It’s that time of year again. It seems like about once a year I get interested designing a schema library. This post is a collection of my latest ideas and design goals, mostly based on what I’ve learnt from the previous three or four implementations.
This topic is probably interesting to a tiny subset of developers, and super boring to everyone else. I’ve tried to write this post in a way that is accessible to a wider developer audience, but you have been warned!
This is a quick look at the recent CI/release automation I wrote for the byebug-skipper gem.
Config for standard-version, which is what bumps the version and updates the changelog using the commit history.
CI step that ensures commit messages have the correct format
I wanted to write down some of my thoughts on end-to-end testing, mainly to help clarify my own thinking.
RSpec examples should have enough detail in the descriptions to rewrite them from scratch.
Start with high-level tests, and step down to lower levels when the implementation is stable.
Why did Apple not test what happens when a user supplies “True” as their last name? Is it incompetence?
https://twitter.com/JPaulGibson/status/1368182852668583948
I don’t think so. I think the explanation is simpler: it’s not feasible. It’s highly impractical, if not computationally impossible, to test every input.