Loic Nageleisen

So, this is WIP: previous work + messy `SAVEPOINT` commit that I'll clean up later, but I needed to push some base to discuss upon. I did it TDD/BDD by...

Might be, it fetches linux-musl when (I assume) Heroku is glibc-based? I really have to find some time to get back on this and finish the PR (should happen soon)

> (should happen soon) Hah, joke's on me. No time for now but I'm not forgetting about this.

I'm going to try to redo this based on [the rubygems 3.0 + bundler 1.17 patch I produced](https://gist.github.com/lloeki/12dcf61324f64a2fa1e8a8b2109c1f00#file-musl-patch), which seems to produce good results. Of course only the specs will...

@virusman not as is. FYI I'm rethinking how it should work and what the expectations of the `platform.version` component should be so that it doesn't break anything. Do you have...

So nice, thanks @deivid-rodriguez. Since comparing a gem platform and the actual ruby platform is asymmetric, I was wondering whether we should introduce a `Ruby::Platform` or `Gem::RubyPlatform` to distinguish from...

You're welcome, there were so many users impacted by this I could don't let that one slide :)

I report the same issue. I produced [libv8-node](https://rubygems.org/gems/libv8-node) builds for the following platforms, published s such: - `x86_64-darwin-19` - `x86_64-darwin-20` - `x86_64-linux` (IOW glibc-based) - `x86_64-linux-musl` - `x86_64-solaris-2.11` On the...

Yup, the hack is only necessary for versions of rubygems before your fix, so thanks for that!

Regarding the "or either has no version" thing, I think this is e.g for darwin or solaris, so that maybe you can have a gem published as `x86_64-darwin` or `x86_64-solaris`...