17:01:20 #startmeeting fpc 17:01:20 Meeting started Thu Mar 19 17:01:20 2015 UTC. The chair is geppetto. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:01:20 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:01:20 #meetingname fpc 17:01:20 The meeting name has been set to 'fpc' 17:01:20 #topic Roll Call 17:01:27 pretty sure we are going to have 4 again 17:01:49 #chair mbooth 17:01:49 Current chairs: geppetto mbooth 17:02:03 orionp: tibbs: ping 17:02:28 hello - FYI Rathann can't make it 17:02:34 yeh, I saw 17:02:38 #chair orionp 17:02:38 Current chairs: geppetto mbooth orionp 17:02:44 OK, I'm around. 17:02:48 why I assume we'll only have have 4 again :( 17:02:52 #chair tibbs 17:02:52 Current chairs: geppetto mbooth orionp tibbs 17:03:21 limburgher racor Rathann SmootherFr0gZ spot tomspur: FPC ping 17:03:24 Sorry about being out earlier. Wasn't planning on being pulled into a meeting. 17:03:33 yeh, no problem … happens to us all 17:03:46 Was Rathann here earlier? 17:03:56 no, had same thing … emergency at work 17:03:59 * SmootherFrOgZ here 17:03:59 Would hate to think I was the one that broke quorum. 17:04:08 #chair SmootherFrOgZ 17:04:08 Current chairs: SmootherFrOgZ geppetto mbooth orionp tibbs 17:04:17 cool, we made it anyway :) 17:04:45 #topic Schedule 17:04:47 https://lists.fedoraproject.org/pipermail/packaging/2015-March/010506.html 17:05:01 #topic #511 Change request: Package Guidelines: DevAssistant 17:05:07 https://fedorahosted.org/fpc/ticket/511 17:05:28 Got my +1 already. 17:06:14 https://fedoraproject.org/w/index.php?title=User%3ATradej%2FDraft%3APackaging%3ADAP&diff=406424&oldid=401659 … is the diff 17:06:33 Looks sensible enough to me 17:06:55 yeh, +1 17:06:56 I think it's nice of them to accommodate. 17:07:15 +1 with changes 17:07:25 +1 17:07:29 +1 17:08:07 #action DevAssistant change for license files (+1:5, 0:0, -1:0) 17:08:09 * orionp wonders about java maven packaging... 17:08:21 #topic #512 Java guidelines change 17:08:25 orionp: That this one? 17:08:30 https://fedorahosted.org/fpc/ticket/512 17:08:45 Or related to dev assistant in some way? 17:08:56 no, wondering about license files 17:09:02 Diff: ​https://fedoraproject.org/w/index.php?title=User%3AMsrb%2FJavaPackagingDraft&diff=406064&oldid=406063 17:09:08 but I think that's handled normally 17:09:54 java folks were pretty solidly behind this change 17:10:06 Also already has my +1. 17:10:22 Yep -- this list is out-of-date and no-one maintains it -- +1 to removal 17:10:23 ok, rm'ing stuff seems like a no brainer +1 … even though I've not much idea what it's about :) 17:10:29 +1 17:11:31 SmootherFrOgZ: vote? 17:11:38 yup, +1 17:11:41 #action Java guidelines remove using EE APIs section (+1:5, 0:0, -1:0) 17:11:52 #topic #513 Use python -Es in shbang 17:11:57 https://fedorahosted.org/fpc/ticket/513 17:12:39 So, it seems like there are more technical discussions to work out 17:13:01 Yeah. 17:13:09 In the end I guess I'd like a %py_build macro that Does The Right Thing (tm) 17:13:15 Indeed. 17:13:20 Which is part of that other stalled ticket. 17:13:28 right 17:13:34 thanks for finding that 17:13:36 If someone wants to revive just that part of it, we could probably move forward. 17:13:51 orionp: Well I think there is general agreement that having -s added would be a good thing to do 17:14:12 yes, that is for sure 17:14:12 Yes, +1 to -s from me. 17:14:39 Honestly I'm leaning towards -E as well, and -I for python 3, but I'm not enough of an expert to advocate for them. 17:15:12 The main questions are: How does -s get added to the she-bang … and do we want -E or -I there too? 17:15:15 I think the absolutely trivial workaround of just sticking "python" or "python3" on the command line before the name of the script kind of overrides any "you broke my workflow" complaints. 17:15:35 geppetto: Without macro-izing the setup.py call, we'd just tell people to do it manually. 17:15:44 yeh 17:15:48 But whipping up a macro for that should take 20 seconds. 17:16:18 See https://bugzilla.redhat.com/attachment.cgi?id=739290 17:16:40 Yeh, it just be cool if there could be two macros and then packagers don't have to do anything if/when we change this again 17:17:34 I think all of the necessary macros are in that attachment, though it needs minor tweaks to add -s/-sE/-I/whatever. 17:17:47 * geppetto nods 17:18:17 Not that I actually understand the nested expansion in py_rt_build and py_rt_install. 