Jeremy Evans

I also don't understand why this was closed and #1932 was opened. As far as I can see, the name `rack.closed` was never discussed prior to this being closed. I...

I think this breaks backwards compatibility too much. I have submitted an alternative backwards compatible proposal that I think offers a nicer API: https://github.com/rack/rack/pull/1881

The plan that this PR sets forth (removing the warnings and breaking code in Rack 3.1) is not backwards compatible. I think if the goal of the PR is to...

> I essentially see this as a bug fix PR where as your proposal is a new feature. Rails is already working around this behaviour. The implementation here should be...

> > I do not see this as a bug fix API. The existing *_header methods are poorly named, but they were never intended to do the type of mangling...

I'm not against a 2.2.4 release. However, if we are going to do a bug-fix 2.2.4 release, we should evaluate other fixes committed since 2.2 and see if any make...

That's because it isn't currently an official rack spec. However, considering the support in falcon, puma, and unicorn on the webserver end and rails, roda, and hanami on the web...

I think I figured out the cause of the segfault. It's at 0xffffffff, so I'm fairly sure `mmap` is returning -1, but we're not catching that and raising an exception....

I think the reason that the example doesn't work on OpenBSD, even after the previous commits, is that OpenBSD does not allow direct system calls for security reasons. Syscalls are...

@64kramsystem Thanks! It would also be worth it to see if this pull request runs correctly on MacOS, to make sure it doesn't break anything. I'm guessing there may be...