| Short Ruby Newsletter | https://shortruby.com |
| Website | https://allaboutcoding.ghinda.com |
What kind of methods should *not* be converted into endless methods? Methods that call `puts`? Methods include `==`? Methods with many args? How complex can a single-line method get before it should not be converted into an endless method?
Personally I think that methods who's body is a single equals comparison (ex: `@foo == 42`) look confusing when converted into an endless method due to the combination of `=` symbols.
def forty_two? = @foo == 42
vs.
def forty_two?
@foo == 42
end
I am working this weekend to finalise the content for the upcoming #GoodEnoughTesting workshop:
- I am trimming a lot of the previous content to keep it to only the essential
- Rethinking the code and exercises I previously used
Testing isn’t about doing everything. It’s about doing the right things and doing enough of them to sleep well at night.
If you're trying to balance quality, speed, and complexity, risk gives structure to your thinking and clarity to your priorities.
Use risk as a spotlight.
After running five sessions of Good Enough Testing workshops last year, in the first months of this year, I decided to pause for a bit. The reason? I needed time to focus on my health.
It was the right call. Now I am ready to come back with fresh energy and a new format.
@lucian If you’re looking for more events for the next edition of #shortruby, the tiny ruby #{conf} CFP is open for submissions.
I am part of the organising team for @friendlyrb - a cozy #Ruby conference happening in Bucharest, Romania in September.
Now is a good time to buy your ticket - the spots for 3rd day trip are limited - in case you are thinking to join it.
Ticket at https://friendlyrb.com
A reminder of two things where I contribute and invite you to join:
I curate @shortrubynews :
> It is a long and comprehensive Monday morning summary of the articles, discussions, and news from the Ruby community
Subscribe at newsletter.shortruby.com
"Rails is still one of the fastest ways to start with a strong foundation that works in production at a global scale"
"Ruby’s commitment to whimsy feels almost revolutionary, punk, even"
Source: https://stevengharms.com/posts/2025-06-07-the-delightful-art-of-rubys-impenetrable-naming
When using LLMs to write tests, you should remain in control of choosing the test cases or at least review them.
Here is an example of prompting Claude Code to identify test cases for my review before asking it to start writing the test cases.
@rahoulb Ah yes, this is pretty similar with that mindset:
I first use LLMs to help choosing the tests that I want to have.
I then review them, sometimes I add or remove cases.
Then I use LLMs to write the tests
And of course then it follows the implementation.
Sometimes I give it also for example the public interface of an object because I know what I want to have at the end.
postmodern
Matias Korhonen
Baz