17:18:24 :) 17:18:42 Never worked with parameterized macros in RPM. 17:19:13 I feel like everytime I look at them I have to learn them again 17:19:42 FWIW - not too much more feedback from python-dev 17:19:45 mbooth: SmootherFrOgZ: You read/understand enough to vote on making -s mandatory? 17:20:23 yeah, was reading comments. So I'm also +1 to -s only 17:23:28 mbooth: ? 17:23:42 -E / -I may be desired for daemons and other security sensitive apps 17:23:50 Sorry it must have been written slowly because I was reading it slowly :-p 17:24:35 +1 -- it makes "sense" to me, but I am a long way from being a python expert 17:24:50 I trust all y'all 17:26:28 #action Mandatory use of -s in she-bang of python programs (+1:5, 0:0, -1:0) 17:26:37 So: a %py_build macro that defaults to --executable "%{__python} -s", but can be changed by defining something like %py_executable 17:27:21 We really need a py_build and py_build3 … unless someone is backporting -I to py-2.7 17:27:51 and mentioning it in the guildlines incase people don't use the macro 17:27:51 Also do we want to vote on that, or ask/wait for the py packager to propose it? 17:27:59 * geppetto nods 17:28:23 Vote on the macros? 17:29:03 I'm +1 in general. 17:29:35 I'm +1 on the macros, but it'd be good to have a concrete proposal 17:29:40 Yes. 17:29:58 Let me see if I can figure out how to tweak them for -s and further args we want to add. 17:30:12 And then run them by the python folks before next week. 17:30:25 that would be great, thanks tibbs 17:30:34 I should be able to find the time for that. 17:30:51 And python packaging needs all the help it can get. 17:31:53 I really want to push an additional macros package to EPEL to get rid of some of the conditionals needed for EPEL support. 17:31:57 +1 on macros 17:32:11 tibbs|w - what bug did that attachment come from? 17:32:22 #action tibbs Will work on macros to vote on next week. 17:32:32 Digging... 17:32:42 https://bugzilla.redhat.com/show_bug.cgi?id=953704 17:33:05 I will also leave the filtering stuff out for now. 17:33:56 ok, cool 17:34:05 Anyone want to say anything else about this? 17:34:57 #topic #325 Temporary bundling exception of yajl library 17:34:59 https://fedorahosted.org/fpc/ticket/325 17:36:23 It was a temporary request. Not sure how temporary, though. 17:36:26 The whole thing is a mess. 17:36:27 "Other option could be to package yajl1 compat package, but this is overkill" … my feeling here is "yes, no" 17:36:48 I don't think it's temporary ganymore given the responses 17:39:44 Wow, and silence from brianmario (of yajl-ruby fame) for 2 years on that PR 17:39:49 I don't know what to do. 17:39:59 Honestly I'd say just fork the damn thing. 17:40:15 tibbs: Can you see any downside to just -1 bundling, but +1 on creating a compat-yajl1 package? 17:41:01 * geppetto nods … I'm not even bothered if they call it just yajl1 or whatever 17:41:22 tibbs|w: The "forks" graph in github has kinda exploded in the last year -- looks like it is no longer maintained 17:41:48 It isn't maintained, except for all of the people who forked it, I guess. 17:41:49 :/ 17:42:08 "maintained" has become kind of a nebulous concept with the existence of social coding. 17:42:23 But, sore, I don't see why we don't have the compat package, even if it's patched. 17:42:39 Would be nice if someone just said "I'm maintaining it now" but whatever. 17:42:41 "sure". 17:43:02 I guess the downside is forcing someone to do more work. 17:45:26 #action No bundling exception of yajl library, someone just create a compat-yajil1 or even yajil1 fork/whatever package. 17:45:59 #info Or become the upstream maintainer :) 17:46:22 sgallagh: I assume you want to skip 506 for another week? 17:46:23 Compat pkg sounds fine to me -- once created it should be low maintenence 17:46:37 * geppetto nods 17:46:49 #topic Open Floor 17:47:10 Well, that's it … first meeting under an hour this year I think :) 17:47:37 Yep, Which is good because I have to leave in 20 minutes. 17:47:41 :) 17:47:52 Anyone have anything to discuss for 25 minutes ?;) 17:48:18 Will probably drop some more needinfo tickets soon. 17:48:22 geppetto: 506 is under active discussion in the ticket and script package review, so yes. Let's keep it there for now. 17:48:36 sgallagh: yeh, cool. 17:49:11 Ok, well I'll close in a couple of minutes then unless someone shouts. 17:51:24 #endmeeting