Perl 6 - the future is here, just unevenly distributed

IRC log for #opentreeoflife, 2015-01-29

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

All times shown according to UTC.

Time Nick Message
01:09 jar286 joined #opentreeoflife
01:22 pmidford2 joined #opentreeoflife
02:05 jar286 joined #opentreeoflife
02:48 ilbot3 joined #opentreeoflife
02:48 Topic for #opentreeoflife is now Open Tree Of Life | opentreeoflife.org | github.com/opentreeoflife | http://irclog.perlgeek.de/opentreeoflife/today
03:28 pmidford2 joined #opentreeoflife
11:44 kcranstn joined #opentreeoflife
14:55 mtholder joined #opentreeoflife
14:56 mtholder josephwb, are you there?
15:09 mtholder joined #opentreeoflife
15:09 josephwb i am here now
15:10 josephwb mtholder ^
16:04 mtholder joined #opentreeoflife
16:16 mtholder joined #opentreeoflife
17:03 pmidford2 joined #opentreeoflife
17:12 jar286 joined #opentreeoflife
17:13 mtholder joined #opentreeoflife
17:13 josephwb mtholder: didjya need something?
17:42 mtholder are there instructions for running gcmdr?
17:43 mtholder josephwb ^
17:43 mtholder (and sorry for the delayed response - I was muted)
17:51 mtholder jar286, do you know the gcmdr commands to use to build the synthetic tree?
17:51 jar286 I would look at gcmdr to find out
17:51 jar286 sorry
17:52 jar286 there might be a script
17:53 mtholder 86 .py files and 142M size repo is intimidating...
17:53 mtholder (I know that many .py  are data not code)
17:54 jar286 I don’t understand why they keep all those files in the repo… they just use the repo as a snapshot of inputs
17:55 jar286 but if you want to replicate, I guess it’s handy to have the source files there.
17:57 josephwb mtholder: nope.
17:57 josephwb what do you want to do?
17:58 mtholder rebuild the synthetic tree in the process of debugging issue-78
17:58 josephwb there is a conf file where you set variables
17:59 josephwb mine: https://github.com/OpenTreeOfLife/gcmdr/blob/master/wopr_conf.py
17:59 josephwb and run a synth with something like this: https://github.com/OpenTreeOfLife/gcmdr/blob/master/run_JWB.py
18:00 mtholder ok thanks. I'll start there.
18:00 josephwb not a great example; studytreelist = [] is empty there.
18:00 josephwb i will push a fresher version
18:01 josephwb oh, wait; it reads in the individual study lists.
18:01 josephwb https://github.com/OpenTreeOfLife/gcmdr/blob/master/run_JWB.py#L18
18:02 josephwb those lists are in the folder: https://github.com/OpenTreeOfLife/gcmdr/tree/master/files_for_submission_v2.0
18:02 josephwb that *should* be it.
18:02 mtholder thanks
18:03 josephwb note stephen and i set our confs differently. if you stick to mine then things should work.
18:03 josephwb r ewe going to do the whole thing?
18:03 mtholder hopefully not.
18:03 josephwb ~24 hours on my fancy pants computer
18:04 mtholder do you know the max memory usage?
18:04 josephwb hmm.
18:04 mtholder very roughly is fine
18:04 josephwb i think i play it safe and allocate 32 GB
18:05 josephwb probably way too much
18:05 josephwb let me check
18:06 mtholder don't sweat it, if you don't have it handy.
18:06 josephwb yup, 32 GB: https://github.com/OpenTreeOfLife/gcmdr/blob/master/wopr_conf.py#L23
18:07 josephwb haven't tried it with less, never had a problem.
18:07 jar286 mtholder, I have a 40G machine you can use
18:08 mtholder thanks. I'll email you if I need it.
18:08 josephwb don't try MLS, tho. we've yet to find a machine with sufficient memory!
18:08 jar286 (with no firewall issues)
18:08 mtholder Diff topic. Are we doing PRs today?
18:42 jar286 yes
18:43 jar286 mtholder ^
18:59 jimallman ready for PRs here
19:00 jimallman https://github.com/pulls?user=OpenTreeOfLife
19:00 mtholder i'm here...
