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

Using timezone: Eastern Standard Time
* dchandekstark joins01:59
* dchandekstark leaves02:04
* thomz joins02:33
* cmmills leaves03:36
* cmmills joins03:37
* tjohnson leaves04:57
* dchandekstark joins05:01
* dchandekstark leaves05:06
* peichman joins08:10
* peichman leaves08:11
* coblej joins08:19
* esm_ joins08:25
* dwilcox joins08:40
* esm_ leaves08:41
* bseeger joins08:56
* whikloj joins08:57
* esm_ joins09:08
* acoburn joins09:11
* ajs6f joins09:27
* dchandekstark joins09:34
* dchandek_ joins09:38
* dchandek_ leaves
* dchandekstark leaves09:39
* apb18 joins09:52
* coblej leaves09:56
<ajs6f>acbourn: Tag me in for any help with that GraphStore thing, although I think it's pretty straightforward.10:09
<acoburn>ajs6f: It's tedious but very straight forward
<ajs6f>acoburn: Yeah, I get the feeling that we could do a lot to factor around in the ITs for kernel-modeshape up, but I just don't care enough to try.10:10
<acoburn>ajs6f: what I've got so far is changing the CloseableGraphStore is now a CloseableDataset, which extends DatasetImpl
ajs6f: so then the CloseableDataset's can be used in try-with blocks10:11
but they need to be cast to DatasetGraphs first via Dataset::asDatasetGraph
it's mostly the http.api.FedoraLdpIT file that needs work10:12
<ajs6f>acoburn: Did you invent CloseableDataset?
<acoburn>ajs6f: yes, in the same way we currently have CloseableGraphStore
ajs6f: my understanding (c.f. the recent msg on the jena list) is that Datasets aren't autocloseable10:13
* mikeAtUVa joins10:15
<ajs6f>acoburn: Hm, okay, I'd go ahead and do what you're doing, but depending on the form of the tests, it might be easier to have a static method Consumer<Consumer<Dataset>> doWithDataset(Dataset d) { return action -> { try { action.accept(d) ; } finally { d.close(); } } }
Sorry for the formatting.
acoburn: Because Datasets indeed are not AutoCloseable, and that's for a reason. I think I actually introduced the CloseableGraphStore, and that was a mistake. I didn't understand Jena very well then.10:16
<acoburn>ajs6f: yeah, there's _a lot_ of refactoring that could be done with the ITs to clean them up. your suggestion make a lot of sense, but I wonder if it might make more sense in the context of a larger refactoring of the ITs10:17
<ajs6f>acoburn: Yeah, probably.10:18
acoburn: With all our extra cycles and free time, let's do that.
ajs6f: mostly I just want to get this jena PR into master with as little work as possible and as soon as possible10:19
awoods:dwilcox: Who is doing the NY Fedora Camp with you?10:22
<dwilcox>ajs6f: Ben, Adam Wead, Trey, and @ruebot. Diego may join us as well.10:23
<ajs6f>dwilcox: Erm. That's a bit Hydra-heavy, but oh, well. I hope Diego can make it, but that's abit early for him, isn't it?
<dwilcox>ajs6f: Not sure what you mean by early?10:24
<ajs6f>dwilcox: I thought he as going to arrive in NY a little late than that, but I'm probably getting confused.
<dwilcox>ajs6f: Ah, no he said he will be there by then
<ajs6f>dwilcox: Cool.10:25
* dchandekstark joins10:26
* dchandekstark leaves
* dchandekstark joins
* mikeAtUVa leaves10:27
* mikeAtUVa joins10:32
* dwilcox leaves10:48
* thomz leaves
* dwilcox joins11:00
* dchandekstark leaves11:12
* dchandekstark joins11:13
* dchandekstark leaves11:17
* dchandekstark joins11:18
* bseeger leaves11:19
* acoburn leaves11:29
* dchandekstark leaves11:51
* tjohnson joins
* mjgiarlo joins11:52
* dchandekstark joins12:52
* bseeger joins12:54
* rfischer joins12:55
* dchandekstark leaves12:57
* dchandekstark joins12:58
* acoburn joins13:00
* dchandekstark leaves13:03
* dchandekstark joins13:12
* ajs6f leaves13:31
* dchandekstark leaves13:33
* dchandekstark joins13:40
* bseeger leaves
* dchandekstark leaves13:44
* github-ff joins13:47
[fcrepo-module-auth-rbacl] acoburn opened pull request #35: Migrate to Jena 3.x (master...fcrepo-1672) https://git.io/vKbUk
* github-ff leaves
* dchandekstark joins13:48
* github-ff joins13:50
[fcrepo-module-auth-xacml] acoburn opened pull request #54: Migrate to Jena 3.x (master...fcrepo-1672) https://git.io/vKbUC
* github-ff leaves
* bseeger joins13:51
* github-ff joins13:54
[fcrepo-module-auth-webac] acoburn opened pull request #71: Migrate to Jena 3.x (master...fcrepo-1672) https://git.io/vKbU9
* github-ff leaves
* github-ff joins13:59
[fcrepo-audit] acoburn opened pull request #38: Migrate to Jena 3.x (master...fcrepo-1672) https://git.io/vKbTi
* github-ff leaves
* dchandekstark leaves14:07
* dchandekstark joins14:08
* bseeger leaves14:11
<acoburn>awoods:ajs6f: the fcrepo-transform module isn't compatible w/ Jena 3.x b/c Marmotta's jena backend expects a 2.x Model14:14
any ideas?
