Jean-Paul Calderone

> @exarkun, if I will try to solve `cancel` semantics difference, does this PR have sense for you? If not, I will close it. I'll leave this to someone else...

What spec does this change conform to? A link to the relevant section might be nice somewhere in here.

> > > The environ parameter is a dictionary object, containing CGI-style environment variables. > > therefore it inherits the CGI spec'd version of this variable. Just checking - there...

> fix the bug in Hyperlink's strategies that generates bogus URLs https://github.com/python-hyper/hyperlink/issues/153 I fixed this almost a year ago with absolutely no response from the maintainers. Any ideas how to...

> > > fix the bug in Hyperlink's strategies that generates bogus URLs [python-hyper/hyperlink#153](https://github.com/python-hyper/hyperlink/issues/153) > > > > > > I fixed this almost a year ago with absolutely no...

Something that flake-utils could offer here is an alternative to `eachSystem` or a helper to go along with it. As `eachSystem systems builder` calls `builder` with values from `systems` and...

If the idea is "who cares about bugs in the URL parser, just let it be broken", then the Hypothesis strategy can be adjusted to avoid those cases.

> > Note that `nix-store --export / --import` is pretty deprecated. Its format is undocumented and it will not have a replacement in the new CLI. It's inefficient because it...

This has been on my mind for a couple days in a general way as I re-acquaint myself with the codebase. I've been thinking that the factoring of many of...

> This will require that callers use `maybeDeferred`. Where would this be? It's typically possible to avoid forcing `maybeDeferred` on users by internally handling the synchronous case with `succeed` and/or...