Time
S
Nick
Message
00:00 cads
pdurbin, LabKey is /very/ impressive!
00:01 cads
extremely specialized for some pretty involved research processes, very awesome example of a domain targeted open data repository system
00:15 pdurbin
nice
00:52 djbrooke joined #dataverse
02:48 djbrooke joined #dataverse
05:13 axfelix joined #dataverse
07:09 jri joined #dataverse
08:57 romainM joined #dataverse
09:33 romainM joined #dataverse
09:47 romainM78 joined #dataverse
10:33 pdurbin
romainM jri: question for you when you have a moment. About the pull request you made yesterday: https://github.com/IQSS/dataverse/pull/3374
10:38 pdurbin
we're basically wondering how to test it
11:10 cads joined #dataverse
11:53 romainM
pdurbin: jri is on vacation for today
11:53 romainM
and I don't know how to test his ... "things" :D
12:00 romainM joined #dataverse
12:06 cads joined #dataverse
12:28 donsizemore joined #dataverse
12:48 pdurbin
romainM: jri added something called ":PiwikAnalyticsHost". I guess I'm wondering if you and jri would be willing to temporarily host an installation of Pwiki that we could test with.
12:49 romainM
I can't tell you till he's back :S
12:51 romainM
oh, and I just made a pull-request for the oai-pmh problem
12:51 pdurbin
romainM: ok, that makes sense. We were hoping to get this into the next release (4.5.1) but we can hold off until a future release.
12:52 romainM
the problem is that I can't "really" test it :S
12:52 romainM
but I mainly made a second xml generation function (as opposed to the only one before, which used both dc definitions, even it wasn't oai_dc kind)
12:53 romainM
and removed the wrong fields
12:53 pdurbin
romainM: thanks for the OAI-PMH pull request! On a related note, can you please read this issue so I can ask you a question about it? Harvest: Investigate OAI-PMH protocol test for registration as a data provider. · Issue #3307 · IQSS/dataverse - https://github.com/IQSS/dataverse/issues/3307
12:53 romainM
reading
12:54 romainM
yes ?
12:55 pdurbin
romainM: is the issue and pull request you opened related? It seems like it to me. Would you add a comment to 3307 saying that a partial fix is in your pull request for your issue (if that's true)?
12:55 romainM
it could
12:55 romainM
I'll add it
12:55 pdurbin
awesome. thanks!
12:55 romainM
well, given the problem, it's probably related yes
12:55 pdurbin
probably
12:56 romainM
given that the xml was wrong ^^
12:57 romainM
hum
12:57 romainM
coverage decreased ?
12:57 pdurbin
it happens
12:58 pdurbin
don't worry about it
12:58 pdurbin
we'd like code coverage to always go up
12:58 pdurbin
but sometimes it goes down, it's ok
12:58 pdurbin
donsizemore: did you notice https://github.com/IQSS/dataverse/pull/3211 got merged? Thanks!
12:59 romainM
was wondering what is the "coverage", and how it is possible to go down with what I've done :'(
12:59 donsizemore
@pdurbin i made two contributions to 4.5.1 =) meager, but contributions nonetheless
13:00 pdurbin
we appreciate them all
13:00 donsizemore
@pdurbin this morning i'm testing leonid's version import fixes with the same database dump / DDI export per https://github.com/IQSS/dataverse/issues/3362
13:01 pdurbin
donsizemore: oh good
13:01 pdurbin
romainM: it's nothing to :'( about
13:01 pdurbin
:)
13:01 romainM
ok :)
13:01 donsizemore
@pdurbin and pending a successful test-run today, i *think* we're going to make the jump on wednesday
13:02 pdurbin
same day we *might* deploy 4.5.1
13:02 pdurbin
but you didn't hear that from me
13:05 romainM
my pull request (if valid) would be in that version ?
13:07 majest1c joined #dataverse
13:08 pdurbin
romainM: nope. It has to be tested first. :)
13:08 romainM
ok
13:08 pdurbin
romainM: but again, the Piwik pull request might get in 4.5.1 if you or jri can host a Piwik server we can test against.
