惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

I
InfoQ
博客园 - 司徒正美
爱范儿
爱范儿
F
Fortinet All Blogs
J
Java Code Geeks
量子位
酷 壳 – CoolShell
酷 壳 – CoolShell
博客园 - 三生石上(FineUI控件)
腾讯CDC
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
S
SegmentFault 最新的问题
Microsoft Security Blog
Microsoft Security Blog
T
The Blog of Author Tim Ferriss
V
V2EX
L
LangChain Blog
aimingoo的专栏
aimingoo的专栏
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
云风的 BLOG
云风的 BLOG
T
Tailwind CSS Blog
Blog — PlanetScale
Blog — PlanetScale
GbyAI
GbyAI
A
About on SuperTechFans
有赞技术团队
有赞技术团队
Y
Y Combinator Blog

DEV Community

Authentication Security Deep Dive: From Brute Force to Salted Hashing (With Java Examples) Why AI Systems Don’t Fail — They Drift Spilling beans for how i learn for exam😁"Reinforcement Learning Cheat Sheet" I Replaced Chrome with Safari for AI Browser Automation. Here's What Broke (and What Finally Worked) How Python Borrows Other People's Work The $40 Architecture: Processing 1 Billion API Requests with 99.99% Uptime Vibe Coding: A Workflow Guide (From Zero to SaaS) Most webhook security guides protect the wrong side. The scary part is delivery. Headless CMS for TanStack Start: Build a Blog with Cosmic EU Age Verification App "Hacked in 2 Minutes" — What Actually Happened Comfy Cloud’s delete function does not actually remove files Running AI Models on GPU Cloud Servers: A Beginner Guide Event-driven media intelligence with AWS Step Functions and Bedrock I scored 500 AI prompts across 8 quality dimensions — here's what broke How to Call Google Gemini API from Next.js (Free Tier, No Backend Needed) The Portal Protocol: Reclaiming Human Connection in the Age of AI How to Fix Your Team's Scattered Knowledge Problem With a Self-Hosted Forum Intro to tc Cloud Functors: A Graph-First Mental Model for the Modern Cloud Designing Multi-Tenant Backends With Both Ownership and Team Access I Built a Neumorphic CSS Library with 77+ Components — Here's What I Learned PostgreSQL Performance Optimization: Why Connection Pooling Is Critical at Scale Cómo construí un SaaS multi-rubro para gestionar expensas en Argentina con FastAPI + Vue 3 🚀 I Built an Ethical Hacking Scanner Tool – Open Source Project I Replaced /usage and /context in Claude Code With a Single Statusline A Pythonic Way to Handle Emails (IMAP/SMTP) with Auto-Discovery and AI-Ready Design I Collected 8.9 Million Polymarket Price Points — Here's What I Found About How Markets Really Move EcoTrack AI — Carbon Footprint Tracker & Dashboard Everyone's Using AI. No One Agrees How. 5 self-hosted ebook managers worth trying in 2026 Building Your First AI Agent with LangChain: From Chatbot to Autonomous Assistant
How I Fixed Bugs in 30+ Open Source Projects (And What I ...
Alex Chen · 2026-06-13 · via DEV Community

Alex Chen

How I Fixed Bugs in 30+ Open Source Projects (And What I Learned)

Over the past few months, I've been contributing to open source as an independent developer. No big company backing, no team — just me, a laptop, and a lot of caffeine. Along the way, I've submitted pull requests to 30+ repositories across the Python, JavaScript, TypeScript, and Rust ecosystems.

Here's what I learned from the process — the good, the bad, and the "I wish someone told me this earlier."

Why Contribute to Open Source?

Let's get the obvious out of the way: it's not about the money (at least not directly). Most bounties pay $50-$500, and you'll spend 10-20 hours on a single PR if it involves deep codebase exploration.