19:00 pmidford2 ready
19:02 jar286 hi
19:02 jar286 https://github.com/OpenTreeOfLife/germinator/pull/9
19:03 pmidford2 This includes some changes to the push scripts and the readme.md that I made this afternoon
19:03 jar286 looking…
19:03 mtholder seems fine. won't break anything..
19:03 pmidford2 Also some related changes to the wiki page
19:04 jar286 is this enabled in a cron job?
19:04 pmidford2 Any suggestions on what the crontab should be
19:04 pmidford2 yes - as on the wiki page
19:04 jar286 what do you mean what it should be?
19:04 jimallman just curious: if there’s “nothing to report” (e.g. no synthesis since last time), is a file created?
19:05 pmidford2 Yes, it just adds the same values, indexed at the later time - maybe we need something other than crontab
19:05 pmidford2 especially for synthesis
19:06 pmidford2 We haven't come to a consensus about when these should run
19:06 pmidford2 I set the cron up so there would be data for jim to play with
19:06 jar286 I’ve suggested daily
19:07 pmidford2 that's what we have now then
19:07 jar286 even if, say, the front end only looks at entries at weekly intervals
19:07 jar286 I think we can merge this
19:07 jar286 any objections?
19:07 jimallman sounds good. if we have lots of visual clutter (redundant data), i can filter it out on the client for now
19:08 jar286 right
19:08 jar286 merged.
19:08 jimallman pmidford2: thanks for linking the wiki here, it always helps to leave these breadcrumbs
19:08 jar286 yep
19:08 jar286 ot-base next
19:08 pmidford2 yes, beats maintaining it in two places
19:08 jar286 https://github.com/OpenTreeOfLife/ot-base/pull/12
19:08 mtholder that looks good to me, but I haven't tested it yet.
19:09 mtholder I need to figure out how to trigger it. Treemachine, presumably...
19:09 jar286 looks good. this only affects taxomachine for now, since treemachine doesn’t use ot-base (yet).
19:10 jar286 objections? need more time?
19:10 mtholder I think that it is fine to merge.
19:10 pmidford2 I think this is ok
19:11 jar286 done.
19:11 jimallman looks OK to me
19:11 jar286 it will go live on production this afternoon
19:11 jar286 reminder to self: we need to merge production into development branch
19:11 jar286 just in case something got forgotten
19:11 jar286 I mean master
19:12 * jimallman nods
19:12 jimallman in fact, it will be interesting to diff this “PR” to see if our current workflow is working
19:12 jar286 let’s see… peyotl next, ok?
19:12 jar286 https://github.com/OpenTreeOfLife/peyotl/pull/109
19:13 jimallman this is gonna take a few minutes...
19:14 jar286 the OTT build goes to some trouble to ensure unicode - this was a serious problem with preottol
19:14 jar286 so glad someone noticed
19:14 mtholder apparently in markdown __init__ becomes bold init
19:14 mtholder (in the description of the PR)
19:15 jimallman yep, **this text** and __this text__ will be bold (_this_ or *this* = italic)
19:15 jar286 (still reading)
19:15 jimallman there’s probably an escape for that.. maybe \* ?
19:15 mtholder or `
19:16 jimallman yeah, that would be best for these paths
19:17 jar286 I idly wish the tree_id for OTT included the version number
19:18 mtholder in the treemachine web services?
19:18 jar286 everywhere
19:18 jar286 I’m not going to be able to read all this code right now.
19:18 jar286 ...
19:18 josephwb it should have that info for treemachine
19:19 mtholder most of it is new tutorials and examples.
19:19 jar286 it looks fine.  merge?
19:19 jar286 we know it will work in production?
19:19 jar286 travis likes it...
19:20 mtholder and the more thorough phylesystem-api and peyotl tests pass
19:20 mtholder (those that we haven't gotten working through travis)
19:20 jar286 if I had to revert it, what would I do?  ‘make’ ?
19:20 jar286 I mean, after the appropriate checkout
19:21 mtholder sorry. I'm confused by that what you mean?
19:21 jar286 We had a problem with treemachine last time, and that makes me think what has to happen if we need to revert to a previous version on the server
19:22 mtholder just redo the deployment for python code.
