19:00:04 <alexsaezm> #startmeeting Go SIG meeting
19:00:04 <zodbot> Meeting started Mon Jan 16 19:00:04 2023 UTC.
19:00:04 <zodbot> This meeting is logged and archived in a public location.
19:00:04 <zodbot> The chair is alexsaezm. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions.
19:00:04 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
19:00:04 <zodbot> The meeting name has been set to 'go_sig_meeting'
19:00:10 <alexsaezm> #topic Roll Call
19:00:16 <alexsaezm> Hi everyone!
19:00:57 <gotmax[m]> .hello gotmax23
19:00:58 <zodbot> gotmax[m]: gotmax23 'Maxwell G' <gotmax@e.email>
19:01:13 <gotmax[m]> @room The Go SIG meeting is starting
19:01:29 <gotmax[m]> Hi alexsaezm!
19:01:36 <copperi[m]> .hello copperi
19:01:37 <zodbot> copperi[m]: copperi 'Jan Kuparinen' <copper_fin@hotmail.com>
19:01:50 <alexsaezm> Hi gotmax (He/Him)
19:02:36 <alexsaezm> we can wait to 05 for everyone :)
19:02:46 <gotmax[m]> +1
19:03:48 <gotmax[m]> I'm working on a script to detect Go leaves
19:04:11 <alexsaezm> uh!
19:04:18 <gotmax[m]> ~13% of packages that BuildRequire golang are leaves
19:04:21 <alexsaezm> wait you add that to the issues right?
19:04:25 <alexsaezm> we can start from that issue :)
19:04:31 <gotmax[m]> SGTM
19:04:50 <alexsaezm> #topic Leaf Go library packages https://pagure.io/GoSIG/go-sig/issue/48
19:05:39 <alexsaezm> 13%?
19:05:42 <gotmax[m]> I'm thinking about making a Change proposal to retire these leaves
19:05:44 <alexsaezm> that sounds like a lot to me
19:05:48 <gotmax[m]> (the last comment on that ticket)
19:06:21 <gotmax[m]> what's a lot? too agressive, too much work, or there's a lot of leaves?
19:07:07 <alexsaezm> a lot of leaves O: didn't expect that
19:07:36 <gotmax[m]> yeah
19:07:46 <alexsaezm> wait, just to be sure, can you clarify to me at least, what exactly quantifies as leaf?
19:07:59 <gotmax[m]> yeah, I had meant to do that :)
19:08:21 <alexsaezm> is it a program that was imported as a library?
19:08:28 <alexsaezm> but not in use anymore?
19:08:43 <gotmax[m]> I consider go library packages that don't contain any binaries and don't have any dependents
19:08:47 <gotmax[m]> alexsaezm: Exactly
19:09:01 <gotmax[m]> See https://git.sr.ht/~gotmax23/fedora-scripts/tree/main/item/go-sig
19:09:33 <alexsaezm> thanks! gotmax (He/Him) ++
19:10:28 <gotmax[m]> binaries are packages that contain binaries and leaves are what I've mentioned. outliers are library only packages that have dependents (or packages that my script erroneously filtered out ;)
19:11:05 <gotmax[m]> I'd prefer to discuss with eclipseo before proposing a change to remove these, but not sure is he's around
19:11:11 <alexsaezm> I'll add comment to the issue, but my vote here is to go ahead and prune those leaves.
19:11:37 <gotmax[m]> Asking individual maintainers is not going to get us too far
19:11:39 <alexsaezm> right
19:11:51 <gotmax[m]> And this feels like a good way to clean up the Go stack
19:11:59 <alexsaezm> talking about that, I think he is busy, I hope he is doing fine, I have a bug that he took that I want to move :D
19:12:04 <copperi[m]> +1
19:12:05 <alexsaezm> (talking about eclipseo)
19:12:19 <alexsaezm> gotmax[m]: ++
19:12:31 <mikelo_m[m]> gotmax (He/Him): I’m seeing some familiar packages, I will check them later. Not in front of my laptop atm
19:13:03 <gotmax[m]> I guess I can type up a draft proposal
19:13:42 <alexsaezm> that guess sounds like an action to me :D (thanks a lot for the work on this!!
19:14:12 <gotmax[m]> #action gotmax to get the ball rolling on a Change proposal to remove Go leaves
19:14:17 <alexsaezm> thanks a lot
19:14:37 <alexsaezm> fancy a new topic?
