Perl 6 - the future is here, just unevenly distributed

IRC log for #opentreeoflife, 2014-06-09

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

All times shown according to UTC.

Time Nick Message
00:18 jimallman mtholder: fyi, i’m troubleshooting some API problems reported a few days ago from Katz lab (failure to import studies from TreeBASE)… looks like there’s some stale code in the API used by dev.opentreeoflife.org (http://ashby.opentreeoflife.org/api/v1/study)
00:45 towodo joined #opentreeoflife
01:01 mtholder joined #opentreeoflife
01:40 jimallman towodo: i’m trying to sort out some weirdness on tree.opentreeoflife.org… got a few minutes?
01:40 jimallman (if not, glad to pick this up tomorrow)
01:41 towodo jimallman, I may have a few minutes, after I do a small task that should take a couple of minutes
01:41 jimallman ok, no rush. i’m here for awhile..
01:45 mtholder joined #opentreeoflife
01:46 mtholder jimallman: sorry I forgot to sign off before walking away from my machine.
01:46 mtholder I have not updated the code on ot10 for awhile. I thought we were testing on ot12 now (for the api)
01:46 jimallman no prob. i’m just trying to figure out our current setup..
01:47 jimallman in deployed-systems, ‘production’ has dev domain names and vice versa. at this point, i’m not sure who’s using what, or where to debug :)
01:48 mtholder let me know if you need a hand.
01:48 mtholder it is confusing, but I think that dev = frozen for the manuscript
01:48 jimallman ah, got it. so i should be testing on tree.opentreeoflife.org (and friends)..?
01:48 mtholder well...
01:49 mtholder this is why I wanted curation on ot3 to redirect to curation on tree.opentre...
01:49 mtholder just so there would would be one point that we knew everyone was using (for curator testing).
01:50 jimallman i thought it had more to do with not messing up the docstore being reviewed. but yes, i get it.
01:50 mtholder it is both. I
01:50 mtholder I think that it would be very unlikely to break the rest of the system if we were to redeploy on ot10.
01:50 mtholder but then I broke the build today, so I'm not sure that I'd listen to me...
01:51 jimallman :) it looks like everyone’s pointing to phylesystem-0 (according to server-config files)… is that right?
01:52 mtholder yes. and that will be a problem of sorts.
01:52 jimallman except for asterales/ot6.config, that is
01:52 mtholder right. good point.
01:52 mtholder multiple instance using phylesystem-0 will mean that only one will be able to push to GH (and nudge oti)
01:53 jimallman right, i recall that restriction. i suppose it’s not a big deal, if we don’t trigger a fresh synthesis. they’re not actually reviewing source studies/trees in the docstore, right?
01:53 jimallman (fyi, the specific bug i’m chasing is failure to import from TreeBASE, found by Katz lab. this is not a known issue, is it?)
01:54 mtholder no.
01:54 mtholder on ot10?
01:55 mtholder or ot3, I guess...
01:55 jimallman just a sec…
01:55 jimallman it was on dev.opentreeoflife.org… so ot3
01:56 mtholder off topic: does readline work for you on ot3? I never get tab completion there...
01:57 jimallman funny. it works for me if i’m ‘admin’, but not as ‘opentree'
01:57 jimallman i never bothered to chase it down
01:57 mtholder no biggie, I'd look at it if it was just me.
01:58 jimallman anyhoo, attempting to import a study from TreeBASE generates an API error in web2py: api/184.3.219.202.2014-06-09.02-56-28.c56183ab-d7f3-4bd1-991a-afcb160c6bc3
01:58 towodo jimallman, i'm back
01:58 jimallman which doesn’t show, of course. if i chase it down on ot3, i see this:
01:58 jimallman <type 'exceptions.TypeError'> __init__() takes at least 4 arguments (3 given)
01:58 jimallman towodo: hi! i was just running some questions by mtholder...
01:59 jimallman he’s already cleared up some of my confusion about domain names and dev vs. production (these are sort of mixed in ‘deployed-system’)
01:59 mtholder I'm trying to find the err file.
01:59 towodo reading
02:00 towodo there's a log directory, and web2py has its own logs, and neo4j has its own logs
02:00 jimallman mtholder: i’ve added a little helper script on ot3, if you’re logged in as ‘admin’
02:00 mtholder I think this err is on ot10...
