Time
S
Nick
Message
07:06 jri joined #dataverse
09:17 juancorr joined #dataverse
11:09 jri joined #dataverse
11:26 donsizemore joined #dataverse
12:23 bjonnh joined #dataverse
16:18 pdurbin joined #dataverse
16:27 pdurbin
poikilotherm: doing some branch cleanup?
16:29 poikilotherm
Yeah
16:29 pdurbin
What's going on?
16:29 poikilotherm
So many stale branches in my fork
16:29 poikilotherm
Those where copy & pasted when forking
16:29 pdurbin
We are seeing every deletion in Slack.
16:30 pdurbin
Are you sure you are only affecting your fork?
16:33 poikilotherm
FUCK
16:33 poikilotherm
Github moved me to the main repo!
16:33 poikilotherm
GNAAAAAAAH
16:33 poikilotherm
That new UI of Github...
16:33 poikilotherm
I'm very sorry
16:33 poikilotherm
It were only stale and old branches not updated for at least 3 years
16:33 poikilotherm
And everyone with a local history can push them back
16:33 poikilotherm
Sorry sorry sorry
16:33 poikilotherm
I was really not trying to mess with your branches
16:34 jri joined #dataverse
16:35 pdurbin
Well, if they really haven't been updated for 3 years, we probably don't need them.
16:38 poikilotherm
As far as I saw most where already merged
16:38 poikilotherm
And they definitly where stale only
16:38 poikilotherm
(I was in the stale view of branches)
16:48 pdurbin
My instinct is always to make a spreadsheet so that's what I did. Here's the list of 133 branches: https://docs.google.com/spreadsheets/d/1lnoMv8EYOD7Gd0Zr91DeP70D8shRegb-GFp477QmV3I/edit#gid=0
16:50 poikilotherm
Had a quick finger...
16:50 poikilotherm
T_T
16:51 pdurbin
You're probably doing us a favor. I should have you come look at my basement and tell me what to throw in the trash.
16:51 poikilotherm
pdurbin you are with this project for a long time. I bet there is a big chance you have a lot of those branches pulled to your box
16:53 pdurbin
All the branches are in every clone. I think that's how git works.
16:53 poikilotherm
Well... I'm really sorry I deleted this stuff. But wether keeping all of those old and stale branches or not is a different discussion :-D I have opinions on that... I just expressed them :-D
16:54 poikilotherm
Not in mine. But I work from a fork and have upstream only as a secondary remote target.
16:54 pdurbin
We *still* have 132 branches. So you killed have of them. This reminds me of a movie I saw.
16:54 poikilotherm
And I removed all those old branches from my fork apparently a while ago :-D
16:55 pdurbin
half*
16:55 poikilotherm
But yes, other forks on Github should still have all those branches
16:55 pdurbin
It's ok. We needed something to talk about.
16:55 poikilotherm
And I killed thim in about 5 minutes :-D
16:55 poikilotherm
Click click click click....
16:57 pdurbin
:)
16:59 poikilotherm
Do you guys want to fix this or shall I write some script to strawl GH forks and recollect the branches? Good idea BTW with that spreadsheet
17:00 donsizemore
this is why I declined write access to IQSS/dataverse
17:01 pdurbin
No need to write scripts. I doubt we'll do much.
17:01 pdurbin
donsizemore: principle of least privilege
17:07 poikilotherm
Most likely https://github.com/BPeuch/dataverse has them all :-D
17:08 poikilotherm
Jim has 2 branches less... https://github.com/qqmyers/dataverse
17:09 donsizemore
@pdurbin principle of IQSS not sending a sock party my way
17:20 jri joined #dataverse
17:29 poikilotherm
donsizemore pdurbin what's beyond the moving of dataverse-ansible to GDCC?
17:29 donsizemore joined #dataverse
17:41 pdurbin
poikilotherm: I don't know. Good question for donsizemore
17:42 donsizemore
@pdurbin @poikilotherm zzz huh whuh?
17:42 * pdurbin
throws a sock at donsizemore
17:49 donsizemore
@pdurbin and in the next episode of don't-give-don-write-access, here's the pull request i just committed to master: https://github.com/GlobalDataverseCommunityConsortium/dataverse-ansible/commit/f11428b5da6368279d14fef10f334478342a4e59
18:19 pdurbin
donsizemore: yay, schemaspy! Thanks!
18:26 donsizemore
@pdurbin http://ec2-35-174-205-21.compute-1.amazonaws.com/schemaspy/develop/
18:28 pdurbin
wow, a fancy new version
18:54 poikilotherm
pdurbin have you noticed? Deployment times are down to ~60 seconds from Payara startup to Deployment done & ready
18:55 poikilotherm
Deployment of the WAR is ~40-50s
18:56 pdurbin
Are you deploying with asadmin?
18:57 poikilotherm
Nope. Using containers :-D
18:57 poikilotherm
Postboot-Command
18:58 pdurbin
so it works with wildfly and liberty too?
18:58 pdurbin
some kind of abstraction?
18:59 poikilotherm
No no no... Using Payara.
19:00 poikilotherm
But different deployment method
19:00 pdurbin
Can you please point to the line in dataverse-kubernetes?
19:04 poikilotherm
I can't, I'm sorry. This is baked into upstream. Let me see if I can find that.
19:04 donsizemore
40-50s seems long?
19:04 pdurbin
Ok. I just don't understand how it works.
19:05 pdurbin
donsizemore: the dream is a deployment time so short it doesn't annoy developers. Developer joy, as the Quarkus website says.
