New release on Pypi ?

[ncrocfer]

Hello,

I was looking for a tool to autoreload my code when I develop and I found this commit : 52c2a84

So it seems Sanic already integrates it since December 07, 2017. But the the latest version on Pypi dates from the day before (1ea3ab7) : https://pypi.org/project/Sanic/#history

Is-it possible to release a new version on Pypi please ? Other features (like the UUID support in routes) are also interesting :)

Thanks in advance !

[c-goosen]

pip install https://github.com/user/repo.git@branch

[ncrocfer]

@c-goosen Yes it's what we are doing now, but I still think pip install sanic is more convenient than pip install https://github.com/channelcat/sanic.git@master 😔

Six months without a new release on Pypi for an awesome project like Sanic... People who read blog posts on this framework will just try pip install sanic, not the full path including the branch.

[ashleysommer]

There is already another thread about requesting a new Sanic release, here:
#1170

[c-goosen]

Well we should discuss integrating with circle CI or other tools with a scheduled release cycle.

[ashleysommer]

@c-goosen
What is your reasoning behind that? How would a scheduled release cycle benefit the Sanic project?

[vlad0337187]

I think, it would be just enough to make new release )

[sjsadowski]

Guess I'll go ahead and +1 this, if only for the ipv6 support.

[leotada]

Who is responsible?

[sjsadowski]

@seemethere seems to be the one to tag and release, not sure if that also carries the responsibility to push to pypi.

[GoodiesHQ]

Personally, I don't like having to specially get the Sanic repo for support for something like nested Blueprints which has been a feature for a while now.
a10d746

[ajutras]

I'm having issues pushing for adoption of Sanic over Flask in my organisation at the moment mostly because of the lack of releases. The lack of a release cycle undermine the confidence in the product.

I've read the few different issues opened about the lack of releases and I still cannot find a good reasoning as to why Sanic stopped doing releases. So what's up?

[sjsadowski]

@channelcat @seemethere Any ideas here?

[leotada]

@r0fls is one of the more active in 2018. Can you make a new release? If no one can make this, maybe we have to migrate for an organization in other repo. @yunstanford @seemethere

[r0fls]

I tried and I get this message 😢 :

HTTPError: 403 Client Error: The user 'r0fls' isn't allowed to upload to project 'Sanic'. See https://pypi.org/help/#project-name for more information. for url: https://upload.pypi.org/legacy/

Seems pypi is hoping to fix this eventually: pypi/warehouse#1506

In the meantime, only @seemethere can do a release. @seemethere any chance you want to change your pypi to a password that you can share with me so that I can do a release and calm the masses?

[ashleysommer]

@r0fls
This small PR #1286 will need to be merged before a release, because auto_reload is currently broken for all users except those on Mac OS.

[henry0312]

@r0fls you don't need to get the password. @seemethere will just add you to collaborators on PyPI :)

Example:

[ashleysommer]

Also would be good to get #1179 and #1278 merged before a release, but I don't want to set a precedent of everyone pleading to get their PRs merged before the release.

[yunstanford]

It'd be great if we have a team/official process for operating release.

I'd also like to see this #1179 merged in so we could get feedbacks.

[ahopkins]

I'll be quite honest..... I've heavily adopted sanic in production, but did so in a way I could easily switch it out if I need to. The lack of release schedule has me and my team worried and I've considered the effort to remove it. I already took it out of my gateway service.

Seeing this move to an organization would give the project a lot. Sanic's biggest weakness (in my opinion) right now is the lack of project direction. This could give some great momentum to an otherwise great framework.

I would be happy to help whatever organizational work would need to be done to accomplish this.

[yunstanford]

@ahopkins agree, moving to an organization helps a lot.

[vltr]

I would be glad to help as well.

[sjsadowski]

I would also be willing to assist, if needed.

[seemethere]

Cutting a release today (@r0fls will handle the release notes), will contact @channelcat to see if we can move this to an organization so we can get more people who can help out with the project.

[seemethere]

Apparently @channelcat hasn't verified his email address on PyPI so uploading to PyPI is currently broken, trying to see if there's a workaround.