02:00 jimallman ~/api-last-error.sh will find and open the latest file in errors/ subdir
02:00 towodo let's see, ot10 = ashby = part of the production system.
02:00 jimallman ! d’oh!
02:01 towodo ot10 also = files.opentreeoflife.org
02:01 mtholder cool jim thanks for that.
02:02 towodo we can point tree.opentreeoflife.org to production instead of development if that would help
02:03 * jimallman has  moved api-last-error.sh to ot10… ssh admin@ot10, then ~/api-last-error.sh
02:04 mtholder is there an issue for this Jim?
02:04 jimallman towodo: it would help, but i don’t want to risk confusing reviewers or screwing up any current config
02:04 mtholder do you know what the treebase ID is
02:04 towodo hmm.  maybe confusion api.opentreeoflife.org vs. phylesystem-api
02:05 jimallman probably not ticket.. just an email bug report from Lindsey in the Katz lab: “Dail picked two studies, one with strain ids: 14246, and one without strain ids: 12157. When I tried the studies, I put just the number in the text box and pressed continue.”
02:05 jimallman i’ve been testing with 12157..
02:05 jimallman reading the web2py error now
02:05 towodo i never run vi, don't know how to exit it
02:06 jimallman :q
02:06 jimallman :q! if it balks
02:06 jimallman specific error is in _gen_hbf_el():  “File "/home/opentree/repo/peyotl/peyotl/nexson_syntax/nexml2nexson.py", line 125, in _gen_hbf_el\n    for i in xrange(att_container.length):\nAttributeError: \'NoneType\' object has no attribute \'length\’”
02:07 jimallman we’ll need to backtrack to see what’s really going wrong. i don’t recall if we have a known-good study ID, but i’ll check old emails..
02:08 mtholder got it, and think that I've fixed it. testing...
02:08 mtholder thanks
02:08 jimallman coolness…
02:08 towodo I deleted the .bashrc on ot3 once upon a time.  you can restore it from another server if you like
02:08 towodo I got sick of the stupid aliases and probably went overboard
02:09 jimallman towodo: what was it doing? is this about readline support?
02:09 towodo that's my hunch.  maybe the .bashrc did something to make readline work
02:09 * jimallman is reading this now, on ot10:/home/opentree
02:11 towodo I am happy to straighten on the domain names.  Need to leave 'dev' and 'api' alone but nothing depends on tree, devtree, or devapi.
02:11 kcranstn joined #opentreeoflife
02:11 towodo I originally thought ot11/12 were going to be our production system very soon. i was wrong i guess
02:12 jimallman towodo: i’m ok for now, just needed to get the lay of the land. it would be cool to set up the new domain names (and i can add/update registered apps on GitHub to recognize them for oauth login), but again, i don’t want to shake the jell-o during review.
02:13 towodo well review doesn't depend on anything other than dev.  so as long as that's steady we can do anything we like with the others
02:13 jimallman OK. if review is ongoing (for more than a few days), we should probably try to keep moving forward with new domain names, etc.
02:13 towodo my guess is probably more than a few days.
02:14 kcranstn I expect so
02:14 towodo I haven't changed ot3 / ot9 / ot10 so I don't understand why there would be any errors there
02:14 mtholder ok. so I just pushed a bug fix to the production branch of peyotl. Dare I deploy?
02:15 mtholder I'll try to test locally. got to do my /etc/hosts song + dance
02:17 towodo if the phylesystem api on ot10 is broken due to peyotl then by all means push.
02:17 towodo pushing phylesystem-api is another story.  would want to look over the commit log on github and the push log on ot10
02:18 towodo for one thing we renamed the phylesystem-api repo since the last time it was pushed.
02:18 jimallman hm, checking server-config files and deploy scripts now.. though as usual, GitHub should insulate us from repo-rename problems
02:19 towodo I see… that would be awfully nice of it
02:20 jimallman there are some literal references to api.opentreeoflife.org in deploy/setup/install-api.sh and setup/install-web2py-apps.sh
02:20 towodo oh yeah. i had to go around and fix a lot of those for the development system
02:21 jimallman these refer to the deployed server’s filesystem, so liable to fail (one is clearing old web2py session files)
02:21 towodo humph.  we could introduce a parameter
02:22 towodo deployed-systems doesn't have a production branch, i didn't think that made sense
02:23 jimallman but deployed-systems sort of transcends the source-code branches, doesn’t it?
