Perl 6 - the future is here, just unevenly distributed

IRC log for #opentreeoflife, 2015-02-26

| Channels | #opentreeoflife index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:01 kcranstn joined #opentreeoflife
01:11 kcranstn joined #opentreeoflife
02:17 jimallman_ joined #opentreeoflife
03:29 pmidford2 left #opentreeoflife
04:22 jimallman_ joined #opentreeoflife
12:34 kcranstn joined #opentreeoflife
15:37 pmidford2 joined #opentreeoflife
15:54 josephwb joined #opentreeoflife
16:13 mtholder joined #opentreeoflife
17:02 mtholder joined #opentreeoflife
17:22 jimallman joined #opentreeoflife
17:27 jimallman kcranstn: i hope to be around for PR review today, but electricity here is touch-and-go.
17:27 kcranstn understood. we’ve been solidly on
18:22 jar286 joined #opentreeoflife
18:22 kcranstn joined #opentreeoflife
18:35 kcranstn joined #opentreeoflife
18:56 kcranstn joined #opentreeoflife
18:58 jar286 joined #opentreeoflife
18:59 pmidford2 joined #opentreeoflife
19:01 mtholder joined #opentreeoflife
19:02 jar286 it’s that time
19:03 jimallman indeed
19:03 pmidford2 here
19:03 jar286 https://github.com/OpenTreeOfLife/opentree/pull/613
19:03 kcranstn here (but also trying to get a talk done for later this afternoon!)
19:03 jimallman full list at https://github.com/pulls?user=OpenTreeOfLife
19:04 jar286 I’m ok with #613
19:04 jimallman drat, this probably still needs attention (in the accompanying #144 above)
19:05 jimallman empty comment results in failure to delete, see https://github.com/OpenTreeOfLife/phylesystem-api/pull/144#issuecomment-75263527
19:05 jar286 humph
19:06 jar286 I’ll add a comment to #613 pointing to #144
19:06 jimallman thanks
19:06 jar286 https://github.com/OpenTreeOfLife/ot-base/pull/13
19:07 jar286 I don’t want to merge this until I’ve heard it doesn’t break devapi
19:08 jar286 we’re done then… no deployment today
19:08 kcranstn how do we go about testing 13?
19:09 * jimallman will follow up with snacktavish re: #144. if she’s super-busy, i can fix this.
19:09 jar286 I’d be satisfied just to know it builds and installs without problems, and my smoke tests pass (germinator/test.sh)
19:09 jar286 I guess I can do that now
19:10 jimallman i’m sure they have some known-bad test data for this (invalid Newick names)
19:11 jar286 although it does say ‘merge please @cohinchliff’ meaning I don’t really want to do the merge
19:11 jar286 but I can do the test
19:16 jar286 it doesn’t build.
19:17 josephwb jar286 it uses a different jade class. it is for our synthesis work. merging *from* master
19:17 josephwb not relevant to any of you
19:17 josephwb cody refactored things...
19:18 josephwb jade branch, that is
19:18 kcranstn ah, I missed the direction of the merge
19:18 josephwb unless i screwed it up?
19:19 jar286 I missed it too
19:19 jar286 that’s too subtle for me… I’m not going to get into the habit of checking that for every PR
19:19 josephwb this is an odd one
19:19 jar286 do we need to split off a deployment branch so we don’t step on one another’s feet?
19:20 jar286 or split off a development branch, either way
19:20 josephwb no
19:20 jar286 I see ot-base commits that didn’t go through the PR review process
19:21 jar286 that’s not the process we agreed… so we need to modify the process
19:23 josephwb sorry about that. a PR for treemachine needed ot-base changes to build (travis).
19:23 josephwb not sure how to do a multi-repo PR
19:23 mtholder git submodules.  but that is a PITA.
19:23 josephwb PITA?
19:24 jar286 the changes build on devapi… so no harm done if the changes work…
19:24 mtholder pain in the ass
19:24 josephwb yes, sounds not fun
19:24 jar286 but in general I would like review before anything goes on its way to deployment
19:24 jar286 I know this can interfere with development…
