18:00:17 <geppetto> #startmeeting fpc
18:00:17 <zodbot> Meeting started Wed Dec 13 18:00:17 2017 UTC.  The chair is geppetto. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:00:17 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
18:00:17 <zodbot> The meeting name has been set to 'fpc'
18:00:17 <geppetto> #meetingname fpc
18:00:17 <zodbot> The meeting name has been set to 'fpc'
18:00:17 <geppetto> #topic Roll Call
18:00:27 <orionp> hello
18:00:41 <mbooth> hallo
18:00:42 <geppetto> #chair orionp
18:00:42 <zodbot> Current chairs: geppetto orionp
18:00:44 <geppetto> hey
18:00:48 <tibbs> Hey, folks.
18:01:00 <geppetto> #chair tibbs
18:01:00 <zodbot> Current chairs: geppetto orionp tibbs
18:02:54 <geppetto> #chair mbooth
18:02:54 <zodbot> Current chairs: geppetto mbooth orionp tibbs
18:03:03 <geppetto> Oh, hey mbooth  … missed you :-o
18:03:44 <tibbs> Just got back from NOLA last night.  Still putting out fires at work.
18:04:03 <geppetto> Yeh, no new tickets and I'm not optimistic we'll get a 5th
18:04:26 <geppetto> So might be some extra time to put them out
18:04:53 <tibbs> We really do need to work through some of the stuff that's there, though.
18:09:14 <geppetto> #topic Open Floor
18:09:27 <geppetto> tibbs: Anything you want to talk about with only 4?
18:09:34 <Rathann> hi
18:09:38 <tibbs> ...
18:09:38 <Rathann> sorry for being late
18:09:43 <geppetto> #chair Rathann
18:09:43 <zodbot> Current chairs: Rathann geppetto mbooth orionp tibbs
18:09:46 <geppetto> Hey!
18:09:51 <geppetto> #topic Schedule
18:10:00 <geppetto> #link https://lists.fedoraproject.org/archives/list/packaging@lists.fedoraproject.org/message/3XJ4QVXTJTT4TGLURHJFENNRYEJLQGYW/
18:10:40 <geppetto> #topic #726 Review for SELinux Independent Policy packaging Draft
18:10:47 <geppetto> .fpc 726
18:10:49 <zodbot> geppetto: Issue #726: Review for SELinux Independent Policy packaging Draft - packaging-committee - Pagure - https://pagure.io/packaging-committee/issue/726
18:11:50 <tibbs> I haven't had time to distill this into something we could use as a guideline.
18:12:15 <tibbs> I'm not sure it's going to be simple at all.
18:12:17 <Rathann> hm *.fedoraproject.org URLs are timing out for me
18:12:52 <Rathann> ok now it responded
18:13:24 <tibbs> I mean, I can take out the extraneous stuff.  But then we still have the question of whether we are concerned with regulating what is actually in the policy, instead of just how you put a policy into a package.
18:13:44 <geppetto> I don't think we want to be doing that
18:13:53 <Rathann> the guidelines should only cover the latter
18:13:59 <tibbs> I don't either, but look at, say, the last comment in the ticket.
18:14:01 <Rathann> i.e. how you put a policy into a package
18:14:07 <tibbs> Sirry, second to last.
18:14:46 <tibbs> Basically, they're handing us a document which is not remotely suitable to be a packaging guideline.
18:15:07 <tibbs> I'm not sure if that isn't understood, or if it's intentional and they expect us to take pieces from it.
18:15:22 <geppetto> Not sure
18:15:46 <Rathann> I don't think any of us told the submitter about that
18:16:27 <Rathann> on another note, setting selinux booleans should be forbidden unless explicitly authorized by FESCo
18:16:32 <geppetto> #info FPC generally agree we don't want to be documenting selinux policy files.
18:16:55 <geppetto> #topic #723 Guidelines for handling deprecated dependencies during review
18:17:00 <Rathann> like enabling services by default
18:17:00 <geppetto> .fpc 723
18:17:01 <zodbot> geppetto: Issue #723: Guidelines for handling deprecated dependencies during review - packaging-committee - Pagure - https://pagure.io/packaging-committee/issue/723
18:19:39 <tibbs> I don't know that I have anything else to add to that one.
18:21:43 <Rathann> right
18:22:02 <Rathann> the proposal seems sensible
18:22:30 <geppetto> Any info. we can add?
