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

Using timezone: Eastern Standard Time
* barmintor leaves01:28
* ksclarke leaves02:39
* kaarefc joins03:08
* kaarefc leaves
* kaarefc joins03:31
* awoods leaves03:40
* kaarefc leaves03:45
* bljenkins leaves
* escowles leaves
* mikeAtUVa leaves
* cbeer leaves
* pivotal-bot leaves
* kaarefc joins03:47
* barmintor joins
* escowles joins
* mikeAtUVa joins
* cbeer joins
* pivotal-bot joins
* bljenkins joins
* kaarefc leaves03:49
* kaarefc joins03:55
* kaarefc leaves04:11
* kaarefc joins04:47
* fasseg joins04:49
* awoods joins05:56
* nbanks joins08:01
* nbanks leaves08:05
* jonathangee joins08:27
* nbanks joins08:50
* github-ff joins09:00
[fcrepo4] mikedurbin closed pull request #148: Improve HTML UI to de-emphasize inlined resources and property 'noise'. ... (master...html-ui) http://git.io/Iu572A
* github-ff leaves
<mikeAtUVa>Hmm.. didn't mean to close.09:01
<awoods>mikeAtUVa: you should be able to unclose09:02
* github-ff joins09:03
[fcrepo4] mikedurbin reopened pull request #148: Improve HTML UI to de-emphasize inlined resources and property 'noise'. ... (master...html-ui) http://git.io/Iu572A
* github-ff leaves
<mikeAtUVa>Yeah... I hit "comment" and nothing happened, so I hit "comment and close" which of course closed the pull request, not the comment box.09:04
I'm probably a poor choice to make UI recommendations;)
* github-ff joins09:06
[fcrepo4] fasseg created repl-timeout (+1 new commit): http://git.io/ncpHDQ
fcrepo4/repl-timeout 1092b65 fasseg: added system property for infinispan replication timeout in clustered configuration
* github-ff leaves
* github-ff joins09:08
[fcrepo4] fasseg opened pull request #150: added system property for infinispan replication timeout in clustered configuration (master...repl-timeout) http://git.io/r0ysaA
* github-ff leaves
<fasseg>are you guys alright with merging https://github.com/futures/fcrepo4/pull/150 into master? It's just an additional system property for infinispan...
* escowles leaves09:09
* escowles joins09:10
<awoods>fasseg: sure, but please add the new parameter to: https://wiki.duraspace.org/display/FF/Deploying+Fedora+409:14
<awoods>fasseg: or maybe there is a new subsection on that page for cluster config parameters
* ermadmix joins
<pivotal-bot>Mike Durbin added "Ensure that the implications of versioning are compatible with policy enforcement" https://www.pivotaltracker.com/story/show/6076942609:17
Mike Durbin added "Review and possibly revise or extend the fcrepo-http-api endpoints relating to versions to ensure they are useful for..." https://www.pivotaltracker.com/story/show/6076951409:18
Andrew Woods accepted "FileSerializer swallows exception" https://www.pivotaltracker.com/story/show/5884534009:19
<awoods>mikeAtUVa: Did you see the note on this item: https://www.pivotaltracker.com/story/show/59056596
<pivotal-bot>feature: Improve the HTML rendering of the fedora 4 rest API to make it more welcoming to new users (finished) / owner: Chris Beer
<awoods>mikeAtUVa: no rush, just checking09:20
<mikeAtUVa>awoods: yeah, that's the pull request on which I just commented... should I comment on the ticket too?
<pivotal-bot>Andrew Woods edited "Ensure that the implications of versioning are compatible with policy enforcement" https://www.pivotaltracker.com/story/show/60769426
<awoods>mikeAtUVa: no
* travis-ci joins09:22
[travis-ci] futures/fcrepo4#1171 (repl-timeout - 1092b65 : fasseg): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/commit/1092b65a1e70
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/13970331
* travis-ci leaves
* github-ff joins09:25
[fcrepo4] fasseg pushed 1 new commit to master: http://git.io/ceVI-w
fcrepo4/master 3521f0c frank asseg: Merge pull request #150 from futures/repl-timeout...
* github-ff leaves
* kaarefc leaves09:26
<bljenkins>Project fcrepo-fixity-corrupter build #422: SUCCESS in 1 min 12 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/422/09:27
Project fcrepo-fedora3-federation-connector build #200: UNSTABLE in 4 min 44 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fedora3-federation-connector/200/09:31
* osmandin joins09:45
<bljenkins>Project fcrepo-fixity-corrupter build #423: SUCCESS in 1 min 31 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/423/09:47
* ksclarke joins09:49
<bljenkins>Yippie, build fixed!09:54
Project fcrepo-fedora3-federation-connector build #201: FIXED in 8 min 21 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fedora3-federation-connector/201/
Project fcrepo-kitchen-sink build #630: STILL FAILING in 7 min 42 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-kitchen-sink/630/09:55
* ajs6f joins10:19
<pivotal-bot>A. "Senphitic" Soroka added comment: "https://wiki.duraspace.org/display/FF/Fedora+3+%3D%3E+Fedora+4+migration" https://www.pivotaltracker.com/story/show/5157176710:42
* ajs6f leaves10:54
<pivotal-bot>Mike Durbin started "Review and possibly revise or extend the fcrepo-http-api endpoints relating to versions to ensure they are useful f..." https://www.pivotaltracker.com/story/show/6076951410:56
* edInCo joins10:59
* ajs6f joins
<awoods>U.S.A/Canada toll free: 866-740-1260, participant code: 2257295
* github-ff joins11:02
[fcrepo4] mikedurbin opened pull request #151: Added basic version creation on all REST API updates commands. (master...master) http://git.io/DVG27w
* github-ff leaves
<pivotal-bot>Mike Durbin added comment: "https://github.com/futures/fcrepo4/pull/151" https://www.pivotaltracker.com/story/show/5961426811:03
<awoods>cbeer, barmintor?
