Spark Sprint GuideFrom Java to your first Apache Spark PR

A 30-hour study plan for developers who know Java but not Python, built around the "Your first PR to Apache Spark" project at PyCon Taiwan 2026 Sprints.

Start with the plan

Event details

Realistic scope

The project lead asks for Python, git and GitHub, with coveragepy as a plus. You do not need to know Spark or PySpark. The plan targets exactly that.

Spark-first order

Environment and the Spark build come first, because PySpark tests need a compiled Spark. Study happens while the build runs.

Java to Python bridge

A cheat sheet that maps what you already know (types, null, interfaces, try-with-resources) to idiomatic Python, plus typing, mypy and coverage.py.

Trilingual

Traditional Chinese (Taiwan), English and Spanish, with the same structure on every page.

How to use this guide

Read The event once (5 minutes) so you know what is expected and what is not.
Do Block 0 of the 30-hour plan tonight: accounts, fork, and start the Spark build in the background.
Follow the blocks in order. Each one ends with a "done when" test so you know when to move on.
Use the Spark setup checklist on the last morning to verify everything works before you leave.
Keep Resources open for the links.

Unofficial. This is a community study guide. It is not affiliated with PyCon TW, the Apache Software Foundation or the Apache Spark project. Dates, requirements and commands were checked against the official pages on 2026-10-08; always re-check the official sources for last-minute changes. Items marked "verify" were not confirmed and should be checked in the repository you cloned.