Log of the #duraspace-ff channel on chat.freenode.net

Using timezone: Eastern Standard Time
* scossu leaves00:20
* scossu joins00:50
* scossu leaves00:59
* fcrepo-bot joins01:03
* ksclarke leaves01:19
* fcrepo-bot leaves02:12
* scossu joins02:55
* scossu leaves03:00
* scossu joins04:56
* scossu leaves05:00
* fasseg joins06:33
* scossu joins06:56
* scossu leaves07:01
* scossu joins07:48
* jgpawletko joins08:11
* scossu leaves08:17
* scossu joins08:28
<pivotal-bot_>Frank Asseg added comment: "Added another guide for a cluster on multiple machines and added some Links to configuration resources: ""08:37
http..." https://www.pivotaltracker.com/story/show/62257150
* scossu leaves08:44
* scossu joins09:15
<pivotal-bot_>Andrew Woods added comment: "https://github.com/futures/benchtool/commit/0dfc1215cdcf3f1c8edea2e752baecf74bf757f0" https://www.pivotaltracker.com/story/show/6475942009:19
* scossu leaves09:33
<pivotal-bot_>Andrew Woods added comment: "@frankasseg, is the above commit correctly the one that resolves this ticket? It seems to be focused on get..." https://www.pivotaltracker.com/story/show/6475942009:41
* ksclarke joins09:50
* ajs6f joins10:07
<pivotal-bot_>Frank Asseg added comment: "@awoods Yes. This resolves the ticket. HttpClient did proper authentication, but getting the triple size was..." https://www.pivotaltracker.com/story/show/6475942010:11
Andrew Woods accepted "Fix Bench-Tool authentication against Fedora 4" https://www.pivotaltracker.com/story/show/6475942010:14
<fasseg>ajs6f: I get a transient dependency to the fcrepo-kernel-api osgi bundle when adding fcrepo-kernel as a dependency to my SCAPE project, but I can't access the bundle and get this error:10:15
Could not find artifact org.fcrepo:fcrepo-kernel-api:bundle:4.0.0-alpha-4-SNAPSHOT in sonatypesnapshots (http://oss.sonatype.org/content/repositories/snapshots/)
Can I somehow install this bundle locally?
*rather: should I install it locally or should the bundle be available on the sonatype repo?10:16
or should I just import the jar?10:17
ah just adding the bundle plugin should help I guess..10:21
yeah adding the maven-bundle-plugin did certainly help :)10:22
<awoods>fasseg: Sometimes the answer just comes.10:23
<fasseg>so helped a lot there: http://stackoverflow.com/questions/9932923/why-cant-maven-find-an-osgi-bundle-dependency?lq=1 although the accepted answer was rubbish it seems ;)10:24
<awoods>fasseg: ...although I would expect the snapshot repo to have the artifact. hmm
<fasseg>it does I guess but maven has no idea what the packaging "bundle" means
<awoods>fasseg: yes, it is there. http://oss.sonatype.org/content/repositories/snapshots/org/fcrepo/fcrepo-kernel-api/4.0.0-alpha-4-SNAPSHOT/10:25
fasseg: So you just added the bundle-plugin to your scape project?
<fasseg>yep, and the plugin is responsible for resolving osgi bundles10:26
<ajs6f>Sorry— was in a conv in real life.10:30
fasseg: I don't know when artifacts get published to our maven repo.
But you should be able to just build locally.
<fasseg>ajs6f: yeah works fine now
<ajs6f>A bundle can be viewed as a jar, so you can always switch the type to jar and go forward.10:31
<fasseg>It might be Jenkins publishing to sonatype
<ajs6f>fasseg: Cool, didn't want to block you.
fasseg: I think it is.
<awoods>ajs6f: our artifacts get published by jenkins on every successful build.
<ajs6f>But you can do it from your workstation.
<fasseg>ajs6f: you didn't block me, I was blocked my own lack of osgi & maven knowlegde, I thought bundles were supported out of the box10:32
<ajs6f>Anyway, given the realtionship between OSGi and Maven, introducing bundle types should _not_ cause any problems.
fasseg: You may need to engage the bundle plugin. But if the bundles in question are just deps, and not the artifacts you are making, you shoukn't.
I had to, beacuse I _was_ making bundles.10:33
But you are just pulling them in.
awoods: Do we want a ticket for JSON-RDF conneg?
<fasseg>As I said I got a transient dep on the osgi bundle from fcrepo-kernel, so as soon as you're importing the kernel as a dep you will have to add the bundle-plugin
<ajs6f>Or JSON-LD, or anyof those other types of things.10:34
fasseg: Hm. Weird. You shouldn't have to… but as long as you're okay now.
<fasseg>I guess we could just change the dep on fcrepo kernel to jar packaging...
<ajs6f>fasseg: Yeah, that's what I meant ^^^.
When you have a chance, can you try that locally?
<fasseg>sure will do10:35
<ajs6f>Much obliged. I want to do the OSGI thing, but I don't want to annoy people while I'm doing it.
<awoods>ajs6f: I thought we are staying away from rdf+json due to limitations in the serialization.
<ajs6f>awoods: I didn't know that. I've never taken a close look at it. But I'm fine with keeping it out unless/until someone actually comes up w/ a use case or a path.10:36
* ajs6f leaves10:45
* github-ff joins
[fcrepo4] fasseg created no-osgi-dep (+1 new commit): http://git.io/wwuFPw
fcrepo4/no-osgi-dep cc24e58 fasseg: Changed bundling to jar...
