Time
S
Nick
Message
10:42 pdurbin
bencomp: wow! https://github.com/bencomp/dataverse-docker ! :)
10:46 bencomp
pdurbin: I don't expect to actually do a lot to make it work soon
10:47 bencomp
just pushed some file stubs
10:51 pdurbin
bencomp: no problem. but I hope you don't mind if I send Docker enthusiasts your way :)
10:52 bencomp
echo "pull requests very welcome" > CONTRIBUTING.md
10:56 pdurbin
:)
10:58 pdurbin
oh, another benefit of having the default branch be the one we are working on is that when we merge in pull requests people will show up under "contributors" right away
10:59 bencomp
is that a reference to my pull requests that you are about to merge? :)
11:54 bencomp
YAPR: https://github.com/IQSS/dataverse/pull/2425
12:05 pdurbin
heh
12:05 pdurbin
wow, 5 open pull requests now
12:26 pdurbin
it's great
12:26 pdurbin
I feel responsible for this one so I just left a comment: https://github.com/IQSS/dataverse/pull/2407#issuecomment-126674732
12:26 pdurbin
bencomp: mentioning you :)
13:39 pdurbin
bencomp: what would be a good URL to use for publishing code coverage reports?
13:39 pdurbin
for now I said "some server" at https://github.com/IQSS/dataverse/issues/775#issuecomment-126687225
14:37 bencomp
pdurbin: I have no clue what a good URL would be. Maybe IQSS can setup a special OpenScholar instance for it?
14:38 pdurbin
bencomp: hmm. I'm thinking just a static site. dump the files in a directory. a new directory for every day maybe. could run it nightly
14:40 bencomp
I was joking
14:42 bencomp
pdurbin: how about dev.dataverse.org
14:43 bencomp
or guides.dataverse.org (less preferred)
14:44 pdurbin
guides I think of as versioned per release. and the rst files get committed into git. this is different. it's dynamic output (reports) generated by commits
14:50 pdurbin
"dev" might work. or "code" or "software"
14:53 bencomp
please discuss this with Gustavo :)
14:53 pdurbin
:)
14:54 pdurbin
this is making me write more tests for edu.harvard.iq.dataverse.search :)
15:02 pdurbin
ok, a start: stub out SolrSearchResultTest · IQSS/dataversehttps://github.com/IQSS/dataverse/commit/3372c77
15:16 axfelix joined #dataverse
15:22 bencomp
Great. Now could you add some documentation to this test? What does it mean for this test to pass?
15:23 bencomp
not trying to sound grumpy, I guess today I'm just a bit grumpy :$
15:34 pdurbin
bencomp: well, my immediate goal was to have the total code coverage here be higher than 0%: http://dvn-vm1.hmdc.harvard.edu/tmp/jacoco/2015-07-31/jacoco/edu.harvard.iq.dataverse.search/SolrSearchResult.html :)
15:34 pdurbin
which is is now. that report was from before the commit
15:36 bencomp
so there is a place to put the reports now?
15:37 pdurbin
not yet, but I sent and email asking if there could be a place
15:47 pdurbin
an*
15:48 pdurbin
bencomp: I left a comment too but I'm not sure if this is the right place: https://docs.google.com/document/d/1Qh-C7dVhy7RqLW2jOJyQeJEL7yH8hjjc3mY-8VNzRYM/edit?usp=sharing . I'm not very involved in this.
15:55 bencomp
pdurbin: I can't see your comment, but I'm sure it's there ;)
15:56 pdurbin
bencomp: oh, it's under "Software"
15:56 pdurbin
I proposed a couple hostnames under dataverse.org. subdomains
15:57 bencomp
ok
16:03 pdurbin
bencomp: some recent comments about Puppet to check out: https://github.com/IQSS/dataverse/pull/2407
16:05 bencomp
I get the notifications
16:06 bencomp
I had wondered about a GitHub org "dataverse community" or something like that
16:06 bencomp
(to myself)
16:08 pdurbin
oh, a separate org? hmm. like https://github.com/openscholar . I think we have a preference to keep everything under https://github.com/IQSS . I'm planning on moving anything important off of https://github.com/dvn
16:09 bencomp
well, up to you. I thought maybe a different org makes clear it doesn't come from IQSS
16:10 pdurbin
oh, I see. hmm
16:10 pdurbin
do other projects on github have this setup?
16:11 pdurbin
certainly when you unpack tarballs you'll sometimes find a "contrib" folder
16:12 pdurbin
bencomp: same offer to you if you want a dataverse-docker repo under the IQSS org!
