Packaging:Minutes20080506

From FedoraProject

Jump to: navigation, search

Contents

Fedora Packaging Committee Meeting of {2008-05-06}

Present

  • JasonTibbitts (tibbs)
  • RalfCorsepius (racor)
  • RexDieter (rdieter)
  • TomCallaway (spot)
  • ToshioKuratomi (abadger1999)
  • XavierLamien (SmootherFrOgZ)

Votes

The following proposals were considered:

Other Discussions

A proposal to change the meeting time to 15:00UTC (two hours earlier) was discussed and will be further discussed via email.

IRC Logs

[12:03]  <tibbs> So, FPC?
[12:04]  <rdieter> spot: ?
[12:04]  * spot is here
[12:05]  <spot> ok, i see rdieter and tibbs are here too
[12:07]  * SmootherFrOgZ is here
[12:07]  <spot> racor, abadger1999?
[12:07]  <abadger1999> 3/4 here.
[12:08]  * abadger1999 is in an ad hoc meeting about pkgdb as well.
[12:08]  <spot> i don't see Rathann or delero
[12:09]  * abadger1999 pings rathannin f-devel
[12:09]  <racor> I am almost absent, this time slot simply doesn't work for me :(
[12:10]  <spot> racor: http://fedoraproject.org/wiki/Packaging/NewMeetingTime doesn't have any entries from you
[12:10]  <spot> (or anyone else, for that matter)
[12:11]  <spot> well, ok, i see abadger1999's blanket acceptance. :)
[12:11]  <abadger1999> spot: I wrote a note that all those times are good for me.
[12:11]  <abadger1999> heh :-)
[12:12]  <tibbs> I don't have significant restrictions, either.
[12:12]  <spot> ok, well, with 3/4 abadger1999 and almost absent racor, that is quorum
[12:12]  <spot> we don't have very much on the agenda today
[12:12]  <spot> http://fedoraproject.org/wiki/PackagingDrafts/PatchUpstreamStatus
[12:12]  <spot> ^ first item
[12:13]  <tibbs> I like the idea; my proposal for extending it to arbitrary metadata didn't gain traction so I guess this one wins.
[12:13]  <abadger1999> Ambivalent.  Comments == Good.  Should items are always kind of wishy-washy, though.
[12:13]  <tibbs> Certainly not a requirement, though.
[12:13]  <spot> if it encourages people to do it, i'm supportive of it.
[12:13]  <racor> spot: I don't see any need to vote. We once had decided to switch meeting time with DST, but this hasn't happened.
[12:14]  <spot> racor: ok, let me be a bit more specific, would 15.00 - 16.00 work better for you?
[12:15]  <racor> 15:00 UTC, that's 17:00 CEST - much better
[12:16]  <spot> i'll send out email proposing that we switch to that time slot, we'll see if everyone can make that work.
[12:16]  <tibbs> 10AM for me; doable most of the time.
[12:16]  <rdieter> good here too
[12:16]  <rdieter> better than now even. :)
[12:16]  <spot> ok, lets vote on http://fedoraproject.org/wiki/PackagingDrafts/PatchUpstreamStatus
[12:16]  <spot> +1 from me
[12:17]  <SmootherFrOgZ> +1
[12:17]  <tibbs> +1 although I will still present a proposal for extending it in the future.
[12:17]  <spot> tibbs: you know how we roll. ;)
[12:18]  <rdieter> +1 , though nervous about slippery slope of wishy-washy common-sense items in guidelines.
[12:18]  <racor> -1, personal preference/stylishness, should not be made part of theFPG
[12:18]  <abadger1999> +1
[12:18]  <rdieter> racor: srsly?  prefer *not* to document patch status?
[12:19]  <tibbs> Well, if we waht to have tools that gather these, then we at least need to document what the tools will accept.
[12:19]  <spot> ok, thats +5 to the draft
[12:19]  <spot> abadger1999: is Javascript ready?
[12:19]  <racor> rdieter: common sense should direct maintainers to document patches at *their* preference.
[12:19]  <tibbs> Honestly I believe that it would be better for the tools to exist first, but nobody but me seems interested so I'll just make them accept what this draft suggests.
[12:19]  <spot> racor: isn't that what a SHOULD means?