02:23 towodo i need to get to sleep.  tomorrow i can change 'tree' to point to production (ot3), and 'devtree' to development (ot11).
02:23 jimallman hm, unless we’re experimenting with deployment tools, i guess
02:23 towodo right, it transcends. that's what i meant by 'doesn't make sense' to create a production branch
02:24 jimallman thanks! that sounds great. we’ll tread lightly and watch for ripples into the reviewer’s domains (dev.opentreeoflife.org)
02:24 towodo the deployment tools are in the opentree repo.  deployed-systems only has config files
02:24 jimallman true! so we’re doing it right :)
02:24 jimallman config-file tinkering (if any) can be on local filesystem
02:25 towodo Not sure what to do about 'devapi'.  I guess 'devapi' = 'api' = ot12. (production has no complete working api)
02:27 towodo mtholder, playing /etc/hosts games shouldn't be necessary… you should be able to just do a local edit to the config files…
02:27 towodo mtholder, sorry I flushed the ot7 config file, i didn't think it was being used.  I can try to recover it if you need it
02:27 mtholder on my machine - to get it to load the curator app
02:27 mtholder no biggie about ot7
02:27 mtholder won't need it for a while
02:28 towodo if not being used we should discard it.  easy to reconstitute later
02:28 mtholder I just got the curator app to load on my machine and confirmed that my fix did let me import 14246 from Treebase
02:28 mtholder that is fine with me.
02:28 jimallman re API, it looks like ot3 (production webapp server) is using ot10.opentreeoflife.org/api/v1
02:29 * mtholder is now running peyotl regressions...
02:29 towodo that sounds right to me.
02:29 jimallman mtholder: sweet, thanks! once this is live, Katz lab will be eager to restart testing
02:30 mtholder OK so 52a05dd1a245b83800bccf714ee5aa151f09cf7e of peyotl fixes it, but I have not deployed yet.
02:30 mtholder I guess I need to be using the production branch of opentree repo when I deploy to ot10?
02:30 mtholder right, towodo?
02:31 jimallman hm, why is that on ot10? is it required to support phylesystem-api on web2py?
02:31 mtholder yup
02:32 mtholder though maybe your right it could be crashing on ot3...
02:32 towodo I think so… I don't think I touched peyotl when I set up development
02:32 mtholder no the error log was on ot10
02:33 jimallman yes, i think you’re on solid ground with that fix. the API errors on ot3 are ancient history!
02:34 towodo wait, were you pushing development peyotl versions to ot10 before?… if you haven't committed to production, and you push production peyotl to ot10, then you'll just be reverting ot10 to its state when production was branched
02:35 * jimallman has cleared the ancient api/errors/ dir on ot3
02:35 mtholder last time I pushed to ot10 was before we had a production branch...
02:36 mtholder I've been pushing recent code changes to ot12
02:36 towodo not according to the log on ot10
02:36 towodo take a log at end of log/messages on ot10
02:36 towodo s/log/look/
02:37 towodo Thu May 29 19:08:17 BST 2014 mholder ot10  Install: peyotl setup.py develop run
02:38 towodo oh.
02:38 mtholder I thought you created the production branches on June 5
02:38 towodo yes that was pre-production.
02:38 towodo sorry
02:38 towodo ok, so no problem.
02:38 mtholder no problem. I don't think this is likely to break anything. here goes...
02:41 towodo it's way past my bedtime. i'll check in tomorrow morning. good night
02:42 jimallman kcranstn: i’ve been chasing bugs tonight, but is there anything i can do to help with API docs?
02:43 jimallman mtholder: import from treebase is working now.. i just grabbed #14246
02:44 mtholder me too ;-)
02:44 mtholder http://dev.opentreeoflife.org/curator/study/edit/ot_25
02:44 jimallman yeah, i need to add some logic somewhere to detect redundant imports and block them
02:44 * jimallman is adding a ticket for this…
02:45 jimallman (or should that be part of the API’s create/import logic?)
