Spring cleaning modules: Looks like we are in a re...
# team-maintainers
f
Spring cleaning modules: Looks like we are in a really good place and handled most PRs already?! Only ones without Status are from @Edmund shakefist
😱 1
Shall we do the issues next @Simon Pearce eyebrows
s
Yeah, just needs following up on the open PRs
And @Edmund needs poking into doing something 😉
f
That's what I meant, we have bumped most PRs and the ones in progress I assume are ones that had updates within last 7 days or so
s
Sure, we can do the issues next. I took a bit of a look yesterday
f
Would be good if we could add the maintainer label if we bumped or checked so we know who looked at what (looking at you @Famke Bäuerle 😉 )
s
I think we should probably close "new module" issues that are more than a few months old
f
ah that was always a bit impractical since i did not go back to the list rip
sadpanda 1
now i would need to check everything that has no maintainer on it - saw too late that not only updating the status was needed
f
No worries, I assume everything without maintainer is either you or Simon 😉
👍 2
Let's tag it when we go back to check the bump or in progress in 2 weeks
f
maybe you can make it a field that can be changed directly from the pr
like the status I think that should be possible
I think we should probably close "new module" issues that are more than a few months old
Sometimes these are also wishlists right? I think I do not completely grasp what these issues are meant for
f
The status field updatets automatically 🙀 😄 ?
s
if you close the PR it'll automatically change the Status to "Closed"
👍 1
(I don't know how that magic occurs, but it seems to)
Hmm, I saw @Edmund start typing but now he has gone quiet...
f
no but you can change it directly on the PR
🙌 1
s
I think we should probably close "new module" issues that are more than a few months old
Thoughts?
f
@Famke Bäuerle You can change maintainer on the PR too!
f
yeah just realized
f
I think we should probably close "new module" issues that are more than a few months old
I did close one of these from myself this morning, so I think I agree! Should maybe ping to see if they plan to work on it during hackathon?
f
I think we should probably close "new module" issues that are more than a few months old
Sometimes these are also wishlists right? I think I do not completely grasp what these issues are meant for @Simon Pearce
I'll add my name iteratively
f
Or if it's more of a wishlist item, can we point them somewhere to put these without clogging issues?
f
Maybe one mega issue where people can add possible tools to a list (link to the tool github repo)?
And then we can close these old issue with a link to that new mega issue that the people should add it there if they still need it but do not implement it themselves?
f
How about a project page?
f
but that would not remove the issues, no?
f
No I meant we close the issues (so they don't clog up issues) and add them to a new projects page that's called something like (modules for the future) or something like that. Depends on what the larger maintainers team think about this. I don't think we have any guidelines for whether you should open issues as "I want this module" but then never work on it...
👍 1
With a projects page, it would still give you the ability to track it, add information but not clog up issues.
👍 1
f
(And link issues in the future - i like it)
this 1
s
I can't really see anyone adding a module just because someone has expressed an interest in having it to be honest
👍 1
The repo generally suggests that you make an issue if you want to a module, and then to tag yourself if you plan on doing it
but I think it is just a mess
I would rather just have issues for bugs/errors etc
👍 1
not for a vague wishlist or "I am going to do this honest"
👍 1
f
So you also on board for modules wishlist project page @Simon Pearce?
That would generally be a lot cleaner I think
And I agree with you. This way, it's impossible to know when there is ACTUALLY something wrong with a module
s
Can do, although I don't really see much value in that either to be honest
f
Maybe we can change the new module template then? Because to me its not clear that if you do not do it yourself nobody else will do it (i mean i can see it as maintainer but not as a random person coming from outside)
💯 1
s
Yes, that is what it currently implies
f
I think the value would be, that you can check this list and see if anyone else has interest in the module you want to make
f
and then i am with you, lets craft a nice message, what do you think about: Closing this issue as it was inactive for a while. New modules and subworkflows are usually not implemented by someone else. If you need this for your pipeline / project just let us know, reopen the issue, try to work on it yourself and ask for help in our slack channel.
f
I did open an issue for a new module and then never got to work on it, so it lignered around, so these things happen
s
Ok, so no issues for new modules?
➕ 1
You can make a wishlist project page if you want
f
Probably should ask the rest of the maintainers?
👍 1
🤷 1
f
Ok, so no issues for new modules?
Probably still needed during Hackathons?
and we tell the people to search for existing issues
Closing this issue as it was inactive for a while. New modules and subworkflows are usually not implemented by someone else. If you need this for your pipeline / project just let us know, reopen the issue, try to work on it yourself and ask for help in our slack channel.
Can I get a thumbs up or down? I would start closing old "new module" issues!
👍 1
Closing the ones from 2022 because this is really driving the numbers up and giving the impression as if we are not doing anything hahaha
🚀 1
s
Probably should ask the rest of the maintainers?
Eh, I think we can just do it personally 😉
👍 1
l
Are we logging somewhere the issues we are going through?
f
I didn't have time to do any issues yet, so from my side no
s
Not sure how best to go through the issues. I'd start with those tagged "new module" and just close them all
f
Will see if I have some time tonight. I guess we are not doing the whole project thing for issues after all?
👍 1
@Famke Bäuerle and @Simon Pearce too fast and in Europe 🐢
s
well, you are welcome to add all the ~600 issues to a project board, 25 at a time 😉
shake no 1
but ~150 are the nf-test migration
f
well i closed a lot of new modulre requests (left those open that are linked somewhere and those that are less than a year old). How do we continue with all the issues ( i know a lot of them are nf-test migration but still its too many)
and I looked at all issues marked with new subworkflow
f
I didn't manage to look yesterday and won't be able to much today, maybe an hour. But yeah would be great to have some directed guidance on what to do because then I can use that hour effectively!
PRs are in a good state with everyone being pinged etc I think right. So only Issues to go. How to divide and conquer?
f
maybe we can do a project board for all of them that are not nf-test migration issues?
and then sort them into bumped, bugs that need fixing and are thus open issues and tracking issues?
f
I could do that today if we agree. I would import all issues and same strategy as for PRs. I know there are many, but seems like the best approach?
👍 3
l
This would be amazing, thanks! I went through some old issues that are not new modules, but I did not tag them in any way - if we have the project I will log them there 🙂
👍 1
f
@Florian Wuennemann will you send the project board when you have one? 🙂 I also reviewd a bunch of issues and would update the status then iteratively
f
Yes, will work on it after lunch a bit and share here!
It seems like views in projects show the same data, so I would need to create a new project for Issues I believe...
s
That's fine, you can make a new project
f
I need an owner of nf-core to make this public please keanu thanks https://github.com/orgs/nf-core/projects/102/views/1
Starting to import issues now loading
@Simon Pearce I actually managed to create an automated way to do this with the gh cli (and some cursor help)!
❤️ 1