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

Using timezone: Eastern Standard Time
* dwilcox joins06:55
* dwilcox_ joins06:58
* dwilcox leaves06:59
* dwilcox joins07:54
* dwilcox leaves07:56
* dwilcox_ leaves07:57
* dwilcox joins07:58
* ksclarke leaves08:16
* dhlamb joins08:34
* dwilcox leaves08:38
* dwilcox joins08:40
* dwilcox leaves08:44
* acoburn joins08:56
* ksclarke joins09:05
* mikeAtUVa joins09:07
* github-ff joins09:19
[ontology] acoburn opened pull request #21: draft fcdm vocabulary (master...fcdm-vocabulary) http://git.io/xIjU
* github-ff leaves
* scossu joins09:33
* dwilcox joins09:35
* scossu leaves09:36
* github-ff joins09:54
[fcrepo-camel] awoods closed pull request #58: add missing dependency; organize list of feature bundles (master...feature-dependencies) http://git.io/xerh
* github-ff leaves
* travis-ci joins09:58
fcrepo4/fcrepo-camel#142 (master - bc952e8 : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4/fcrepo-camel/compare/3d7127772afd...bc952e88cb76
Build details : http://travis-ci.org/fcrepo4/fcrepo-camel/builds/52428459
* travis-ci leaves
* osmandin joins10:04
* github-ff joins10:26
[fcrepo-camel] awoods pushed 2 new commits to master: http://git.io/xLu2
fcrepo-camel/master e3233c4 Aaron Coburn: improved support for cached data streams...
fcrepo-camel/master fa18184 Andrew Woods: Merge pull request #57 from acoburn/fcrepo-1380...
* github-ff leaves
* ajs6f joins10:28
* travis-ci joins10:32
fcrepo4/fcrepo-camel#143 (master - fa18184 : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4/fcrepo-camel/compare/bc952e88cb76...fa1818472982
Build details : http://travis-ci.org/fcrepo4/fcrepo-camel/builds/52432850
* travis-ci leaves
* ajs6f1 joins10:40
* ajs6f leaves
* barmintor joins10:47
mikeAtUVa: if I push a branch for that bug you reported, are you in a position to test it?
I’m almost done w/ the IT for it.10:48
<mikeAtUVa>barmintor: yeah, I could do that.10:51
I may not be able to run all the tests, but I could build the artifacts and deploy it locally to the same data directory and see if the export works.
<barmintor>mikeAtUVa++ // I’ll ping you
awoods: dumb ?: what time are the standups now? Trying to fix my caledar for next week.10:52
Caledar, of course, is my Elfquest LARP name10:54
<osmandin>barmniator: the standup calls at 11 were removed in favor of irc updates before 1110:55
* scossu joins
<osmandin>sorry, barmintor10:57
<barmintor>osmandin: let me assure you, I might as well have changed my name to ‘barminator’ at this point10:58
<awoods>barmintor: yes, the stand-ups are IRC updates before 11am11:00
<barmintor>awoods: thanks!11:01
* dwilcox leaves11:05
* github-ff joins11:31
[fcrepo4] ajs6f force-pushed Java8 from e8c4a5e to 1124100: http://git.io/F98X
fcrepo4/Java8 cef145f ajs6f: Step to Java 8
fcrepo4/Java8 ba5b16e Aaron Coburn: updated travis-ci configuration
fcrepo4/Java8 efb73b9 ajs6f: Removing obsolete PermGem flag
* github-ff leaves
<barmintor>ajs6f1++11:34
<ajs6f1>barmintor: Thanks, but waht for?
<barmintor>JDK8 heap flag cleanup
and, you know, being nice11:35
<ajs6f1>barmintor: Oh, cool. That's actually from weeks ago— that's a rebase. Check out the branch to see what's up. I'm in a holding pattern now, because I'm several thousand line changes ahead of master now and I don't want to get too far away.
barmintor: And I think you're nice too.
