17:00:53 #startmeeting fpc 17:00:53 Meeting started Wed Oct 4 17:00:53 2017 UTC. The chair is geppetto. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:53 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:00:53 The meeting name has been set to 'fpc' 17:00:53 #meetingname fpc 17:00:53 The meeting name has been set to 'fpc' 17:00:53 #topic Roll Call 17:01:02 hi 17:01:05 hey 17:01:08 #chair mboddu 17:01:08 Current chairs: geppetto mboddu 17:01:12 #undo 17:01:12 Removing item from minutes: 17:01:24 #unchair mboddu 17:01:24 Current chairs: geppetto 17:01:30 #chair mbooth 17:01:30 Hi 17:01:30 Current chairs: geppetto mbooth 17:01:33 #chair tomspur 17:01:33 Current chairs: geppetto mbooth tomspur 17:05:08 Hm 17:07:55 IIRC tibbs said he probably couldn't make it this week 17:08:12 Sorry I wasn't here last week btw 17:09:12 hi 17:09:14 sorry got distracted 17:09:28 hello. Here, barely 17:09:29 #chair Rathann 17:09:29 Current chairs: Rathann geppetto mbooth tomspur 17:09:33 #chair orionp 17:09:33 Current chairs: Rathann geppetto mbooth orionp tomspur 17:09:41 Woo … 5 peeps :) 17:09:48 \o/ 17:09:56 #topic Schedule 17:10:26 https://lists.fedoraproject.org/archives/list/packaging@lists.fedoraproject.org/message/TO325LMDAHN4N5SSI2PPESSLOZJU5GLT/ 17:12:06 #topic #708 Allocating a static uid and gid for openvswitch 17:12:12 .fpc 708 17:12:14 geppetto: Issue #708: Allocating a static uid and gid for openvswitch - packaging-committee - Pagure - https://pagure.io/packaging-committee/issue/708 17:13:27 So, I'm not sure if there is precedent for this -- I'm almost sure tibbs would know :-) 17:13:51 Also … pretty much every application would hit this, no? 17:14:15 Yes 17:14:24 yeah, most backups use names 17:14:45 restoring from backup case is not a valid reason, but having to match between host and VMs could be 17:15:40 Hmm, yeah. They give no reason why ID must survive a backup/restore 17:16:26 Why does it matter if openvswitch picks a new ID, etc? 17:19:02 Further, if files he needs to backup are containing the ID then this seems like bug in openvswitch, right? It should be storing and using the name? 17:19:12 #action We would need to know why it matters if the ID changes. Backup software understand ID's can change, and so save names/etc. 17:19:41 Yeh, but good luck getting upstream to fix that 17:20:24 #topic #710 Ruby packaging guidelines update 17:20:27 .fpc 710 17:20:28 geppetto: Issue #710: Ruby packaging guidelines update - packaging-committee - Pagure - https://pagure.io/packaging-committee/issue/710 17:21:06 Diff: 17:21:07 https://fedoraproject.org/w/index.php?title=User%3AVondruch%2FDraft_RubyGuidelines&diff=501566&oldid=500202 17:23:18 "F26 and older" in the warning box also includes EPEL? 17:23:45 presumably 17:23:54 I'd like an explanation why the paths changed from current dir to ../ 17:24:15 tomspur: I guess it's really RPM < 4.14 17:24:37 The older warning box removed F19/20 and EPEL, so maybe it works on EPEL now 17:25:05 ah, I think it's because the unpacked gem lands the .gemspec file in topdir 17:25:14 ok, no question then 17:25:17 Rathann: was not creating a build directory before I think - probably should be a source macro instead of relative path 17:26:09 no, actually the ../ is fine 17:28:11 ok, the changes look fine to me 17:28:13 +1 17:28:54 The minutes from the last meeting say these guidelines only work on F27+ 17:29:02 From the log " it should be modified to "Different Guidelines for Fedora up to including 26 and EPEL6/7" 17:29:17 tomspur: To answer your Q 17:29:41 mbooth: thanks 17:30:02 I guess we should ask if the situation changed or if the warning should change to reflect reality :-) 17:31:06 well there's an {{admon saying that for F26 and older you need to keep using the previous code snippet 17:31:23 I'd just add EPEL6/7 there too then 17:31:36 yes 17:31:59 I also think we should notify the EPEL SIG when introducing new guidelines incompatible with EPEL 17:32:01 This was in the previous warning box "In Fedora 19/20 and EPEL 6/7 the Ruby guidelines were slightly different." 