Development Guidelines#

This section provides information for developers who wish to contribute to the C++ codebase.


Since most of the project’s developers work on Linux or macOS, not all features or developer tools are uniformly supported on Windows. If you are on Windows, have a look at Developing on Windows.

Compiler warning levels#

The BUILD_WARNING_LEVEL CMake option switches between sets of predetermined compiler warning levels that we use for code tidiness. For release builds, the default warning level is PRODUCTION, while for debug builds the default is CHECKIN.

When using CHECKIN for debug builds, -Werror is added when using gcc and clang, causing build failures for any warning, and /WX is set with MSVC having the same effect.

Running unit tests#

The -DARROW_BUILD_TESTS=ON CMake option enables building of unit test executables. You can then either run them individually, by launching the desired executable, or run them all at once by launching the ctest executable (which is part of the CMake suite).

A possible invocation is something like:

$ ctest -j16 --output-on-failure

where the -j16 option runs up to 16 tests in parallel, taking advantage of multiple CPU cores and hardware threads.

Running benchmarks#

The -DARROW_BUILD_BENCHMARKS=ON CMake option enables building of benchmark executables. You can then run benchmarks individually by launching the corresponding executable from the command line, e.g.:

$ ./build/release/arrow-builder-benchmark


For meaningful benchmark numbers, it is very strongly recommended to build in Release mode, so as to enable compiler optimizations.

Code Style, Linting, and CI#

This project follows Google’s C++ Style Guide with these exceptions:

  • We relax the line length restriction to 90 characters.

  • We use the NULLPTR macro in header files (instead of nullptr) defined in src/arrow/util/macros.h to support building C++/CLI (ARROW-1134).

  • We relax the guide’s rules regarding structs. For public headers we should use struct only for objects that are principally simple data containers where it is OK to expose all the internal members and any methods are primarily conveniences. For private headers the rules are relaxed further and structs can be used where convenient for types that do not need access control even though they may not be simple data containers.

  • We prefer pointers for output and input/output parameters (the style guide recommends mutable references in some cases).

Our continuous integration builds on GitHub Actions run the unit test suites on a variety of platforms and configuration, including using Address Sanitizer and Undefined Behavior Sanitizer to check for various patterns of misbehaviour such as memory leaks. In addition, the codebase is subjected to a number of code style and code cleanliness checks.

In order to have a passing CI build, your modified Git branch must pass the following checks:

  • C++ builds with the project’s active version of clang without compiler warnings with -DBUILD_WARNING_LEVEL=CHECKIN. Note that there are classes of warnings (such as -Wdocumentation, see more on this below) that are not caught by gcc.

  • Passes various C++ (and others) style checks, checked with the lint subcommand to Archery. This can also be fixed locally by running archery lint --cpplint --clang-format --clang-tidy --fix.

  • CMake files pass style checks, can be fixed by running archery lint --cmake-format --fix. This requires Python 3 and cmake_format (note: this currently does not work on Windows).

On pull requests, the “Dev / Lint” pipeline will run these checks, and report what files/lines need to be fixed, if any.

In order to account for variations in the behavior of clang-format between major versions of LLVM, we pin the version of clang-format used. You can confirm the current pinned version by finding the CLANG_TOOLS variable value in .env. Note that the version must match exactly; a newer version (even a patch release) will not work. LLVM can be installed through a system package manager or a package manager like Conda or Homebrew, though note they may not offer the exact version needed. Alternatively, binaries can be directly downloaded from the LLVM website.

For convenience, C++ style checks can run via a build, in addition to Archery. To do so, build one or more of the targets format (for clang-format), lint_cpp_cli, lint (for cpplint), or clang-tidy. For example:

$ cmake -GNinja ../cpp ...
$ ninja format lint clang-tidy lint_cpp_cli

Depending on how you installed clang-format, the build system may not be able to find it. In that case, invoking CMake will show errors like the following:

-- clang-format 12 not found

Or if the wrong version is installed:

-- clang-format found, but version did not match "^clang-format version 12"

You can provide an explicit path to the directory containing the clang-format executable and others with the environment variable $CLANG_TOOLS_PATH, or by passing -DClangTools_PATH=$PATH_TO_CLANG_TOOLS when invoking CMake. For example:

# We unpacked LLVM here:
$ ~/tools/bin/clang-format --version
clang-format version 12.0.0
# Pass the directory containing the tools to CMake
$ cmake ../cpp -DClangTools_PATH=~/tools/bin/
-- clang-tidy found at /home/user/tools/bin/clang-tidy
-- clang-format found at /home/user/tools/bin/clang-format

To make linting more reproducible for everyone, we provide a docker-compose target that is executable from the root of the repository:

$ docker-compose run ubuntu-lint

Alternatively, on an open pull request, the comment bot can format C++ code for you (it will push a commit to the branch that can then be pulled). Just comment the following:

@github-actions autotune

Cleaning includes with include-what-you-use (IWYU)#

We occasionally use Google’s include-what-you-use tool, also known as IWYU, to remove unnecessary imports.