02:45 mtholder could be API, but it would be easier if you could check OTI and block...
02:45 mtholder (easier for me ;-)  that is)
02:46 mtholder does OTI search on the treebase ID?
02:46 jimallman hm, maybe a belt-and-suspenders approach. for those using the “raw” API
02:46 jimallman yes, i’m pretty sure oti will do this. i should check on creation and in the curation UI (when setting DOIs, references, etc) as well.
02:48 mtholder at some point, I'd like to get a crontab to just import new studies from treebase. Then the "new study from treebase" could morph into a "curate a study that was automatically imported from TB, but no curator has worked on" interface.
02:49 jimallman i like it! presumably TreeBASE studies would show up with no trees marked “candidate for synthesis”, so no adverse impact on synthesis.. or the validator could sideline them.
02:50 mtholder yes.
02:50 mtholder speaking of, can we turn the "reject if invalid" feature back on?
02:50 mtholder I think that it was the studyYear int vs string that was getting us.
02:54 jimallman oh yeah.. i believe we fixed that in the UI, but let me make sure...
02:54 jimallman meanwhile, here’s a ticket on blocking duplicate studies: https://github.com/OpenTreeOfLife/opentree/issues/300
02:56 jimallman yes, studyYear is coerced to int in the UI:  https://github.com/OpenTreeOfLife/opentree/blob/f2ddb1b208c75eafd9bbcabe9b499be6b6310d2b/curator/static/js/study-editor.js#L1196
02:57 jimallman …along with a few similar issues. so yes, let’s restore the block-if-invalid feature
02:57 jimallman hey, don’t let me keep you here! i’m back in action.
02:59 mtholder cool. let me see if I can implement "config-dependent branch on GitHub to push to" that would let both ot10 and ot12 use the phylesystem-0 repo.
02:59 mtholder then we can try testing of the validation w/o having to set up another repo...
03:00 jimallman nice
03:00 jimallman s/nice/noice
03:23 mtholder jimallman. I may have it, but I'll have to wait until tomorrow until I can deploy it and test it. bootstrapping that is going to be a bit of a pain, but I may punt on that and configure the git repo on ot12 manually.
03:26 jimallman understood. let me know if i can help testing this.
06:31 scrollback joined #opentreeoflife
07:29 20WAAAKL8 joined #opentreeoflife
07:39 scrollback joined #opentreeoflife
07:51 92AABCLEJ joined #opentreeoflife
10:42 towodo joined #opentreeoflife
13:33 kcranstn joined #opentreeoflife
15:04 kcranstn joined #opentreeoflife
15:09 77CAAAUAT joined #opentreeoflife
15:11 20WAAASSD joined #opentreeoflife
15:55 blackrim joined #opentreeoflife
15:56 blackrim what are we going to talk about on the call today, anyone know
15:59 towodo joined #opentreeoflife
16:04 kcranstn I haven’t seen an agenda
16:04 kcranstn @towodo?
16:05 towodo t
16:05 towodo right. unexpectedly preoccupied with car idiocy
16:06 towodo kcranstn, I would say API, server aliases, and next round of synthesis are on my agenda
17:03 mtholder joined #opentreeoflife
17:05 towodo joined #opentreeoflife
18:02 towodo traffic
18:03 mtholder more traffic
18:08 kcranstn joined #opentreeoflife
18:14 travis-ci joined #opentreeoflife
18:14 travis-ci [travis-ci] OpenTreeOfLife/phylesystem-api#555 (branchpush - d4026fa : Mark T. Holder): The build passed.
18:14 travis-ci [travis-ci] Change view : https://github.com/OpenTreeOfLife/phylesystem-api/commit/d4026fa51394
18:14 travis-ci [travis-ci] Build details : http://travis-ci.org/OpenTreeOfLife/phylesystem-api/builds/27152531
18:14 travis-ci left #opentreeoflife
18:17 josephwb joined #opentreeoflife
18:18 josephwb y'all, it seems like the proposed phylografter block is going to be a major problem for synthesis 2.0.
18:18 kcranstn explain?