19:22 jar286 most hands-off would be to create a branch on github, and redeloy pointing to it
19:22 jar286 ok. no matter, we’ll burn that bridge when we come to it. :)
19:22 jar286 any objections to merge?
19:22 jimallman nope
19:22 pmidford2 no
19:23 jar286 done.
19:23 jar286 https://github.com/OpenTreeOfLife/peyotl/pull/110
19:23 mtholder we use "python setup.py develop" for peyotl. So we install a link in the python libs
19:23 josephwb jar286: do you mean this treemachine issue: https://github.com/OpenTreeOfLife/treemachine/issues/152
19:23 mtholder so if the source tree changes, the next python process will use the new code.
19:23 jar286 yes
19:23 josephwb i thought it was not a treemachine issue.
19:24 josephwb deployed fine
19:24 pmidford2 I'm fine with peyotl 110 - it works as advertised
19:24 jar286 it is a treemachine issue, but it got me thinking, what if something like that happened in other repos?
19:24 jar286 ok, merging
19:24 mtholder (note that what I said about the next python process above means restarting web2py for the web apps)
19:25 jar286 merged.
19:25 jimallman that (#152) was a transient bug, with no satisfying explanation iirc. i re-deployed master branches and the plugin *poof* just appeared.
19:25 pmidford2 re:110 great
19:25 jar286 https://github.com/OpenTreeOfLife/peyotl/pull/112
19:26 josephwb right. but seems to be independent of the treemachine code.
19:26 jimallman josephwb: true
19:26 josephwb i'd fix things, but nothing has changed.
19:27 jar286 josephwb, there were 4 commits, I think, that might have caused the problem.  (maybe we should discuss this on github)
19:27 jar286 oh right. jim said it was transient.
19:28 jar286 jim & I will talk about this after PR review is over.
19:28 josephwb none of the commits touched anything relevant.
19:28 josephwb ok
19:28 jar286 (right, I saw that… later)
19:28 jar286 verbose.  that’s ok though
19:29 jar286 ok, looks good.  objections to merge?
19:29 jimallman looks good to me so far
19:29 pmidford2 looks ok, first file looks like a dup of 110
19:29 jar286 merged.
19:30 mtholder yes pmidford2 that is correct.
19:30 mtholder not sure why...
19:30 jar286 skipping reference-taxonomy and phylografter per usual
19:30 jar286 oh wait.  let’s do https://github.com/OpenTreeOfLife/reference-taxonomy/pull/121
19:31 jar286 some day I need to learn about pickles
19:31 pmidford2 This is a initial commit - there are problems that I have noted.
19:32 mtholder cool. is this used with the daily snapshots or entire releases?
19:32 pmidford2 entire releases for now - I just wanted a accession file I could work with
19:33 jar286 this is just silva seqs, right?  not all of genbank?
19:33 pmidford2 No, actually this is all of genbank
19:33 jar286 oh wow
19:34 jar286 jimallman, let’s get this hooked into the UI within the next 2 months (before Katz lab departs the grant)
19:34 jimallman yes, i’m hoping to hit this hard in the coming week
19:34 pmidford2 Would it make sense to try and filter to just silva?
19:35 jar286 I’ll make an April 15 milestone and attach the existing issue to it, we can talk later
19:35 mtholder note that db_xref='taxon:... can occur multiple times in a FEATURES table
19:35 jar286 I don’t know.  depends on load time and memory footprint
19:35 mtholder we might want to store all of those.
19:35 jar286 filtering to silva is just an optimization
19:35 mtholder taxa that is.
19:35 pmidford2 mtholder - good point, I should make sure those are caught
19:36 mtholder just made a line note.
19:36 pmidford2 Got it
19:36 jimallman jar286: my head was in the stats pages, sorry… this would appear in the general “tell me about this taxon” info?
19:36 mtholder we probably want to avoid accesssions that map to multiple taxa in automated OTT ID mapping...
19:37 jimallman (can’t find the issue #s, sorry)
19:37 jar286 no, this is about genbank ids in OTU original labels
19:37 jar286 oh, I didn’t know there were such accessions.