<pivotal-bot>Mike Durbin finished "Simple Versioning" https://www.pivotaltracker.com/story/show/59614268
<cbeer>awoods: there.
<barmintor>Gaah, this call is not on G+11:04
<ajs6f>barmintor: No one is on G+, right?
That's what everyone says.
<barmintor>ajs6f: it was just me
<awoods>barmintor: U.S.A/Canada toll free: 866-740-1260, participant code: 225729511:05
<barmintor>awoods++ // thanks
<ajs6f>An army of one.
* github-ff joins11:16
[fcrepo4] ajs6f created KillWarnings (+1 new commit): http://git.io/r2F64g
fcrepo4/KillWarnings d551158 ajs6f: Simple fixes for type warnings
* github-ff leaves
* github-ff joins11:18
[fcrepo4] ajs6f opened pull request #152: Simple fixes for type warnings (master...KillWarnings) http://git.io/pNa_kA
* github-ff leaves
<cbeer>ajs6f: when we go back to your streaming thing, i'm not sure how we're supposed to handle HTML responses that way.11:21
<ajs6f>cbeer: DIfferently.
barmintor: we have machinery for that.
<ajs6f>cbeer: Given that HTML generally isn't huge, I figured we'd just roll up the RDF stream into a Jena-style in-heap structure and give that to the templates.11:22
<cbeer>barmintor: https://github.com/futures/fcrepo4/blob/master/fcrepo-kernel/src/main/java/org/fcrepo/kernel/TxAwareSession.java
scary black magic.11:23
awoods: should we table this discussion for later?11:25
<awoods>cbeer: as long as it will happen later11:26
<barmintor>we're always covering about half the remaining distance to JAXRS-211:27
<ajs6f>Zeno's library.
<cbeer>hm. sounds like my next ruby gem name.
scope is fixed, time is fixed, and resources are out of our control.11:30
<ajs6f>Well, if everything is fixed, then we are bug-free. yay!
<barmintor>cbeer: are you suggesting we have not recognized the iron bounds of our triangle?11:31
<pivotal-bot>A. "Senphitic" Soroka added "Make property values in HTML UI editable" https://www.pivotaltracker.com/story/show/6078128011:39
<cbeer>huh. when did our test running data start ending up in $TMPDIR?11:40
that's probably why we started having all those PID conflicts and whatnot11:41
<escowles>cbeer: pretty recently -- noticed when mvn clean didn't reset repo state
<ajs6f>Bad substitution?
<bljenkins>Project fcrepo-fixity-corrupter build #424: SUCCESS in 1 min 38 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/424/
<ajs6f>Oh, wait, you mean system tmp, like /tmp?
<pivotal-bot>Chris Beer added "Don't write data from running tests into $TMPDIR" https://www.pivotaltracker.com/story/show/60781524
<cbeer>ajs6f: or, on OS X, /var/folders/v3/......../T/
<barmintor>Dockerized clusters.11:43
<ajs6f>barmintor: want a ticket?
<barmintor>ajs6f: more generic than docker, but yes: I think we should deliver a node-image recipe ASAP.11:45
<ajs6f>barmintor: Like something that runs in VirtualBox or the ike?
<bljenkins>Project fcrepo-kitchen-sink build #631: STILL FAILING in 4 min 23 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-kitchen-sink/631/
<ajs6f>"scope fence".11:46
<cbeer>it's a very big fence.
<ajs6f>And full of barbed wire.
"beta fence"11:48
Zeno's beta.
<barmintor>We could go down this route: http://docs.oracle.com/javaee/7/api/javax/persistence/EntityManager.html11:58
<ajs6f>JavaEE, the Nurse Rachet of our little world.11:59
* barmintor shrugs
it insulates http-commons and http-api from jcr*
<ajs6f>Truth. And it's a well-known, commonly-used set of gear.
Lots and lots of frameworks happily sit over javax.persistence right out of the box.12:00
<mikeAtUVa>there goes my versioning pull request;)12:01
<barmintor>one problem with javax.persistence is that it's very jdbc-centric12:02
though it also has a graph notion
<ajs6f>It has a graph notion? I didn't know that. That's _something_.12:04
<ajs6f>I've seen graphstores (e.g. neo4j) support it, to some extent.
<barmintor>properties and subgraphs
<nbanks>The latest build isn't passing tests for me on OSX unless I do it as a separate step like Travis does "mvn test -B" is this acceptable?
<ajs6f>barmintor: "for a find operation or query." kinds of weirds me out. Using Dataset from Jena was weird (IMHO) partly because it was intended for query, and we were/are using it to pass around mutations.12:05
* barmintor 's shoulders are tired from all the helpless shrugging he does
<ajs6f>barmintor: what about combining a graph api and a content api? e.g. clerezza + a few bitstream-centric methods...12:07
and take only the xaction-related stuff from JavaEE.12:08
<barmintor>ajs6f: TX and versions?
versions… dunno. I can't think offhand of a really nice API I liked for that.
