-RUN apt-get update &&  apt-get -y install python3-dev python3-pip python-is-python3 cmake nodejs jq curl apt-transport-https gnupg sudo git libarrow-dev libparquet-dev
+RUN apt-get update && apt-get -y install python3-dev python3-pip python-is-python3 cmake nodejs jq curl apt-transport-https gnupg sudo git
 RUN pip3 install requests psutil
 
 RUN mkdir -p $CSV_DIR $SERIALIZED_DIR 

diff --git a/external/CMakeLists.txt b/external/CMakeLists.txt
@@ -0,0 +1,21 @@
+cmake_minimum_required(VERSION 3.11)
+
+project(Kuzu_external VERSION 0.0.1 LANGUAGES CXX)
+
+find_package(Threads REQUIRED)
+
+set(CMAKE_FIND_PACKAGE_RESOLVE_SYMLINKS TRUE)
+set(CMAKE_CXX_STANDARD 20)
+set(CMAKE_CXX_STANDARD_REQUIRED True)
+set(CMAKE_POSITION_INDEPENDENT_CODE ON)
+find_program(CCACHE_PROGRAM ccache)
+if (CCACHE_PROGRAM)
+    set_property(GLOBAL PROPERTY RULE_LAUNCH_COMPILE "${CCACHE_PROGRAM}")
+else ()
+    find_program(CCACHE_PROGRAM sccache)
+    if (CCACHE_PROGRAM)
+        set_property(GLOBAL PROPERTY RULE_LAUNCH_COMPILE "${CCACHE_PROGRAM}")
+    endif ()
+endif ()
+
+include(arrow/apache_arrow.cmake)
diff --git a/external/arrow/apache_arrow.cmake b/external/arrow/apache_arrow.cmake
@@ -0,0 +1,29 @@
+include(ExternalProject)
+
+set(CMAKE_POSITION_INDEPENDENT_CODE ON)
+
+ExternalProject_Add(apache_arrow
+        GIT_REPOSITORY "https://github.com/apache/arrow"
+        GIT_TAG 89f9a0948961f6e94f1ef5e4f310b707d22a3c11
+        PREFIX "${CMAKE_BINARY_DIR}/arrow/"
+        INSTALL_DIR "${CMAKE_BINARY_DIR}/arrow/install"
+        CONFIGURE_COMMAND
+        ${CMAKE_COMMAND} -G${CMAKE_GENERATOR} -DCMAKE_BUILD_TYPE=Release
+        -DBUILD_WARNING_LEVEL=PRODUCTION -DARROW_DEPENDENCY_SOURCE=BUNDLED
+        -DCMAKE_CXX_COMPILER_LAUNCHER=${CMAKE_CXX_COMPILER_LAUNCHER}
+        -DCMAKE_CXX_FLAGS=${CMAKE_CXX_FLAGS}
+        -DCMAKE_CXX_STANDARD=${CMAKE_CXX_STANDARD}
+        -DCMAKE_C_COMPILER_LAUNCHER=${CMAKE_C_COMPILER_LAUNCHER}
+        -DCMAKE_INSTALL_PREFIX=${CMAKE_BINARY_DIR}/arrow/install
+        -DCMAKE_MODULE_PATH=${CMAKE_MODULE_PATH}
+        -DCMAKE_TOOLCHAIN_FILE=${CMAKE_TOOLCHAIN_FILE} -DARROW_ALTIVEC=OFF
+        -DARROW_USE_CCACHE=ON -DARROW_BOOST_USE_SHARED=OFF -DARROW_BUILD_SHARED=OFF
+        -DARROW_BUILD_STATIC=ON -DARROW_COMPUTE=OFF -DARROW_CSV=ON -DARROW_IPC=ON -DARROW_JEMALLOC=OFF -DARROW_JSON=OFF
+        -DARROW_PARQUET=ON -DARROW_SIMD_LEVEL=NONE -DARROW_RUNTIME_SIMD_LEVEL=NONE -DARROW_WITH_BROTLI=OFF
+        -DARROW_WITH_LZ4=ON -Dlz4_SOURCE=BUNDLED -DARROW_WITH_PROTOBUF=OFF -DARROW_WITH_RAPIDJSON=OFF
+        -DARROW_WITH_SNAPPY=ON -DSnappy_SOURCE=BUNDLED -DARROW_WITH_ZLIB=ON -DZLIB_SOURCE=BUNDLED
+        -DARROW_WITH_ZSTD=ON -Dzstd_SOURCE=BUNDLED -DThrift_SOURCE=BUNDLED -DBOOST_SOURCE=BUNDLED -DARROW_WITH_RE2=OFF
+        -DARROW_WITH_UTF8PROC=OFF -DARROW_BUILD_BENCHMARKS=OFF -DARROW_BUILD_EXAMPLES=OFF -DARROW_BUILD_INTEGRATION=OFF
+        -DARROW_BUILD_UTILITIES=OFF -DARROW_BUILD_TESTS=OFF -DARROW_ENABLE_TIMING_TESTS=OFF -DARROW_FUZZING=OFF
+        <SOURCE_DIR>/cpp
+        UPDATE_COMMAND "")
diff --git a/scripts/dockerized-ci-tests-runner/Dockerfile b/scripts/dockerized-ci-tests-runner/Dockerfile
@@ -1,31 +1,21 @@
-FROM python:3.9
-
-ENV RUNNER_ALLOW_RUNASROOT=1
+FROM ubuntu:22.04
 
