Log of the #fcrepo channel on chat.freenode.net

Using timezone: Eastern Standard Time
* awoods leaves06:58
* awoods joins07:05
* acoburn joins08:25
* dwilcox joins08:38
* dwilcox leaves08:39
* dwilcox joins08:43
* dhlamb joins08:47
* awead joins08:48
* dwilcox leaves09:10
* whikloj joins09:29
* dhlamb leaves09:40
* bseeger joins10:00
* dhlamb joins10:36
* bseeger leaves10:44
<f4jenkins>Yippee, build fixed!10:53
Project fcrepo4-T2 build #317: FIXED in 5 min 32 sec: http://jenkins.fcrepo.org/job/fcrepo4-T2/317/
awoods: Remove dependency from kernel-api on fcrepo-mint
* github-ff joins12:03
[fcrepo4] acoburn opened pull request #873: Replace guava's Function and Predicate with Java8 idioms (master...fcrepo-1681) http://git.io/vOxRD
* github-ff leaves
* github-ff joins12:07
[migration-utils] mikedurbin opened pull request #34: https://jira.duraspace.org/browse/FCREPO-1680 (master...fcrepo-1680) http://git.io/vOxuB
* github-ff leaves
* dhlamb leaves12:11
* github-ff joins12:27
[fcrepo-message-consumer] awoods opened pull request #90: React to recent Jena version updates (master...fcrepo-1682) http://git.io/vOxo4
* github-ff leaves
* github-ff joins12:28
[fcrepo4-client] awoods opened pull request #29: Update Jena version (master...fcrepo-1682) http://git.io/vOxoo
* github-ff leaves
* dhlamb joins12:48
* github-ff joins12:51
[fcrepo-message-consumer] acoburn pushed 2 new commits to master: http://git.io/vOx91
fcrepo-message-consumer/master fbf56af Andrew Woods: React to recent Jena version updates...
fcrepo-message-consumer/master 2cd8fa8 Aaron Coburn: Merge pull request #90 from awoods/fcrepo-1682...
* github-ff leaves
* github-ff joins
[fcrepo4-client] acoburn pushed 2 new commits to master: http://git.io/vOx9S
fcrepo4-client/master a179e1b Andrew Woods: Update Jena version...
fcrepo4-client/master c5e14b9 Aaron Coburn: Merge pull request #29 from awoods/fcrepo-1682...
* github-ff leaves
* travis-ci joins12:55
fcrepo4-labs/fcrepo4-client#80 (master - c5e14b9 : Aaron Coburn): The build passed.
Change view : https://github.com/fcrepo4-labs/fcrepo4-client/compare/5d8141a0bcd2...c5e14b9e95bc
Build details : https://travis-ci.org/fcrepo4-labs/fcrepo4-client/builds/74601221
* travis-ci leaves
<acoburn>awoods: one thing I noticed with that java8 PR was the extensive use of Iterators in the kernel13:12
awoods: I wonder whether the java8 Stream API wouldn't be a better fit?13:13
awoods: we'd still get the lazy/stream-based evaluation, but without having to do gymnastics with Iterables
<awoods>acoburn: possibly... although didn't ajs6f have an opinion on "Iterators" being lighter weight?13:14
<acoburn>awoods: they are certainly lighter weight when compared directly, but for everything we do with them, we have to run them through guava's API13:15
awoods: so in the end, I'm not sure how much lighter they actually are
* github-ff joins
[migration-utils] awoods pushed 3 new commits to master: http://git.io/vOxxZ
migration-utils/master a1d11cc Mike Durbin: Added pid to migrated properties with configurable mapping
migration-utils/master 97ab5cb Mike Durbin: Fixed a bug that prevented datastream properties from being migrated.
migration-utils/master 02762e2 Andrew Woods: Merge pull request #34 from mikedurbin/fcrepo-1680...
* github-ff leaves
<acoburn>awoods: something to discuss with ajs6f when he's back
<awoods>acoburn: moving towards Java8 constructs seems constructive if the comparison is marginal. Would you add it to next week's agenda as a reminder? https://wiki.duraspace.org/display/FF/2015-08-13+-+Fedora+Tech+Meeting13:19
* awead leaves13:20
* awead joins13:21
<acoburn>awoods: here's guava's impl: https://github.com/google/guava/blob/master/guava/src/com/google/common/collect/Iterators.java#L661-L67713:25
awoods: it's light weight, but I'm not sure how less light weight a Stream would be
awoods: and a Stream would, potentially, buy us the ability to do parallel maps and filters13:26
<awoods>acoburn: is the use case you see in the kernel mostly around "filtering"?13:27
<acoburn>awoods: that's just one example, but most of the use cases would relate to maps and filters13:28
awoods: it was the common use of Iterators.filter() that caught my eye
<awoods>acoburn: there is strong appeal for one standard approach over two... but let's discuss next week.13:30
<acoburn>awoods: sounds good
<awoods>acoburn: all things considered, it is hard to imagine the heavy/light weightedness in these case having much of a performance impact.13:31
acoburn: all things considered, it is hard to imagine the heavy/light weightedness in this case having much of a performance impact.