Myabe just translate the jCR-native APi?
afk bbl
<barmintor>we could keep exposing the JCR apis, but claim that kernel has a different implementation, and that the exposed JCR is provided as an abstraction over HEY WHAT'S THAT BEHIND YOU (runs)12:09
<cbeer>ajs6f: i've just run into an unfortunate limitation about JCR properties, and hope you have some brilliant insight.
(and i confirmed with rhauch that this is expected behavior :/)12:11
http://www.day.com/specs/jcr/2.0/3_Repository_Model.html#3.6.3 Single and Multi-Value Properties declares that the values of a multivalued property have to be a single type.
which means if you have something like this in your CND:12:12
- dc:title (*) multiple
you can't assign both a string and a URI to dc:title (dc:title was a bad example, i guess.)
<barmintor>cbeer: that makes sense from the impl side12:13
<cbeer>i guess DCES just pretends everything is a literal string?
it just means if we did the whole dcterms vocab, if we said something was a URI, we really mean it.
it also means no mixing and matching URIs and REFERENCEs12:14
<barmintor>reference == node ref?
<cbeer>yeah. i guess we could translate URIs to nodes into some UUID-based URI
<barmintor>not being able to have a property be a ref or a URI troubles me much more than not being bale to mix strings and URIs12:15
but we could hack that.
<cbeer>it's just kinda weird from RDF and SPARQL-update where you can assert something we can't persist12:16
<barmintor>what does Sparql advise if you assert something in an update that is out of range/domain for a predicate?
* ksclarke leaves12:17
<cbeer>hm. i assumed it wouldn't care. maybe i should check that
* github-ff joins12:18
[fcrepo4] cbeer force-pushed sparql from 13faf53 to 1e052a1: http://git.io/Rl-hQw
fcrepo4/sparql d782f23 Chris Beer: define all the Dublin Core elements as pre-defined properties, and use the dc/elements/1.1 namespace instead of dcterms.
fcrepo4/sparql 49b8098 Chris Beer: Implement basic SPARQL-based administrative search
fcrepo4/sparql 1e052a1 Chris Beer: wire in a sparql search form.
* github-ff leaves
<awoods>Coming from nbanks comment: Is there anyone who can not currently build the fcrepo4: "mvn clean install"?12:20
<cbeer>awoods: what comment?
<awoods>IRC @12:04
I, for one, require the following command to build, which is unfortunate: "mvn clean install -Djacoco.skip"12:21
osmandin, nbanks, escowles, mikeAtUVa, ermadmix, fasseg, cbeer, barmintor: Could you please give a "build +1" or "build -1" if you can or can not successfully run "mvn clean install"12:24
<cbeer>build +1
<barmintor>awoods: from master?
<fasseg>build-1 jacoco plugin hangs quite regular for me12:25
<nbanks>build -1
<cbeer>(well, no, -1. i suppose I do something with heap and permgen.)
<fasseg>and permgen blows up, though i got -XX:MaxPermSize=1024m which in my mind is too much anyway ;)
<awoods>cbeer: what do you do with heap and permgen?
<cbeer>same thing i've always had to do. MAVEN_OPTS="-Xmx1g -XX:MaxPermSize=512M"12:26
<nbanks>I have the same problems with permgen and jacoco
<barmintor>nbanks: try tagging https://github.com/futures/fcrepo4/blob/master/fcrepo-http-api/src/test/java/org/fcrepo/integration/http/api/html/FedoraHtmlResponsesIT.java with @Ignore and re-running12:27
<fasseg>I thikn there was fix using the <jvmFlags> property in a maven plugin, telling the invoked jvms to use a larger permgen...12:28
<nbanks>capital "I" @Ignore?
<barmintor>nbanks: yes
<nbanks>how does tagging the entire class work, do I place it right after the class declaration?12:29
<awoods>nbanks: try adding @Ignore to line 4212:30
<barmintor>nbanks: before
<awoods>is there any value in running jacoco on local builds? Is this something we can add to a profile that gets invoked on the CI server?12:32
<barmintor>awoods: Good luck.12:33
<barmintor>eddies and I probably spent a week each trying to get something along those lines to work
and still get propagated to sonar
<awoods>barmintor: that is too bad12:34
<fasseg>Oh I think the build problem might be a bug shouldn't this read: ${failsafe.argLine}: https://github.com/futures/fcrepo4/blob/master/pom.xml#L817
<barmintor>I will freely admint my maven amateurism
<awoods>fasseg: That does look like a typo. Does your build work with the typo fixed?12:36
<barmintor>fasseg: trying with that change now
<fasseg>just running the build will let you know
<barmintor>that's ominous: https://github.com/futures/fcrepo4/blob/master/pom.xml#L90012:37
<nbanks>baramintor: success!
<barmintor>nbanks: with the @Ignore?12:38
<nbanks>barmintor: yup
<barmintor>damn you, httpUnit
<fasseg>build+ with the change12:39
<barmintor>I saw some threads that suggested certain versions of Java make that http response testing library blow up permgen, which probably doesn't play well with the default heap settings
<escowles>awoods: build +1
<awoods>escowles: you can build without argLine updates or @Ignore updates?12:40
<escowles>awoods: yes, tho my opts are: -server -Dfile.encoding=UTF-8 -Djava.awt.headless=true -Xms512m -Xmx2048m -XX:NewSize=256m -XX:MaxNewSize=256m -XX:PermSize=256m -XX:MaxPermSize=256m -XX:+DisableExplicitGC
<awoods>escowles: Those are MAVEN_OPTS or JAVA_OPTS?12:41
<escowles>awoods: both -- i set them to the same thing to make sure fcrepo3 and fcrepo4 were using the same mem settings12:42
<barmintor>build +1 with the pom fix and without the @Ignore
and with cbeer's heap settings
<awoods>can someone commit the argLine pom.xml fix?12:44
<barmintor>awoods: that line in the pom is obviously a bug, can we just push it to master?
