Floris Bruynooghe

Hi, just letting you know I'm not forgetting about this PR. Am still waiting for a round of feedback from @gretchenfrage before doing followup changes.

(Still massaging commits to take feedback into account, these newly pushed commits are not yet ready for a round of review. Will re-request review once ready again.)

> * Why replace "cancel-safe" with "cancellation safe"? This mirrors how tokio talks about it. They use a section header named `Cancel safety` and in prose use "This method is...

I assume what happens is that on dropping the `Connecting` a CONNECTION_CLOSE frame is sent, even though only the initial packet was ever sent. Since CONNECTION_CLOSE is ack-eliciting you then...

I was looking at this because of our multipath branch. I think I can do a fairly easy and clear patch for this based on #2168 and #2195 without too...

pytest-timeout already announces itself in the reporting header like most other pytest plugins. I'm not sure I'm in favour of adding something like this, it adds maintenance complexity for very...

Yeah, if people think the default message and exception types aren't clear enough that's probably more reasonable to change. Feel free to propose something, but I'll put some parameters around...

I guess it is reasonable to not give an error, after all the peer already knows no more data will be written on the stream if it is closed and...

> I am ready to tackle this. > > Please ping me if the team wants to add `--color auto/always/never/` and respect `NO_COLOR` env var. @azzamsa sure, we'd accept a...

Nice! I guess this can be closed then. Feel free to re-open if that's a mistake though.