19:14:55 <gotmax[m]> I'll have to think a little more about how to got about this
19:15:08 <alexsaezm> gotmax (He/Him): If I can help, just say it :)
19:15:09 <gotmax[m]> mainly how to notify maintainers and allow them to opt out
19:15:42 <gotmax[m]> but I'd like to get more feedback from the Go SIG before bring this to the wider community/FESCo
19:15:44 <mikelo_m[m]> Force orphan and wait 6 weeks? </troll>
19:15:56 <mikelo_m[m]> (Just trolling)
19:16:05 <gotmax[m]> :D
19:16:24 <alexsaezm> lol
19:17:46 <gotmax[m]> Mass filling Bugzillas and retiring the package's whose bugs are still open after a month might work or maybe just make a simple opt out list in a hackmd or wiki page or something
19:17:58 <gotmax[m]> Anyways, I'll give it more thought and we can probably move on :)
19:18:17 <gotmax[m]> mass filling bugs might be a problem for maintainers with a lot of packages
19:19:01 <alexsaezm> also, not sure if spamming people with emails will do any favor
19:19:46 <gotmax[m]> I think sending an email to devel-announce and BCCing maintainers
19:20:01 <gotmax[m]> but make sure to deduplicate the BCC list
19:20:21 <gotmax[m]> * BCC list :)
19:21:31 <alexsaezm> as you said, you can always think about any kind of approach you feel comfortable and discuss that in the next go sig meeting
19:21:43 <alexsaezm> * feel comfortable with and discuss
19:22:51 <gotmax[m]> +1
19:23:21 <alexsaezm> For the next topic, we have two you gotmax (He/Him) created long ago, want to talk about any of those? (talking about docker containers and ix86)
19:23:33 <gotmax[m]> Next topic should probably be the long term FTBFS packages that will be retired next month
19:23:38 <gotmax[m]> https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/message/54KWSXRNUO5R6UCGBKWDTJZCUIBOM4MC/ is the link
19:23:46 <gotmax[m]> but it's not loading for me...
19:24:09 <alexsaezm> oh sorry, didn't have that in the issues. Absolutely (it loads here)
19:24:33 <gotmax[m]> I think it's just Firefox...
19:24:46 <gotmax[m]> #topic Long term FTBFS packages to be retired in February
19:24:51 <gotmax[m]> #link https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/message/54KWSXRNUO5R6UCGBKWDTJZCUIBOM4MC/
19:25:39 <alexsaezm> Is anything there that we should be worried about? I see a lot of golang packages
19:25:48 <gotmax[m]> Yeah, there's 33 go packages
19:26:31 <gotmax[m]> When I looked at this earlier, 5 or 6 of the go ones were leaves
19:26:41 * alexsaezm freaked out because he searched for golang in the link and got 2k results
19:27:16 <gotmax[m]> I need https://bugzilla.redhat.com/show_bug.cgi?id=2160202 reviewed to fix one of the FTBFS packages
19:27:31 <gotmax[m]> I'm worried that this will tank the entire Go stack...
19:28:16 <alexsaezm> you mean if it doesn't get approve?
19:28:19 <alexsaezm> your bug?
19:28:40 <gotmax[m]> No :). I mean the mass retirement in general
19:29:55 <alexsaezm> I see a lot of packages that look interesting
19:29:55 <alexsaezm> I thought they were in good shape
19:29:55 <gotmax[m]> helm has been broken for a while
19:29:56 <alexsaezm> like httpdump and helm
19:30:09 <gotmax[m]> well, I guess that was a dumb comment
19:30:31 <gotmax[m]> they all have been :)
19:30:54 <gotmax[m]> * well, I guess my comment was dumb
19:31:02 <mikelo_m[m]> I fixed some of the FTBFS for packages that were most depended, but had no time to continue :/
19:31:14 <gotmax[m]> yeah, there's a lot
19:31:28 <davide> I was looking at helm yesterday
19:31:54 <davide> I saw there's reviews up for the missing deps needed to update it (Oras)
19:32:13 <davide> I meant to review tonight, and once they're in I'll try to update it again
19:32:49 <gotmax[m]> Updating golang-github-docker to v22 might also be a probably for helm
19:32:51 <gotmax[m]> https://src.fedoraproject.org/rpms/golang-github-docker/c/235c3e3ffc356037b4b3796241f87eee0aa7b14c?branch=rawhide
19:33:24 <davide> Good to know, thanks
19:33:48 * alexsaezm has a bugzilla search for the FTBFS but it only returns two bugs :-/
19:35:37 <gotmax[m]> I had a magic FTBFS Bugzilla query bookmarked somewhere, but I think that was before I reinstalled my laptop :/
19:37:16 <alexsaezm> If I can retrieve a list of all of the packages I can review the packages, my concern is, if the list is too long, how we can tell which ones are more relevant (to focus on them)
19:38:14 <davide> For my packages i was planning to use the packager dashboard to try and find the ones in need of work
19:38:53 <gotmax[m]> Looping over the packages and seeing which have the most dependents is an option
19:38:59 <alexsaezm> oh right, the packager dashboard lists them
19:39:11 <gotmax[m]> You can use my new repoquerying tool if you'd like :) https://git.sr.ht/~gotmax23/fedrq
19:39:17 <gotmax[m]> </selfplug>
19:39:38 <davide> Oh excellent, I will try that, thanks!
19:40:25 <gotmax[m]> Should we move on to https://pagure.io/GoSIG/go-sig/issue/49 ?
19:40:53 <alexsaezm> if you want to... I created two issues
19:41:10 <alexsaezm> that's one of them
19:41:24 <alexsaezm> #topic Merging specfiles
19:42:09 <gotmax[m]> Tangentially related question: can RHEL take Fedora's patch to disable the GOPROXY?