To begin using IWYU, you must first build it by following the instructions in the project’s documentation. Once the include-what-you-use executable is in your $PATH, you must run CMake with -DCMAKE_EXPORT_COMPILE_COMMANDS=ON in a new out-of-source CMake build directory like so:

mkdir -p $ARROW_ROOT/cpp/iwyu
cd $ARROW_ROOT/cpp/iwyu

In order for IWYU to run on the desired component in the codebase, it must be enabled by the CMake configuration flags. Once this is done, you can run IWYU on the whole codebase by running a helper script:


Since this is very time consuming, you can check a subset of files matching some string pattern with the special “match” option


For example, if you wanted to do IWYU checks on all files in src/arrow/array, you could run

./$IWYU_SH match arrow/array

Checking for ABI and API stability#

To build ABI compliance reports, you need to install the two tools abi-dumper and abi-compliance-checker.

Build Arrow C++ in Debug mode, alternatively you could use -Og which also builds with the necessary symbols but includes a bit of code optimization. Once the build has finished, you can generate ABI reports using:

abi-dumper -lver 9 debug/ -o ABI-9.dump

The above version number is freely selectable. As we want to compare versions, you should now git checkout the version you want to compare it to and re-run the above command using a different version number. Once both reports are generated, you can build a comparison report using

abi-compliance-checker -l libarrow -d1 ABI-PY-9.dump -d2 ABI-PY-10.dump

The report is then generated in compat_reports/libarrow as a HTML.

API Documentation#

We use Doxygen style comments (///) in header files for comments that we wish to show up in API documentation for classes and functions.

When using clang and building with -DBUILD_WARNING_LEVEL=CHECKIN, the -Wdocumentation flag is used which checks for some common documentation inconsistencies, like documenting some, but not all function parameters with \param. See the LLVM documentation warnings section for more about this.

While we publish the API documentation as part of the main Sphinx-based documentation site, you can also build the C++ API documentation anytime using Doxygen. Run the following command from the cpp/apidoc directory:

doxygen Doxyfile

This requires Doxygen to be installed.

Apache Parquet Development#

To build the C++ libraries for Apache Parquet, add the flag -DARROW_PARQUET=ON when invoking CMake. To build Apache Parquet with encryption support, add the flag -DPARQUET_REQUIRE_ENCRYPTION=ON when invoking CMake. The Parquet libraries and unit tests can be built with the parquet make target:

make parquet

On Linux and macOS if you do not have Apache Thrift installed on your system, or you are building with -DThrift_SOURCE=BUNDLED, you must install bison and flex packages. On Windows we handle these build dependencies automatically when building Thrift from source.

Running ctest -L unittest will run all built C++ unit tests, while ctest -L parquet will run only the Parquet unit tests. The unit tests depend on an environment variable PARQUET_TEST_DATA that depends on a git submodule to the repository apache/parquet-testing:

git submodule update --init
export PARQUET_TEST_DATA=$ARROW_ROOT/cpp/submodules/parquet-testing/data

Here $ARROW_ROOT is the absolute path to the Arrow codebase.

Arrow Flight RPC#

In addition to the Arrow dependencies, Flight requires:

  • gRPC (>= 1.14, roughly)

  • Protobuf (>= 3.6, earlier versions may work)

  • c-ares (used by gRPC)

By default, Arrow will try to download and build these dependencies when building Flight.

The optional flight libraries and tests can be built by passing -DARROW_FLIGHT=ON.


You can also use existing installations of the extra dependencies. When building, set the environment variables gRPC_ROOT and/or Protobuf_ROOT and/or c-ares_ROOT.

We are developing against recent versions of gRPC, and the versions. The grpc-cpp package available from is one reliable way to obtain gRPC in a cross-platform way. You may try using system libraries for gRPC and Protobuf, but these are likely to be too old. On macOS, you can try Homebrew:

brew install grpc