13:09 romainM
I can't, I have no rights on our server
13:09 romainM
he's our admin god :(
13:09 romainM
but on vacation (yes, god do have that)
13:09 pdurbin
admin gods are worth their weight in gold
13:10 romainM
cats, in our case
13:10 pdurbin
everyone deserves a day of rest
13:10 romainM
:D
13:23 pdurbin
cads: I've been thinking... you should write a blog post about your use case. :)
13:43 djbrooke joined #dataverse
13:44 djbrooke_ joined #dataverse
13:47 djbrooke joined #dataverse
13:52 djbrooke joined #dataverse
14:12 djbrooke joined #dataverse
14:22 pdurbin
telnoratti: I just left you some feedback at https://github.com/IQSS/dataverse/pull/3292
14:26 telnoratti
pdurbin: I actually noticed that too independantly about 2 hours ago, I was going to ask if you wanted me to include that fix in the same commit
14:27 pdurbin
telnoratti: wow! You must be reading our minds. Yes, if you would push another commit to your pull request to fix this it would be much appreciated!
14:28 telnoratti
cool, I'll probably get to it today, but might not. there are a number of fires that need putting out here
14:30 pdurbin
telnoratti: no rush
14:31 pdurbin
telnoratti: it doesn't even have a milestone yet at https://waffle.io/IQSS/dataverse
14:32 pdurbin
cads: you could also ask at https://lists.okfn.org/mailman/listinfo/okfn-labs
14:33 telnoratti
heh alright, by the way our aws stack is pretty close to production. We're testing upgrade paths right now and beginning to release the images. You can probably expect to see something on that soon
14:34 pdurbin
telnoratti: you're making AWS images of Dataverse?
14:34 telnoratti
no, we're using docker
14:34 pdurbin
oh, ok
14:35 pdurbin
telnoratti: you're running Docker on AWS?
14:36 telnoratti
yeah
14:36 telnoratti
we're using coreos images for most everything
14:36 telnoratti
then running docker on that, it makes it easier to upgrade the application without having to upgrade the cloudformation
14:37 pdurbin
telnoratti: what if I wanted to spin up an arbitrary branch from https://github.com/IQSS/dataverse/branches on AWS so that I could hit that branch of Dataverse with integration tests? Hard? Easy?
14:39 telnoratti
it's not tooled for that, but it wouldn't be many modifications, you would just have to build a docker image of it and specify that as a parameter when building the stack
14:40 telnoratti
In fact I'm not sure any modifications would be necessary, other than the docker images we built wouldn't work
14:40 telnoratti
they hard code the release url on github
14:41 pdurbin
cool, ideally Travis would kick off a deployment to AWS
14:42 djbrooke joined #dataverse
14:43 telnoratti
I'd be happy to help make that happen
14:43 djbrooke joined #dataverse
14:43 telnoratti
We don't currently have Two Ravens or rserve included in the stack, but it shouldn't be much of a reach to include those
14:45 pameyer
sounds interesting
14:52 djbrooke joined #dataverse
14:56 djbrooke joined #dataverse
15:00 pdurbin
telnoratti: I'd *love* some help with this. I'll definitely ping you.
15:20 djbrooke joined #dataverse
15:25 djbrooke joined #dataverse
15:26 djbrooke joined #dataverse
15:28 donsizemore joined #dataverse
15:37 djbrooke_ joined #dataverse
15:55 donsizemore
@pdurbin: you saw Craig Willis' stuff at https://github.com/nds-org/ndslabs-dataverse/tree/master/dockerfiles ?
15:59 djbrooke joined #dataverse
16:04 pdurbin
donsizemore: oh yes. I sort of assume telnoratti has seen it as well
16:04 pdurbin
Dataverse 4.2.3?!? We're about to release Dataverse 4.5.1!
16:04 pdurbin
well, soon, I hope
16:06 pdurbin
I should probably try to help them update it.
