Date: Sat, 24 Feb 2024 03:24:49 +0000 From: bugzilla-noreply@freebsd.org To: python@FreeBSD.org Subject: [Bug 256885] [NEW PORT] www/py-homeassistant: Open-source home automation platform Message-ID: <bug-256885-21822-TxXgPsNvc0@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-256885-21822@https.bugs.freebsd.org/bugzilla/> References: <bug-256885-21822@https.bugs.freebsd.org/bugzilla/>
next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D256885 --- Comment #25 from cgqix35t4@mozmail.com --- I was working on this and updating it, and frankly it's pretty gnarly. I'm = not entirely sure it's feasible. I also went a bit further than the OP for bett= er or worse as I'd really prefer not to have to include all the build-time dependencies in my jail. They've become a pain to manage. In no particular order, issues I ran into: 1. Upgraded to 2024.1.3 (monthly release cycle) 2. Required Python 3.11 (it would be a port only) 3. Project pins to exact versions at aggressively updates (making it churny= to keep in sync) 4. HA installs packages at runtime with the `default_config` via pip (not listed directly in requirements) Ideally, I'd like to see the default install not need to reach for pip (whi= ch means a fair amount of ports). Even better if I could also support common m= ajor integrations. So I got through most of this into #4 before I hit the blocker. It ended up pulling in python-matter-server which uses the https://github.com/project-chip/connectedhomeip SDK. The SDK implements the Matter protocol and doesn't support being built on FreeBSD at the moment. I= 'm sure there could be interested dev's on that, it's out of my league. This is where I hit my 'crap.' point, as that's writing C interfaces to the FreeBSD networking stack. It's possible and I think there would be interest by that community, but it made me pause. I'll work on cleaning up what I did get up to and putting it on GH to share= . I took the poudriere+portshaker approach to this because of how many ports we= re necessary. P.S. I also ended up heavily hacking up pytoport heavily to support modern python packaging requirements and recursively follow dependencies to facili= tate the process. But that project seems inactive, so I'm not sure I could get a= ny of the useful behaviors cleaned up and upstreamed. Unfortunately, at the end of the day I'm not sure if a port could be mainta= ined without a significant and ongoing effort. At this time the project is focus= ed on their currently supported platforms and hasn't been interested in porter= s. I'm not sure how much interest/engagement it'd take on the FreeBSD side to = move that needle. Even if it's as a "Tier 2" target in FreeBSD's parlance. --=20 You are receiving this mail because: You are on the CC list for the bug.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-256885-21822-TxXgPsNvc0>