Hey,
nodens(nodens@debian.org) and me are currently packaging the new version of onionshare for Debian and stumbled over the dependency cepa[1], what is a fork of stem. After digging deeper into it, I found out, that the main reason why they do so is the support for Client Auth v3 onions[2]. I really would like not to package a fork of stem, if possible. That's why I ask, when it is planed to release a new version of stem? It hasn't seen any release since 2019 and there are a lot of changes in master branch too. Maybe you can just release the current state of the maint branch, that is actually what is released as cepa 1.8.3.
I some how need to solve this till January, when Debian is freezing (not accepting new versions).
Regards
hefee
[1] https://github.com/onionshare/cepa [2] https://github.com/torproject/stem/commit/ 2f3bbf64b4ce3b63160a3c56b15748ba626de4a0
Hefee:
Hey,
nodens(nodens@debian.org) and me are currently packaging the new version of onionshare for Debian and stumbled over the dependency cepa[1], what is a fork of stem. After digging deeper into it, I found out, that the main reason why they do so is the support for Client Auth v3 onions[2]. I really would like not to package a fork of stem, if possible. That's why I ask, when it is planed to release a new version of stem? It hasn't seen any release since 2019 and there are a lot of changes in master branch too. Maybe you can just release the current state of the maint branch, that is actually what is released as cepa 1.8.3.
I some how need to solve this till January, when Debian is freezing (not accepting new versions).
That is tricky as stem is not maintained anymore and therefore deprecated. There are no further releases planned and talking to atagar last time he said the master branch was not ready for release. We could think about minor point releases for the 1.8 series but those would just contain small bug fixes (like Python 3.10 compatibility) but no new features.
Hope this helps, Georg
Hey,
That is tricky as stem is not maintained anymore and therefore deprecated.
That is a pitty. It would be nice if you can actually make this obvious on the git repo like "currently stem is unmaintained and therefore deprecated" and maybe search for people in the community to take over - maybe ask people from onionshare? Especially as stem is still marked as the way to interact with tor via python. Like it was done on this ml in August: https://lists.torproject.org/pipermail/tor-dev/2022-August/014768.html
There are no further releases planned and talking to atagar last time he said the master branch was not ready for release. We could think about minor point releases for the 1.8 series but those would just contain small bug fixes (like Python 3.10 compatibility) but no new features.
Having a way to work with Python 3.10+ is still great but will not help for our use case as onionshare using the feature of client auth v3.
regards,
hefee
That is tricky as stem is not maintained anymore and therefore deprecated.
That is a pitty. It would be nice if you can actually make this obvious on the git repo like "currently stem is unmaintained and therefore deprecated"
Hiya. Just a minor terminology correction: Stem is unmaintained but not deprecated. The later implies that there's a replacement. There's other options [1] but for most of Stem's features it's the only game in town.
maybe search for people in the community to take over - maybe ask people from onionshare?
If someone's serious about it we can talk but properly maintaining Stem is a lot of work [2][3].
Especially as stem is still marked as the way to interact with tor via python.
As far as I'm aware it is.
Cheers! -Damian
[1] https://stem.torproject.org/faq.html#are-there-any-other-controller-librarie... [2] https://github.com/torproject/stem/issues/97 [3] https://github.com/torproject/stem/issues/77
Hefee:
Hey,
That is tricky as stem is not maintained anymore and therefore deprecated.
That is a pitty. It would be nice if you can actually make this obvious on the git repo like "currently stem is unmaintained and therefore deprecated" and maybe search for people in the community to take over - maybe ask people from onionshare?
We don't have access to the official repository on Github. We've had done it otherwise. For the taking over part it seems atagar replied already, great.
Especially as stem is still marked as the way to interact with tor via python. Like it was done on this ml in August: https://lists.torproject.org/pipermail/tor-dev/2022-August/014768.html
There are no further releases planned ahttps://lists.torproject.org/pipermail/tor-dev/2022-August/014768.htmlnd talking to atagar last time he said the master branch was not ready for release. We could think about minor point releases for the 1.8 series but those would just contain small bug fixes (like Python 3.10 compatibility) but no new features.
Having a way to work with Python 3.10+ is still great but will not help for our use case as onionshare using the feature of client auth v3.
That's right. atagar: what about this: it seems you already created a separate branch with the change needed by the onionshare folks (nice!). Would you mind creating a point release (or 1.9) so the Debian folks can pick that up, including the Python 3.10 compatibility fix as well? I think that would be the easiest short-term solution to the problem hefee came up with.
Thanks, Georg
regards,
hefee
tor-dev mailing list tor-dev@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-dev
Hello,
A friendly reminder that txtorcon exists and supports v3 onions, (even if developed has slowed somewhat in recent times).
It probably doesn't help onionshare a lot, as txtorcon is a Twisted-based async API so it works best with other Twisted-using code. That said, you _can_ inter-operate with asyncio or even with threaded code if you want. However, it's not going to be a drop-in Stem replacement (and that has never been txtorcon's goal).
https://github.com/meejah/txtorcon/
Documentation available via onion (which self-hosts on the Twisted webserver with txtorcon for Tor support):
http://fjblvrw2jrxnhtg67qpbzi45r7ofojaoo3orzykesly2j3c2m3htapid.onion/
Hi hefee,
On 9/17/22 11:06, Hefee wrote:
There are no further releases planned and talking to atagar last time he said the master branch was not ready for release. We could think about minor point releases for the 1.8 series but those would just contain small bug fixes (like Python 3.10 compatibility) but no new features.
Having a way to work with Python 3.10+ is still great but will not help for our use case as onionshare using the feature of client auth v3.
we've released stem 1.8.1 [1] including v3 onion services patches [2]. There'd be a pypi package soon (in aprox. 1 day) and stem's website update.
Please, let us know whether that works for you,
Cheers,
juga.
Hi hefee,
we've released stem 1.8.1 [1] including v3 onion services patches [2]. There'd be a pypi package soon (in aprox. 1 day) and stem's website update.
Please, let us know whether that works for you,
Thanks a lot. It looks fine for our use case. And good that you now communicate it openly that there is active development for the moment.
Regards,
hefee
Hefee:
Hi hefee,
we've released stem 1.8.1 [1] including v3 onion services patches [2]. There'd be a pypi package soon (in aprox. 1 day) and stem's website update.
Please, let us know whether that works for you,
Thanks a lot. It looks fine for our use case. And good that you now communicate it openly that there is active development for the moment.
Just to get that part less confusing for others following along: I guess you meant that there is *no* active development at the moment, which is right. :)
Georg
Regards,
hefee
tor-dev mailing list tor-dev@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-dev
-
Damian Johnson -
Georg Koppen -
Hefee -
juga -
meejah