19:37 jar286 ouch.
19:37 mtholder chimeric sequences
19:37 mtholder bacteria with inserts...
19:38 jar286 oh foo.  will they end up as OTUs?
19:38 mtholder we can just drop those accesssions
19:38 mtholder so the curator will have to decide how to map them.
19:38 jar286 right.  let curator find taxon manually.
19:38 mtholder they aren't too comon.
19:38 pmidford2 are they likely to be in silva?
19:38 jar286 ok, objections to merging? this is great
19:38 mtholder looks good to me. very nice.
19:39 jar286 done.
19:39 jar286 https://github.com/OpenTreeOfLife/opentree/pulls
19:39 jar286 https://github.com/OpenTreeOfLife/opentree/pull/536
19:39 jar286 Karen is busy this week, yet another workshop or meeting
19:40 jimallman this PR is (as usual) embryonic.. lots of work to do building release/version pages, un-faking data, etc.
19:40 jimallman at this point i have a lot to work with, should have more progress next week
19:41 jar286 then I think we should skip it, maybe set up a separate meeting or make another email call for review
19:41 jimallman yep, no change here in the last week in any case
19:41 jar286 https://github.com/OpenTreeOfLife/opentree/pull/565
19:41 jimallman but many questions answered (thanks all!)
19:41 jimallman this is the incomplete SSL cert chain, and some other SSL vulnerabilities that turned up along the way
19:42 jar286 just as a matter of good practice, it’s best to not discuss security problems in public fora until they’re fixed.  I guess this one doesn’t matter since it’s about certs
19:42 jimallman noted, thanks
19:42 jimallman person-to-person email, then, or hangout?
19:42 jar286 yes
19:43 mtholder god this looks complex. but the https issues are fixed for me. thanks for wading into this hellhole, jimallman.
19:43 jar286 gmail-to-gmail mail, or PGP
19:43 jar286 I haven’t been following this too closely, it’s daunting
19:43 jimallman OK with putting public certs in the repo? i did a lot of reading, and this is safe (as far as i know) provided private keys are not included.
19:43 jar286 yes
19:43 jar286 but sort of fragile
19:43 jimallman yeah, it was gnarly but quite educational
19:44 pmidford2 I'll sign off on this, not having followed this closely enough to say anything more
19:44 jimallman i’ve tried to leave lots of explanation here: https://github.com/OpenTreeOfLife/opentree/blob/apache-ssl-fixes/deploy/setup/ssl-certs/README.md
19:44 jar286 ahh… certs from ~/.ssh - that’s great
19:45 jar286 oh wait
19:45 jimallman and more explanation here: https://github.com/OpenTreeOfLife/opentree/issues/553
19:45 pmidford2 jimallman - thanks for the pointer
19:45 jar286 looks like certs aren’t in .ssh, maybe they should be?? I don’t know
19:45 jar286 on the deploying machine, not on aws
19:46 jimallman the resolver script (added here) fetches the entire chain, using our wildcard cert as a starting point.
19:46 jimallman very clever. it now works as the entire chain, see modified apache config
19:46 jar286 looking at line 24
19:46 jar286 of readme
19:47 jimallman the resulting single file (STAR_opentreeoflife_org.pem) is all we need on the server, along with the private key of course.
19:48 jar286 and that’s transferred manually when the server is set up, right?
19:48 jimallman right. as it turns out, the .pem needs to be placed manually in any case, along with the private key.
19:48 jimallman so that deployment scripts know whether to enable HTTPS
19:49 jar286 ahh.
19:49 jar286 but that could be a configuration option.
19:49 jimallman alternately, we could place the .pem programmatically, and check for the private key...
19:49 jar286 I was just typing that.
19:49 jimallman true, we could use an explicit flag in our server-config file
19:49 jar286 but put on to-do list.
19:50 jar286 hmm.  putting public certs in github seems weird.
19:51 jar286 github admins could clobber them.
19:51 jar286 I wonder what other people do.
19:51 jimallman hm, i can see why. you need to look inside to see whether they contain private keys.
19:51 jar286 one key could be swapped out for another
19:51 jimallman ssl is so funky. file extensions are almost meaningless, a .pem file can hold this OR that OR both PLUS other stuff...