18:23:07 <geppetto> I don't mind asking people to add a deprecated provides … or something … even if we don't MUST NOT depend on them.
18:23:21 <tibbs> Well, it just needs a draft we can use.  The single sentence from the original draft wasn't really useful.
18:23:57 <tibbs> (because there was no way for people to know what was actually deprecated).
18:24:03 <geppetto> #info We generally agree with adding the provides, at least. Need a draft though.
18:24:11 <Rathann> true
18:24:14 <tibbs> So the solution using the Provides: was proposed, and I guess we all agreed.
18:24:21 <geppetto> #topic #720 Easy way of changing/removing shebangs
18:24:22 <Rathann> yes, I'm +1 to that, too
18:24:24 <tibbs> But that needs to be turned into a draft and then voted on.
18:24:35 <geppetto> .fpc 720
18:24:37 <zodbot> geppetto: Issue #720: Easy way of changing/removing shebangs - packaging-committee - Pagure - https://pagure.io/packaging-committee/issue/720
18:24:59 <geppetto> Nothing changed since the last update.
18:25:27 <tibbs> I don't think anything happened upstream in RPM, either.
18:25:50 <tibbs> Honestly I'm happy to drop the macros into redhat-rpm-config but.....
18:26:28 <tibbs> it comes down to the old question of how much we should be documenting these things in the guidelines.
18:26:29 <geppetto> #topic #719 Simplify packaging of forge-hosted projects
18:26:34 <geppetto> .fpc 719
18:26:37 <zodbot> geppetto: Issue #719: Simplify packaging of forge-hosted projects - packaging-committee - Pagure - https://pagure.io/packaging-committee/issue/719
18:26:43 <tibbs> This is another "fancy macro" topic.
18:27:10 <geppetto> This actually looks like a new ticket
18:27:17 <geppetto> Just confused by the old number
18:27:27 <tibbs> Well, it's two months old.
18:27:41 <tibbs> I honestly just do not like the proposal, honestly.
18:28:36 <tibbs> I mean, I agree with the idea of this being made simpler, but I guess I just don't like the implementation.
18:29:16 <tibbs> %forge_metadata should be a macro with parameters instead of having three magic values defined.
18:29:34 <tibbs> And I don't understand why it would have a leading underscore.
18:30:21 <tibbs> I'm not sure that it handles multiple downloaded tarballs at all, though I will accept that case is rare.
18:30:55 <Rathann> I don't see a leading underscore anywhere...
18:31:09 <Rathann> ah
18:31:20 <Rathann> it's there in the initial draft, but gone in the latest
18:31:53 <mbooth> "It does not work in EL 7 because github urls require the use of #" is not true :-)
18:32:07 <mbooth> Github no longer requires the #
18:32:49 <mbooth> I will add a comment
18:33:18 <Rathann> I'm +1 to this
18:33:47 <geppetto> I'm a 0 I think … I don't really mind packagers doing this … but how much better is it than what they'd normally do?
18:33:56 <geppetto> And it's a lot of text
18:35:11 <tibbs> The problem is that it's going to grow without bounds if it tries to accommodate everything.
18:35:22 <tibbs> I understand that too well.
18:36:06 <Rathann> I mean I'm +1 to including the macros
18:36:15 <tibbs> I don't think the effort is misguided, but I don't think that it's really helpful.
18:36:30 <tibbs> I'm certainly -1 to including them as they are currently.
18:37:04 <geppetto> tibbs: Any specific advise you'd give?
18:37:07 <Rathann> tibbs: even the macros?
18:37:15 <tibbs> Especially the macros.
18:37:29 <tibbs> The problem is that once you include them, they're baked in there forever.
18:38:08 <tibbs> We can change a guideline if it's not particularly good.  Macros are basically not changeable once they're in.
18:38:40 <tibbs> And this one needs way more thought.
18:40:00 <Rathann> dp you see any specific issues here?
18:40:21 <tibbs> I mean, the general issue is problematic enough for me to -1 this.
18:41:02 <tibbs> I think it's trying to be too smart.  I'm not sure that amount of smart is going to be supportable long term.
18:41:36 <tibbs> And I personally vastly prefer the "macro with parameters" way of doing this rather than the "define magic globals" approach.
18:42:02 <geppetto> #info Macros need more thought, cleaned up and how to handle corner cases. Maybe look at using parameters.
18:42:34 <Rathann> fair enough
18:42:36 <tibbs> %autosetup and %configure certainly work with parameters; I'm curious whether there's some reason to break with those examples.
