<@james:fedora.im>
16:01:20
!startmeeting fpc
<@meetbot:fedora.im>
16:01:21
Meeting started at 2024-08-01 16:01:20 UTC
<@meetbot:fedora.im>
16:01:21
The Meeting name is 'fpc'
<@james:fedora.im>
16:01:24
!topic Roll Call
<@tibbs:fedora.im>
16:01:49
Hey.
<@james:fedora.im>
16:01:59
!hi
<@zodbot:fedora.im>
16:02:00
James Antill (james)
<@limb:fedora.im>
16:02:13
!hi
<@zodbot:fedora.im>
16:02:14
Gwyn Ciesla (limb) - she / her / hers
<@carlwgeorge:matrix.org>
16:02:28
!hi
<@zodbot:fedora.im>
16:02:31
Carl George (carlwgeorge) - he / him / his
<@conan_kudo:matrix.org>
16:02:50
!hi
<@zodbot:fedora.im>
16:02:52
Neal Gompa (ngompa) - he / him / his
<@decathorpe:fedora.im>
16:03:18
hey 👋🏼 just wanted to let you know, I just got home from the dentist and am barely functional with pain killers, so I'll skip this one unless you need me specifically
<@salimma:fedora.im>
16:03:51
!hi
<@zodbot:fedora.im>
16:03:52
Michel Lind (salimma) - he / him / his
<@james:fedora.im>
16:03:57
We already have 5, so go lie down. :)
<@salimma:fedora.im>
16:04:10
oh interesting, do you get the checkmark if you're one of the chairs? that helps counts attendance
<@james:fedora.im>
16:04:22
Yeh
<@nhanlon:beeper.com>
16:05:25
!hi
<@zodbot:fedora.im>
16:05:27
Neil Hanlon (neil) - he / him / his
<@carlwgeorge:matrix.org>
16:05:29
gonna be double dipping with the infra meeting going on right now, wish me luck
<@james:fedora.im>
16:05:35
The old bot used to have a chair command, the checkmark helps me now there isn't one.
<@jonathanspw:fedora.im>
16:06:57
!hi
<@zodbot:fedora.im>
16:06:59
Jonathan Wright (jonathanspw)
<@james:fedora.im>
16:07:39
!topic FPC PR#1354 https://pagure.io/packaging-committee/pull-request/1354
<@james:fedora.im>
16:07:53
I think we can just merge this ... any objections?
<@limb:fedora.im>
16:08:23
None from me.
<@conan_kudo:matrix.org>
16:08:58
none from me
<@tibbs:fedora.im>
16:09:36
Not from me; I was +1 a long time ago.
<@conan_kudo:matrix.org>
16:11:19
the note about must not owning the directory makes sense as well
<@james:fedora.im>
16:11:42
Okay.
<@james:fedora.im>
16:11:54
!topic FPC PR#1378 https://pagure.io/packaging-committee/pull-request/1378
<@james:fedora.im>
16:12:12
Again, speak or I click merge soon :)
<@carlwgeorge:matrix.org>
16:12:30
+1
<@carlwgeorge:matrix.org>
16:12:50
it's weird we didn't have this page sooner
<@salimma:fedora.im>
16:12:52
unofficial +1
<@conan_kudo:matrix.org>
16:12:57
+1
<@carlwgeorge:matrix.org>
16:12:58
but that's how docs go
<@tibbs:fedora.im>
16:13:45
I guess.
<@tibbs:fedora.im>
16:14:22
It's odd that it was already clearly indicated in the guidelines.
<@limb:fedora.im>
16:14:38
do it
<@tibbs:fedora.im>
16:15:01
But organizing this stuff is hard.
<@james:fedora.im>
16:16:36
!topic FPC PR#1379 https://pagure.io/packaging-committee/pull-request/1379
<@conan_kudo:matrix.org>
16:17:14
this one is reasonable too
<@conan_kudo:matrix.org>
16:17:18
+1
<@james:fedora.im>
16:17:34
Carl George: Did you want wording changes here before it gets merged?
<@carlwgeorge:matrix.org>
16:17:50
ideally yeah, seems miro agreed but hasn't made the tweak yet
<@tibbs:fedora.im>
16:17:52
From a grammar standpoint I have always preferred "build dependency" or "build-time dependency" over something like "build require", but it doesn't make much difference.
<@conan_kudo:matrix.org>
16:18:18
I would prefer not using the phrase "build require" and use "build-time dependency"
<@carlwgeorge:matrix.org>
16:18:26
i like that, would further enforce that it doesn't have to be a literal `BuildRequires`
<@conan_kudo:matrix.org>
16:18:33
in my experience, it confuses people :)
<@conan_kudo:matrix.org>
16:18:41
so making it a different phrase is clarifying
<@carlwgeorge:matrix.org>
16:18:45
we could also merge and i can send that tweak as a follow up
<@james:fedora.im>
16:19:46
Yeh, I'm not sure if it's easy to tweak it via. the web UI.
