IQSS logo

IRC log for #dataverse, 2018-08-09

Connect via chat.dataverse.org to discuss Dataverse (dataverse.org, an open source web application for sharing, citing, analyzing, and preserving research data) with users and developers.

| Channels | #dataverse index | Today | | Search | Google Search | Plain-Text | plain, newest first | summary

All times shown according to UTC.

Time S Nick Message
00:00 jri joined #dataverse
01:02 jri joined #dataverse
02:54 djbrooke joined #dataverse
03:18 jri joined #dataverse
04:02 jri joined #dataverse
05:03 jri joined #dataverse
06:04 jri joined #dataverse
07:04 jri joined #dataverse
07:06 jri joined #dataverse
08:35 djbrooke joined #dataverse
12:58 djbrooke joined #dataverse
13:50 djbrooke joined #dataverse
13:59 djbrooke joined #dataverse
14:01 pdurbin andrewSC bricas_ jri: Dataverse 4.9.2 is out: https://github.com/IQSS/dataverse/releases/tag/v4.9.2
14:16 djbrooke joined #dataverse
14:17 djbrooke joined #dataverse
14:21 djbrooke joined #dataverse
14:24 pameyer joined #dataverse
14:31 djbrooke joined #dataverse
14:48 andrewSC pdurbin: nice!! I'm still on 4.8.5 however once I've gotten the OAuth2 uri collision resolved with our proxy/env I'll be cruising right up to latest :)
14:49 pdurbin Yikes. Good luck.
14:49 andrewSC hahah
14:49 andrewSC thanksss
14:51 pdurbin If there's something we can do to help, please let us know.
14:52 djbrooke joined #dataverse
14:55 pdurbin pameyer: thanks for linking to the new 4.9.2 feature of importing datasets with existing DOIs: https://groups.google.com/d/msg/dataverse-community/ax0ytpGFMck/Hj7Bb6YuDgAJ . It was on my list to mention this. :)
14:57 pameyer pdurbin: I'd been about to mention that it was merged not released... then I heard 4.9.2 was immanent
14:57 pameyer I might've jumped the gun on the release announcement thought - guides doesn't have `/latest` pointing to 4.9.2
14:57 djbrooke joined #dataverse
14:59 pdurbin I always link to specific release numbers anyway. In the Guides. Because we may rearrange pages in the future.
15:12 djbrooke joined #dataverse
15:12 pdurbin pameyer: in docker-dcm, what port does DCM run on? And can I get at it from my Mac?
15:16 pameyer pdurbin: 80 and 22, and no
15:16 pameyer that's why there's a client container
15:20 pameyer if you wanted to, you could probably change the forwarding in the compose file.  why?
15:29 djbrooke joined #dataverse
15:29 pdurbin pameyer: I'd like to be able to use my normal development workflow of editing files in ~/NetbeansProjects/dataverse on my Mac, autodeploy to Glassfish running on my Mac (on save). The only difference from normal is that I want/need to configure :DataCaptureModuleUrl to point at either the mock running on my Mac or (localhost:5000) or docker-dcm on whatever port Glassfish on my Mac can reach. I hope that
15:29 pdurbin makes sense!
15:31 pameyer sounds like just the API port, right?
15:31 pdurbin yeah, ur.py or whatever
15:32 pameyer adding a `ports` line with `$your_favorite_local_port:80` under dcmsrv should do it
15:32 pameyer won't get you a shared filesystem though
15:44 pdurbin pameyer: in the compose file?
15:45 pameyer conf/docker-dcm/docker-compose.yml
15:46 djbrooke joined #dataverse
15:46 djbrooke joined #dataverse
15:51 pdurbin cool, thanks!
15:52 djbrooke joined #dataverse
15:55 djbrooke joined #dataverse
16:04 djbrooke joined #dataverse
16:17 djbrooke joined #dataverse
16:33 pameyer oddness - integration tests pass, trying to create a dataset with the native API fails
16:36 pdurbin pameyer: on "develop"? d77cb88?
16:38 pameyer d77cb88321f6910cfeae889203f6b0225e59f9ac
16:38 pameyer aka - yeah
16:38 djbrooke joined #dataverse
16:39 pameyer saw it initially w\ docker-dcm; wiped things out and saw it without the DCM configs
16:39 pameyer I'm still testing w\ ezid; and first pass at the logs suggests its sending a bad target url
16:40 pameyer still investigating though
16:40 pdurbin Ok. I haven't built that commit on the phoenix server. I just kicked off a build. Thanks for the heads up.
16:43 pameyer integration tests, which presumably test creating a dataset with the native API, are passing for me
16:50 djbrooke joined #dataverse
16:52 pdurbin pameyer: integration tests are passing on phoenix as well: https://build.hmdc.harvard.edu:8443/job/phoenix.dataverse.org-apitest-develop/223/
16:53 pameyer so that's consistent
16:54 pdurbin pameyer: you are welcome to try to create a dataset via native api on the phoenix server if you like
16:56 pameyer pdurbin: usual test API key?
16:57 pameyer or do I need to setup an account first?
16:57 pdurbin you'd need to create an account
16:58 pdurbin there is no usual API key. new keys are created on every build. dropping the database happens early
16:59 pameyer thanks - I may tried that depending on what the debugger tells me
17:00 djbrooke joined #dataverse
17:00 pdurbin ok
17:00 pameyer ahh - `https://ce470b6229d4/citation?persistentId=doi:10.5072/FK2/EHXOLK`
17:03 pdurbin what is https://ce470b6229d4 ?