+ENV DEBIAN_FRONTEND=noninteractive
 # Install dependencies
-RUN apt-get update
-RUN apt-get install -y apt-transport-https curl gnupg
-RUN apt-get install -y cmake
-RUN apt-get install -y clang
-RUN apt-get install -y clang-format
-RUN apt-get install -y clang-13
-RUN apt-get install -y nodejs
-RUN apt-get install -y jq
-RUN apt-get install -y ca-certificates lsb-release wget
-RUN wget https://apache.jfrog.io/artifactory/arrow/$(lsb_release --id --short | tr 'A-Z' 'a-z')/apache-arrow-apt-source-latest-$(lsb_release --codename --short).deb
-RUN apt-get install -y ./apache-arrow-apt-source-latest-$(lsb_release --codename --short).deb
-RUN rm ./apache-arrow*
-RUN apt-get update
-RUN apt-get install -y libarrow-dev libparquet-dev
+RUN apt-get update && apt-get install -y --no-install-recommends apt-utils
+RUN apt-get update && apt-get install -y g++ gcc clang-14 python3-dev python3-pip python-is-python3 cmake nodejs jq curl apt-transport-https gnupg sudo git clang-format-11 ca-certificates lsb-release wget
+
+RUN useradd --create-home runner
+USER runner
 
-# Install GitHub action runner
-RUN mkdir /actions-runner
-WORKDIR /actions-runner
-RUN curl -o actions-runner-linux-x64-2.298.2.tar.gz -L https://github.com/actions/runner/releases/download/v2.298.2/actions-runner-linux-x64-2.298.2.tar.gz
+RUN mkdir /home/runner/actions-runner
+WORKDIR /home/runner/actions-runner
+RUN curl -o actions-runner-linux-x64-2.298.2.tar.gz -L\
+    https://github.com/actions/runner/releases/download/v2.298.2/actions-runner-linux-x64-2.298.2.tar.gz
 RUN echo "0bfd792196ce0ec6f1c65d2a9ad00215b2926ef2c416b8d97615265194477117  actions-runner-linux-x64-2.298.2.tar.gz" | shasum -a 256
 RUN tar xzf ./actions-runner-linux-x64-2.298.2.tar.gz
 
-COPY start.sh start.sh
+COPY --chown=runner:runner start.sh start.sh
 RUN chmod +x start.sh
 
 ENTRYPOINT ["./start.sh"]
diff --git a/scripts/dockerized-ci-tests-runner/start.sh b/scripts/dockerized-ci-tests-runner/start.sh
@@ -1,6 +1,6 @@
 #!/bin/bash
 
-cd /actions-runner
+cd /home/runner/actions-runner
 
 # Get registration token
 REG_TOKEN=$(curl \

diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
@@ -11,7 +11,8 @@ add_subdirectory(storage)
 add_subdirectory(transaction)
 
 add_library(kuzu STATIC ${ALL_OBJECT_FILES})
-target_link_libraries(kuzu PUBLIC antlr4_cypher antlr4_runtime utf8proc Threads::Threads Arrow::arrow_shared parquet_shared)
+target_link_libraries(kuzu PUBLIC antlr4_cypher antlr4_runtime utf8proc 
+        Threads::Threads parquet_lib arrow_lib arrow_deps)
 target_include_directories(
         kuzu PUBLIC $<BUILD_INTERFACE:${CMAKE_CURRENT_SOURCE_DIR}/include>
         $<INSTALL_INTERFACE:${CMAKE_INSTALL_INCLUDEDIR}>)