19:11 poikilotherm
pdurbin there you go https://github.com/payara/Payara/blob/4a3fdaf3d62ad3643fd0d7c2c4f01bd324aac30b/appserver/extras/docker-images/server-full/src/main/docker/bin/init_1_generate_deploy_commands.sh
19:12 pdurbin
Thanks. I would still consider this to be asadmin.
19:15 poikilotherm
https://docs.payara.fish/community/docs/5.201/documentation/payara-micro/asadmin/pre-and-post-boot-scripts.html
19:16 poikilotherm
Yeah. Kind of.
19:18 pdurbin
:)
19:18 poikilotherm
OK I just tested https://github.com/IQSS/dataverse/issues/6819 again. Works flawless.
19:18 poikilotherm
Still need to test for the timers :-/
19:19 pdurbin
nice
19:20 poikilotherm
Ha. Looks like it work with timers, too.
19:20 poikilotherm
PSQL complains the timer database table already exists ;-)
19:27 poikilotherm
Looking for all appearences of postgres.jar in the repo now.
19:27 poikilotherm
pdurbin stumbling over https://github.com/poikilotherm/dataverse/blob/develop/scripts/setup/asadmin-setup.sh
19:27 poikilotherm
Shouldn't this be deleted?
19:28 pdurbin
Is that the one Michael Bar-Sinai? added?
19:28 poikilotherm
Searching for the script name in the repo reveils no usages
19:29 pdurbin
Yeah, at this point, it's cruft. Unmaintained. If you feel like it, you can create a pull request to delete it.
19:33 poikilotherm
Looks like
19:33 poikilotherm
Dead glassfish stuff
19:33 poikilotherm
Using Postgres JDBC 9.3 drivers
19:33 poikilotherm
It resides as the only thing under scripts/setup. I remember scratching my head what this thing is...
19:33 poikilotherm
Separated PR?
19:33 poikilotherm
Or part of my PR for #6819?
19:33 pdurbin
separate PR please
19:51 poikilotherm
Yosh!
19:52 poikilotherm
(よし)
19:54 poikilotherm
donsizemore do you want me to create an issue in dataverse-ansible about JDBC handling linked to issue 6918 and it's PR?
19:56 donsizemore
@poikilotherm certainly
20:07 poikilotherm
donsizemore pdurbin looks like djbrooke is cutting the edges...
20:08 poikilotherm
Plz review https://github.com/IQSS/dataverse/pull/7048
20:08 poikilotherm
:-)
20:09 poikilotherm
BTW donsizemore you might want to set the IQSS/dataverse-ansible repo to archived status. Generates a big fat warning for users landing there. See https://github.com/payara/docker-payaraserver-full for an example
20:14 pdurbin
poikilotherm: I took a look. Seems like a cleaner way to do it. It's like you're trying to put the installer out of a job. In these tough times, even. We'll probably want Leonid to take a look but he's out all week.
20:15 poikilotherm
donsizemore please poke me if you want me to write more at https://github.com/GlobalDataverseCommunityConsortium/dataverse-ansible/issues/6
20:27 poikilotherm
pdurbin more work for you.... https://github.com/IQSS/dataverse/pull/7049
20:27 poikilotherm
Should I add anyone else for a review?
20:29 poikilotherm
Right :-)
20:29 pdurbin
I just asked Michael for a review.
20:29 poikilotherm
Testing a new postgres driver version boils down to building with mvn package -Dpostgresql.version=42.2.x, deploy and test :-)
20:30 poikilotherm
Patty cake patty cake...
20:31 poikilotherm
pdurbin do you think I should push for a database configuration by annotation, too?
20:32 pdurbin
by @Annotation or do you mean some kind of microprofile config api thing?
20:32 poikilotherm
Ah that might offend Leonid... Better open an issue for this first ;-)
20:36 pdurbin
I'm just not sure what you mean. I assume you mean Java annotation.
20:36 poikilotherm
That would be a step towards this
20:36 poikilotherm
Instead of manual configuration by the installer / script, you would define the database connection in an annotation.
20:37 poikilotherm
Or in glassfish-resources.xml
20:37 poikilotherm
Now that we are a Java EE 8 App, an annotation seems a better fit
20:37 pdurbin
Whatever makes people's lives easier.
20:37 poikilotherm
That can be configured from outside, for example using MicroProfile Config API . But it's also usable with usual things like env vars, which is a Payara feature.
20:38 poikilotherm
Better go for standards :-)
20:39 poikilotherm
The installer will not be meaningless. You folks will one day need to think about how you want configuration of dataverse to continue.
20:39 poikilotherm
Maybe a fancy /etc/dataverse/config.properties? ;-)
20:40 pdurbin
:)
20:40 pdurbin
sounds good to me
20:41 pdurbin
I've been hoping you keep pushing the microprofile stuff. Or whatever we should do to modernize.
20:49 poikilotherm
Heh.
20:49 poikilotherm
Small steps :-D
20:49 poikilotherm
Packaging the JDBC driver is one of those
20:49 poikilotherm
So many places where code can be dropped :-D
20:49 poikilotherm
Oh now I remember: you wanted that review from me...
20:50 poikilotherm
Let me check that
21:00 donsizemore
@poikilotherm I'm totally on board from an Ansible perspective
21:01 pdurbin left #dataverse
21:01 poikilotherm
donsizemore: +1
21:13 poikilotherm
Heh. pdurbin left. I bet he had to run.
21:13 poikilotherm
pdurbin when you read this later on, review done. https://github.com/IQSS/dataverse/pull/7025#pullrequestreview-441188392