19:52 jar286 this is real paranoia. I don’t know a better way.  maybe we could print fignerprints on initialization
19:52 jar286 s/gn/ng/
19:52 jar286 and keep the fingerprints on 3x5 cards
19:53 jimallman oh, you mean someone could switch the certs on us (or on a hapless later user)?
19:53 jar286 right
19:53 jimallman hrm
19:53 jar286 could be slipped into a PR
19:53 jimallman we can remove them from version control, but i kinda like having the cert and HOWTO and resolver script in one place.
19:54 jar286 like i say this is true paranoia. let’s merge and deal with it later. I’ll put on my to-do list
19:54 jimallman i assume anyone with a real deployment would have their own domains, certs, keys… right?
19:54 jimallman so it’s mostly a hazard for future-us
19:54 jar286 they would if they knew what was good for them
19:55 jar286 the installation step is still manual, so I wouldn’t worry too much
19:55 jimallman …or whoever’s running *.opentreeoflife.org, i mean
19:55 jimallman ok
19:55 jar286 merge?
19:55 mtholder sure
19:55 pmidford2 fine
19:55 jar286 done. next…
19:56 jar286 https://github.com/OpenTreeOfLife/opentree/pull/566
19:56 jar286 how do I see this?  just go to any tree?
19:57 jimallman yes, but it works best with one with an unconfirmed root… looking for a link here: https://github.com/OpenTreeOfLife/opentree/issues/524
19:57 jimallman that issue has screnshots of all changes in one place
19:57 jimallman scratch that, latest text is just on devtree
19:58 jimallman whoa, devtree is NOT showing this code.
19:59 jimallman we should be seeing it here: https://devtree.opentreeoflife.org/curator/study/edit/694?tab=trees&tree=tree1247
19:59 jimallman my bad, not sure what’s up on devtree
19:59 jar286 branch names not in server details
20:00 jimallman doh! i’m looking at my local server. the url above should work for you!
20:00 jimallman jar286: noted
20:00 jimallman :)
20:00 jar286 I guess I could file an issue for that
20:00 jimallman here’s the “view” (non-edit) for the same tree: https://devtree.opentreeoflife.org/curator/study/view/694?tab=trees&tree=tree1247
20:00 mtholder looks good.
20:01 jar286 doesn’t work for me, click on tree - weirdness
20:01 jimallman jar286: already an issue for this: https://github.com/OpenTreeOfLife/opentree/issues/561
20:01 jar286 ah. your link works
20:01 jar286 ok
20:01 jimallman weirness, you say?
20:01 jimallman weirdness?
20:02 jar286 no problem now. can’t reproduce
20:02 jar286 the tree popup popped up and immediately disappeard.
20:02 jimallman ah, i’m getting a weird one now: “Sorry, there was an error inferring the context.” i’ll chase this on my own, later (probably a phantom)
20:02 jar286 walk me through this. I’m a curator and I’m alerted to need for root check how?
20:03 jar286 ok, in the tree list.
20:03 jar286 so, I select the tree...
20:03 jimallman yes, but i don’t flag it as an error per se.
20:03 jar286 umm… then what.
20:03 jimallman we set the default root-state as “unconfirmed”, then it’s up to the curator to confirm it as they see fit.
20:04 jar286 how do I confirm?
20:04 jar286 i.e. how do I choose the root?
20:04 jimallman i could highlight it in red (in the tree list, and “quality” prompts above), but then i’ll need to be sure that the root is not confirmed by inattention, and not on purpo;se.
20:04 jar286 but how do I choose it now?
20:04 jimallman in the editor, the tree popup has a Properties tab with this (radio buttons)
20:05 jimallman if you click the “tree root” state in the trees list, it will be highlighted in the popup.
20:05 jimallman whoa, i just saw the disappearing popup! just once, now it works. will investigate
20:06 jar286 ok, the message is good… but don’t see ‘tree root’ state
20:06 jar286 oh there
20:06 jimallman yeah, it’s not red or underlined.. just clickable for the curious.
20:06 jar286 no, I mean, how to I pick a particular node, and confirm it as the root?