16:07 pdurbin
or *someone* should :)
16:12 djbrooke joined #dataverse
16:15 djbrooke joined #dataverse
16:17 telnoratti
yeah I saw it, they had a lot of custom install scripts iirc, and I wanted to avoid deviating far from the offical scripts for compatibility
16:17 pdurbin
telnoratti: interesting. So do you have your own Dockerfile for Dataverse?
16:19 telnoratti
yeah
16:20 telnoratti
I expect to be able to share it out soon, we need to get approval
16:20 telnoratti
made a dataverse specific solr one too
16:23 djbrooke joined #dataverse
16:28 pdurbin
interesting
16:28 pdurbin
kids today with their dockers and clouds
16:54 donsizemore
i bet they don't even dial their phones with pencils
16:56 pdurbin
:)
16:56 pdurbin
telnoratti: do please let me know when share your Dockerfiles
16:57 djbrooke joined #dataverse
16:57 telnoratti
will do
17:10 djbrooke_ joined #dataverse
17:32 pdurbin
thanks
17:35 djbrooke joined #dataverse
17:43 djbrooke_ joined #dataverse
17:48 djbrooke joined #dataverse
18:00 djbrooke joined #dataverse
19:14 djbrooke_ joined #dataverse
19:29 donsizemore joined #dataverse
19:33 djbrooke joined #dataverse
20:02 axfelix joined #dataverse
20:05 djbrooke_ joined #dataverse
20:14 donsizemore joined #dataverse
20:14 donsizemore
@pdurbin dumb shib question?
20:15 donsizemore
@pdurbin (because it's friday afternoon and my brain is fried)
20:19 pdurbin
donsizemore: hit me
20:19 djbrooke joined #dataverse
20:22 donsizemore
@pdurbin "urn:mace:incommon:unc.edu" is a perfectly good entityID for a unc.edu Shib group, yes? the sphinx examples contain URLs
20:23 donsizemore
@pdurbin: same thing for Carleton College (urn:mace:incommon:carleton.edu) but that's what was in the DiscoFeed
20:23 pdurbin
donsizemore: yes, perfectly good
20:24 pdurbin
donsizemore: go look at your authenticateduserlookup table: http://phoenix.dataverse.org/schemaspy/latest/tables/authenticateduserlookup.html
20:24 pdurbin
persistentuserid field
20:25 donsizemore
933 peeps there
20:25 pdurbin
probably a lot of urn:mace:incommon:unc.edu in there
20:26 pdurbin
we have users from MIT in our production and their entityID doesn't look like a URL either
20:26 donsizemore
@pdurbin only one on today's migration run, but yes =) thank you!
20:27 pdurbin
the URL style is newer and what's recommended these days
20:27 pdurbin
but entityID is a string, not a URL
20:27 pdurbin
the newer ones just happen to look like URLs
20:27 axfelix joined #dataverse
20:29 donsizemore
excellent. thank you!
20:29 pdurbin
donsizemore: I made you a pull request by the way, sort of by accident: https://github.com/IQSS/dataverse/pull/3380
20:29 pdurbin
it was going to be a quick test for https://waffle.io/IQSS/dataverse
20:30 pdurbin
but I had the code somewhat written so I went ahead and pushed it
20:30 donsizemore
@pdurbin I saw! Thu-Mai and Sophia are looking over the wreckage of today's migration. for now, I'm just editing Bundle.properties post-deploy
20:30 pdurbin
huh. well, send us a pull request for your edits
20:35 axfelix joined #dataverse
20:39 pdurbin
ok, I'm out of here. have a good weekend everyone
20:39 djbrooke joined #dataverse
20:39 djbrooke joined #dataverse
20:42 pameyer
have a good weekend
21:18 cads joined #dataverse
21:22 djbrooke joined #dataverse
21:24 djbrooke joined #dataverse
21:27 djbrooke joined #dataverse
21:59 axfelix joined #dataverse
22:34 axfelix joined #dataverse
23:28 cads joined #dataverse