19:25 jar286 but we can get noninterference by using different branches
19:26 jar286 I don’t care what the branch names are, you could use ‘master’ and deployment could use ‘production’ if you want… if you think we should go this route
19:26 jar286 ?
19:26 jar286 or ‘deployment’ rather than ‘production’
19:27 kcranstn or we could stick with the existing, which is that master is always deployable and everything pushed to master must be reviewed
19:27 jar286 that’s my preference, but I was giving josephwb the option
19:27 josephwb i think we discussed the change and rejected it
19:28 josephwb i will do my best not to pollute master again
19:28 kcranstn thanks, josephwb
19:28 josephwb i feel we need a way to make simple bug fixes without PRs, tho.
19:29 josephwb but i will play nice
19:29 kcranstn we are all fairly regularly online if you need something urgent reviewed
19:29 kcranstn or you can work off a branch until we can review
19:30 jar286 right.  the review doesn’t have to wait until thursday at 2
19:30 mtholder I think that the main problem with our current system is that we have very few tests (unit or integration). So the PRs can be pretty cursory and superficial.
19:31 kcranstn I agree with that
19:45 kcranstn joined #opentreeoflife
19:46 jar286 joined #opentreeoflife
19:48 pmidford2 left #opentreeoflife
19:59 mtholder joined #opentreeoflife
20:03 jar286 joined #opentreeoflife
20:21 pmidford2 joined #opentreeoflife
20:52 jar286 joined #opentreeoflife
22:13 kcranstn joined #opentreeoflife
22:15 jar286 joined #opentreeoflife
22:22 kcranstn joined #opentreeoflife
22:30 jimallman jar286: got a sec for a tiny code review?
22:30 jimallman https://github.com/OpenTreeOfLife/phylesystem-api/pull/144/files
22:30 jar286 sure
22:31 jimallman i just added a check for empty commit msg here
22:32 jar286 looks reasonable although a bit paradoxical… I assume code further on does something interesting in the commit_msg == None case
22:33 kcranstn joined #opentreeoflife
22:33 jar286 there’s a second place that does kwargs.get(‘commit_msg’)
22:33 jimallman yes, phylesystem.delete_study should substitute a generic message (will check this now)
22:34 jar286 line 691, does it need a ,’’ ?
22:34 jimallman hm, i think you’re right. we should add this check there as well.
22:35 jar286 i.e. should 691 look like 798
22:36 jimallman yes, i’m adding the whole try/except block
22:37 jimallman it’s there now (now checking downstream behavior)
22:41 jimallman here’s the default message (if commit_msg is None):  https://github.com/OpenTreeOfLife/peyotl/blob/c5457ee292afc10bdcbc460410e8de1c3b339a51/peyotl/phylesystem/git_actions.py#L341-L342
22:48 travis-ci joined #opentreeoflife
22:48 travis-ci OpenTreeOfLife/phylesystem-api#688 (comment-on-delete-study - 1ba3918 : Jim Allman): The build passed.
22:48 travis-ci Change view : https://github.com/OpenTreeOfLife/phylesystem-api/compare/8771de330c84...1ba39181f802
22:48 travis-ci Build details : http://travis-ci.org/OpenTreeOfLife/phylesystem-api/builds/52347179
22:48 travis-ci left #opentreeoflife
22:49 jimallman jar286: if you don’t mind, i’d like to merge and test this on dev (or switch dev deployment to this branch for a quick test)
22:49 jar286 dev is yours
22:50 jimallman ok, thanks! merging now
23:02 travis-ci joined #opentreeoflife
23:02 travis-ci OpenTreeOfLife/phylesystem-api#690 (master - 5211d69 : Jim Allman): The build passed.
23:02 travis-ci Change view : https://github.com/OpenTreeOfLife/phylesystem-api/compare/b4d5a3f4d767...5211d6924ce2
23:02 travis-ci Build details : http://travis-ci.org/OpenTreeOfLife/phylesystem-api/builds/52348881
23:02 travis-ci left #opentreeoflife
23:12 josephwb joined #opentreeoflife

| Channels | #opentreeoflife index | Today | | Search | Google Search | Plain-Text | summary