16:13 bencomp
fc-repo-labs, DANS-labs
16:13 bencomp
I won't be able to maintain the Docker stuff
16:19 pdurbin
bencomp: ok, no worries. but you want to link to your repo from the Installation Guide?
16:19 metamattj joined #dataverse
16:19 lwo joined #dataverse
16:20 * pdurbin
eyes lwo
16:20 lwo
Hello all
16:20 pdurbin
here comes trouble
16:20 pdurbin
:)
16:20 lwo
Where ?
16:20 pdurbin
heh
16:20 lwo
How is your cold ?
16:20 pdurbin
hee hee!
16:20 pdurbin
my (cough) lovely shell scripts
16:21 lwo
Yes they are..... A good job.....
16:21 pdurbin
heh. A for effort ;)
16:21 pdurbin
maybe a C+
16:21 lwo
And if where weren't any ideosyncratic shell scripts, you would not have generic puppet solutions replacing them.
16:22 pdurbin
yeah
16:22 lwo
Anyway... what you propose about the repo looks very reasonable.
16:23 pdurbin
ok! great! it's worked out very well with Center for Open Science folks (/me looks at jeffspies_ and LyndsySimon)
16:23 lwo
Indeed
16:23 pdurbin
lwo: you're fine with the repo being called dataverse-puppet?
16:24 bencomp
pdurbin: maybe IQSS/dataverse-docker is better after all, at least for linking from the guides
16:25 lwo
A typical puppet module is named AUTHOR-NAMEOFMODULE... so it would be more appropriate to name it likewise in the repo: IQSS-dataverse for example.
16:25 lwo
But on second thougts lets not over complicate things now... renaming can be done later.....
16:25 lwo
Your proposal is fine I think
16:26 lwo
But my college gordan-cupac also works on puppet\Cont. Integration issues .... he should also be able to push.
16:26 pdurbin
ok, so https://github.com/gordan-cupac
16:26 lwo
YEs... you recognize him by his characteristics
16:26 lwo
https://github.com/gordan-cupac
16:26 lwo
Match
16:27 lwo
Good.
16:27 pdurbin
lwo: hmm GPL at https://github.com/IISH/iish-iqss/blob/v0.1.0/LICENSE but Apache at https://github.com/IQSS/dataverse/blob/4.2/LICENSE.md . Are you ok with Apache?
16:27 pdurbin
bencomp: same question, since you went with GPL .
16:28 lwo
We do use GPL as a standard.... let me check if Apache is ok.
16:28 pdurbin
this one is apache too: https://github.com/IQSS/dataverse-client-python/blob/master/LICENSE
16:28 pdurbin
I am not a lawyer.
16:29 bencomp
Apache is fine
16:30 pdurbin
bencomp: ok, so you want a repo under IQSS too? Even if you're concerned you won't have time to maintain it?
16:32 pdurbin
bencomp: if you want it could just be an issue for now. It's really up to you.
16:32 bencomp
I'll get back to it later then
16:37 pdurbin
lwo: I just checked and you are welcome to use whatever license you wish
16:37 lwo
Tell them thank you
16:37 pdurbin
we do prefer Apache for the main Dataverse code since it's more flexible
16:38 lwo
Then it is the GPL 3
16:38 lwo
Flexible, in what way ?
16:38 lwo
Flexible, in what way ? (As a side note )
16:39 pdurbin
oh, uh
16:39 pdurbin
people are more free to do with the code what they wish, I guess
16:40 lwo
I though GPL3 was already quite free... I'll guess I'll do some reading during my holiday
16:40 lwo
http://www.apache.org/licenses/GPL-compatibility.html
16:41 pdurbin
lwo: ok, empty repo created: https://github.com/IQSS/dataverse-puppet
16:41 pdurbin
git remote add origin git
16:41 lwo
Lets see....
16:41 pdurbin
to push your code there
16:52 lwo
Do we have PUSH permissions ?
16:58 lwo
@pdurbin: I am afraid I am out of time. My holiday will begin...
16:59 bencomp
lwo: lekkah! :)
16:59 lwo
Ah Ben.... still working ?
16:59 bencomp
yup
16:59 lwo
Yes a fine rest for me
16:59 lwo
Finally time to do the things I want to research
17:00 lwo
But I will be seeing you the 12th
17:00 bencomp
enjoy! (just had mine)
17:00 bencomp
indeed, I look forward to our meeting
17:00 lwo
Or what is the 15th ?
17:00 bencomp
I think 12
17:00 bencomp
yes, 12 in my calendar
17:01 lwo
Excellent
17:01 lwo
Phill: thanks for the join link.
17:01 lwo
I am signing of noww
17:01 lwo
See you all !
17:01 pdurbin
lwo: bye! thanks!