maybe swap out the jena backend for the linkeddata or sesame backend?
waiting for Marmotta to make a new release (w/ Jena 3.x) would be a bad idea14:15
b/c we may be waiting for a very long time14:16
<awoods>acoburn: ajs6f has mentioned moving off of the jena backend anyways...14:17
acoburn: this may be one more reason to actually do that.
<acoburn>awoods: should this be a blocker for moving to jena 3.x? (I hope not)14:18
awoods: b/c I don't have much more time to work on this
<awoods>acoburn: I don't think we can just drop fcr:transform
<acoburn>awoods: that's not what I'm suggesting
<awoods>acoburn: what are you suggesting?14:19
<acoburn>awoods: I'm suggesting moving forward with the jena 3.x migration and have fcrepo-transform catch up next week
awoods: we can't have an extension module blocking the forward movement of the core codebase
If you need fcrepo-transform, don't upgrade yet
<awoods>acoburn: does fcr:transform still work if the rest of the codebase is on jena 3.x?
<acoburn>awoods: no it doesn't
awoods: that's why I'm suggesting moving off of the jena backend to some other backend14:21
awoods: that's the fix — it's clear and straight forward, but I can't do it today
<awoods>acoburn: is there a particular urgency in getting jena 3.x into master?14:22
<acoburn>awoods: if only our API used RDF-commons…
awoods: it is an enormous PR, and I don't want to have to deal with rebasing
awoods: meaning that all the other "big" PRs are waiting on this to be merged14:23
<awoods>acoburn: as far as I know, you are the only one working on other big PRs.
<acoburn>awoods: right, meaning I'd need to rebase, which I don't want to do
awood: if you want to wait, that's fine, but I'll be waiting, too14:24
<awoods>acoburn: I don't like the idea of having the codebase in a state that is somewhat broken.
<acoburn>awoods: it's an extension module, not the core code base14:25
awoods: that's a _huge_ difference
awoods: we can discuss it tomorrow
* acoburn leaves
<awoods>whikloj: on tomorrow's call, mention that I am willing to do the rebase if that is the only reason to push for jena 3.x sooner than later.14:34
<whikloj>awoods: ok
* barmintor joins14:37
* barmintor leaves14:38
* rfischer leaves14:45
* bseeger joins14:59
* esm_ leaves15:05
* esm_ joins15:11
* ajs6f joins15:13
* apb18 leaves15:30
* bseeger leaves15:42
* dchandekstark leaves15:44
<whikloj>esm_: You are testing/using Postgres with Fedora right? Do you find it slower to start the first time?15:47
* dchandekstark joins
<esm_>whikloj: as far as i know we (JHU) aren't using Postgres with Fedora 4, but I could be mistaken :)15:56
<whikloj>esm_: oops, my mistake15:57
<esm_>haha no worries
* ajs6f leaves15:58
<esm_>we aren't officially in production with FF, but will be at some point soon-ish...
<whikloj>cool, I was just trying it out and it took 10 minutes for Tomcat to come up
I wondered if it was normal for the initial database creation or something15:59
I'll see if a restart is faster16:00
<esm_>whikloj: yikes :)16:02
* dwilcox leaves16:12
* dchandekstark leaves16:16
* dchandekstark joins16:22
* dchandekstark leaves16:26
* dwilcox joins16:33
* dwilcox leaves16:37
* dchandekstark joins16:43
* mikeAtUVa leaves16:53
* whikloj leaves17:02
* esm_ leaves17:07
* bseeger joins17:11
* bseeger leaves17:13
* barmintor joins17:24
* dchandekstark leaves
* dchandekstark joins17:47
* dchandekstark leaves17:59
* mjgiarlo leaves20:19

Generated by Sualtam