Time
S
Nick
Message
00:20 axfelix joined #dataverse
02:14 metamattj joined #dataverse
03:14 axfelix joined #dataverse
03:59 garnett joined #dataverse
06:01 bencomp1 joined #dataverse
07:08 jri joined #dataverse
07:42 bencomp joined #dataverse
13:06 bencomp joined #dataverse
13:07 bencomp
pdurbin: what is the passcoe
13:07 bencomp
?
13:18 * bencomp
is in the call
13:24 pdurbin
bencomp: http://dataverse.org/community-calls
13:25 bencomp
I hoped this was faster, as I had restarted my computer and Firefox and outlook had yet to start up
13:27 iamtimmo
pdurbin: Want to just give me edit access?
13:27 pdurbin
iamtimmo: I just did. Tried to.
13:28 iamtimmo
yep. worked. I had to reload to see it.
13:28 pdurbin
ah
13:28 pdurbin
iamtimmo: thanks so much for taking notes!
13:28 iamtimmo
thanks. and np
13:29 iamtimmo
pdurbin: I’m not a great note taker, but at least there’s something there ;-0
13:29 iamtimmo
;-)
13:40 pameyer joined #dataverse
13:59 pdurbin
I'm always glad to have *some* notes at least.
13:59 pdurbin
here they are for anyone who's interested: https://docs.google.com/document/d/1TOPTG6PEKfL-bh7R2EbdVhNyFS3aDFzkYMD9l7wSfVY/edit?usp=sharing
14:02 iamtimmo
bencomp: Thanks for the illuminating comment here: https://docs.google.com/document/d/1idh4wvCcoxSo7rPkolOkTD5Wvn6PQO__uXqkf7eQujA/edit?disco=AAAAAmyWJ9w&usp=comment_email_document
14:03 bencomp
iamtimmo: np :) just hope that you can do without me in a few months, when I start my next job ;)
14:08 iamtimmo
bencomp: I will definitely miss you!
14:09 pdurbin
we all will. don't forget us, bencomp!
14:10 bencomp
awww
14:10 bencomp
of course I won't
14:11 bencomp
:)
14:24 pdurbin
iamtimmo: it seems like that tsv doc is coming along
14:25 bmckinney joined #dataverse
14:36 iamtimmo
pdurbin: Yes. Still need to get Gustavo to take a pass at it, since he seems to be the expert on this. Can you help/nudge?
14:37 iamtimmo
pdurbin: After loading a new mdblock, is it necessary to trigger reindex?
14:38 pdurbin
sigh
14:38 pdurbin
That's the part that I'm the expert on... you need to create a new Solr schema.
14:38 iamtimmo
A-ha!
14:38 pdurbin
bmckinney: are you still interested in documenting the workflow of adding a new metadata block?
14:40 pdurbin
iamtimmo: "We update the Solr schema.xml file by..." https://github.com/IQSS/dataverse/blob/v4.3/src/main/java/edu/harvard/iq/dataverse/search/SearchFields.java
14:41 majest1c joined #dataverse
14:41 pdurbin
"We use the output of this method to generate our Solr schema.xml" https://github.com/IQSS/dataverse/blob/v4.3/src/main/java/edu/harvard/iq/dataverse/api/Index.java#L454
14:42 pdurbin
that's where I check for uniqueness... but you have to load up *all* the metadata blocks
14:43 iamtimmo
pdurbin: I just added a warning line to the mdblock doc. Do you have an instructions or scripts that would guide us through this process?
14:44 pdurbin
not really. but bmckinney expressed interest in writing down the steps, I think
14:45 pdurbin
after a new metadata block is added and I do my thing, I add a commit to the branch that looks something like this: new Solr schema.xml for HBGDki Custom Metadata #2646 · IQSS/dataversehttps://github.com/IQSS/dataverse/commit/d1d3705a08e440165f23d87f17b491164b9b4fe7
14:46 axfelix joined #dataverse
14:55 iamtimmo
pdurbin: Thanks. That’s helpful. Maybe I can effect a workaround while the write-up is forthcoming.
14:56 iamtimmo
pdurbin: Is it possible to back out a metadata block once it has been loaded, but before any content references it?
14:57 pdurbin
iamtimmo: it's a point of no return, I think ;)
14:57 iamtimmo
okay. thanks.
14:58 pdurbin
that's why I think there should be a "preview" mode: https://github.com/IQSS/dataverse/issues/2551
14:58 bmckinney
pdurbin: i might be able to contribute something but i'm hopeful that a new json-schema driven version would make it obsolete
14:59 pdurbin
bmckinney: I'd get an estimate on when the json thing would be ready. Is it a year or more out?
14:59 bmckinney
i have some new marching orders so i'm not sure how much time i can spend on it
15:01 bmckinney
pdurbin: this morning i've been trying to discover how to expose custom metadata blocks via the datasets API (only citation is given currently)
15:04 pdurbin
bmckinney: only citation? That ain't right. All metadata blocks should be exposed in the JSON . It's supposed to be complete. Lossless.
15:09 bmckinney
pdurbin: i stand corrected, it is coming through - thanks!
15:10 pdurbin
oh sure
15:10 pdurbin
I'm glad *something* works. ;)
15:17 majestic joined #dataverse
15:20 bmckinney51 joined #dataverse
15:20 bmckinney51 left #dataverse
15:23 bmckinney joined #dataverse
17:21 metamattj joined #dataverse
17:56 bmckinney joined #dataverse
18:30 bencomp joined #dataverse
19:11 bmckinney
pdurbin: hey, do you know if you can specify the dataverseType (aka category) when creating a dataverse with json via api?
19:16 pdurbin
bmckinney: https://github.com/IQSS/dataverse/blob/v4.3/src/test/java/edu/harvard/iq/dataverse/api/SearchIT.java#L543
19:18 bmckinney
pdurbin: got it, thanks
19:18 pdurbin
which it to say that when I wrote that test it's a GUI -only feature :(
19:19 pdurbin
this proposal of mine to allow it via API never got merged: https://github.com/IQSS/dataverse/commit/5d5f60e36a51613a6edf3976f5b3c9e4ffe605a3
19:21 pdurbin
here's where it gets set to UNCATEGORIZED if null: https://github.com/IQSS/dataverse/blob/v4.3/src/main/java/edu/harvard/iq/dataverse/engine/command/impl/CreateDataverseCommand.java#L68
19:21 pdurbin
and it's always null via API
19:23 metamattj joined #dataverse
19:27 bmckinney
pdurbin: thanks, i may need to cherry-pick that for our migration. setting types for 50+ labs via the UI is a bit silly
19:30 pdurbin
amen
19:31 pdurbin
bmckinney: you should probably create a GitHub issue for it. It's a real issue.
23:08 metamattj joined #dataverse
