!JiiOHXrIUCtcOJsZCa:matrix.org

nio

378 Members
The nio matrix python library | Latest release 0.6 https://pypi.org/project/matrix-nio/ | Documentation: https://matrix-nio.readthedocs.io/en/stable/153 Servers

Load older messages


SenderMessageTime
10 Aug 2024
@dkasak:termina.org.ukdkasak
In reply to @anton:rendezvous.anton.molyboha.me
I can compile them -- but they seem to pin the version of vodozemac to an old commit. There a few mentions of "bugfixes" and "security fixes" in the release notes for the intervening versions, though nothing looks super scary on the surface.
Yep, nothing super scary from memory. Should be pretty easy to update to a newer commit though.
16:51:05
@dkasak:termina.org.ukdkasak
In reply to @anton:rendezvous.anton.molyboha.me

I am actually looking at the c++ bindings at the moment, but I come from not knowing rust at all. Could end up being a nice intro into the language for me -- or a complete failure :)

By any chance, is there an appropriate room for me to join, if I need help with this?

Hmm there isn't a vodozemac specific room though maybe that's something we should consider.
16:52:05
@dkasak:termina.org.ukdkasak For now, jumping into #e2e:matrix.org could work. If you need something specific, feel free to ping me or poljar 16:52:58
@anton:rendezvous.anton.molyboha.meanton
In reply to @dkasak:termina.org.uk
For now, jumping into #e2e:matrix.org could work. If you need something specific, feel free to ping me or poljar
Thank you!
16:53:56
@me:shahpaarth.comPaarth Shah - Github Maintainer I ended up messaging that channel just now as well: link 17:57:52
11 Aug 2024
@sfink:mozilla.orgsfink changed their display name from sfink|backAug10 to sfink|covid.20:52:10
12 Aug 2024
@me:shahpaarth.comPaarth Shah - Github Maintainer To follow up, and to provide visibility: after discussing with poljar, we decided to go ahead and create https://pypi.org/project/vodozemac-python/, which is currently a straight copy of the the the vodozemac-bindings/python dir et al. It being under the matrix-nio umbrella is going to be a convenience for us (mainly me, as I don't work for the matrix foundation) to be able to maintain the python bindings freely, and publish updated versions. 17:47:43
@me:shahpaarth.comPaarth Shah - Github Maintainer *

To follow up, and to provide visibility: after discussing with poljar, we decided to go ahead and create https://pypi.org/project/vodozemac-python/, which is currently a straight copy of the the the vodozemac-bindings/python dir et al. It being under the matrix-nio umbrella is going to be a convenience for us (mainly me, as I don't work for the matrix foundation) to be able to maintain the python bindings freely, and publish updated versions.

This is a copy of https://matrix.to/#/!vlnjqGLpLJlFmBSkfQ:matrix.org/$AoZA8ZjCgjt7tw7a5Qz9PPcfUWoNzD9DOQHX8n8w4TY?via=matrix.org&via=privacytools.io&via=envs.net sent in #e2e:matrix.org

17:48:33
13 Aug 2024
@me:shahpaarth.comPaarth Shah - Github Maintainer removed the room topic "The nio matrix python library | Latest RC: 0.25.0rc3 Latest stable release 0.24.0 https://pypi.org/project/matrix-nio/ | Documentation: https://matrix-nio.readthedocs.io/en/stable/".18:28:19
@me:shahpaarth.comPaarth Shah - Github Maintainer set the room topic to "The nio matrix python library | Latest stable release 0.25.0 https://pypi.org/project/matrix-nio/ | Documentation: https://matrix-nio.readthedocs.io/en/stable/".18:28:36
@me:shahpaarth.comPaarth Shah - Github Maintainer @room matrix-nio 0.25.0 has been published, which is just a minor dependency bump from 0.25.0rc.3.
A downstream project (homeassistant/core) was getting soft-blocked by matrix-nio's aiofiles, so I figured it was a good enough excuse as any to finally hit the button on this 🙂
Changelog is in the usual place, and as always, please file a github issue if you see any problems!
18:31:11
@me:shahpaarth.comPaarth Shah - Github Maintainer * @room matrix-nio 0.25.0 has been published, which is just a minor dependency bump from 0.25.0rc.3.
A downstream project (homeassistant/core) was getting soft-blocked by matrix-nio's aiofiles, so I figured it was a good enough excuse as any to finally hit the button on this 🙂
Changelog is in the usual place, and as always, please file a github issue if you see any problems!
18:31:30
@nex:nexy7574.co.uknex (she/it)oh thank god, I didn't know if it was worth opening an issue about dependency blocks lol18:45:18
@nex:nexy7574.co.uknex (she/it) @Paarth Shah - Github Maintainer What's breaking about #489? The changes aren't described 19:07:24
@me:shahpaarth.comPaarth Shah - Github Maintainer
In reply to @nex:nexy7574.co.uk
@Paarth Shah - Github Maintainer What's breaking about #489? The changes aren't described

Darn I knew I was missing something in that changelog lol.
I'll update it later to be clearer, but #489 resolves this issue
Which was technically a Matrix spec v1.1 support issue, relating to the use of /r0/ vs /v3 in endpoints (and for newer endpoints such as threading, /v2... which was very fun to parse through lol)

TLDR: if you have a matrix homeserver that isn't compatible with Matrix v1.1 of the matrix spec, matrix-nio 0.25.0 (nor 0.25.0rc1 or beyond) would work for you. If you've already been using an RC without issue, then there's no problem. If you aren't using a spec-noncompliant homeserver, then there's also no problem.

