Update supported versions of Ruby

[kotp]

• description in README
• version for test in CI

[kotp]

The impetus was reviewing someone new to Ruby that was using Ruby 2.6 the other day.

[meatball133]

Would it be possible to update the track to Ruby 3.4? thinking Ruby 4.0 should be released quite soon

[kotp]

Would it be possible to update the track to Ruby 3.4? thinking Ruby 4.0 should be released quite soon

https://www.ruby-lang.org/en/downloads/branches/ for reference on EOL schedule.

[meatball133]

Yes I know that 3.3 is supported for a while longer, but thinking most people likely uses the latest version would it be neat to not be super far behind the latest, thinking breaking changes can be introduced as we can see by the ci failing.

[meatball133]

But I still think testing 3.3 makes sense, it just that I think all the tooling and the "main" version should be brought to 3.4

[kotp]

But I still think testing 3.3 makes sense, it just that I think all the tooling and the "main" version should be brought to 3.4

We can avoid deprecated and dropped syntax and test for 3.4, knowing that (some) students will be using prior versions, though.

We could update it to 3.3 though, knowing that we have a end of life coming up, and moving it up at that time, though.

I did conscientiously say "want" rather than "need" to indicate the version of 3.4, since it should not be a necessity currently, but an indication that joy will be more likely had with the newer version.

So exercise-tests.yml could be:

ruby-version: [3.3, 3.4]

While leaving the documentation as it is changed here. What do you think?

[kotp]

The currently failing test is "normal", though undesirable.

[meatball133]

I did conscientiously say "want" rather than "need" to indicate the version of 3.4, since it should not be a necessity currently, but an indication that joy will be more likely had with the newer version.
So exercise-tests.yml could be:
ruby-version: [3.3, 3.4]
While leaving the documentation as it is changed here. What do you think?

Yes I think that makes sense

[meatball133]

The currently failing test is "normal", though undesirable.

I am a bit afraid if that tests fails all the time thinking it is an exercise test though

[kotp]

The currently failing test is "normal", though undesirable.

I am a bit afraid if that tests fails all the time thinking it is an exercise test though

Absolutely this, but looking at using minitest-allow to not have CI fail because of the current "Ractors is experimental" and the Ruby 3.4 consistent failing, while 3.3 does pass (usually).

We would still get the report of the failures for 3.4, while not failing CI.