Use CMake for Build System

.circleci/setup_env.sh

@@ -98,9 +98,10 @@ if [ "$(conda list -f qt --json)" != [] ]; then
 fi
 
 echo "Build extensions"
-python setup.py build_ext -q -j3
+cmake .
+cmake --build . --config Release --parallel
 
 echo "Install pandas"
-python -m pip install --no-build-isolation --no-use-pep517 -e .
+python -m pip install --no-build-isolation -e .
 
 echo "done"

.github/actions/build_pandas/action.yml

@@ -12,8 +12,9 @@ runs:
 
     - name: Build Pandas
       run: |
-        python setup.py build_ext -j $N_JOBS
-        python -m pip install -e . --no-build-isolation --no-use-pep517 --no-index
+        cmake .
+        cmake --build . --config Release --parallel
+        python -m pip install -e . --no-build-isolation  --no-index
       shell: bash -el {0}
       env:
         # Cannot use parallel compilation on Windows, see https://github.com/pandas-dev/pandas/issues/30873

.github/workflows/32-bit-linux.yml

@@ -35,9 +35,10 @@ jobs:
           /opt/python/cp38-cp38/bin/python -m venv ~/virtualenvs/pandas-dev && \
           . ~/virtualenvs/pandas-dev/bin/activate && \
           python -m pip install --no-deps -U pip wheel 'setuptools<60.0.0' && \
-          pip install cython numpy python-dateutil pytz pytest pytest-xdist pytest-asyncio>=0.17 hypothesis && \
-          python setup.py build_ext -q -j2 && \
-          python -m pip install --no-build-isolation --no-use-pep517 -e . && \
+          pip install cmake cython numpy python-dateutil pytz pytest pytest-xdist pytest-asyncio>=0.17 hypothesis && \
+          cmake . && \
+          cmake --build . --parallel && \
+          python -m pip install --no-build-isolation -e . && \
           export PANDAS_CI=1 && \
           pytest -m 'not slow and not network and not clipboard and not single_cpu' pandas --junitxml=test-data.xml"
 

.github/workflows/sdist.yml

@@ -43,11 +43,16 @@ jobs:
 
     - name: Install dependencies
       run: |
-        python -m pip install --upgrade pip setuptools wheel
+        python -m pip install --upgrade pip cmake setuptools wheel
 
         # GH 39416
         pip install numpy
 
+    - name: Build pandas
+      run: |
+        cmake .
+        cmake --build . --parallel
+
     - name: Build pandas sdist
       run: |
         pip list

.gitignore

@@ -37,6 +37,10 @@
 *.so
 .build_cache_dir
 MANIFEST
+Makefile
+CMakeCache.txt
+CMakeFiles/
+cmake_install.cmake
 
 # Python files #
 ################

CMakeLists.txt

@@ -0,0 +1,29 @@
+cmake_minimum_required(VERSION 3.18)
+
+set(CMAKE_OSX_DEPLOYMENT_TARGET "10.9")
+set(CMAKE_C_STANDARD 99)
+set(CMAKE_C_STANDARD_REQUIRED True)
+set(CMAKE_CXX_STANDARD 11)
+set(CMAKE_CXX_STANDARD_REQUIRED True)
+
+project(pandas)
+
+if(WIN32)
+  if(NOT CMAKE_BUILD_TYPE)
+    set(CMAKE_BUILD_TYPE
+        Release
+        CACHE STRING "Build type" FORCE)
+  endif()
+  find_package(Python3 REQUIRED COMPONENTS Interpreter Development NumPy)
+  set(CMAKE_LIBRARY_OUTPUT_DIRECTORY_RELEASE "")
+  link_directories(${Python3_LIBRARY_DIRS})
+else()
+  # we only choose Development.Module to support virtual environments where
+  # libpython may not be available see
+  # https://github.com/pypa/manylinux/issues/484
+  find_package(Python3 REQUIRED COMPONENTS Interpreter Development.Module NumPy)
+endif()
+
+add_compile_definitions(NPY_NO_DEPRECATED_API=0)
+add_subdirectory("pandas/_libs")
+add_subdirectory("pandas/io/sas")

Dockerfile

@@ -2,7 +2,7 @@ FROM quay.io/condaforge/miniforge3
 
 # if you forked pandas, you can pass in your own GitHub username to use your fork
 # i.e. gh_username=myname
-ARG gh_username=pandas-dev
+ARG gh_username=WillAyd
 ARG pandas_home="/home/pandas"
 
 # Avoid warnings by switching to noninteractive
@@ -18,7 +18,7 @@ RUN apt-get update \
     && dpkg-reconfigure -f noninteractive tzdata \
     #
     # Verify git, process tools, lsb-release (common in install instructions for CLIs) installed
-    && apt-get -y install git iproute2 procps iproute2 lsb-release \
+    && apt-get -y install build-essential git iproute2 procps iproute2 lsb-release \
     #
     # cleanup
     && apt-get autoremove -y \
@@ -28,10 +28,18 @@ RUN apt-get update \
 # Switch back to dialog for any ad-hoc use of apt-get
 ENV DEBIAN_FRONTEND=dialog
 
