Roadmap to version 1.0

[pgjones]

I would like to propose that we shortly release 0.14, which feature wise includes the handshake separation and the advice around HTTP/2 WebSockets. Then we wait ~3 months to see if there are any bugs or issues to fix (especially around the HTTP/2 part). Then I think we target the 0.15 release to drop Python 2 support (I'd like to add type hinting throughout). Then again wait ~3 months for any bugs or issues and then release version 1.0. Timing wise I see this as, 0.14 in a few weeks, 0.15 in June/July and 1.0 in October.

This is based on my view that the project is roughly 1.0 ready, which I think is a shared view.

[pgjones]

Looking into this some more, I'd like to drop support for Python 3.5 at the same time. Doing so would allow the events to be converted to dataclasses (there is a backport to 3.6) and hence allow the entire codebase to be typed. The Python 3.5 EoL is September 2020 so we would be dropping support about a year early, based on the October 1.0 target.

[Kriechi]

Fine by me - here a quick reference to the mitmproxy, which dropped py3.5 support mid last year!
mitmproxy/mitmproxy#2266

[pgjones]

See #116.

[pgjones]

So with the merge of #116 we are close to the 0.15 release, are there any pressing features desired for that release? (Otherwise I'd target a release in early June).

[pgjones]

0.15 released (a little later than planned). Assuming it works well I think a bump to 1.0 later this year makes sense - anything missing?

[Kriechi]

@pgjones I think we are pretty close to a v1.0.0 - anything you want to see included but not yet in a PR?
I could cut a release today or next weekend.

[pgjones]

Go for it 🎉

[Kriechi]

fixed by dafe0ca

🎉 👍 💯 🥇