18:42:39 <geppetto> #info It might just be trying to be too smart in general though.
18:42:51 <mbooth> Also, sometimes the "URL" tag should not be the github repo
18:42:59 <mbooth> As a project may have dedicated website
18:43:08 <tibbs> Yes, that's rather common.
18:43:16 <tibbs> It would be nice to know how this is targeted.
18:43:37 <Rathann> I'm sure the nim would like to hear about specific issues
18:43:41 <tibbs> I'm OK with something that's simple and simplifies even half of the spec.
18:43:51 <tibbs> These comments are public, so....
18:44:08 <Rathann> mbooth: %{forgeurl}“ use as “URL:”
18:44:09 <Rathann> You do not have to use %{forgeurl} as URL: value if the packaged project has a better customized home page. It is a convenience, nothing more.
18:44:16 <Rathann> it's there in the draft
18:44:21 <geppetto> yeh, when I comment the summary from the meeting it should be easy to see all these comments
18:44:36 <tibbs> But we have a way to specify the project URL.  It's called "URL:".
18:44:59 <Rathann> I don't get your point, tibbs
18:45:04 <mbooth> Rathann: Ah yes.
18:45:25 <tibbs> If you want to specify the project URL, you should just add a URL: tag.  I mean, that's how it's always worked.
18:45:41 <Rathann> when URL: and Source: share part of the URL, why not factor out the common part into a macro?
18:45:59 <tibbs> But it's when they don't which was the topic of discussion.
18:45:59 <Rathann> I see nothing wrong here
18:46:20 <Rathann> but it's already covered, the forgeurl macro is optional, you don't have to use it
18:46:31 <tibbs> But that's exactly my point.
18:46:46 <Rathann> which I'm still not getting, sorry
18:47:07 <tibbs> With these macros, you now have this other way to specify a URL.  You can't just use "URL:".
18:47:18 <Rathann> what? why do you say that?
18:47:24 <Rathann> sure you can
18:47:37 <tibbs> So why does %forgeurl exist, then?
18:47:58 <Rathann> for the common case when project url and source urls share part of the url
18:48:13 <Rathann> like most projects on github
18:48:38 <Rathann> sorry, but I don't see any issue here
18:49:05 <Rathann> %global forgeurl is used by the macros to build %{forgesource}
18:49:16 <Rathann> you can optionally use it in URL: %forgeurl
18:49:20 <Rathann> that's it
18:50:14 <tibbs> I will have to experiment more but I'm just not convinced.
18:50:57 * Rathann shrugs
18:51:14 <Rathann> I'd like to see this applied to Go packaging, though
18:52:28 <geppetto> I'm not sure about "URL: %forgeurl" which you sometimes use … seems like a source of confusion, and not much of a win over the current
18:53:02 <Rathann> geppetto: where's the confusion?
18:53:19 <geppetto> Again, I'd be happy with a group of packagers reusing macros to make packaging a bunch of Go code easier … but I'm much less happy about mandating that using in policy.
18:53:53 <geppetto> Rathann: People using it when they shouldn't, or not when they could … just general N ways of doing something
18:54:11 <tibbs> I frequently open the spec and click the URL.  It's generally annoying when that isn't actually a URL (such as when it includes %name) , but more annoying when it completely hides what the actual URL is.
18:54:37 <Rathann> well that's one point, but you can say the same about Source:
18:55:00 <tibbs> I very rarely open source files in a browser.  Especially tarballs.
18:55:35 <tibbs> That said, I could easily just write something that grabs the true URL from the parsed spec and opens a browser.
18:57:24 <Rathann> alright, let's table this for now and move on
18:57:51 <geppetto> yeh, esp. as %name is so common
18:58:48 <geppetto> #topic Open Floor
18:59:06 <geppetto> Ok, we technically only have a couple of minutes left
18:59:36 <geppetto> Any tickets we really need to talk about?
19:00:34 <geppetto> No real updates on any of the older tickets that I can see
19:00:58 <geppetto> So one thing is that I doubt we'll have another meeting until Jan 2018
19:01:04 <tibbs> True.
19:01:29 <tibbs> I have to be in court next Thursday so I will most likely not be able to be around.
19:01:39 <geppetto> :-o
19:01:42 <geppetto> good luck
19:02:33 <geppetto> Ok, thanks for turning up everyone
19:02:42 <geppetto> #endmeeting