meta: Status update and v2.0.0 holy grail requests

ファイル、テスト、エントリーポイントは指定されていません。まず、ここで参照されている未解決のIssueと要求された機能を確認し、その後、具体的なv1の変更または最終化されたv2 API提案を特定してください。完了とするには、このIssueの広範なロードマップではなく、明確に定義されたスコープとフィードバックに裏付けられた計画が必要です。

I've been gone from this library for a while now but I'm back now. I'm going to be working on it full time till the end of the year.

I'm going to start by fixing all the minor outstanding issues and adding all currently requested features without breaking any backwards compatibility. I want v1 of this library to be rock solid and feature complete to the extent it makes sense.

But after that, I want to work on a v2 where I change the API a bit here and there to entirely perfect it. I don't anticipate major changes as I quite like the API. It has stood the test of time. My main irk with the API is that's a little too constrained. I don't think it should be necessary to sacrifice any performance whatsoever in using my library vs gorilla or gobwas.

I want the comparison to say there is absolutely no rational reason to use any other library except this one if you're writing WebSockets with Go except for stability (though v1 of my library is quite widely used now too though still about 10x less than Gorilla).

The aim is to strike an even better balance between performance and being idiomatic. I have a few ideas but nothing's finalized. I'm also a massively better engineer than I was in 2018 (5 years ago) so there's lots of code quality to improve too.

If anyone has suggestions, please share them here. Ideally those rooted in the long term use of my library. As I fix up v1 and my ideas become very concrete, I'll publish them here first to get feedback before I implement.

cc @tailscale @coder