20:06 jimallman again, because i didn’t want to cause undue alarm, in case a curator says “this is fine as-is"
20:07 jar286 oh it says at the top
20:07 jimallman oh! open the tree popup, click on any node (or edge) and choose that option in the menu
20:07 jimallman right
20:07 jar286 Click any node or edge to re-root the entire tree.   -> click, but tree isn’t rerooted.
20:08 jimallman click to select, should show a menu. menu includes rooting, if possible.
20:08 jar286 no menu. what does ‘if possible’ mean?… it’s outside the ingroup?
20:09 jimallman are you in the editor? or a read-only view? if the latter, i should modify the help text to omit these features.
20:09 jar286 ah. read-only.  I always get that wrong
20:09 jimallman (if tree label above the phylogram is editable, you’re in the editor)
20:09 jar286 I’ll pursue this later.
20:09 jar286 merge?
20:10 jimallman hm, let’s hold back.
20:10 jimallman i’m having trouble if i go directly to the trees list: https://devtree.opentreeoflife.org/curator/study/edit/694?tab=trees
20:10 jar286 ok.
20:10 jimallman i can’t get a consistent tree popup, it keeps disappearing immediately.
20:10 jimallman sorry, thought i had thoroughly tested this. :-/
20:10 mtholder left #opentreeoflife
20:10 jar286 add a comment to the PR?
20:10 jar286 so we can pick up again next time?
20:11 mtholder joined #opentreeoflife
20:11 jimallman yes, or sooner if we want to (shouldn’t take long to fix)
20:11 jar286 ok
20:11 jar286 https://github.com/OpenTreeOfLife/opentree/pull/567
20:11 jimallman adding a comment now
20:12 jimallman best described (with screenshots) here: https://github.com/OpenTreeOfLife/opentree/issues/481
20:12 mtholder looks good to me
20:12 jimallman a simple change, moving this message from an intrusive popup to the sidebar
20:13 jimallman working example on devtree: https://devtree.opentreeoflife.org/curator/study/edit/pg_2584/?tab=trees
20:13 jimallman also the Files tab
20:15 jimallman back in a sec…
20:15 jar286 what am I looking at?...
20:15 jar286 found it.
20:16 jar286 lower right of trees tab.
20:16 mtholder and files tab
20:16 jar286 looks fine to me
20:16 mtholder me too.
20:16 jar286 merge without waiting for jim?
20:17 jimallman i’m back!
20:17 pmidford2 ok with me
20:17 jimallman fyi, the link to deposit will have sensible formatting (if treebase vs. others)
20:17 jar286 done.
20:17 jar286 we should stop at 3:30… I didn’t send out a downtime reminder, i’ll do that now. you can carry on without me
20:18 jimallman https://github.com/OpenTreeOfLife/opentree/pull/568
20:18 jimallman this one is a simple CSS fix, to push OTU mapping choices above the right sidebar.
20:19 jimallman see screengrabs here: https://github.com/OpenTreeOfLife/opentree/issues/544
20:19 jimallman it’s literally a one-liner, no likely side effects.
20:19 jimallman (i realize this is not a beautiful solution, but i’m not sure how else to resolve this space crunch)
20:19 mtholder one line with a comment. a PR, and a GH issue. that is quite a line :-)
20:19 mtholder it looks good to me.
20:20 mtholder I'll merge...
20:20 jimallman danke
20:20 pmidford2 yes, fine
20:20 jar286 back
20:20 mtholder https://github.com/OpenTreeOfLife/opentree/pull/569
20:21 jimallman we’re doing some easy ones..
20:21 jimallman again, best explained by screenshots in the issue: https://github.com/OpenTreeOfLife/opentree/issues/549
20:21 jimallman this was just a cosmetic update to the tree-view legend
20:21 mtholder purty.
20:21 jimallman to show the blue color and “i”, versus the old red “?”
20:21 jimallman :D
20:21 jar286 we’ve seen this. looks fine. hilmar will be pleased
20:21 pmidford2 yes, looks better to me
20:21 mtholder i'll merge.
20:22 jimallman https://github.com/OpenTreeOfLife/opentree/pull/570
20:22 jimallman another tiny one, just reconciling the tree “Inference method” labels for now.