<acoburn>awoods: I agree13:32
* github-ff joins13:49
[fcrepo4] awoods pushed 2 new commits to master: http://git.io/vOpqb
fcrepo4/master 218b9a3 Aaron Coburn: Replace guava's Function and Predicate with Java8 idioms
fcrepo4/master e42ffa2 Andrew Woods: Merge pull request #873 from acoburn/fcrepo-1681...
* github-ff leaves
<awoods>afk13:53
* dhlamb leaves14:28
<awoods>acoburn: as a heads-up, more is coming from impacts of: https://jira.duraspace.org/browse/FCREPO-167514:29
<acoburn>awoods: thanks
awoods: a good reason to increment dependencies early in the dev cycle14:30
* acoburn leaves14:46
* acoburn joins14:47
* acoburn1 joins14:58
* acoburn1 leaves
* acoburn leaves
* acoburn joins15:07
* acoburn1 joins15:32
* acoburn leaves15:34
* github-ff joins15:36
[fcrepo-karaf] awoods pushed 4 new commits to master: http://git.io/vOpF0
fcrepo-karaf/master 9577f68 Aaron Coburn: remove fcrepo-kernel-api dependency on fcrepo-mint
fcrepo-karaf/master 66dbf88 Aaron Coburn: update dependencies
fcrepo-karaf/master e038cb4 Aaron Coburn: remove dependency injection for fcrepo-kernel-api
* github-ff leaves
* travis-ci joins15:38
fcrepo4-labs/fcrepo-karaf#30 (master - 7e00919 : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4-labs/fcrepo-karaf/compare/f03409f5dd7d...7e00919b87e2
Build details : https://travis-ci.org/fcrepo4-labs/fcrepo-karaf/builds/74622141
* travis-ci leaves
* jgpawletko joins15:58
* jgpawletko leaves16:00
* github-ff joins16:03
[fcrepo4-oaiprovider] awoods opened pull request #22: Address impact of moving from Guava to Java8 (master...fcrepo-1681) http://git.io/vOhJI
* github-ff leaves
* dwilcox joins16:04
<awoods>acoburn1: not Jena, the issue in oaiprovider from from Guava to Java8: https://jira.duraspace.org/browse/FCREPO-1681
<acoburn1>awoods: I can review that one
* dwilcox leaves16:07
* github-ff joins16:19
[fcrepo4-oaiprovider] acoburn pushed 2 new commits to master: http://git.io/vOhqJ
fcrepo4-oaiprovider/master 9c54f23 Andrew Woods: Address impact of moving from Guava to Java8...
fcrepo4-oaiprovider/master 8a44d38 Aaron Coburn: Merge pull request #22 from awoods/fcrepo-1681...
* github-ff leaves
<acoburn1>awoods: I thought there would be some fallout from the Guava-Java8 change.16:21
<awoods>acoburn1: fortunately, it was minimal16:22
<acoburn1>awoods: I had tested fcrepo-audit and the various auth modules, but didn't try the oaiprovider
<awoods>acoburn1: I have some local scripts for running through builds of all of the projects, which is sometimes helpful... but Jenkins also works.16:23
<acoburn1>awoods: jenkins is great for that
* travis-ci joins
fcrepo4-labs/fcrepo4-oaiprovider#90 (master - 8a44d38 : Aaron Coburn): The build passed.
