Time
S
Nick
Message
02:19 axfelix joined #dataverse
02:23 garnett joined #dataverse
02:52 garnett joined #dataverse
04:10 metamattj joined #dataverse
05:43 metamattj joined #dataverse
10:00 bencomp joined #dataverse
12:56 rliebz joined #dataverse
13:14 pdurbin
rliebz: yesterday I noticed this. awesome: Feature/select dataverse by rliebz · Pull Request #2700 · CenterForOpenScience/osf.io - https://github.com/CenterForOpenScience/osf.io/pull/2700
13:17 rliebz
pdurbin: Yep! Responding to code review today
13:18 rliebz
pdurbin: Right now we have dataverse.harvard.edu, dataverse-demo.iq.harvard.edu, and apitest.dataverse.org as pre-populated Dataverses repositories to select from. Are there any other that have been upgraded to 4.0?
13:19 * pdurbin
thinks
13:19 pdurbin
rliebz: I can think of one more but I might take it down or change the hostname. that's a good list
13:21 pdurbin
rliebz: for testing, apitest is the best. sometimes people give demos using dataverse-demo and they get bothered by weird data
13:21 rliebz
pdurbin: Awesome. Users can input their own, but when other Dataverses like Odum become compatible, we can add those too.
13:22 rliebz
pdurbin: I've documented that, but I'll make sure to communicate it to QA
13:22 pdurbin
awesome
15:04 axfelix joined #dataverse
15:11 axfelix joined #dataverse
15:51 metamattj joined #dataverse
17:23 rliebz joined #dataverse
18:26 rliebz
pdurbin: What exactly is `key` supposed to be for creating a built-in user?
18:27 rliebz
pdurbin: Right now I'm just passing in "burrito" since that's what the tests do, which seems to work, but I'd be more comfortable using it if I knew what it was
18:29 pdurbin
rliebz: heh. I *just* sent an email about this internally.
18:29 pdurbin
I think it's a little weird.
18:29 pdurbin
I dunno.
18:29 pdurbin
maybe it's fine
18:29 pdurbin
rliebz: do you plan to implement user creation via API ?
18:30 pdurbin
I know we talked about it here: https://github.com/IQSS/dataverse-client-python/issues/10#issuecomment-97798073
18:31 rliebz
pdurbin: I've got some spare time today while I wait for code review, so I was going to try to hash together user creation for automated testing
18:31 pdurbin
cool. as I've said, that's what I do (and then delete the user at the end)
18:31 pdurbin
(have to delete all their stuff first)
18:32 rliebz
pdurbin: I'm not really sure how to go about interfacing it in the client, since the `Connection` object that we start with requires an existing user. So just implementing it in tests for now
18:32 rliebz
pdurbin: Will the key always be burrito, or is it just a nonsense field?
18:32 pdurbin
rliebz: do you mean in production or apitest?
18:33 rliebz
I'd like the tests to be able to run on apitest, dataverse-demo, or any other non-production server
18:34 pdurbin
rliebz: it'll be "burrito" on apitest :)
18:35 rliebz
pdurbin: And on dataverse-demo?
18:35 pdurbin
rliebz: I kind of doubt anyone changed it but I have less control over that server.
18:35 pdurbin
I'm super excited about the automated testing!
18:36 pdurbin
Hmm, I haven't done a build on apitest for a while. Maybe I should.
18:38 rliebz
pdurbin: I'm wondering if it's better to have separate sets of tests for the python client. One set that tests that it's sending the requests we expect it to (a test of the client itself using mocking), and a separate set that can test against a server
18:41 pdurbin
rliebz: I'm all for mocking. We'd have both, ideally.
20:39 rliebz
pdurbin: Trying to get some Travis tests running. When you have a second, could you hop on https://travis-ci.org/profile/IQSS and enable IQSS/dataverse-client-python
21:14 rliebz
pdurbin: I'm off. Since it's my last day at COS, I probably won't be on IRC much, but feel free to email me when you've set up Travis, or if you need anything else from me
21:15 rliebz
pdurbin: https://github.com/IQSS/dataverse-client-python/pull/23 solves a few issues, and is just waiting on setting up Travis for verification
21:43 garnett joined #dataverse
23:35 axfelix joined #dataverse
23:56 pdurbin
shoot, I missed him. thanks, Robert!
23:58 * pdurbin
had wandered off to this talk: ABCD WWW » Jeff Winkler – React, a JavaScript library for building user interfaces - http://w3.abcd.harvard.edu/ai1ec_event/jeff-winkler-reactjs/
23:58 pdurbin
which was fantastic