20:22 jar286 this will be very helpful I think, even in its present state
20:23 jimallman yes, it should avoid some confusion while we figure out the bigger picture of tree properties
20:23 mtholder agreed.
20:23 pmidford2 yes
20:23 * jar286 waiting for mtholder to merge. hand MC role back to me if you like
20:24 jimallman this is also the start of hiding more of the default HTML that appears briefly as our pages load.. more of this to come.
20:24 mtholder merged
20:24 jimallman (thus some of the more cryptic diffs)
20:24 jimallman https://github.com/OpenTreeOfLife/opentree/pull/571
20:24 jimallman this was done a few weeks back, but never made it into a PR for some reason
20:26 jimallman we now can have more useful sharable URLs, like https://devtree.opentreeoflife.org/curator?match=Joseph+Brown&order=Oldest+publication+first
20:26 jar286 could this be related to the glitch we saw earlier?
20:26 jar286 ‘intermittent "tree not found" error before showing the tree’
20:26 jimallman this code fixes that (found the bug as I was building this)
20:27 jimallman see coment here: https://github.com/OpenTreeOfLife/opentree/commit/8f30141aadb14b8218c21249789456b6924fee81
20:27 jar286 ok. this looks good to me.
20:27 jimallman since the page is more disciplined about waiting for all data to load, it always finds the tree (assuming it’s present)
20:27 mtholder very nice to have the state in the urls.
20:27 pmidford2 yes
20:27 mtholder looks good. OK to merge
20:27 mtholder merging...
20:28 jimallman almost done, 3 minutes to go!
20:28 jimallman https://github.com/OpenTreeOfLife/opentree/pull/572
20:28 jar286 can’t make things worse, right?
20:28 jimallman this is for mtholder’s recent finding of single-child nodes..
20:28 mtholder it was a big help to me today.
20:28 mtholder thanks.
20:28 jimallman cool. we could alternately treat the parent-of-one-child as an unnamed node, if that makes more sense
20:29 jimallman (would be a quick change)
20:29 mtholder let me check one really tough case. hold on...
20:29 jar286 should be ‘recur’ not ‘recurse’ :)
20:29 jar286 like incur, incursion
20:29 jar286 recur, recursion
20:29 jimallman ! good to know (makes sense)
20:30 jimallman occur, occursion… no wait
20:30 mtholder yes. works on https://devtree.opentreeoflife.org/opentree/argus/otol.draft.22@3843724
20:30 jimallman the square brackets are an unfortunate side effect
20:30 jimallman since my code also doesn’t expect to find these situations.
20:30 mtholder hopefully it won't seem them. soon...
20:31 mtholder I'll merge....
20:31 jimallman i can fix this, but if we don’t expect this tree structure it doesn’t seem worth the effort
20:31 mtholder I think it is fine as is. folks will be weirded out by the out-degree one nodes, but the naming is helpful..
20:31 mtholder merged.
20:32 mtholder I think that we're done.
20:32 jar286 done! thanks all
20:32 mtholder ciao
20:32 jar286 now for deployment
20:32 jimallman cheers
20:32 jimallman (oh, i’m staying)
20:32 pmidford2 good luck
20:32 jar286 can we deploy master to devtree and devapi, and do a smoke test?
20:33 * jar286 I need a biobreak, brb
20:34 jimallman i’m OK with testing on devtree and devapi. we’ll just need to restore the development branches later, or i’ll confuse myself. :-/
20:36 jar286 right
20:36 jar286 i’ll go ahead with that, shouldn’t take long
20:38 jar286 devapi was pointing to master !
20:39 jar286 that is right though
20:39 jar286 how confusing.
20:39 jimallman ah, maybe we don’t have a development branch there? haven’t been messing around there lately
20:40 jimallman yep, no such branch here: https://github.com/OpenTreeOfLife/phylesystem-api/branches
20:40 jimallman nore in oti, etc
20:40 jimallman nor
20:41 jar286 but devapi didn’t have branch development for opentree (which is used by something or other)
20:41 jimallman that’s all on devtree
20:41 jar286 nope. there’s a copy of the opentree repo on devapi / api. I forget why
20:42 jimallman maybe a mistaken push?