18:18 mtholder why
18:18 josephwb ALL trees in synthesis are from phylografter. chances are we may need to edit a lot of them (e.g. with taxonomy updates)
18:18 mtholder you can still edit in pg
18:18 mtholder just don't edit the *same* study in both
18:19 josephwb there is no way in phylografter to designate a "Preferred" tree
18:19 josephwb e.g. http://dev.opentreeoflife.org/curator/study/view/2926
18:19 josephwb i uploaded this to phylografter
18:19 josephwb curator says it is incomplete
18:19 josephwb how do i designate "preferred" now?
18:20 josephwb or anywhere in phylografter
18:20 josephwb ?
18:20 mtholder jimallman^
18:20 * jimallman was elsewhere, sorry.. catching up now.
18:20 mtholder sorry he was asking about pg.
18:20 mtholder nm
18:21 josephwb pg just used freeform tags (yes: ugh)
18:21 josephwb but that is all we had to work with
18:21 kcranstn once you log in, there is a checkbox for preferred
18:21 kcranstn in the tree list
18:21 josephwb i'm logged in, but don't see that
18:21 jimallman yes, but that won’t be accessible once the block is in place
18:21 josephwb all i see is: NO
18:21 mtholder nor do I (chrome on ubuntu)
18:22 josephwb Safari on Mac here
18:22 jimallman josephwb: go to the Trees tab
18:22 josephwb there
18:22 mtholder oops click "edit study", too
18:22 kcranstn did you click EDIT and log in?
18:22 mtholder (that always gets me.
18:22 josephwb ooph
18:22 kcranstn :)
18:22 josephwb ok, see it now
18:23 towodo joined #opentreeoflife
18:23 josephwb ok, i believe i did it
18:23 jimallman that label should be more clear, but i haven’t come up with better verbiage that is short enough to fit comfortably. suggestions welcome!
18:23 mtholder would be nice to have the edit tools visible, but inactive + provide a "Click Edit Study" message if someone tries to focus on them...
18:23 josephwb but, if the pg block is implemented, i wouldn't be able to, yes?
18:24 mtholder yes
18:24 jimallman josephwb: correct
18:24 kcranstn you would have to edit those in pg
18:24 jimallman or vi   :)
18:24 kcranstn (sucks, but temporary until we work out the workflow)
18:24 josephwb ok, so *every* pg study will be incomplete then
18:24 josephwb but you can't edit them in pg
18:25 kcranstn because we aren’t storing ‘preferred’ in the nexsons?
18:25 josephwb pg has no "Preferred" field
18:25 mtholder what is the tag that it uses?
18:25 josephwb at all
18:25 josephwb freeform tag
18:25 mtholder this info has been in the gcmdr list not the studies.
18:25 mtholder so we could stick to that.
18:25 josephwb yes
18:25 mtholder until things are smoother.
18:25 jimallman could we mark all trees as preferred on import from pg? then let the validator make the call?
18:26 kcranstn can we do a one-time update of the nexsons based on gcmdr info?
18:26 josephwb no
18:26 mtholder no it needs expertise
18:26 josephwb some studies have a dozen trees
18:26 mtholder one time update is possible
18:26 mtholder (but we
18:26 mtholder 'd need it not to be clobbered by pg export
18:27 mtholder or I suppose I could add the "incorporate info from gcmdr" into the sync-with-pg script...
18:27 mtholder ugh
18:28 mtholder ugh (for added emphasis and more traffic)
18:28 josephwb i am thinking about bryan here: he has curated hundreds of studies in pg. how do we get his studies to "complete"?
18:29 josephwb i mean, there is no way, right (if the block is there)
18:31 josephwb another issue: Tree root is coming up as "Arbitrary (not biologically correct)". How is that fixed?
18:31 mtholder click on the tree
18:31 mtholder radio btns
18:32 josephwb d'oh. ok.
18:32 josephwb thanks.
18:33 josephwb so everything from pg will also be considered arbitrarily rooted?
18:35 josephwb uh oh. http://dev.opentreeoflife.org/curator
18:35 kcranstn ?
18:35 josephwb is the Mitchell duplicated?
18:36 mtholder that is a bug.