<barmintor>“several thousand line changes ahead of master now and I don't want to get too far away”11:36
“several thousand line changes ahead of master now and I don't want to get too far away”
“several thousand line changes ahead of master now and I don't want to get too far away"
<ajs6f1>barmintor: Have you used Java 8?
<barmintor>no
<ajs6f1>barmintor: Most of those line changes are deletions of massive swathes of -kernel-impl that all collapse into a few lines.
<barmintor>are the changes vast?
oh, that’s nice.11:37
<ajs6f1>barmintor: Not so much, much what's there lends itself to producing concision: simple lambdas, a streaming data API, etc.
barmintor: Think about how much of -kernel-impl is about producing and manipulating streams of RDF triples and you'll see why a lot of it evaporated.11:38
* scossu leaves11:59
* scossu joins12:02
* dwilcox joins12:08
* scossu leaves12:51
<barmintor>mikeAtUVa: https://github.com/barmintor/fcrepo/tree/fcrepo-138412:56
* barmintor goes to a meeting
<mikeAtUVa>barmintor: I'm on it...
* barmintor goes to a meeting grumbling about how hard it is to test this crap
* scossu joins12:57
* scossu1 joins13:02
* scossu leaves13:04
* awead leaves13:14
* awead joins13:15
<mikeAtUVa>barmintor: built the installer, installed it locally, started it, stopped it, replaced the objectStore and datastreamStore, rebuild the sql index, started it again, did an export?context=archive, voila... fixed.13:20
* dwilcox leaves13:49
* dwilcox joins13:50
<barmintor>mikeAtUVa++14:14
<mikeAtUVa>barmintor, thanks for the rapid fix... is there some sort of bat signal in the sky for you when anyone creates an issue referencing fedora 3.8.0?14:15
<barmintor>I get a red-phone call from Commissioner Woods14:16
mikeAtUVa: if you have a chance, please comment that you tested on the issue at https://jira.duraspace.org/browse/FCREPO-138414:19
* acoburn leaves14:28
* mikeAtUVa leaves14:36
* ajs6f1 leaves14:37
* ajs6f joins
* ajs6f1 joins14:38
* ajs6f leaves
* mikeAtUVa joins14:40
* acoburn joins14:54
* osmandin leaves15:05
* dwilcox leaves15:09
* osmandin joins15:31
* github-ff joins15:33
[fcrepo4] osmandin opened pull request #739: Change problematic SPARQL update statement namespace prefix URI (master...1381) http://git.io/xqjq
* github-ff leaves
* awoods leaves15:46
* awoods joins15:49
acoburn: I was not under the impression that fcrepo-camel has a built-in reindexing feature. Does it? Or do you have some sort of external script for reindexing?15:51
<acoburn>awoods: one would have to build it15:52
awoods: but I have code that does that
<awoods>acoburn: scossu1 was mentioning something to the effect that the capability existed...
acoburn: It sounds like that capability was something you created on the side.15:53
<acoburn>awoods: I'll send you a link in just a sec...
<awoods>acoburn: would your code be of general interest?
<acoburn>awoods: probably
awoods: I've been trying to put most of it in the examples/ projects in fcrepo-camel15:54
<awoods>acoburn: that gets us one step closer to deprecating fcrepo-message-consumer
ruebot: how was the Islandora IG call?
<acoburn>awoods: https://github.com/acoburn/artic-camel/blob/master/src/main/java/edu/artic/camel/IndexRouter.java15:55
* dwilcox joins15:56
<ruebot>awoods: very good i think - https://github.com/Islandora/Islandora-Fedora4-Interest-Group/blob/master/meetings/04.md15:57
<awoods>ruebot: it sounded like there was some FCDM discussion15:58
<ruebot>awoods: lots of positive feedback and reinforcement from attendees
<acoburn>awoods: that what the last PR about StreamCaching was about (streaming a root node with a lot of children and pushing each onto a separate queue for processing)
<ruebot>awoods: yeah, a little bit. i talked about it w/r/t where we're at with it. then how folks can get involved in the discussion if they have thoughts/opinions on it15:59
<awoods>ruebot: diego jumped in the fray
<ruebot>awoods: excellent. he is a good egg.
