How does Bundler work, anyway?
summary
History shows why Bundler exists and how it works: it runs an install-time dependency resolver to pick exact versions (RubyGems adds multiple versions so conflicts are possible), writes Gemfile.lock, downloads and installs those exact gems, and bundle exec replaces the load path so only locked gems load. Resolver heuristics cut pathological cases to under ten seconds for most projects. It also offers binstubs to avoid bundle exec and a local caching server gem stash with an initial release candidate due tonight.
AI-written summary · about
chapters (11)
- 0:00Intro
- 0:15how does bundler work, anyway?
- 1:03a brief (?) history of dependency management
- 3:13let's share ruby code
- 10:53How do setup.rb? ruby setup.rb setup ruby setup.rb config ruby setup.rb install
- 11:14use ruby libraries 1. Find a cool library 2. Download the library 3. Untar the library
- 12:29no versions no uninstall
- 14:07maybe a little bit incredibly tedious
- 18:54Welcome to the team, here's your machine. We expect setting up to take a week.
- 25:20The moral: runtime resolution, install-time resolution
- 34:16or if you're too busy, fund this work.