[TheGrandmother]

I totally agree with @ahopkins. I really like Sanic but it seems like @channelcat haven't been that active in the project for a while (at least from what I can tell) and that kinda scares me.

[sjsadowski]

Has anyone tried to contact @channelcat directly via email to try and get these changes made?

I hate to ask this but is it time for those of us who have availability and desire to do a community fork of this in order to make it sustainable and build some redundant leadership? I've been using sanic for some time now and I think a number of us on this thread want to be able to keep the project going and give back.

[TheGrandmother]

@sjsadowski I think forking should be a last alternative.

[vltr]

@sjsadowski I had this idea a million times and discussed with @ahopkins a million times as well, but we also agree with @TheGrandmother's opinion that forking should be the last alternative here. We all have the feeling that @channelcat looks abandoned, but yet this might not be true, so we have to wait a little bit more and see (and yes, this is rather frustrating, I can't deny that).

[ahopkins]

I agree that maybe now is not the time to fork. Sounds like @r0fls may be making some progress (hopefully). Hopefully we can get this moved to an organization. If not..... might I suggest pip install tailz 😂

[sjsadowski]

I don't disagree with @TheGrandmother, @vltr, or @ahopkins - but there is clearly a lot of concern about release cadence and PR management. The fact that this issue has been open for 2 months with @channelcat referenced multiple times and zero response.

I'd rather keep pushing this project forward, personally, with increased community involvement to alleviate the pressures that come from the situation we're in now.

[seemethere]

We're looking into moving the project into an organization with a larger team of maintainers so changes are incoming but the current team has been extremely busy with other things.

Sorry for the wait guys, this has been frustrating at both ends.

[c-goosen]

Need volunteers and maintainers to help? Think we can gain more from helping than criticising.

[c-goosen]

Also can we move some of the sanic-jwt/sanic-openapi etc under the same organization?

[ahopkins]

@c-goosen I can only speak for sanic-jwt. If there is enough interest from the community, I am not opposed to it. However, I currently intend to continue maintaining it as a third party package. I use it myself everyday at least you can feel comfortable that it is and will continue to be active so long as the sanic project is active. There are a decent group of contributors to the package, and you are welcome to join.

I am sure many of you would agree with me, one of the greatest strengths of sanic is the barebones nature of it, and that the framework does not force you to develop in one particular way. I strongly believe that options and flexibility in a framework are good. Therefore, my biggest hesitation in moving the project would be minimizing this.

[TheGrandmother]

@ahopkins Why do you think that moving to an organization would affect the nature of sanic?

[c-goosen]

@ahopkins I agree 100%. I use it every day as well and I love it thus far. I also love the plugin/modular nature of sanic and the ecosystem.

What I propose is a organisation with multiple projects, so you can still choose what you wish to bolt onto sanic.

[ahopkins]

Not saying it would change the nature of sanic per se.... Just that I know (for example) I am generally more inclined to use something from the same organization than not. I'm not so sure that sanic-jwt is or should be the "sanctioned" way of doing authentication. On the other hand.... I am happy to share the code, adopt new users, and if there is sufficient interest, move it to an organization.

[yunstanford]

Any update or any plan ?

[ahopkins]

What about moving some of this discussion about the planning of how and when over to the Gitter channel. I mean ... its not like much else is going on over there.

[yunstanford]

i feel like issue pages have better visibility. Thoughts ?

[TheGrandmother]

Since the Gitter channel seems fairly stale, continuing the discussion on github might be prudent.

[ahopkins]

Fair enough.... was just a suggestion 😉

I withdraw my statement.

[TheGrandmother]

@ahopkins It's okay. The council of the elders of the internet forgives you ^-^

[dmitry-sobolev]

Any updates on this issue? Version 0.8.0 still not in PyPI :(

[ahopkins]

Hey everyone ... I started a mailing list to bring together a "task force" of like minded developers wanting to push this into a community backed organization. I urge everyone that has any inclination to help push Sanic forward into the future to join the conversation:

https://groups.google.com/forum/#!forum/sanic-dev

The goal is to continue this discussion with a set of specifics on how we can achieve this for the betterment of the Sanic community.

[abuckenheimer]

8,1 now available on pypi 🎆