18:36 mtholder we have a # aliasing pg_#
18:36 jimallman not in oti (just filtered the list for ‘mitchell’)
18:36 mtholder it is showing both..
18:36 mtholder I'll look at oti.
18:36 mtholder and see if I can add that.
18:36 mtholder Or I could stop supporting aliases.
18:37 jimallman ? i see Mitchell, 2014 and Mitchell A., 1999. where’s the duplicate?
18:37 mtholder http://dev.opentreeoflife.org/curator/study/view/pg_2926 and http://dev.opentreeoflife.org/curator/study/view/2926 are the same in the db
18:37 mtholder but in the study list they both show up
18:37 jimallman my bad, i’m looking at my local server (damn you HOSTS file!)
18:38 josephwb sorry, was away
18:38 josephwb talking with stephen here
18:38 josephwb issue with synthesis 2.0 is that >99% of studies are pg studies.
18:39 josephwb some need re-rooting, some need name fixing
18:39 josephwb it would be nice if we could do this in the curator
18:39 josephwb at least for the duration of the ncurrent push
18:41 jimallman can we reverse the block, putting it on phylografter instead?
18:41 josephwb yes!
18:41 jimallman i guess this would mean pg would ping oti to check for a study, or explicitly mark a study as having “gone to the other side”...
18:41 jimallman across the rainbow bridge…
18:41 josephwb i don't know how often that block will get used, tho
18:42 jimallman i don’t think that matters, right? just a heads-up that the study should now be edited Over Here (linkie)
18:42 josephwb e.g. the curator has fields (like "Preferred") that pg just doesn't have
18:43 jimallman right. do we know if pg has fields that curation app doesn’t? i know it has TOOLS that aren’t shared, but i think curation-app’s Nexson is a superset of its data ..
18:44 josephwb yes, that is how i understand it
18:44 jimallman so it’s tempting to put the block on the pg side.
18:44 kcranstn what are the technical barriers to that?
18:44 josephwb anywho: it would be nice to be able to fix studies in the curator app during the final push
18:45 jimallman i would think it’s easy, but probably requires a quick check against oti…
18:45 josephwb can we hold off on it for then?
18:45 kcranstn hold off on what?
18:45 josephwb the block
18:45 josephwb the blocking-pg-studies block
18:45 josephwb on the curator side
18:46 kcranstn ah, yes
18:46 jimallman let’s check with pmidford and see how hard the block-on-pg would be… i can help him implement this, or probably do it myself since it’s web2py
18:46 jimallman o
18:46 josephwb it will be a bit tedious, but we can click our way through at least the synthesis source studies
18:46 mtholder don't do that.
18:47 josephwb but there is really no other way, as pg lacked the relevant fields anyway
18:47 josephwb no?
18:47 mtholder we can script addition of the synth info into the nexsons then push to git.
18:47 josephwb ok
18:48 josephwb but when you say don't do that, you mean it is a waste of time rather than harmful, right?
18:48 mtholder waste of time
18:48 josephwb gotcha
18:48 mtholder waste of time = harmful ;-)
18:48 josephwb ha
18:49 josephwb so, this will handle the synthesis trees
18:49 josephwb what information do you need?
18:49 josephwb just the tree id?
18:49 mtholder i think so.
18:50 josephwb ok
18:50 josephwb the other thing is that many (most?) of new synthesis source trees are already in phylografter, but have not yet been fixed.
18:50 jimallman i’ve made a ticket (question/enhancement) in pg: https://github.com/OpenTreeOfLife/phylografter/issues/126
18:51 josephwb when they are all processed, i guess i can just give you a list of IDs to process.
18:52 josephwb but re-rooting would be so much easier in the curator
18:52 josephwb in fact, re-rooting is not available in pg at all
18:53 josephwb what has to be done is 1) delete the offending tree, and 2) upload a proper one
18:53 kcranstn I vote for trying to come up with a workflow that allows josephwb et al to work in the new curator app for the next round of synthesis
18:53 josephwb thanks. stephen has left, but he wants this as well
18:54 josephwb at least for the short term
18:55 jimallman hm, i guess blocking on pg is pretty extreme.. since virtually all studies are shared, it almost disables pg from any editing at all.