<scossu1>awoods: acoburn and I have also been talking about how to check index consistency periodically. We came up with some interesting ideas.16:00
<awoods>acoburn: It is not immediately clear to me how the IndexRouter.java iterates... but maybe I just need to play with it some.
<acoburn>awoods: it's recursive: starting on line 60 (pull from a queue)16:01
awoods: find children (lines 69-72)
awoods: push each child back onto the same queue (75)
<awoods>acoburn: and what pushes the root onto the queue in the first place?16:02
<acoburn>awoods: a POST request, specifying where to start
https://github.com/acoburn/artic-camel/blob/master/src/main/resources/OSGI-INF/blueprint/camel-context.xml#L45
<awoods>acoburn: thanks16:03
<acoburn>awoods: and that is handed off to the "direct:index" route: https://github.com/acoburn/artic-camel/blob/master/src/main/java/edu/artic/camel/IndexRouter.java#L41
<awoods>acoburn: got it16:04
<acoburn>awoods: which cleans headers and translates the path to a fcrepo-ready path
<awoods>acoburn: I can see this addition being useful in the general fcrepo-camel work... is there any reason you did not include it there?16:05
<acoburn>awoods: it's in the examples directory (though that project needs some more updates)
awoods: the issue is that this is a concrete implementation — the camel component doesn't implement anything per se16:06
<awoods>acoburn: ideally we can get to the point of pulling together a fcrepo-camel package that is immediately usable/deployable.16:07
<acoburn>awoods: yes, it would be something that *uses* fcrepo-camel, but would be separate from it16:08
<awoods>acoburn: yes
<acoburn>awoods: for instance, one could reimplement fcrepo-message-consumer using fcrepo-camel
<ajs6f1>awoods/acoburn: fcrepo-camel-toolbox?16:09
<awoods>acoburn: Would a thin webapp wrapper make sense?
<acoburn>ajs6f1: you mean a set of implemented routes that people could use?16:10
<ajs6f1>acoburn: More or less. it would also serve as a handy box of examples.
<acoburn>ajs6f1: that was sort of the idea of fcrepo-camel/examples, but it would be nice to move those somewhere else
<ajs6f1>acoburn: The "-toolbox" phrasing would be meant to indicate that what's inside is ready-usable stuff.16:11
<acoburn>awoods: a webapp wrapper would make sense — there is a maven artifact that will generate most of the scaffolding for this16:12
<awoods>acoburn: a maven [archetype]?
<acoburn>awoods: yes, an archetype16:13
awoods: there's an example at the bottom of: https://wiki.duraspace.org/display/FEDORA4x/Setup+Camel+Message+Integrations16:14
awoods: under "deployment"
<awoods>acoburn: ++, again16:15
<acoburn>ajs6f1: perhaps fcrepo-camel-toolbox could be added to fcrepo4-labs?16:16
<ajs6f1>acoburn: Makes sense to me.16:17
<awoods>acoburn/ajs6f1: +1
<ajs6f1>awoods: Can only you create a repo, or can I? Or for that matter, can acoburn just do that for himself?16:18
<acoburn>ajs6f1 / awoods: I believe that I can create a repo
<ajs6f1>Cool.
<awoods>acoburn: hmm
ajs6f1: I as I see it, only ecowles, mikeAtUVa, and I can do it at the moment... other Fedora committers should be added to that list.16:19
<acoburn>awoods: I see a green "New Repository" button in fcrepo4-labs (not in fcrepo4)16:20
<ajs6f1>awoods: Whatever you want. I don't think we make new repos so frequently as to be a pain.