The real value is:

  1. Reputation — Each merged PR is a public signal that you can read, understand, and improve other people's code
  2. Learning — You'll see how major projects are structured, tested, and maintained
  3. Network — Maintainers remember helpful contributors. Jobs come from these relationships
  4. Scratching your own itch — Fix a bug that annoys you? Everyone benefits

My Process: Finding Good Issues

Step 1: Pick the Right Projects

Not all projects are equally welcoming to new contributors. Here's my filter:

Signal Good ✅ Bad ❌
Response time < 7 days > 30 days or never
Issue labels good first issue, help wanted None
CI/CD Green, fast builds Broken, 30min+ builds
PR merge rate > 60% of open PRs merge < 20% merge

Step 2: Find Issues You Can Actually Fix

I look for:

  • Bug reports with clear reproduction steps — Someone already did the hard work of identifying what's wrong
  • Issues labeled easy-fix or similar — The maintainer thinks it's approachable
  • Issues in domains I know — Don't pick a C++ compiler bug if you've never written C++

Step 3: Before Writing Code

This is where most beginners fail. Don't start coding yet!

  1. Read the CONTRIBUTING.md — Every project has different style, commit message format, and PR requirements
  2. Look at recent merged PRs — What do good PRs in this project look like?
  3. Comment on the issue first — Say "I'm working on this" so others don't duplicate effort
  4. Set up the dev environment — Make sure tests pass before you change anything

Real Examples from My Contributions

Example 1: One-Line Fix (Pygments)

In pygments/pygments#3127, the CUDA lexer was inheriting from the C lexer instead of the C++ lexer. Since nvcc is a C++ compiler, modern CUDA code uses C++ features like constexpr and class.

The fix: One line change

# Before
class CudaLexer(CLexer):
# After  
class CudaLexer(CppLexer):

Sometimes the smallest fixes have the biggest impact.

Example 2: TLS Socket Leak (Tornado)

tornadoweb/tornado#3614 was trickier. When a TLS handshake times out in TCPClient.connect(), the underlying socket was leaking because ownership transferred to a new SSLIOStream that was then discarded on timeout.

The fix required:

  1. Understanding Tornado's stream ownership model
  2. Adding proper cleanup in the error path
  3. Writing a test that reproduces the race condition

Total time: ~3 hours. Lines changed: ~15. But the learning about async socket handling was invaluable.

What Goes Wrong (And How to Handle It)

1. Your PR Gets Ignored 😴

This happens. A lot. Some maintainers are overwhelmed, some projects are understaffed, sometimes your PR just gets lost in the shuffle.

What to do:

  • Wait 2 weeks before pinging
  • When you ping, add value: "I noticed X merged recently, wondering if there's anything else needed for this one"
  • If another month passes, it might be time to move on

2. Your PR Gets Rejected 🙅

Don't take it personally. Code review is about the code, not about you.

Common reasons:

  • Missing tests — Always include tests when possible
  • Style mismatch — Read the project's style guide
  • Scope creep — Keep PRs focused on one thing
  • Performance concerns — Your "fix" might be slower than the original!

3. You Get Blocked From a Project 🚫

Yes, this happened to me. Some projects have aggressive anti-spam filters, and submitting multiple PRs quickly can trigger them.

What to do:

  • Don't panic
  • Move on to other projects — there are millions of repos
  • Maybe reach out to maintainers via other channels (Discord, email) to understand why

Tools That Helped Me

  • GitHub CLI (gh) — Essential for everything: creating PRs, checking statuses, managing forks
  • git-worktree — Work on multiple issues simultaneously without stashing
  • pytest / jest / go test — Whatever the project uses, master it
  • IDE with good git integration — I use VS Code with GitLens

By The Numbers (So Far)

Metric Value
PRs submitted 30+
Repos contributed to 15+
Merged (so far) Still waiting on most...
Average wait time for review 2-8 weeks
Biggest lesson Patience matters more than code quality

What's Next?

I'm continuing to contribute to projects I use daily. The goal isn't to hit some number — it's to become the kind of developer who can jump into any codebase and make meaningful improvements.

If you're thinking about contributing but haven't started: open an issue right now. Find a project you use, look at their issue tracker, and pick something small. Your future self will thank you.


Have you contributed to open source? What's your experience been? Let me know in the comments!