<fasseg>lemme build agina to verify, then ill commit it
<awoods>@Ignore on FedoraHtmlResponsesIT successfully built for me.12:45
<barmintor>fasseg: done
<fasseg>kk thx
* github-ff joins
[fcrepo4] barmintor pushed 1 new commit to master: http://git.io/AnRW3w
fcrepo4/master 4e16031 Benjamin Armintor: fix a property spec bug in root pom
* github-ff leaves
<bljenkins>Project fcrepo-fixity-corrupter build #425: SUCCESS in 1 min 44 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/425/12:46
<barmintor>awoods: that test is a permgen hotspot, but it ran fine for me if jverify got the correct argLine
<nbanks>barmintor: thanks
<barmintor>ok, now somebody tell me who is logging out my jcr sessions12:47
<awoods>barmintor: I just updated master with your commit... running "mvn clean install"... results pending
<barmintor>awoods: MAVEN_OPTS="-Xmx1g -XX:MaxPermSize=512M" mvn clean instal
MAVEN_OPTS="-Xmx1g -XX:MaxPermSize=512M" mvn clean install
<awoods>my MAVEN_OPTS="-Xmx1024m -XX:MaxPermSize=1024m"12:48
should be fine
* ksclarke joins
* ksclarke leaves12:54
<fasseg>MAVEN_OPTS="-Xmx1024m -XX:MaxPermSize=256m" mvn clean install works now for me too
-Xmx 512m gives test failures though
* ksclarke joins12:55
<awoods>build +1
<barmintor>fasseg++ // nice catch
<awoods>Next: cbeer mentioned issue with test content being stored in $TMPDIR12:57
<escowles>awoods: i don't care if it's in TMPDIR or somewhere else, but it's nice to be able to specify dir for all fcrepo content without having to specify a bunch of different mode/ispn properties12:58
<awoods>There are three situations that we need to cover: Integration tests, jetty/tomcat deploy with user-defined SystemProperties, and jetty/tomcat deploy with not defined SystemProperties
escowles: agreed, although I think that is a separate issue
escowles: actually, can you put in a ticket for that
which of the three situations were you flagging, cbeer?12:59
<cbeer>awoods: tests.
* travis-ci joins13:00
[travis-ci] futures/fcrepo4#1178 (master - 4e16031 : Benjamin Armintor): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/3521f0c7a5a4...4e1603136f5e
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/13982166
* travis-ci leaves
<awoods>cbeer: hmm, I thought that was resolved since the test system properties are defined in our failsafe pom config.
cbeer: which project/module in particular is test content going to tmp with?13:01
<pivotal-bot>Esme Cowles added "Single system property to specify directory for all fcrepo content" https://www.pivotaltracker.com/story/show/60788492
<awoods>escowles: thanks
<cbeer>awoods: i don't remember. probably fcrepo-transform
<awoods>cbeer: I will start there...13:02
<ajs6f>awoods: With MAVEN_OPTS=' -Xmx2048M -XX:MaxPermSize=256M ', plain mvn clean install worked fine for me.13:03
<osmandin>awoods: +113:04
<awoods>ajs6f: great, I think the build is now working for everyone (at least: fasseg, nbanks, ajs6f, cbeer, awoods, barmintor, and escowles)
thanks for the vote of encouragement, osmandin.
<bljenkins>Project fcrepo4 build #1374: UNSTABLE in 19 min: http://ci.fcrepo.org/jenkins/job/fcrepo4/1374/13:05
armintor: fix a property spec bug in root pom
<awoods>osmandin: is your build working?
* github-ff joins
[fcrepo4] ajs6f force-pushed KillWarnings from d551158 to 51558ee: http://git.io/l_ehEg
fcrepo4/KillWarnings 51558ee ajs6f: Simple fixes for type warnings
* github-ff leaves
* ermadmix leaves13:06
<ajs6f>barmintor: are you working on the "session lifecycle" thing you mentioned earlier?13:11
<osmandin>awoods: yes
<barmintor>ajs6f: yes
ajs6f: for some reason, the session-in-the-transaction is closed when it gets to FedoraNodes
<ajs6f>Are there other Jersey2 things I can help with? I'm asking because the things that're on my plate should wait for Jersey2...13:12
wait, closed when it _gets_ to FedoraNodes?
After it was injected but before it was used?
<barmintor>ajs6f: I put a log stmt in the finally block of FedoraNodes.describe, and it suggests the sesison is no longer live13:13
<ajs6f>lemme get your branch...
I don't remember what FedoraNodes looks like anymore.
<pivotal-bot>Esme Cowles estimated "Make property values in HTML UI editable" as 3 points https://www.pivotaltracker.com/story/show/60781280
Esme Cowles started "Make property values in HTML UI editable" https://www.pivotaltracker.com/story/show/60781280
<osmandin>afk (lunch)13:17
* osmandin leaves
<pivotal-bot>Esme Cowles added comment: "The SPARQL Update statements to add and delete properties are: ""13:18
insert data { <http://localhost:8080/rest/ob..." https://www.pivotaltracker.com/story/show/60781280
Esme Cowles edited "Make property values in HTML UI editable" https://www.pivotaltracker.com/story/show/60781280
* travis-ci joins
[travis-ci] futures/fcrepo4#1179 (KillWarnings - 51558ee : ajs6f): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/d55115815281...51558ee93ae3
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/13983270
* travis-ci leaves
<ajs6f>barmintor: just trying to catch up with you: where is the Jersey2 injection stuff for sessions? I can't find it in http-commons or -api?
