arrow

This example shows how to use foga to build, test, and document Apache Arrow, the columnar data platform. It combines system package installation, a native CMake build, Python installs, pytest, and Sphinx docs in one working configuration.

The example pins Arrow to commit 36c8c9a (apache-arrow-24.0.0.dev-318-g36c8c9a24a) to keep the demo reproducible.

Typical workflow

Start the example container:

examples/arrow/run-in-docker.py

Once inside the container:

1. Validate the configuration before running any workflow:

   foga validate
   

2. Build the Arrow C++ targets:

   foga build cpp
   

3. Install the Python test environment and run tests:

   foga install --target test-env
   foga test
   

4. Build the arrow documentation through the configured docs workflow:

   foga docs
   

What this example demonstrates

• one config coordinating system setup, native build, Python install, tests, docs, and cleanup

• a credible upstream integration instead of a toy repository

• heavier docs handling, where foga docs rebuilds Arrow C++ with a docs profile and reinstalls docs-env in pre-hooks before running Sphinx