Realistically, I'm not even expecting anyone to be affected by this, hence why I was lazy about the full cause of the "break"... but it seemed intellectually honest to include it.

19:53:49
@nex:nexy7574.co.uknex (she/it)I think even Dendrite supports v1.2, so afaik all four major homeserver implementations are unaffected by this19:55:05
@me:shahpaarth.comPaarth Shah - Github Maintainer
In reply to @nex:nexy7574.co.uk
oh thank god, I didn't know if it was worth opening an issue about dependency blocks lol
Funny thing is that the original reason I became a maintainer was so that I could get in and fix dependency blocks... so maybe I need to think about a way to make the dependencies a lot more flexible to make this stop being a problem.
Maybe this is my excuse to get rid of poetry :^)
19:55:43
@me:shahpaarth.comPaarth Shah - Github Maintainer
In reply to @nex:nexy7574.co.uk
oh thank god, I didn't know if it was worth opening an issue about dependency blocks lol
*

Funny thing is that the original reason I became a maintainer was so that I could get in and fix dependency blocks... so maybe I need to think about a way to make the dependencies a lot more flexible to make this stop being a problem.
Maybe this is my excuse to get rid of poetry :^)

Anyway, always feel free to make an issue for that, it's one of the easier things to do and I'm happy to handle it/accept PRs for it

19:56:07
@nex:nexy7574.co.uknex (she/it)Don't you make use of GitHub's dependabot? that would probably make most dependency updates an one-click solution in most cases19:56:29
@nex:nexy7574.co.uknex (she/it)I'll open an issue for it so it doesnt get lost19:56:44
@nex:nexy7574.co.uknex (she/it) Opened #515 19:59:51
@me:shahpaarth.comPaarth Shah - Github Maintainer
In reply to @nex:nexy7574.co.uk
Don't you make use of GitHub's dependabot? that would probably make most dependency updates an one-click solution in most cases

I do, and there's even a PR it opened as recently as 3 days ago.
I think the issue with it is twofold:

  • it only automatically modifies rtd-requirements.txt, while actual package requirements are set within pyproject.toml
  • even with modifying pyproject.toml, we'd still need to bump the package version + add a git tag to trigger an automatic publish to pypi (which is still a vast improvement on what we had to do pre 0.25.0rc1, which was manually do python -m build . and manually upload to pypi).

Basically, I have it on my personal todo list to configure dependabot to work better, and to add a github action to automatically bump versions and do things. Once that's in place, this all gets a lot easier to do on a whim

20:00:04
@nex:nexy7574.co.uknex (she/it)Oh yeah, I forgot there's a bunch of different requirements files in the repo20:01:08
@me:shahpaarth.comPaarth Shah - Github Maintainer Perfect world solution is that the pyproject.toml itself gets modified by dependabot/used by readthedocs, and all is well 20:02:11
@nex:nexy7574.co.uknex (she/it)Doesn't rtd require a seperate requirements.txt file in its configuration?20:03:54
@nex:nexy7574.co.uknex (she/it)Oh actually20:04:44
@nex:nexy7574.co.uknex (she/it)https://docs.readthedocs.io/en/stable/config-file/v2.html#packages20:04:45
@nex:nexy7574.co.uknex (she/it)
version: 2

python:
  install:
    - method: pip
      path: .
      extra_requirements:
        - docs
Looks like you can install an extra from pyproject.toml, which might make that a lot easier
20:05:07
@me:shahpaarth.comPaarth Shah - Github Maintainer
In reply to @nex:nexy7574.co.uk
version: 2

python:
  install:
    - method: pip
      path: .
      extra_requirements:
        - docs
Looks like you can install an extra from pyproject.toml, which might make that a lot easier
And immediately I know exactly how I'm going to do this lol
20:15:21
@me:shahpaarth.comPaarth Shah - Github Maintainer
In reply to @nex:nexy7574.co.uk
version: 2

python:
  install:
    - method: pip
      path: .
      extra_requirements:
        - docs
Looks like you can install an extra from pyproject.toml, which might make that a lot easier
* And immediately I know exactly how I'm going to do this lol
Ty for finding this!
20:15:36

Show newer messages


Back to Room ListRoom Version: 4