<barmintor>yes you can13:19
<ajs6f>I' glad you believe in me.
<barmintor>ajs6f: https://github.com/futures/fcrepo4/blob/jersey2/fcrepo-http-commons/src/main/java/org/fcrepo/http/commons/session/AuthenticatedSessionProvider.java
<ajs6f>ah, thx!
<barmintor>ajs6f: also https://github.com/futures/fcrepo4/blob/jersey2/fcrepo-http-commons/src/main/java/org/fcrepo/http/FedoraApplication.java13:20
<ajs6f>barmintor: So we're not "discovering" components any more?13:21
We're calling the out?
(meaning just the exceptionmappers)
<barmintor>ajs6f: that's not by design
that's a bug to be worked out later
<barmintor>ajs6f: See lines 46-47 in FedoraApplication13:22
<awoods>cbeer: mvn clean install -pl fcrepo-transform did not leave anything in $TMPDIR
cbeer: I will try a full build
<ajs6f>ajs6f: Oh, somehow my eye skipped that. Shall I do a little research on classpath scanning vs. Maven vs. Jersey2?
<bljenkins>Yippie, build fixed!
Project fcrepo4 build #1375: FIXED in 17 min: http://ci.fcrepo.org/jenkins/job/fcrepo4/1375/
<ajs6f>On it.
<bljenkins>Project fcrepo-fixity-corrupter build #426: SUCCESS in 1 min 6 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/426/13:24
<barmintor>Hmm. When the session is injected, it is live. But it is closed when FedoraNodes exits13:27
<bljenkins>Project fcrepo-kitchen-sink build #632: STILL FAILING in 3 min 31 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-kitchen-sink/632/
<ajs6f>Are you sending a no-tx-session through a method with a finally{} block?
<fasseg>MAVEN_OPTS="-Xmx1024m -XX:MaxPermSize=128m" is enough it seems for building
<barmintor>ajs6f: the second time through it's whatever session comes from Transaction.getSession13:28
<ajs6f>barmintor: That ought to return a tx-aware-session. Weird.13:29
<barmintor>sidenote: if we care about performance, we will eventually replace those dynamic proxies with delegating wrappers. Reflection is slooooow.13:30
like, YourKit points at you and laughs
<ajs6f>barmintor: We would have if we could have, but maybe we can now, with the scouring power of Jersey2?
<barmintor>c'est posible
<awoods>fasseg: https://wiki.duraspace.org/pages/diffpages.action?pageId=34654261&originalId=39027858
<ajs6f>Cleaner, brighter, and type-safe!
<awoods>fasseg: just to give a little margin13:31
<fasseg>wiggle room is always a good thing
<awoods>fasseg: that is what we had document here as well: https://wiki.duraspace.org/display/FF/Java+HotSpot+VM+Options+recommendations13:32
<ajs6f>barmintor: I want to know one thing: why did you have to explicitly register the RdfProvider but not the HtmProvider… I think that's a clue to this classpath scanning thing.
<barmintor>b/c RdfProvider is in http-commons, I hypothesize13:34
<awoods>cbeer: fcrepo-kernel is leaving some artifacts in $TMPDIR. Other projects may be as well... still looking.13:35
<cbeer>barmintor: i'm thinking about solving the URI <=> Reference issue by adding a special reference property for any URIs that point to fcrepo4 objects13:40
that way we can pick them up in queries and the like as expected
<ajs6f>cbeer: a "well-known" name?13:41
<cbeer>ajs6f: yup. propertyName + _ref right now
<ajs6f>Modified Reverse Hungarian Notation.
<cbeer>it's not great, but i'm out of ideas.
<barmintor>cbeer: My only otheridea would be to store all references as URIs, and hack the protocol bit with a handler we provide13:44
<ajs6f>barmintor: we could lose some perf on queries. References are much faster than storing literal paths.13:45
<barmintor>this is for situations where the range of a prop is {uri, ref}13:47
<ajs6f>yeah. I just mean that we want to use ref when we can.
<barmintor>I think it's acceptable to say: "We'll resolve URIs in this form as paths if you must have mixed content, but it's faster to distinguish reference properties"13:49
did we ever switch to a longer line length? I'm still getting the 80 character warnings (on barmiintor's jersey2 branch)13:55
<barmintor>ajs6f: my branch will be rebased when it works13:57
<ajs6f>barmintor: cool. I just wanted to make sure we actually did decide to do that.
<awoods>ajs6f: I update the build-helper rules to create a checksum error at 120 characters...13:58
<ajs6f>awoods: Checksum error?
<awoods>ajs6f: Yes, but since the artifact was -SNAPSHOT, it was not available in maven-central13:59
* ermadmix joins
<awoods>ajs6f: so it will need to be released as a non-SNAPSHOT version... or we make it available in another maven repository.
<ajs6f>I'm talking about building from source. Am I missing something? Shouldn't it just be an adjustment to the checkstyle config?