<@carlwgeorge:matrix.org>
16:20:10
it's also no rush if we want to just wait for miro to adjust it
<@james:fedora.im>
16:23:40
!topic FPC PR#1368 https://pagure.io/packaging-committee/pull-request/1368
<@james:fedora.im>
16:23:52
I think we are maybe ready to merge this one too, now?
<@tibbs:fedora.im>
16:25:53
I think so. The exceptions are going to be unsatisfying but dumb decisions far upstream don't give us much choice.
<@james:fedora.im>
16:29:16
Okay, I merged it ... let the complaints begin
<@james:fedora.im>
16:29:28
!topic FPC PR#1300 https://pagure.io/packaging-committee/pull-request/1300
<@conan_kudo:matrix.org>
16:30:32
I agree with that rationale. +1
<@james:fedora.im>
16:30:39
I think this is fine, but I'm not sure ... I remember talking about how macros should be used for install'ing only, but I'm not sure if it was about this ticket or another one.
<@james:fedora.im>
16:31:20
The given benefit isn't really a thing in Fedora, but it still seems good.
<@carlwgeorge:matrix.org>
16:32:54
might be good to have a flatpak-focused individual like yselkowitz weigh in
<@yselkowitz:fedora.im>
16:33:12
!hi
<@conan_kudo:matrix.org>
16:33:12
we do have SCLized packages in Fedora too
<@zodbot:fedora.im>
16:33:13
Yaakov Selkowitz (yselkowitz)
<@conan_kudo:matrix.org>
16:33:25
some of the compiler toolchain packages are like this
<@conan_kudo:matrix.org>
16:34:24
look at the gdb spec if you want your eyes to bleed
<@yselkowitz:fedora.im>
16:34:36
`Requires: %{_bindir}/foo` breaks flatpak builds all the time
<@yselkowitz:fedora.im>
16:34:47
same for BRs btw
<@carlwgeorge:matrix.org>
16:35:24
so if i understand you correctly, we should keep the guideline, i.e. reject this pr to revert adding it
<@yselkowitz:fedora.im>
16:36:03
imo we should get away from file dependencies entirely
<@carlwgeorge:matrix.org>
16:36:27
that's a bigger conversation that the scope of this meeting, and the topic of whether to merge this or not
<@james:fedora.im>
16:36:39
I read it was don't use macros in any requires, so apply the revert.
<@yselkowitz:fedora.im>
16:36:59
re-permitting installation path macros in file dependencies will make flatpaks harder
<@yselkowitz:fedora.im>
16:37:21
I've been using the current policy to fix a lot of them
<@carlwgeorge:matrix.org>
16:38:00
the current guidelines say using macros in path deps MUST NOT be used. this pr removes that paragraph.
<@yselkowitz:fedora.im>
16:39:02
wrt flatpaks I would ask that the current policy remain
<@yselkowitz:fedora.im>
16:39:39
SCL macros I don't have a problem with, they're not relevant to flatpaks afaics
<@yselkowitz:fedora.im>
16:39:54
but regular installation path macros are a big problem
<@james:fedora.im>
16:40:17
I'm confused now ... I thought the PR was to revert removing the MUST NOT.
<@carlwgeorge:matrix.org>
16:40:47
the original commit added the MUST NOT https://pagure.io/packaging-committee/c/35d89f0f3ff118359f1fef1c328570a944e0ddf8
<@yselkowitz:fedora.im>
16:40:50
yes, and I would urge you to reject the PR
<@carlwgeorge:matrix.org>
16:41:04
this pr proposes reverting that commit, removing the MUST NOT
<@james:fedora.im>
16:41:28
Damn it ... I thought you were agreeing and merged it. Going to have to revert the revert now.
<@carlwgeorge:matrix.org>
16:41:30
admittedly confusing with the double/triple negative parsing
<@tibbs:fedora.im>
16:42:03
I think this is going to require more extensive discussion that isn't going to be possible as a part of this meeting.
<@yselkowitz:fedora.im>
16:42:18
why would you merge it while discussion is ongoing?!?
<@tibbs:fedora.im>
16:43:44
At the very least, any final guideline is going to have to include some discussion of why doing this kind of thing will cause problems for some systems.
<@james:fedora.im>
16:46:30
Maybe we could have more explanation? Seems like another issue/PR.
<@james:fedora.im>
16:46:30
I'll fix up the bad merge in an hour or so, unless someone wants to help me out. No need to discuss it, just re-propose the initial commit and merge referencing my comment in this PR . Sigh.
<@james:fedora.im>
16:46:30
<@james:fedora.im>
16:46:36
!topic Open Floor
<@james:fedora.im>
16:46:44
Anything else anyone has?
<@conan_kudo:matrix.org>
16:50:19
nope
<@tibbs:fedora.im>
16:50:21
Nothing from me.
<@james:fedora.im>
16:52:27
!endmeeting