<awoods>acoburn: but if you can create a repo, that would be good to know.16:21
<acoburn>awoods: well github tells me that I can. Should that not be so?16:22
<awoods>acoburn: go for it16:23
acoburn: there are three owners of fcrepo4-labs, at the moment: myself, escowles, and mikeAtUVa16:24
acoburn: I would expect that only owners can create new repos
<acoburn>awoods: https://github.com/fcrepo4-labs/fcrepo-camel-toolbox16:26
awoods: ajs6f1 and I are "members" of the fcrepo4-labs organization, which must give us the ability to create repositories16:28
<awoods>acoburn: that is a little surprising. Maybe anyone who is a member of fcrepo4-labs can create new repos
agreed
acoburn: good to know
* dwilcox leaves16:29
<acoburn>ajs6f1: on the topic of repositories, you suggested moving the indexing ontology into fcrepo4-labs16:34
ajs6f1: any thoughts on a repo name for that?
<ajs6f1>acoburn: how about indexing-ontology?
<acoburn>ajs6f1: sounds good16:35
<ajs6f1>acoburn: cool.
<awoods>acoburn/ajs6f1: in which github organization the indexing ontology lives seems less important than where it is host: http://fedora.info/definitions/v4/indexing
<acoburn>ajs6f1: and this hasn't been raised yet, but the oai-pmh ontology is also in the fcrepo4/ontology repo
ajs6f1: does it belong there? I don't know anything about it
<awoods>acoburn: you may want to wait on moving things around16:36
<ajs6f1>acoburn: I have no idea what that is doing there. Perhaps git blame can tell us?
<acoburn>awoods: I'm in no hurry
<ajs6f1>awoods: I disagree, but I don't disagree that the namespace is important. I have no problem with the current namespace, since the indexing ontology is specific to Fedora.
<awoods>ajs6f1: where does your disagreement lie?16:37
<ajs6f1>awoods; I think the code repository is just as important. It represents another kind of responsibility that is just as real as that represented by the hostname in an URL.
<awoods>ajs6f1: I believe the semantics of fcrepo4-labs is "experimental, but a candidate for the main show"16:38
<ajs6f1>awoods: Hm. I didn't see it that way. My understanding was that an fcrepo4-labs home indicates an ancillary product of interest to the Fedora community, but which may or may not receive anything like the same support as the core and for which the whole community specifically does not take responsibility.16:41
awoods: I'm fine either way— if fcrepo4-labs is "migrational"
awoods: then a new home is needed for such things as the indexing ontology.
awoods: Which may very well be the case for other things as well.16:42
<acoburn>awoods/ajs6f1: fcrepo-extras?16:43
<ajs6f1>acoburn: That sounds pretty good.
<acoburn>awoods/ajs6f1: apache has a repository of projects that don't conform to the apache license (GPL and such) and those are put there: apache-extras16:44
<awoods>acoburn/ajs6f1: fcrepo4-extras could be a reasonable option... let's raise it next Thursday, or on the list beforehand if you are daring.16:45
<ajs6f1>awoods: Sounds good16:46
<acoburn>awoods: so I suspect the indexing-ontology would go there?16:47
awoods: and possibly also the fcrepo-camel-toolbox? (following some sort of graduation process?)16:48
<awoods>acoburn: do you think fcrepo-camel-toolbox would have a different status than fcrepo-message-consumer?... or would that also go there?16:49
<acoburn>awoods: no strong opinion either way16:50
<awoods>acoburn: lots of gray area16:51
<acoburn>awoods: perhaps this could be part of the discussion on Thursday (and possibly before on the mailing list)
<awoods>acoburn +1
<acoburn>awoods: in terms of what belongs in fcrepo4/ and what belongs in fcrepo4-extras/16:52
* osmandin leaves16:53
<awoods>acoburn: and how fcrepo4-labs relates to both16:54
<acoburn>awoods: exactly
* mikeAtUVa leaves
* acoburn leaves17:16
* ajs6f1 leaves17:29
* barmintor leaves
* ksclarke leaves18:30
* ksclarke joins18:45
* dhlamb leaves18:46
* scossu1 leaves18:50
* dwilcox joins20:15
* dwilcox leaves21:03
* dhlamb joins22:01
* dhlamb leaves22:10

Generated by Sualtam