* osmandin joins14:00
<awoods>ajs6f: yes14:05
ajs6f: have you traced how that machinery works?
ajs6f: here is the update: https://github.com/futures/fcrepo-build-tools/commit/c688b372077e0affff80f7e3dd25b14f30bc02cd14:06
<ajs6f>awoods: yeeeesss. That's what I would have thought. So what has SNAPSHOT and Maven releases got to do with anything? Can't I just build fcrepo-checkstyle locally and… that's it?14:09
{whine whine whine}
<awoods>ajs6f: yes, if you build fcrepo-build-tools locally, and change your fcrepo4/pom.xml to build-tools-version to 1.4-SNAPSHOT: https://github.com/futures/fcrepo4/blob/master/pom.xml#L2414:11
it will work
ajs6f: however, if you commit the fcrepo4/pom.xml update, our ci builds will fail.
<ajs6f>awoods: I'm not going to do that.14:12
<ajs6f>awoods: The only things in fcrepo-build-tools are a couple of XML configs, right?
awoods: Why wouldn't we just put them in fcrepo4/
Why a separate artifcat?
<awoods>ajs6f: good question
<ajs6f>Well, there ya go.14:13
afk bb i 5
* ajs6f leaves14:17
* ajs6f joins14:19
<barmintor>resource.getPropertiesDataset(subjects, offset, realLimit) closes the session somewhere along the way14:41
<ajs6f>barmintor: when you said that the classpath scanning thing didn't work in Maven, did you mean while running tests, or while trying jetty-run, or both?
<barmintor>jetty-run is a thing that happens after mvn clean install works14:42
<ajs6f>ok, so during tests?
<ajs6f>barmintor: getPropertiesDataset() should just call JcrRdfTools, which should just hand off to the RDF stream generators, which should _never_ close the session.14:43
I would think we would have anoticed by now if they accidentally didi.
lots of "should"s there.
<barmintor>right before that line, session is live. that line throws an exception b/c it is no longer live
<ajs6f>time to trace14:44
* nbanks leaves14:48
<pivotal-bot>Mike Durbin added "Add versioning documentation/walkthrough to the wiki" https://www.pivotaltracker.com/story/show/60798758
Mike Durbin edited "Add versioning documentation/walkthrough to the wiki" https://www.pivotaltracker.com/story/show/60798758
<cbeer>ajs6f: do you know how i can do something like this with mockito?
verify(mockNode, times(1)).setProperty("mockPropertyName_ref", new Value[] { });
verify(mockNode, times(1)).setProperty("mockPropertyName_ref", asList(mockRefValue).toArray(new Value[0]), WEAKREFERENCE);
i see stuff about ArgumentCaptors, but that seems like a massive pain14:49
* ajs6f leaves14:52
* nbanks joins14:57
* github-ff joins
[fcrepo4] escowles created editable-views (+1 new commit): http://git.io/uhVXGA
fcrepo4/editable-views 7e50933 Esmé Cowles: Basic add/delete property functionality
* github-ff leaves
<escowles>cbeer: yes?
<cbeer>escowles: can you rebase that on top of the html-ui branch?
<escowles>cbeer: sure -- it's not ready for merging yet
<cbeer>ok. html-ui changes the structure in #triples and maybe some other stuff14:59
just waiting for awoods to get around to merging it
* ajs6f joins
<awoods>which ticket, cbeer?
<pivotal-bot>Mike Durbin edited "Ensure that the implications of versioning are compatible with policy enforcement" https://www.pivotaltracker.com/story/show/6076942615:00
<pivotal-bot>feature: Improve the HTML rendering of the fedora 4 rest API to make it more welcoming to new users (finished) / owner: Chris Beer
<cbeer>awoods: ^
<awoods>cbeer: That ticket looked good to me, if mikeAtUVa approved... which he did.15:01
which line?
I think the process generally strains when you have a valuable, expensive task15:12
* travis-ci joins15:13
[travis-ci] futures/fcrepo4#1180 (editable-views - 7e50933 : Esmé Cowles): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/commit/7e50933cd263
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/13989115
* travis-ci leaves
<cbeer>s/worked on/made a mess of before giving up on/?
<ajs6f>+1 to that.15:14
<barmintor>ajs6f: somewhere the underlying Jcrsession is getting logged out15:16
<ajs6f>barmintor: I believe you. I just can't think of any good candidates for where it might be happing15:17
I think PropertiesRdfContext is the bulk of executing code underneath that call.
barmintor: wait, where exactly is the error being thrown from. Like who is trying to use the dead session?
<bljenkins>Project fcrepo-fixity-corrupter build #427: SUCCESS in 1 min 4 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/427/15:18
<barmintor>ajs6f: at org.fcrepo.kernel.FedoraResource.getPropertiesDataset(FedoraResource.java:255)
<ajs6f>final Dataset dataset = DatasetFactory.create(propertiesStream.asModel());15:19
From that _exact_ line?
somewhere in that calchain
<ajs6f>Yeah, I'm trying to figure out where in that call chain— you don't get a stack trace?
Here's the ting:
asModel rolls up the iterators. It actually pulls all the triples from them.
<barmintor>ajs6f: https://gist.github.com/barmintor/4bde06156e2c18aa411d15:21
<ajs6f>So if you assemble the iterators, and then the session closes, and then you call asModel, you could blow it out.
But that shoudn't ever happen here.
<barmintor>no, this error is coming from the instantiation of things
<ajs6f>No, it;s not. It's coming from the iterators as they are being pulled. See all those "hasNext"s? That's asModel pulling all of the data out of the iterators.15:22
Because of asMOdel.
<barmintor>oh, what I mean is: it's happening when the dataset is instantiated15:23
<ajs6f>Yeah, true.
And that data is being pulled from properties, and from the hierarchy.
<bljenkins>Project fcrepo-kitchen-sink build #633: STILL FAILING in 5 min 58 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-kitchen-sink/633/15:24
<ajs6f>Where is it being called _from_ (Like where in the HTTP layers)?