[12:20]  <tibbs> racor: How does this guideline not do that?
[12:20]  <abadger1999> spot: No.  I haven't had time to package another library  up.
[12:20]  <abadger1999> tibbs: I'd be interested in seeing your patch metadata in these comments.
[12:20]  <rdieter> tibbs: orthogonal, this doesn't need to block on tools existing or not (tho tools would certainly be preferable)
[12:20]  <racor> rdieter: I see the SHOULD, ...
[12:20]  <tibbs> abadger1999: BTW, it was indicated to me that it's spelled "JavaScript".  Not that I particularly care.
[12:21]  <abadger1999> tibbs: Yah. I need to do that and include the comments from the last meeting.
[12:21]  * abadger1999 fixes spelling now since it's easy
[12:21]  <spot> okay. thats all i have on the agenda for today.
[12:21]  <tibbs> I recall an on-list proposal that people were discussing.
[12:22]  <spot> tibbs: which one?
[12:22]  <tibbs> I forgot to tick it so the article expired from my folder.  There was no draft, so....
[12:22]  <rdieter> caillon had posted awhile back wondering if desktop-file-validate could be added, anyone's feelings change since that was discussed way back when?
[12:22]  <tibbs> Maybe that was it.
[12:23]  <spot> rdieter: oh, yes, i remember that
[12:23]  <rdieter> can't pull up the archive, redhat.com is down.  heh.
[12:24]  <abadger1999> rdieter: I'm +1 to that proposal.
[12:24]  <spot> Since the purpose of this guideline is to validate, I propose to amend the section of the packaging guidelines on desktop-file-install usage[1]  as follows:
[12:24]  <spot> * Rename the sub-heading from "desktop-file-install" to ".desktop file installation and validation"
[12:24]  <spot> * Change the first sentence to:
[12:24]  <spot> <<
[12:24]  <spot> It is not simply enough to just include the .desktop file in the
[12:24]  <rdieter> he wanted this added as an example of d-f-* usage:    desktop-file-validate %{buildroot}/%{_datadir}/applications/foo.desktop
[12:24]  <spot> package, one MUST run desktop-file-install OR desktop-file-validate in
[12:24]  <spot> %install (and have BuildRequires: desktop-file-utils), to help ensure
[12:24]  <spot> .desktop file safety and spec-compliance. desktop-file-install MUST be
[12:24]  <spot> used if the package does not install the file or there are changes
[12:24]  <spot> desired to the .desktop file (such as add/removing categories, etc).
[12:24]  <spot> desktop-file-validate MAY be used instead if the .desktop file's content/location does not need modification.  Here are some examples of usage:
[12:24]  <spot> >>
[12:24]  <spot> * Add the following example:
[12:24]  <spot> <<   desktop-file-validate %{buildroot}/%{_datadir}/applications/foo.desktop
[12:24]  <spot> (sorry for the mini flood)
[12:25]  <spot> seems pretty obviously correct to me.
[12:25]  <rdieter> I'm ok with that too: +1
[12:25]  <spot> hans had an onlist +1
[12:26]  <SmootherFrOgZ> hm, that could be a feature for rpmlint (?)
[12:26]  <abadger1999> +1
[12:27]  <spot> ok, i see +4 total on this proposal
[12:27]  <tibbs> Actually it could be a rpmbuild script as well, I guess.
[12:27]  <SmootherFrOgZ> i'm ok with that too
[12:28]  <tibbs> This is yet another thing that probably shouldn't have to be done explicitly, but until we get there I have no problem requiring, so +1.
[12:28]  <racor> I got to go, bye ...
[12:28]  <spot> ok, thats +5. anyone else want to get a vote for the record?
[12:29]  <SmootherFrOgZ> +1 from me
[12:29]  <spot> alright, any other issues for today?
[12:29]  <tibbs> Can someone write this up into an actual draft?
[12:29]  <spot> i will.
[12:30]  <spot> ok, i think we're done for today.
[12:30]  <spot> thanks all
[12:31]  <abadger1999> Thanks spot