19:42:33 <alexsaezm> TL;DR: RHEL specfile for go and for Fedora are like day and night. I want to address this and make RHEL as similar as I can to the Fedora one, without changing the functionality (mostly due the macros)
19:42:57 <alexsaezm> gotmax (He/Him): As far as I can tell, yes, but need to check it
19:44:25 <alexsaezm> I don't have anything done yet, but I wanted to rise this idea that was requested by a coworker, in case the Go SIG had any concerns
19:44:40 <gotmax[m]> I'm a bit hesitant merging the specfiles
19:44:40 <gotmax[m]> I think merging parts, but trying to keep them exactly the same won't work
19:45:06 <alexsaezm> to be honest, I'm not sure how exactly will look like, I got the request last week and I didn't give it too much thought
19:45:08 <gotmax[m]> For example, the Source lines will have to conflict due to RHEL using the FIPS fork/patches
19:45:39 <alexsaezm> yeah, there I need to do some trick with flags
19:46:17 <alexsaezm> I mean, an approach to that is to make FIPS patches available in Fedora too (not forcing them by any means)
19:47:01 <alexsaezm> Needless to say, I won't submit anything without the whole Go SIG checking it first.
19:47:17 <alexsaezm> and, I might be wrong here, but I do recall that LLVM is looking into the same idea
19:50:55 <alexsaezm> As always, any critics, concerns are more than welcome :)
19:51:08 <gotmax[m]> I'll probably have more to say once I see the implementation :)
19:52:53 <alexsaezm> Absolutely
19:52:59 <alexsaezm> thanks
19:53:23 <alexsaezm> I opened another issue for today
19:53:44 <alexsaezm> #topic golang-race package relevance
19:53:47 <alexsaezm> #link https://pagure.io/GoSIG/go-sig/issue/50
19:54:39 <alexsaezm> So my question is: do we want the golang-race package? I don't see anything using it and it's going to mess up go1.20
19:55:54 <gotmax[m]> I don't have much of an opinion here :)
19:57:04 <alexsaezm> As I don't understand exactly who would want the golang-race package, I'll send an email to devel, but my idea is to remove it, I can't make 1.20 work with it
19:57:17 <gotmax[m]> Ah, so it's breaking the build?
19:57:34 <alexsaezm> yes, well kinda
19:57:39 <gotmax[m]> If you decide to go ahead with it, I'd add it to the Change Proposal
19:57:46 <alexsaezm> good idea
19:57:48 <alexsaezm> so for the record
19:57:52 <gotmax[m]> and communicate to devel@ and golang@
19:58:49 <alexsaezm> we build go libraries with the -race, that used to put the output in a folder under pkg/, then we move it, and create the golang-race, easy peasy.
19:59:21 <alexsaezm> now what happens is that the output of the build is not store in that folder anymore, it uses the cache so it makes the whole thing way more complicated
19:59:40 <alexsaezm> hence my question of how much work worth this
19:59:54 <alexsaezm> hope my explanation makes sense
20:01:12 <alexsaezm> I'll write an email to devel@ and golang@
20:01:21 <gotmax[m]> Sounds good
20:01:36 * gotmax[m] notes the time
20:02:19 <gotmax[m]> Re the Go proxy issue, I filled a RHEL 9 bugzilla https://bugzilla.redhat.com/show_bug.cgi?id=2161390 :)
20:04:02 <alexsaezm> gotmax (He/Him): thanks, I'll follow it up (probably take it from David)
20:04:46 <alexsaezm> do we want to jump into the open floor? at least for a couple of minutes
20:04:54 <gotmax[m]> +1
20:05:23 <mikelo_m[m]> gotmax (He/Him): thanks for reviewing my go2rpm PR
20:05:23 <alexsaezm> #topic Open floor
20:06:24 <gotmax[m]> mikelo_m: Sure. I need to finish that.
20:07:05 <gotmax[m]> I remember having two minor nitpicks left. I think one was a missing trailing comma and the other was a redundant is None.
20:07:40 <gotmax[m]> * I remember having two minor nitpicks left. I think one was a missing trailing comma and the other was a redundant is not None.
20:08:08 <mikelo_m[m]> I’ll check it later, the comma might be there still, but is none should not. Will check later
20:08:24 <gotmax[m]> Thanks
20:08:31 <gotmax[m]> I left a comment with what I just said
20:08:38 <mikelo_m[m]> Ack
20:09:01 <gotmax[m]> I'll probably cut a release after your change gets merged
20:10:22 <mikelo_m[m]> awesome
20:11:11 <alexsaezm> From y side, nothing really relevant: 1.18.10 and 1.19.5 are on bodhi (in testing)
20:11:17 <alexsaezm> s/y/my/
20:13:34 <alexsaezm> if nothing else needs to be discuss... we can call it for today :)
20:14:58 <gotmax[m]> SGTM
20:16:11 <alexsaezm> Thanks everyone for another awesome meeting :)
20:16:12 <alexsaezm> #endmeeting