Get started with Component-based Rails applications!

on YouTube ↗01h 28m

" summary

Claims that componentizing a Rails monolith actually reduces cognitive load and test time, and shows how to do it: extract a domain into a predictor gem by moving models into lib, require_relative the classes, add rspec as a dev dependency, and run tests inside‑out. Uses mountable engines for persistence, keeps one migration per table, exposes test helpers from engines, and fixes bundler pitfalls with a path do block. Extraction finishes in minutes and yields faster, smaller test suites.

AI-written summary · about

" chapters (23)

1. 0:00Introduction
2. 1:30Who is Stefan Hagen
3. 2:47Benefits of componentbased Rails
4. 7:49Example App
5. 8:31App Structure
6. 13:10Unpublished Gems
7. 17:46Prediction Engine
8. 28:23Predictor Component
9. 31:18Fixing the namespace
10. 34:23Fixing the folder structure
11. 36:23Creating teams
12. 44:23Clean up
13. 45:23Add predictor
14. 47:23What happened here
15. 48:23How I shouldnt use Texico
16. 49:23Test breaking
17. 50:23Removing predictor
18. 51:23Main app
19. 53:23Remove dependencies
20. 55:23Dont entangle
21. 56:23Smash it together
22. 57:23Separate UI
23. 58:23Components

" related talks

• 33mRefactoring Towards Component-based Rails ArchitecturesStephan Hagemann · RailsConf 2014
• 32mComponent-based Architectures in Ruby and RailsStephan Hagemann · MountainWest RubyConf 2013
• 34m"Taming your Rails monolith"Mugurel Chirica · RubyC 2019
• 38mAll you need is Rails (Engines): Compartmentalising your MonolithJulián Pinzón Eslava · RubyConf AU 2023
• 35mWrangling Large Rails CodebasesStephan Hagemann · Rocky Mountain Ruby 2012
• 41mMorning KeynoteStephan Hagemann · Birmingham on Rails 2020
• 40mMonoliths Between MicroservicesVladimir Dementyev · RailsConf 2020 CE
• 34mViewComponents in the Real WorldJoel Hawksley · RailsConf 2021