<ajs6f>Is there still a finally{ session.logout()} in there?15:25
Or did you use the FedoraVersions "fix"?
<ajs6f>no finally{} there?
<barmintor>neither, technically
<barmintor>I just took the logout stmt out
<ajs6f>I'm seeing it, so can you push or somehow show me what you are actually working with?
<barmintor>ugh, there's a ton of nonsense in my debugging stuff here15:26
<ajs6f>(As long as they are nasty remarks abot awoods and not me, I don't mind.
<cbeer>nbanks: https://www.pivotaltracker.com/story/show/5453060615:28
<pivotal-bot>feature: Leverage Event.getMixinNodeTypes() in DefaultFilter (unscheduled) / owner:
<nbanks>cbeer: thanks!
<cbeer>nbanks: sorry i didn't think of it earlier.
<nbanks>cbeer: no worries15:29
* github-ff joins
[fcrepo4] barmintor pushed 1 new commit to jersey2: http://git.io/yVEAHQ
fcrepo4/jersey2 3c9fc57 Benjamin Armintor: fix the wrapping runtime exceptions to inherit from WebApplicationException, use a real delegate for TxAwareSession
* github-ff leaves
<barmintor>ajs6f: ^^15:30
That session.isLive() is returning false?15:31
<ajs6f>I'm going to run the build and stare intently at the failure. That should help. If necessary, I will even stroke my chin while blinking slowly.15:34
<barmintor>ajs6f: I've been doing that all day, it doesn't work
* github-ff joins
[fcrepo4] barmintor pushed 1 new commit to jersey2: http://git.io/5CnteQ
fcrepo4/jersey2 4a26e39 Benjamin Armintor: clean up the TxAwareSession delegate
* github-ff leaves
<ajs6f>Goodness. I may have to go as far as leaning back in my chair, staring at the ceiling, and sighing.15:35
<nbanks>cbeer: The spec for jcr 2.1 seems to still be in it's draft stage.
cbeer: Do you know what the timeline would be for it to be finalized?15:36
<cbeer>nbanks: nope.
<nbanks>cbeer: k15:37
<ajs6f>barmintor: after that pull, I just get compile errors:
<barmintor>Oh, damn. Just delete that line in FedoraApplication (and the import)15:38
<ajs6f>k, np.
I'll comment out the other errors.15:40
* travis-ci joins
[travis-ci] futures/fcrepo4#1181 (jersey2 - 3c9fc57 : Benjamin Armintor): The build has errored.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/606fc612e7a9...3c9fc57973d2
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/13990478
* travis-ci leaves
* travis-ci joins
[travis-ci] futures/fcrepo4#1182 (jersey2 - 4a26e39 : Benjamin Armintor): The build has errored.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/3c9fc57973d2...4a26e39876f0
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/13990724
* travis-ci leaves
<nbanks>cbeer: I think I could handle the different event generation if I used many event handlers rather than just the one. I think I'll take that route for now since JCR 2.1 may be a long time away.15:41
cbeer: each event listener could be limited to certain mixin:types and as such could be aware of what was removed at the time of removal.
<ajs6f>barmintor: there's a couple of other things missing here (compile errors) in the kernel, in Transaction and TxAwareSession. Maybe I should get out of your face until you get to a good pausing place to look at the "suicide session" thing?15:43
<barmintor>ajs6f: The Spring scheduled runnable is logging it out
<ajs6f>Oh, Spring, Spring, honey. I know you were only trying to help, but that's a big-boy job. So please put the session down and go play with your XML wiring.15:44
Seriously, that's a hell of a timing fluke.15:45
<pivotal-bot>Chris Beer added "Propose a Modeshape Property API e.g. Property#getNodes() for retrieving multi-valued referenced nodes" https://www.pivotaltracker.com/story/show/6080602415:48
Chris Beer edited "Propose a Modeshape Property API e.g. Property#getNodes() for retrieving multi-valued referenced nodes" https://www.pivotaltracker.com/story/show/6080602415:49
<barmintor>ajs6f: soooort of
<barmintor>re: hk2
<ajs6f>sooooort of what?
<barmintor>as a Spring alternative
<pivotal-bot>A. "Pholeotic" Soroka finished "Document runtime wiring options for F4 modules" https://www.pivotaltracker.com/story/show/56894946
<ajs6f>I think it is definitely an alternative. It's not a _replacement._15:53
<bljenkins>Project fcrepo-fixity-corrupter build #428: SUCCESS in 1 min 25 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/428/
<pivotal-bot>Chris Beer added comment: "Remember that some properties are not user-editable. Maybe we need to inject that information into the RDF st..." https://www.pivotaltracker.com/story/show/60781280
A. "Pholeotic" Soroka edited "Self-Hosted Configuration" https://www.pivotaltracker.com/story/show/6025714815:54
A. "Pholeotic" Soroka unstarted "Self-Hosted Configuration" https://www.pivotaltracker.com/story/show/60257148
A. "Pholeotic" Soroka edited "Self-Hosted Configuration" https://www.pivotaltracker.com/story/show/6025714815:55
<cbeer>nbanks: there's this ticket for doing a better JMS serialization: https://www.pivotaltracker.com/story/show/4562943915:57
<pivotal-bot>feature: Make a JMS serialization that isn't annoying to parse (unscheduled) / owner:
A. "Pholeotic" Soroka added comment: "Via meta-double-extra-reification? {grin} Actually, there's another facet-- an individual user may..." https://www.pivotaltracker.com/story/show/6078128015:58
<nbanks>cbeer: Perhaps I could tackle that as part of this.
