Jonathan Hefner

I can also confirm this bug when in Bash, using [wwtd](https://github.com/grosser/wwtd) which shells out to chruby-exec. And also with the IRB examples mentioned above: ``` $ irb > `chruby-exec 2.6...

@flavorjones Oh wow, I forgot about this one! I am fine with waiting until after #2147 — no rush. Still, I was curious to see if I could get it...

I think this is the same issue as #39431, which was not completely resolved.

@ghiculescu Thanks! :smiley: If my understanding is correct, I think #45413 is closer to `goldiloader`. In this PR, I wanted to take a more conservative approach, to try to avoid...

@shalvah Would you mind also [squashing the commits](https://guides.rubyonrails.org/contributing_to_ruby_on_rails.html#squashing-commits)? :pray:

Doing a squash with `git` followed by a merge is a bit different than GitHub's "squash and merge" button. The former creates a merge commit (which we prefer), while the...

> The version of this change that I had pictured after seeing hotwired/turbo#84 (via #45383) had us retaining the legacy 302 status for "legacy HTTP actions", rather than safe/idempotent ones......

Thank you, @fatkodima! (And @simi, and @kddnewton!) :bow:

Something you might also consider is CPU instruction cache. Although I'm not an expert in the black art of optimizing code for the CPU cache, I do know cached instructions...

As @ixti said, sub vs slice is a bit odd. It would be helpful to know what end result you're trying to achieve before deciding which methods to compare. Also,...