Time
S
Nick
Message
07:25 jri joined #dataverse
09:56 jri joined #dataverse
12:36 cdsp-rmo joined #dataverse
13:09 cdsp-rmo joined #dataverse
14:03 pameyer joined #dataverse
14:09 donsizemore joined #dataverse
14:16 donsizemore
@pdurbin knock knock?
14:18 pdurbin
who's there?
14:19 donsizemore
@pdurbin don and akio. he's asking if we know of an external tool which accesses either multiple datasets or cross-dataset
14:19 donsizemore
@pdurbin i was looking at the binderverse stuff but it still wants a github repo
14:20 pdurbin
external tools can't even operate on a single dataset. external tools only operate on a single tabular file
14:21 pameyer
*yet*
14:21 donsizemore
@pdurbin whoops, correct. so in theory akio could simply call multiple datafile ids. @pameyer: yet?
14:24 pameyer
@donsizemore dataset-level external tools, and further out multiple dataset tools, are things I've been making small amounts of noise about for a while
14:25 pameyer
no PRs planned soon :( some stuff needs more simmering / experimenting
14:26 pameyer
dataset-level probably relatively soon-ish post-migration
14:32 donsizemore
@pameyer I showed Akio my silly python scripts which pull the file metadata from the dataset api endpoint
14:37 pameyer
@donsizemore if it works, it's not silly
14:38 donsizemore
@pameyer oh, it's silly!
14:39 pdurbin
pameyer: let's see some PRs. :)
15:09 pameyer
pdurbin: migration first
15:09 pameyer
and possibly dataset page leak too
15:09 pdurbin
yeah. let's get you on the map
15:34 jri joined #dataverse
16:00 pameyer
every now and then, I wonder why that map isn't a dataset in dataverse
16:02 pdurbin
Yeah. It should be. The data should be a tabular file. The map should be created with the Dataverse "tabular mapping" feature.
16:05 pdurbin
pameyer: I've got Python on the brain because I'm writing the metrics aggregator in it.
16:07 pameyer
pdurbin: that's not necessarilly a bad thing
16:08 pameyer
I've never been tempted to use a compute node to analyse a memory leak from a python program
16:08 pdurbin
pameyer: well, now I have all sorts of questions. For https://github.com/sbgrid/data-capture-module or rsal have you ever considered hooking up the repo to coveralls or similar?
16:09 pameyer
not for dcm or rsal
16:10 pameyer
for the django stuff, I definately do static and coverage analysis
16:10 pdurbin
Ok. Basically at this point I have two scripts that work (download.py and aggregate.py) but there are no tests.
16:10 pdurbin
Just two standalone scripts.
16:11 pdurbin
Using a common config file.
16:11 pameyer
well, no tests simplifies the coverage analysis :)
16:11 pdurbin
yeah
16:12 pameyer
my usual approach is to have requirements.txt and requirements-dev.txt; with the test/analysis/coverage stuff in requirements-dev.txt
16:13 pdurbin
That's fine but my only requirements are Python 2.7 or Python 3.
16:13 pdurbin
I'm using the included batteries.
16:13 pameyer
nosetests seems to work relatively well for running tests; I usually go through a script to call from coverage - either use the html reports, or send them to sonar
16:17 pameyer
do you care about tooling, or badges?
16:18 pdurbin
A badge that said "Coverage 80%" or whatever would be nice.
16:19 pameyer
no ideas on badges
16:20 pdurbin
a non-empty folder called "tests" or whatever would be nice
16:21 pameyer
you can usually use a regex - `tests.py` or `tests/test_$foo.py` usually get picked up by the defaults
16:21 pameyer
https://github.com/sbgrid/data-capture-module/blob/master/api/test_mock.py
16:24 pdurbin
ok, so you have some integration tests. any unit tests?
16:27 pameyer
both - about to head for food; but feel free to ping me after
17:26 pameyer
pdurbin: back now
17:26 pameyer
no unit tests in that project, but others have them
17:27 pameyer
usually simpler
17:27 pameyer
but only slightly
17:28 pdurbin
ok, other projects. thanks
17:49 pdurbin
it does feel good to use vi again
17:51 pameyer
vi does work for java dev :)
17:51 pdurbin
heh
18:50 pdurbin
pameyer: ok, I made a pull request: https://github.com/IQSS/dataverse/pull/4973
18:50 pdurbin
two actualy, but that's a good starting point
19:13 pameyer
pdurbin: I assumed you knew about http://docs.python-requests.org/en/master/
19:24 pdurbin
yep
20:42 pdurbin
have a good weekend, y'all
20:42 pdurbin left #dataverse
21:28 pameyer left #dataverse
22:04 jri joined #dataverse