17:32:34 Unsure if EPEL behaves now as F20 or F26 17:33:35 I think I'm other happy with the guidelines if we get that clarification 17:34:14 s/other/otherwise/ 17:35:47 mbooth: AFAIK some of the macros were backported to EL7, but I think some differencies still exist (I'm not sure at all about EL6) 17:36:14 Sure -- it's just the admon has different wording to what vondruch said it should be :-) 17:36:44 Why should some of the test frameworks be avoided? 17:36:56 "Rake almost always draws in some unnecessary dependencies like hoe" 17:37:26 unnecessary runtime dependencies might be slow, but I don't fully get why it should be avoided 17:39:01 tomspur: it's harder to package if gem has a lot of dependencies, esp. circular, and most of the time it can be easily avoided 17:40:22 tomspur: f.e. in most cases rake just executes the test suite, so it can be executed by simillar means, without the redundant dependencies for build 17:40:51 ok 17:41:04 pvalena: thakns 17:41:14 tomspur: yaw 17:41:34 One regex for the tests has changed. I guess that should all be ./test/test_*.rb, right? 17:42:02 The rest looks fine to me 17:42:47 tomspur: no, the latter is the correct one, as the tests can be nested in multiple folder levels 17:43:06 ** expands those 17:44:22 I'd use a consistent regex on both examples. The one with ** is also fine 17:46:52 tomspur: well I'd say it's fine this way, as in the text you're just trying to reference to the test files while mentioning the regex may be different 17:47:45 (it's also actually not a regex) 17:48:51 (in case you're interested, ref: https://ruby-doc.org/core-1.9.3/Dir.html#method-c-glob ) 17:49:10 I think geppetto fell asleep :-) 17:49:19 I added my concern to the ticket 17:49:24 Ruby API docs will do that to me :) 17:49:58 I will have to leave at the top of the hour 17:50:17 * geppetto nods 17:51:30 pvalena: I see. So this doesn't work for jruby and therefore another pattern is used there? 17:51:51 * tomspur doesn't do anything with ruby usually :) 17:52:53 Yeh, me either 17:53:01 tomspur: I do not really know about jruby, as we do not use or maintan it, but I think the behavior should be consistent 17:53:36 it's more dependent on the 'test/' file structure 17:53:59 ok 17:54:23 (f.e. in some rare cases the _test.rb suffix may be different) 17:57:22 ok, before the end of the hour … is there anything specific we want as an action? 17:57:38 So that people will be able to vote 17:57:58 The clarification from mbooth is fine for me 17:58:29 I have to go now... 18:00:18 #action Q: Has the situation changed or should the warning should change to reflect reality: 18:00:24 Q: Has the situation changed or should the warning should change to reflect reality: 18:00:35 Is that good? 18:01:10 * Rathann sees too many "should" 18:01:19 * geppetto nods 18:01:22 #undo 18:01:22 Removing item from minutes: ACTION by geppetto at 18:00:18 : Q: Has the situation changed or should the warning should change to reflect reality: 18:02:14 #action Q: Has the situation changed or should the admon/warning change to reflect reality? 18:03:04 #topic Open Floor 18:03:15 Ok, anything anyone wants to discuss? 18:04:03 nothing from me this week 18:04:12 I'm still catching up after vacation 18:04:16 ok 18:05:10 Ok, see you next week 18:05:19 #endmeeting