Change view : https://github.com/fcrepo4-labs/fcrepo4-oaiprovider/compare/37d58ba4129d...8a44d386282d
Build details : https://travis-ci.org/fcrepo4-labs/fcrepo4-oaiprovider/builds/74627641
* travis-ci leaves
<acoburn1>awoods: btw, I don't think i have access to https://github.com/fcrepo4-exts
<awoods>acoburn1: invited16:24
<acoburn1>awoods: thanks16:25
* github-ff joins16:43
[fcrepo-camel-toolbox] acoburn opened pull request #48: update scm information (master...fcrepo-1683) http://git.io/vOhcC
* github-ff leaves
<acoburn1>awoods: don't merge this yet ^^^16:44
* acoburn joins16:47
* dwilcox joins16:49
<acoburn>awoods: github should have sent you a request related to travis-ci access to fcrepo4-ext; you'll need to take care of that16:51
* dwilcox leaves
* acoburn1 leaves
* dwilcox joins
* dwilcox leaves16:55
<awoods>acoburn: thanks. granted.16:56
* whikloj leaves17:00
<acoburn>awoods: I can't get https://travis-ci.org/fcrepo4-exts to display anything. Is there something you need to enable on the travis side?17:08
* awoods looking17:10
acoburn: you probably need to have a build run first.17:13
<acoburn>awoods: I thought that, but I just pushed a new commit (after enabling the service), but no dice
<awoods>acoburn: https://travis-ci.org/fcrepo4-exts/fcrepo-camel-toolbox17:15
<acoburn>awoods++
awoods: seems to be working now: https://github.com/fcrepo4-exts/fcrepo-camel-toolbox/pull/4817:20
<awoods>acoburn: nice
* github-ff joins17:47
[fcrepo-camel-toolbox] acoburn opened pull request #49: update urls in gh docs (gh-pages...fcrepo-1683-docs) http://git.io/vOh9s
* github-ff leaves
* escowles leaves17:49
* github-ff joins17:57
[fcrepo-camel] acoburn opened pull request #85: Github Shuffle (master...fcrepo-1683) http://git.io/vOh5Y
* github-ff leaves
* github-ff joins18:04
[fcrepo-camel] acoburn opened pull request #86: Github Shuffle (gh-pages...fcrepo-1683-docs) http://git.io/vOhNl
* github-ff leaves
* github-ff joins18:10
[fcrepo-audit] acoburn opened pull request #21: Github Shuffle (master...fcrepo-1683) http://git.io/vOhhJ
* github-ff leaves
* github-ff joins18:13
[fcrepo-audit] acoburn opened pull request #22: Github Shuffle (gh-pages...fcrepo-1683-docs) http://git.io/vOhjE
* github-ff leaves
* github-ff joins18:20
[fcrepo-webapp-plus] acoburn opened pull request #21: update scm location (master...fcrepo-1683) http://git.io/vOjv5
* github-ff leaves
* github-ff joins18:29
[fcrepo-message-consumer] acoburn opened pull request #91: update scm location (master...fcrepo-1683) http://git.io/vOjTk
* github-ff leaves
* dwilcox joins18:38
* dwilcox leaves18:39
* dwilcox joins18:40
* github-ff joins
[fcrepo-message-consumer] acoburn opened pull request #92: Github Shuffle (gh-pages...fcrepo-1683-docs) http://git.io/vOjL7
* github-ff leaves
* acoburn1 joins18:41
awoods: ping18:48
<awoods>acoburn118:49
* acoburn leaves
<acoburn1>awoods: about maven…
awoods: we talked about projects that inherited from the fcrepo4 POM
awoods: but perhaps we need an fcrepo4-parent artifact that defines all of the core needed for any java project18:50
then, fcrepo4 would inherit from that, as would, for instance, fcrepo-camel
<awoods>acoburn1: "all of the core needed"??18:51
acoburn1: are you talking about dependency-management?
<acoburn1>awoods: what I mean is that the fcrepo4-parent would define a certain minimum set of requirements
<awoods>acoburn1: what kind of requirements?18:52
<acoburn1>awoods: for instance, checkstyle
but not dependency versions
that way, e.g. an inheritance chain might look like this: fcrepo-webapp-plus -> fcrepo4 -> fcrepo4-parent18:53
or: fcrepo-camel -> fcrepo4-parent
or: fcrepo-camel-toolbox -> fcrepo4-parent
am I making any sense?
<awoods>acoburn1: fcrepo4-parent would still have to inherit from: <parent>
<groupId>org.sonatype.oss</groupId>
<artifactId>oss-parent</artifactId>
<version>9</version>
</parent>
<acoburn1>exactly18:54
that's the absolute minimum that it would need to do
<awoods>acoburn1: and fcrepo4-parent/pom.xml would only define plugins?
<acoburn1>awoods: maybe also <organization>, <licenses>, <distributionManagement>18:55
* dwilcox leaves
<acoburn1>awoods: <scm> <repositories> and some required plugins
awoods: well, it can't actually enforce plugins, but it can define a minimum configuration18:56
<awoods>acoburn1: sounds reasonable.