17:04 pameyer that's the container internal hostname
17:05 pameyer that puzzles me less than the `citation` in the target URL
17:05 pdurbin I'm completely puzzled now.
17:07 pameyer I'm pretty confused too, and starting to think that I won't get it sorted before lunch.
17:07 pameyer integration tests passing suggest that target URLs for DOIs are valid from within docker
17:08 pdurbin I thought you couldn't create a dataset via native api.
17:08 pameyer non-interactive install in docker shows invalid target URLs being created (resulting in a dataset creation failure), and somehow "citation" gets inserted into the target url instead of dataset.xhtml
17:08 pameyer pdurbin: *I* can't, but the *integration tests* can
17:09 pameyer so yeah; puzzeling
17:09 pdurbin ¯\_(ツ)_/¯
17:11 pameyer http://phoenix.dataverse.org/dataset.xhtml?persistentId=doi:10.5072/FK2/Q1516N&version=DRAFT
17:12 pameyer hmm.... setupIT vs normal installation; and DataCite vs EZID
17:13 djbrooke joined #dataverse
17:15 pdurbin What's a target URL?
17:15 djbrooke joined #dataverse
17:16 pameyer what the DOI resolves to
17:16 pdurbin ah, thanks
17:16 pameyer aka - `curl -I https://doi.org/$doi` shows a redirect to that URL
17:18 pameyer still puzzled; but leaning towards the site URL
17:18 pdurbin man, I'm gonna start using -I instead of -i. thanks
17:19 pdurbin I don't think I control any servers that create real target URLs. Both of mine are test servers.
17:20 djbrooke joined #dataverse
17:20 pdurbin test DOIs with FK2 shoulders and all that
17:22 pameyer ezid test has the target you give it, datacite test dois always resolve to the same identifier
17:22 djbrooke joined #dataverse
17:24 pameyer ... and yeah, will not get un-confused before lunch
17:24 pdurbin yeah https://doi.org/10.5072/FK2/IT02CG redirects to http://www.datacite.org/testprefix so I can't see if "citation" is being added or not.
17:25 pameyer > stop in edu.harvard.iq.dataverse.DOIEZ​IdServiceBean.createIdentifier Set breakpoint edu.harvard.iq.dataverse.DOIEZ​IdServiceBean.createIdentifier >  Breakpoint hit: "thread=jk-connector(4)", edu.harvard.iq.dataverse.DOIEZId​ServiceBean.createIdentifier(), line=249 bci=0 249            logger.log(Level.FINE, "createIdentifier");  jk-connector(4)[1] next >  Step completed: "thread=jk-connector(4)", edu.harvard.iq.dataverse.DOIEZIdServiceBean.cr
17:26 pameyer that had newlines before I pasted it :(
17:28 djbrooke joined #dataverse
17:31 pdurbin Is it a problem that "citation" is in the target URL?
17:38 djbrooke joined #dataverse
17:41 djbrooke joined #dataverse
17:42 djbrooke joined #dataverse
17:56 djbrooke joined #dataverse
17:59 pameyer not sure; leaning towards not the source of the problem
18:03 pdurbin As I mentioned in Slack, it's been there since Dataverse 4.4: https://github.com/IQSS/dataverse/commit/96c8bdfb
18:03 pameyer yeah - thanks
18:04 pameyer I'd thought it was there to facilitate resolving file DOIs vs dataset DOIs, but it doesn't appear to be doing that
18:04 djbrooke joined #dataverse
18:05 pdurbin It's hard to tell. The pull request is linked to many issues.
18:08 pdurbin pameyer: I'm using the DCM mock, by the way. Seems to work fine for my needs so far.
18:10 pameyer pdurbin: good to hear.  and siteUrl was the problem
18:12 pameyer joined #dataverse
18:13 pdurbin pameyer: did you see https://github.com/IQSS/dataverse/pull/4887/files ? :)
18:14 pameyer one of the reasons why it should've occured to me sooner
18:23 djbrooke joined #dataverse
18:27 djbrooke joined #dataverse
18:55 djbrooke joined #dataverse
19:13 djbrooke joined #dataverse
19:51 djbrooke joined #dataverse
19:56 pdurbin pameyer: oh, do you feel like exposing that port in 4940 while you're in there?
19:57 pameyer what's that one do?
19:58 pdurbin it's docker-dcm so related. somewhat :)
19:59 pameyer ?
19:59 pdurbin nevermind
20:00 pdurbin if i ever try to use docker-dcm maybe I'll expose the port like you advised above. for now I'm using the mock
20:00 pameyer whatever works
20:00 pdurbin yeah
20:01 pameyer I'll probably be back into docker-dcm at somepoint, but pretty much done for today
20:05 pdurbin ok, no worries
20:08 djbrooke joined #dataverse
20:12 djbrooke joined #dataverse
20:20 djbrooke joined #dataverse
20:41 djbrooke joined #dataverse
21:35 djbrooke joined #dataverse
21:36 djbrooke_ joined #dataverse
21:42 djbrooke joined #dataverse
21:47 djbrooke_ joined #dataverse
21:48 djbrooke joined #dataverse
22:04 pameyer was taking a quick look at https://github.com/IQSS/dataverse-ansible/issues/20; figured I'd try re-running that role in vagrant
22:04 pameyer `Box 'sjoeboo/centos-7-1-x86-ansible' could not be found` :(
22:54 pameyer left #dataverse

| Channels | #dataverse index | Today | | Search | Google Search | Plain-Text | plain, newest first | summary

Connect via chat.dataverse.org to discuss Dataverse (dataverse.org, an open source web application for sharing, citing, analyzing, and preserving research data) with users and developers.