Steven L

Yes please. I mostly like the new styling, but it's now _extremely_ hard to tell if a reply is part of a thread or not. Particularly on some devices, since...

to stick it in here too: overall looks pretty good. simpler and the overall goal (and why it achieves it) is clearer too. seems like just minor tweaks (many optional)...

> Regarding the atomics, each goroutine is intended to have its own `Buffer`. In other words, a single buffer is not meant to be accessed concurrently. The docs mention the...

re CLA... ... tbh I don't think I can agree with a license that requires someone to share their mailing address. That seems like an absurd overreach.

> There's also some chance you may have attempted to go through the "corporate" license workflow instead of the "individual" one. Asking around, nobody was aware that you'd need to...

Oops. Looks like I'm on an older one, this might be resolved now. I'll dive through the code now / that's released and close this if so.

Nope. It's much-improved by the addition of the testing.T injection (thank you @posener !), but non-test-injected ones still behave like this. I mean, I'm going to upgrade and put testing.T...

In that case! The places I've seen it include: - https://archivebox.io/ in the "Comparison to Other Projects" collapsed section - which, separately, is rendering as plain text instead of the...

A slightly-fragile possibility just occurred to me, and I figured I'd share: Since there really isn't a way in Go to say "resume my execution when a read from X...

I'm not sure if synctest can accommodate all mock-time cases that I need, tbh. Broadly it looks excellent and is almost certainly preferred when possible, but e.g. it's unusable if...