20:42 jimallman here’s the component list from devapi.config:
20:42 jimallman OPENTREE_COMPONENTS='phylesystem-api oti treemachine taxomachine'
20:42 jar286 no, I remember mark I think saying he needed it.  no, the opentree *repo*, not the component
20:42 jimallman it does sound strangely familiar… for ncl, perhaps?
20:43 jimallman or peyotl?
20:43 jar286 don’t remember. not important, but it needs the right branch.
20:43 jar286 *** Failed: studies/find_studies {"property":"ot:studyId","value":"pg_41","verbose":true} Cariaga
20:43 jar286 {
20:43 jar286 "matched_studies" : [ ]
20:43 jar286 }
20:43 jar286 Successes: 9 Failures: 1
20:43 jimallman ok, but i’d like to understand the dependency
20:44 jar286 you think that’s an indexing issue.
20:44 jimallman perhaps, or difference from phylesystem-0 vs phylesystem-1?
20:45 jimallman https://devtree.opentreeoflife.org/curator/?match=cariaga
20:45 jar286 I believe I chose an example that worked in all three phylesystems.
20:45 jimallman i see it in the devtree study list, but the ids there have no ‘pg’ or ‘ot’ prefix
20:45 jimallman (filtered URL for the win)
20:45 jar286 ah!
20:46 jar286 that would do it.
20:46 jar286 ok.
20:46 jimallman yeah, i keep starting to file bug reports, then remembering this difference.
20:46 jar286 now for devtree.
20:46 jimallman maybe we should pave over phylesystem-0 one of these days, with the latest phylesystem-1?
20:47 jar286 yes
20:47 jar286 ok, time for smoke tests on devtree
20:49 jar286 annoying that my name matches ‘trees’
20:49 jar286 ok, I declare it working.  will do api now.
20:52 jar286 Successes: 10 Failures: 0
20:53 jimallman better!
20:56 jar286 ok, we’re up.
20:56 jar286 the ‘i’ bubble on nodes is sticky
20:56 jar286 mouseover, then move off to the right
20:56 jimallman hm, i got that error “Sorry, there was an error inferring the context.” again, on production. will pursue this ASAP.
20:57 jimallman re: sticky, in the synth-tree viewer, or curation app?
20:57 jimallman oh, must be synth tree
20:58 jar286 synth tree viewer.
20:58 jimallman i see. yeah, these linger if they don’t detect the mouse-out event
20:58 jar286 mouseover a rightmost node then move mouse off of it to the right
20:58 jar286 maybe that’s old behavior. not too important
20:59 jimallman this is old news. a new highlight will always clean these up, so i haven’t made it a priority
20:59 jar286 if I click on a focal clade in the study list, I get nothing
20:59 jar286 not sure what to expect there, but didn’t expect nothing
20:59 jimallman might need to clear cache
21:00 jimallman yes, that fixed it for me (inline script is using clade name, linked JS is using ottid)
21:00 jar286 ok, works now.
21:03 jar286 jimallman, where did you learn your way with words?
21:04 jimallman hmm. undergrad in scriptwriting helped, i guess? design education is all about communication too.
21:04 jimallman and it’s fun!
21:05 jar286 can’t think of examples right now but you keep coming up with gems
21:05 jar286 thanks for the PR and deployment help
21:05 jimallman :) thanks. i like my language loose, except in code.
21:05 jimallman this project has been a great experience all around for me
21:05 jar286 oh good
21:05 jar286 sometimes seems like a slog
21:06 jimallman yes, definitely a marathon
21:06 jimallman i get so involved in the building, i need to look up and see if we’re getting engagement.
21:06 jar286 we’re not.
21:07 jimallman “traction”, as the startup folk would say. always a challenge. it’s another major effort, and someone needs to carry the ball.
21:08 jimallman i know we’re talking about it in all-hands etc, but i’ll try to put more serious thought into promotion and incentives.
21:08 jimallman if you build it, they will… wander off and watch American Idol.
22:09 jar286 oops! forgot to do the glibc patch
22:09 jar286 doing it now

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