18:58 kcranstn would require manual intervention to merge
18:58 kcranstn which could be scripted
18:59 kcranstn pg writing to github needs to happen at some point - we can’t keep putting it off indefinitely
19:06 mtholder joined #opentreeoflife
19:06 josephwb jimallman: is the curator app currently working?
19:07 josephwb e.g. i made changes, but when i come back they are not there.
19:07 jimallman last i checked,yes.. where did you make changes
19:07 jimallman ?
19:07 mtholder and did you click "save study"
19:07 josephwb yes
19:08 josephwb the problem is that it was going to the pg_2926 page
19:08 josephwb http://dev.opentreeoflife.org/curator/study/view/pg_2926
19:08 josephwb instead of http://dev.opentreeoflife.org/curator/study/view/2926
19:08 jimallman i just made a minor change to the first listed ‘delete me’ study, and it was saved
19:09 jimallman ah, interesting! so we’re not handling study id “prefixes” consistently!
19:09 mtholder the db thinks of them as the same, I thought. checking...
19:09 jimallman ah, now i see the duplicates y’all were talking about.
19:09 jimallman oti has two entries, anyway
19:10 jimallman hm, two different curator names too (Joseph Brown vs. Joseph W. Brown)
19:10 josephwb i saw that
19:10 josephwb ok, here: http://dev.opentreeoflife.org/curator/study/view/2926
19:10 mtholder i prefer to call him pg_joseph_w_brown
19:10 josephwb i set the preferred tree, but it has "taken"
19:11 mtholder did you mean "has not 'taken'" ?
19:11 jimallman hasn’t, you mean?
19:11 josephwb if i click "edit study" i see the box clicked
19:11 josephwb erg. yes. has not taken
19:11 jimallman hm, no preferred trees in http://dev.opentreeoflife.org/curator/study/view/pg_2926 either
19:11 mtholder the commit is https://github.com/OpenTreeOfLife/phylesystem-0/commit/9b845a82076341ccfb7014d3ea38f53143890709
19:12 jimallman i don’t see the preferred flag in these changes
19:12 josephwb i saved it twice
19:13 mtholder sorry https://github.com/OpenTreeOfLife/phylesystem-0/commit/068d6f144bd2b8bbb571d653f7a238e51f471203
19:13 mtholder (I'm not seeing the change going by commit msg)
19:14 jimallman yes, the change is there. preferred tree(s) indicated by:^ot:candidateTreeForSynthesis": ["tree6757"]
19:15 josephwb hmm, doesn't appear here: http://dev.opentreeoflife.org/curator/study/view/2926
19:15 josephwb but rooting does
19:16 mtholder jimallman. that used to be syntax like ot:candidateTreeForSynthesis:{candidate : ["tree6757"]} we changed that somewhat recently
19:16 mtholder perhaps that is a clue
19:18 kcranstn fyi, posted a summary of this discussion on the software list (this = editing in curator rather than pg)
19:21 pmidford joined #opentreeoflife
19:26 jimallman mtholder: thanks, maybe a glitch (master vs. production branch) in opentree...
19:27 jimallman YES, it was just a big (old JSON structure) in the View (vs. Edit) display
19:28 jimallman josephwb: you should see accurate tree status in the Edit display (checkbox) even if the View display says “NO” for all trees
19:28 * jimallman is fixing this now…
19:38 pmidford left #opentreeoflife
19:46 pmidford2 joined #opentreeoflife
19:46 mtholder kcranston: the curl calls now on https://github.com/OpenTreeOfLife/opentree/wiki/Open-Tree-of-Life-APIs work
19:47 kcranstn w00t!
19:47 mtholder I'll probably break them soon...
19:47 lcoghill joined #opentreeoflife
21:01 kcranstn joined #opentreeoflife
21:32 josephwb joined #opentreeoflife
21:50 kcranstn joined #opentreeoflife
21:54 kcranstn is this temporary? http://files.opentreeoflife.org/ott/
21:56 kcranstn where did the page go with all of the detail about OTT?
22:04 mtholder joined #opentreeoflife
22:43 towodo joined #opentreeoflife
23:33 towodo jimallman, traffic
23:33 towodo I've been thinking about how to disable the curator app on the production server
23:44 josephwb joined #opentreeoflife

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