Time
S
Nick
Message
01:41 axfelix joined #dataverse
01:44 pdurbin
nice post: http://savageminds.org/2015/04/20/dataverse-an-open-source-solution-for-data-sharing/
01:49 axfelix joined #dataverse
01:51 axfelix joined #dataverse
13:11 rliebz joined #dataverse
13:51 rliebz
pdurbin: Is there a way to get the full citation of a dataset programatically? We'd like to prefer published citations to draft citations, but I believe SWORD always returns the latest version
13:51 pdurbin
rliebz: you can get it from the Search API
13:52 pdurbin
(you're right that SWORD does always return the latest version)
13:56 rliebz
pdurbin: I'll look into that, thanks
13:58 pdurbin
rliebz: sure. it sounds like something we'd want the "datasets" endpoint of the "native" API to be able to do so please feel free to make a ticket
14:01 pdurbin
rliebz: did someone from COS create this? https://dataverse.harvard.edu/dataverse/sitetesting1
14:01 rliebz
pdurbin: Not sure. I messaged the testing team earlier, but no one has claimed it yet.
14:02 pdurbin
ok. I wish there was an easy way for me to tell who created it
14:02 pdurbin
without digging into the production database I mean
14:03 rliebz
pdurbin: Did they not use a real email?
14:03 pdurbin
well, we don't expose people's email addresses
14:03 pdurbin
hmm, there's a viewDatavese method...
14:04 pdurbin
viewDataverse*
14:04 rliebz
At the very least there is an "email dataverse contact" button
14:05 pdurbin
rliebz: yeah, it was one of your guys. I can tell from this: curl "https://dataverse.harvard.edu/api/dataverses/sitetesting1/?key=$PDURBINKEY "
14:06 pdurbin
give that curl command a try if you like
14:07 rliebz
pdurbin: That makes it easy. I'll message him
14:08 pdurbin
rliebz: I added him to cc on the email thread
14:10 pdurbin
rliebz: I can upload a file to Dataverse via OSF!
14:11 rliebz
pdurbin: Great news!
14:11 pdurbin
rliebz: sounds like you made a config change rather than a code change
14:13 pdurbin
rliebz: oh, and yesterday we patched Weld to the latest (2.2.10) on https://apitest.dataverse.org so pleas do notify us right away if you see any more scrambled sessions!
14:13 rliebz
pdurbin: Yep. Right now, we have the Dataverse host specified once for our OSF server and once for waterbutler (which does files)
14:13 pdurbin
searchbot: lucky waterbutler python
14:13 searchbot
pdurbin: https://github.com/CenterForOpenScience/waterbutler
14:14 pdurbin
"WaterButler is a Python web application for interacting with various file storage services via a single RESTful API , developed at The Center for Open Science." https://waterbutler.readthedocs.org/en/latest/
14:14 pdurbin
neat!
14:14 pdurbin
rliebz: any plans to add Dataverse to the list at https://waterbutler.readthedocs.org/en/latest/providers.html ?
14:16 pdurbin
:)
14:18 rliebz
pdurbin: We sure can. I'm not sure if you noticed, but the documentation is a bit incomplete
14:20 pdurbin
great!
14:21 pdurbin
rliebz: hey so, are you still using https://github.com/swordapp/python-client-sword2 ?
14:23 rliebz
pdurbin: We are not.
14:23 pdurbin
gotcha. that's fine. just curious. I still link to it from http://guides.dataverse.org/en/latest/api/sword.html#client-libraries
14:24 pdurbin
rliebz: am I right in thinking that the OSF Dataverse plugin used to be able to create a dataset on the Dataverse side?
14:25 rliebz
pdurbin: We never exposed dataset creation on the OSF plugin—just in the python client
14:25 pdurbin
rliebz: ah, ok. I haven't really played with this addon much.
14:26 pdurbin
and I don't want to publish a fake dataset but I'll assume publishing works :)
14:26 pdurbin
rliebz: any future development planned for this addon? if so, what's the best way to track it? search for "dataverse" in the OSF issue tracker?
14:27 rliebz
pdurbin: The only difference in publishing functionality is that the OSF offers to publish your Dataverse as well if it is unpublished
14:27 pdurbin
rliebz: ah, that's a nice feature
14:28 rliebz
pdurbin: I noticed that's what the Dataverse does, so I thought we should include it too
14:28 pdurbin
yep! great stuff
14:29 rliebz
pdurbin: I think we're looking into allowing different Dataverse hosts (at least evaluating it in the next month), but feel free to open issues for any functionality you want to see
14:29 rliebz
pdurbin: Searching dataverse is probably the best way to see what's going on
14:29 pdurbin
sounds good!
14:30 pdurbin
rliebz: I guess I could also ask a bit on the OSF mailing list
14:30 pdurbin
maybe once things settle down a bit here :)
15:23 rliebz joined #dataverse
15:26 rliebz
pdurbin: So we're running into an issue with pre-existing datasets. If they don't have all of the newly-required fields (e.g., subject), you can't upload files
15:27 rliebz
pdurbin: The error message: "Couldn\'t update dataset edu.harvard.iq.dataverse.engine.command.exception.IllegalCommandException: Validation failed: Subject is required."
16:34 pdurbin
rliebz: ah, cool, you made an issue: Pre-existing datasets cannot be uploaded to through the API · Issue #2062 · IQSS/dataverse - https://github.com/IQSS/dataverse/issues/2062
16:34 pdurbin
thanks!
16:42 pdurbin
rliebz: I just left a comment
16:54 metamattj joined #dataverse
17:04 rliebz
pdurbin: Thanks!
17:05 rliebz
pdurbin: Will it be the case that we will have to fill in contact email as well? I believe that there was a dataset where this was the case
17:06 pdurbin
rliebz: yes, it's called "datasetContactEmail": https://github.com/IQSS/dataverse/blob/master/scripts/search/tests/data/dataset-finch1.json
17:09 rliebz
pdurbin: I suppose this is probably an odd request, but is there a way for me to generate datasets without subjects or emails for testing purposes?
17:12 pdurbin
rliebz: for "Subject" anyway, can you try leaving dcterms:subject out of the XML when creating a dataset via SWORD? that should do it, I think
17:17 rliebz
pdurbin: Looks like that worked
17:17 pdurbin
sweet
17:43 rliebz
pdurbin: Looking into the solution you recommended—do you have an example of what the metadata should look like?
17:44 rliebz
pdurbin: I'm pulling from http://$SERVER/api/datasets/$id/versions/$versionId/metadata?key=$apiKey and trying to use that, but getting 500s
17:45 pdurbin
rliebz: you might need to wait a few minutes. One of our servers went down :(
17:55 rliebz
pdurbin: I don't think what I'm seeing is server related
17:56 rliebz
pdurbin: Well, server being down related, I mean
17:57 pdurbin
oh
17:58 pdurbin
rliebz: oh, duh. you're using apitest, right?
17:58 rliebz
pdurbin: Yep
17:59 rliebz
pdurbin: Just found https://github.com/IQSS/dataverse/blob/master/scripts/search/tests/data/dataset-finch1.json
17:59 pdurbin
rliebz: I linked to that a bit ago :) ^^
18:01 rliebz
pdurbin: Last time was for Dataverses, I think. But I used that as a starting point!
18:01 pdurbin
rliebz: try this: https://apitest.dataverse.org/api/datasets/11/versions/:draft?key=8f73354d-35d8-4d41-9323-3ed30bc44afa
19:55 bshambaugh joined #dataverse
21:23 bshambaugh joined #dataverse