+# Without this versioneer will not be able to determine the pandas version.
+# This is because of a security update to git that blocks it from reading the config folder if
+# it is not owned by the current user. We hit this since the "mounted" folder is not hit by the
+# Docker container.
+# xref https://github.com/pypa/manylinux/issues/1309
+RUN git config --global --add safe.directory "$pandas_home"
+
 # Clone pandas repo
 RUN mkdir "$pandas_home" \
     && git clone "https://github.com/$gh_username/pandas.git" "$pandas_home" \
     && cd "$pandas_home" \
+    && git checkout cmake-build \
     && git remote add upstream "https://github.com/pandas-dev/pandas.git" \
     && git pull upstream main
 
@@ -49,5 +57,6 @@ RUN . /opt/conda/etc/profile.d/conda.sh \
     && conda activate base \
     && cd "$pandas_home" \
     && export \
-    && python setup.py build_ext -j 4 \
+    && cmake . \
+    && cmake --build . --parallel \
     && python -m pip install --no-build-isolation -e .

MANIFEST.in

@@ -42,6 +42,9 @@ global-exclude *~
 global-exclude .DS_Store
 global-exclude .git*
 global-exclude \#*
+recursive-exclude **/CMakeFiles *
+global-exclude *.cmake
+global-exclude CMakeCache.txt
 
 global-exclude *.c
 global-exclude *.cpp

doc/source/development/contributing_environment.rst

@@ -42,7 +42,8 @@ Run Container::
     If you bind your local repo for the first time, you have to build the C extensions afterwards.
     Run the following command inside the container::
 
-        python setup.py build_ext -j 4
+        cmake .
+        cmake --build . --config Release --parallel
 
     You need to rebuild the C extensions anytime the Cython code in ``pandas/_libs`` changes.
     This most frequently occurs when changing or merging branches.
@@ -169,7 +170,8 @@ We'll now kick off a three-step process:
    source activate pandas-dev
 
    # Build and install pandas
-   python setup.py build_ext -j 4
+   cmake .
+   cmake --build . --config Release --parallel
    python -m pip install -e . --no-build-isolation --no-use-pep517
 
 At this point you should be able to import pandas from your locally built version::
@@ -216,7 +218,8 @@ You also need to have ``setuptools`` 51.0.0 or later to build pandas.
    python -m pip install -r requirements-dev.txt
 
    # Build and install pandas
-   python setup.py build_ext -j 4
+   cmake .
+   cmake --build . --config Release --parallel
    python -m pip install -e . --no-build-isolation --no-use-pep517
 
 **Unix**/**macOS with pyenv**
@@ -240,7 +243,8 @@ Consult the docs for setting up pyenv `here <https://github.com/pyenv/pyenv>`__.
    python -m pip install -r requirements-dev.txt
 
    # Build and install pandas
-   python setup.py build_ext -j 4
+   cmake .
+   cmake --build . --config Release --parallel
    python -m pip install -e . --no-build-isolation --no-use-pep517
 
 **Windows**
@@ -266,5 +270,6 @@ should already exist.
    python -m pip install -r requirements-dev.txt
 
    # Build and install pandas
-   python setup.py build_ext -j 4
+   cmake .
+   cmake --build . --config Release --parallel
    python -m pip install -e . --no-build-isolation --no-use-pep517

doc/source/development/debugging_extensions.rst

@@ -8,11 +8,12 @@ Debugging C extensions
 
 Pandas uses select C extensions for high performance IO operations. In case you need to debug segfaults or general issues with those extensions, the following steps may be helpful.
 
-First, be sure to compile the extensions with the appropriate flags to generate debug symbols and remove optimizations. This can be achieved as follows:
+First, be sure to compile the extensions with the appropriate flags to generate debug symbols and remove optimizations. This can be achieved on Unix-like systems as follows:
 
 .. code-block:: sh
 
-   python setup.py build_ext --inplace -j4 --with-debugging-symbols
+   cmake . -DCMAKE_BUILD_TYPE=Debug
+   cmake --build . --parallel
 
 Using a debugger
 ================

environment.yml

@@ -13,6 +13,7 @@ dependencies:
 
   # building
   # The compiler packages are meta-packages and install the correct compiler (activation) packages on the respective platforms.
+  - cmake>=3.18.0
   - c-compiler
   - cxx-compiler
   - cython>=0.29.30

pandas/__init__.py

@@ -28,7 +28,8 @@
     raise ImportError(
         f"C extension: {_module} not built. If you want to import "
         "pandas from the source directory, you may need to run "
-        "'python setup.py build_ext --force' to build the C extensions first."
+        "'cmake . && cmake --build . --config Release --parallel' "
+        "to build the C extensions first."
     ) from _err
 else:
     del _tslib, _lib, _hashtable