<acoburn1>awoods: so to my question: where would that live? build-tools?18:57
awoods: or in fcrepo4? (fcrepo4/fcrepo-parent)
<awoods>acoburn1: Yes, I was thinking fcrepo4/fcrepo-parent18:58
<acoburn1>awoods: I see quite a few projects that follow that pattern
<awoods>acoburn1: go for it18:59
<acoburn1>awoods: ok, I'll see what I can come up with19:00
awoods: btw, I did not include fcrepo4-client in https://jira.duraspace.org/browse/FCREPO-168319:01
awoods: I don't feel that it's ready for graduation out of -labs, especially as there are questions about whether to do a full rewrite on it19:02
<awoods>acoburn1: ok, but most of the projects on your list are in various states of half-bakedness
acoburn1: some criteria would be helpful19:03
<acoburn1>awoods: that's why I stopped where I did
<awoods>acoburn1: the community is submitting PRs against fcrepo4-client
<acoburn1>awoods: yes, but it's api is still all about datastreams and objects
<awoods>acoburn1: leaving it out is fine19:04
acoburn1: I am just saying some objective criteria would be nice for projects that get into fcrepo4-exts
<acoburn1>awoods: I agree19:05
awoods: how do you feel about the projects that are currently in -exts?
awoods: the basic criterion I have is this: is it listed on the download page?19:07
https://wiki.duraspace.org/display/FF/Downloads
awoods: and following that, if the download is in fcrepo4, all of its dependencies should also be in fcrepo4
awoods: and if the artifact lives in -exts, all of its dependencies should be in either fcrepo4/ or fcrepo4-exts19:08
awoods: that is, if it's in -exts, it shouldn't depend on something in -labs
awoods: the current exception is fcrepo4-vagrant19:09
awoods: which I excluded simply because it wasn't on the list from the tech call
awoods: the other exception is the message consumer, which is its own special case, as we discussed19:10
<awoods>acoburn1: maybe we should include fcrepo4-vagrant, and fcrepo-karak, but leave fcrepo4-release-tests in labs.19:11
<acoburn1>awoods: I am definitely in favor of including fcrepo4-vagrant.19:12
<awoods>acoburn1: message-consumer should not be in the core org, but exts, no?
<acoburn1>awoods: I think the fcrepo-karaf project is lacking some maturity
<awoods>acoburn1: re karaf, ok
<acoburn1>awoods: I'd like to see pax-exam added as a testing framework first
awoods: and greater participation than just me19:13
awoods: that way I'm not just manually testing it all the time.
<awoods>acoburn1: fcrepo-karaf only has a features.xml
<acoburn1>awoods: correct, but we should have a mechanism for testing that19:14
awoods: and that's what pax-exam does
<awoods>acoburn1: sure
<acoburn1>awoods: basically, it starts up an entire OSGi runtime, and then uses the features file to provision the bundles
<awoods>acoburn1: what were you saying about message-consumer... where are you putting it?19:15
<acoburn1>awoods: it's in -exts because it shouldn't be in fcrepo4
<awoods>acoburn1: good
<acoburn1>awoods: there was discussion that it shouldn't be in -labs
<awoods>acoburn1: correct
<acoburn1>awoods: but there might be a fcrepo4-archive organization coming soon?19:16
<awoods>acoburn1: correct
acoburn1: and are you ok with leaving release-tests in labs?
<acoburn1>awoods: and after some period of time (6 mos?) it would go there
<awoods>acoburn1: correct
<acoburn1>awoods: yes, let's leave release-tests in labs — it currently uses the message consumer in its tests19:17
awoods: so that means for projects left to migrate: fcrepo-vagrant19:18
awoods: migration-utils?
awoods: fcrepo4-upgrade-utils?
<awoods>acoburn1: upgrade-utils -> exts19:19
acoburn1: migration-utils can stay in labs
acoburn1: it is temporary
<acoburn1>awoods: ok, I'll change the ticket to list fcrepo-vagrant and upgrade-utils (and remove the other items)19:20
<awoods>acoburn1: yes. Is it safe to assume that you are moving the github projects by simply changing the github ownership?
<acoburn1>awoods: yes, I'm assigning ownership to the "committers" group in -exts19:21
awoods: it's actually a very simple process
<awoods>acoburn1: I mean, how are you moving the projects? with a click of a button in github?
<acoburn1>awoods: yes, under "settings"19:22
<awoods>acoburn1: good
acoburn1: thanks
* acoburn1 leaves19:27
* the_mgt leaves19:51
* the_mgt joins19:55
* jgpawletko joins22:11