* github-ff leaves
* github-ff joins10:46
[fcrepo4] fasseg opened pull request #225: Changed bundling to jar (master...no-osgi-dep) http://git.io/WkZ_8g
* github-ff leaves
<fasseg>awoods: you wann review: https://github.com/futures/fcrepo4/pull/225 ? it's really *REALLY* straightforward ;)
<awoods>fasseg: sure10:47
fasseg: I am not sure we want to do that. ajs6f?10:48
<fasseg>kk, I talked about this with adam, and this was the idea, yeah
<awoods>fasseg: From above, yes, it sounds like ajs6f is on-board.
fasseg: So the question I have is, what is the reason for not removing all of our "bundle" packaging definitions?10:49
<fasseg>lets wait for adam and CI.....no rush there.
there might be osgi bundles only available on an osgi repo, which can not be resolved by maven10:50
and there are no other deps using the bundle packaging10:52
all the other <packaging>bundle<packaging> directives are part of the artifact definitions, not the deps10:53
and I think that's what Adam's trying to achieve: generate osgi bundles, but don't use them ad deps
*as deps
* ajs6f joins
<fasseg>here he's back !!
here's a pullie: https://github.com/futures/fcrepo4/pull/225 there are no other deps using bundle packaging...10:55
<ajs6f>Yeah, _create_ bundles, so that OSGi-aware builds can use them as bundles, but in non-OSGi builds, _use_ them as jars,
so that the artifacts _Have_ the OSGi metadata and can be used as such, but no one is _required_ to do so.
<fasseg>kk so the PR should be fine, builds nicely for me and I can use fcrepo-kernel w/o the transient osgi bundle dep
<ajs6f>fasseg: but the artifacts are still being _produced_ as bundles, right?10:56
Just referenced as jars?
<fasseg>yep....only 1 line changed :)
<fasseg>well removed really
<ajs6f>Good enough. As longas we have the OSGi metadata, that is all we should need to care about at this stage.
<fasseg>and I can build my SCAPE stuff directly from master ;)10:57
* fasseg is too lazy to tag ...
* scossu joins10:58
<ajs6f>awoods: will be two-three minutes late to standup10:59
* edInCo joins11:01
<fasseg>awoods: not allowed to join :/
<awoods>fasseg: and you are using your same gmail account?11:02
<fasseg>only have this one
* mikeAtUVa can't join either...
<scossu>awoods: Is it OK for me to join?
<ajs6f>Don't knwo what awoods thinks, but I think it's _awesome_ that you want to join.
<scossu>ajs6f: That's very flattering.11:03
<ajs6f>We're a friendly bunch.
afk bbs
* ajs6f leaves
<awoods>now try
* travis-ci joins11:04
[travis-ci] futures/fcrepo4#1502 (no-osgi-dep - cc24e58 : fasseg): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/commit/cc24e5801f22
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/17914035
* travis-ci leaves
<scossu>I'm not allowed to join either.
<fasseg>awoods: no cigar
<awoods>clearly we need another solution.
<fasseg>not rdytalk though plleeeeeeaaase11:05
<awoods>the 15-person google hangouts are not going to work... since they are limited to DuraSpace.
we can do a 10-person hangout.
<fasseg>fine by me
* ajs6f joins11:08
<awoods>try this one: https://plus.google.com/hangouts/_/event/c1glu6soq43r1rr6ou17qtobug8
<ajs6f>I'm in a call with Stefano— is that the right one?11:09
Wait— I'm not allowed to join that one.
I guess not the right one.
awoods: I've got to roll at 11:30 ET.11:11
Nothing makes me happy.
awoods: As a sidebar: we've known from the beginning that we wanted to maintain the formal published ontos as artifacts out of the codebase (or, for that matter, vice versa) but they are still completely independent in construction. We should make that not so. At some point. When we have time. And resources.11:21
scossu: In your Google Hangout picture, are you buying vegetables?11:25
awoods/all: Gonna have to leave momentarily (using that word in the American sense).11:35
* ajs6f leaves11:39
* ajs6f joins12:55
<scossu>ajs6f: Yes, at a farmer's market in Flint MI
<ajs6f>scossu: Makes me hungry.12:56
awoods: Did you and fasseg get straight on the OSGi metadta thing?
* ajs6f leaves13:48
* ajs6f joins13:50
* scossu leaves14:21
<ajs6f>awwods: Check out:14:28
Good stuff. Helping me work through the indexing questions.
* ajs6f leaves14:36
* mikeAtUVa leaves14:47
* scossu joins15:09
* scossu leaves15:25
* scossu joins15:37
* scossu1 joins15:47
* scossu leaves15:51
* jgpawletko leaves16:21
* ajs6f joins16:28
Ach! Mac OS 10.8 is horrible! I think it may be the year of Linux on the desktop. For ME.
* ajs6f leaves16:46
* edInCo leaves18:01
* ksclarke leaves18:06
* fasseg leaves18:46
* scossu1 leaves19:20
* github-ff joins19:50
[fcrepo4] awoods closed pull request #225: Changed bundling to jar (master...no-osgi-dep) http://git.io/WkZ_8g
* github-ff leaves
<pivotal-bot_>Andrew Woods added "Tomcat restart hanging" https://www.pivotaltracker.com/story/show/6487512019:52
* travis-ci joins20:10
[travis-ci] futures/fcrepo4#1504 (master - 0aa3b52 : Andrew Woods): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/c9e6388e9f52...0aa3b52ec53e
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/17948329
* travis-ci leaves
* ksclarke joins21:53
* scossu joins22:39
* scossu leaves22:41