17:01 lwo left #dataverse
17:09 pdurbin
bencomp: wow. long email you sent me. thanks! is it ok for me to forward it to our internal dataverse list?
17:10 bencomp
pdurbin: could you wait until I confirmed some things with Peter and Marnix next week?
17:10 pdurbin
bencomp: absolutely. thank for your thoughts either way!
17:11 bencomp
these are my ramblings, there will be higher-level emails
17:11 pdurbin
ok
17:11 pdurbin
:)
17:35 bencomp
pdurbin: that second email that I just sent, should I have sent it to the community list?
17:36 pdurbin
I haven't gotten it yet.
17:38 pdurbin
maybe if I go for a quick walk it will be there
17:55 pdurbin
bencomp: still nothing. can you please send it again?
17:57 bencomp
pdurbin: I had put you in CC in a mail to Gustavo
17:57 bencomp
but it turns out Outlook (@#$@$%@) used notifications
17:58 bencomp
even though that is not in my address book
17:59 pdurbin
oh
18:00 pdurbin
bencomp: oh good, you sent it to Gustavo too
18:01 bencomp
this was something I specifically discussed with him
18:01 pdurbin
gotcha. yeah, he handled the migration. I only know about it at a high level
18:02 pdurbin
bencomp: but the password stuff will probably be different for you because you're already using Shibboleth
18:03 pdurbin
but I see this is in your email
18:03 pdurbin
also...
18:03 bencomp
:)
18:03 pdurbin
"Please note that you will have to enter a new password when logging in for the first time due to stricter password requirements." Dataverse 4.0 is here! - Google Groups - https://groups.google.com/forum/#!topic/dataverse-community/J1djJwY_7bc
18:04 pdurbin
that's how we worded it
18:04 pdurbin
that was for Update password encryption to use BCrypt · Issue #1034 · IQSS/dataverse - https://github.com/IQSS/dataverse/issues/1034
18:06 bencomp
can you (make Gustavo) stick it in an email? :)
18:10 pdurbin
I can ask that when he replies all that he fix my email address. :)
18:48 pdurbin
which he did :)
18:48 pdurbin
bencomp: I'm suggesting architecture diagrams that look like http://plantuml.com/imgp/component_004.png based on the text at http://plantuml.com/component.html#Component which could be put under version control
19:02 mjturk joined #dataverse
19:07 pdurbin
MVP: https://github.com/IQSS/dataverse/commit/2ed629b :)
20:24 garnett joined #dataverse
21:33 axfelix joined #dataverse
23:15 cnk joined #dataverse
23:19 pdurbin
cnk: welcome! nice chatting with you at https://botbot.me/freenode/django/msg/46103134/ !
23:20 pdurbin
we have Django apps too: https://github.com/IQSS/milestone-reader :)
23:21 cnk
LOL I am not that much of a Django / Python proponenet - but the problem domain seems pretty interesting
23:21 cnk
and I like your example code for writing clients
23:22 pdurbin
oh? sorry, which code
23:22 metamattj joined #dataverse
23:22 cnk
dataverse-client-python
23:22 cnk
Python library for writing clients that use APIs from Dataverse
23:23 pdurbin
oh yes, great stuff that. contributed. love it
23:23 cnk
hmmm and I may have to dive into your selenium helper repo. I have been avoiding it like the plague but may have to use ti for integration testing
23:24 pdurbin
shout out to Robert at http://guides.dataverse.org/en/latest/api/client-libraries.html
23:25 pdurbin
I've been meaning to dive into it too. Trying to test as much as I can via APIs first though. It's easier than driving a browser. :)
23:25 pdurbin
and in practice, I just find that I add more API endpoints if there's something that is previously untestable via API :)
23:25 pdurbin
so learning selenium well is on the back burner for me
23:27 cnk
hmmmmm I am just getting to the point of testing my API stuff. This has been an exploritory week so am doing test last instead of test first. But if I can get my last serialization thing working, time to start wrigint some tests
23:28 cnk
What is the spread of fields that have deposites into a Dataverse repo?
23:29 pdurbin
cnk: well, here are the metadata schemas we currently support: http://guides.dataverse.org/en/latest/user/appendix.html
23:30 pdurbin
the idea is that we can easily add more though: https://github.com/IQSS/dataverse/tree/master/scripts/api/data/metadatablocks
23:30 cnk
Thank you. More and more, I think I need to look into this further
23:31 pdurbin
oh sure. we just had our first community meeting, which was pretty awesome: http://projects.iq.harvard.edu/dataverse-community-meeting
23:41 pdurbin
cnk: but yeah, I don't often do test first. I'm just happy to write tests
23:41 pdurbin
bencomp: thanks again for the encouragement :)