<pivotal-bot>Andrew Woods edited "Review and possibly revise or extend the fcrepo-http-api endpoints relating to versions to ensure they are useful f..." https://www.pivotaltracker.com/story/show/60769514
<nbanks>cbeer: Do you have an idea or recommendation for the format of the message body?15:59
<cbeer>nbanks: no body. just a set of key-value headers?
<pivotal-bot>Mike Durbin unstarted "Review and possibly revise or extend the fcrepo-http-api endpoints relating to versions to ensure they are useful..." https://www.pivotaltracker.com/story/show/60769514
<ajs6f>How about a genuine Atom feed? Not the half-way thing we inherited from fcrepo3.16:00
<cbeer>ajs6f: i don't know. i'm pretty lazy.16:01
<nbanks>cbeer: So like (NODE_REMOVED, /path/to/node) or (PROPERTY_CHANGED, /path/to/properties, new_property_value)
cbeer: In the header, that is.
<ajs6f>cbeer: Me too. That's why I brought a parser to the party.
cbeer: actually, anything would be better.
We could even make it pluggable.16:02
Or did we do that already?
<nbanks>cbeer: oh wait I get what your saying.16:03
ajs6f: In what way pluggable?
<ajs6f>Select your desired format for event publication via JMS.
Legacy, KVP, true Atom, etc.
<nbanks>ajs6f: ah ic16:04
<ajs6f>Just use our pleasant modularization framewo— oh, right.
<awoods>thanks, escowles for the scrum-master work!
<nbanks>ajs6f: Spring I presume? Still seems like magic to me.16:05
<ajs6f>nbanks: We'd have to do with it with Spring now. And Maven. So if I wanted to to write a nice extension module to do a decent Atom feed, fcrepo4's method is, 'Check out fo the
"source code and build it yourself." :(16:06
But better than nothing, I suppose.
<nbanks>ajs6f: Would the hk2 framework allow us to do that sorta extension at deployment/runtime instead. Or is it just a nicer way of doing it at compile time?
<ajs6f>Because we don't _have_ a modularization framework, is the point. Have I mentioned that?
<cbeer>ajs6f: PATCHES WELCOME!16:07
<nbanks>ajs6f: No I don't believe you ever have :P
<ajs6f>A little of both. HK2 (or Spring, for that matter) allows us to offer options. You still need a way to select from those options. Our current method is Spring XML that gets packaged at compile time.
We, in theory, write a config layer on top of Spring that runtime builds contexts. That doesn't sound like such a good thing.16:08
s/We/We could/
<pivotal-bot>Esme Cowles added comment: "So right now i've got basic property addition and deletion working. So if a user tries to delete a property..." https://www.pivotaltracker.com/story/show/60781280
<ajs6f>cbeer: I'd seriously like to, but for once, that's not a realistic thing. That would take weeks for one person to do and it would probably never get done because master would shift so much while the work happened.16:09
cbeer: That's one thing that was in my mind when I said what I did about trying to break things up.16:10
* github-ff joins16:11
[fcrepo4] cbeer pushed 1 new commit to sparql: http://git.io/c4OhmA
fcrepo4/sparql 993bb63 Chris Beer: When storing a URI property that points to a fcrepo4 resource, also store a WEAKREFERENCE node to the same, so we can e.g. query against it easily.
* github-ff leaves
<pivotal-bot>A. "Pholeotic" Soroka added comment: "I think "just fial" is a great place to be this quickly. Besides, the general problem of "what are..." https://www.pivotaltracker.com/story/show/60781280
<nbanks>I'm taking off for supper, see ya's in a bit
* nbanks leaves
* nbanks joins16:13
<bljenkins>Project fcrepo-fixity-corrupter build #429: SUCCESS in 1 min 19 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/429/
Project fcrepo-fedora3-federation-connector build #207: UNSTABLE in 5 min 8 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fedora3-federation-connector/207/16:17
Project fcrepo-jms-indexer-pluggable build #212: UNSTABLE in 5 min 52 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-jms-indexer-pluggable/212/16:18
* ajs6f leaves
<awoods>mikeAtUVa: When you get around to review others' work, could you give the "one-click" a good looking-over and generate some polish tickets as a result?16:21
* ajs6f joins16:22
cbeer: is the idea to lock the uri property and weakreference property together to get the benefits of storing both?16:23
<pivotal-bot>Eric James added comment: "Implementation and work up to date: ""16:25
For planned d..." https://www.pivotaltracker.com/story/show/60276468
Eric James finished "Define needs of FileSystem Federation" https://www.pivotaltracker.com/story/show/6027646816:26
* travis-ci joins
[travis-ci] futures/fcrepo4#1183 (sparql - 993bb63 : Chris Beer): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/1e052a13eb4b...993bb634d808
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/13992420
* travis-ci leaves
<mikeAtUVa>awoods: sure
* github-ff joins
[fcrepo4] awoods pushed 1 new commit to master: http://git.io/2sahLg
fcrepo4/master e1e7bd0 Andrew Woods: Merge pull request #148 from futures/html-ui...
* github-ff leaves
<pivotal-bot>Andrew Woods delivered "Improve the HTML rendering of the fedora 4 rest API to make it more welcoming to new users" https://www.pivotaltracker.com/story/show/59056596
<awoods>cbeer/escowles: html-ui branch has been merged.16:27
<ajs6f>awoods: We've got a new Jenkins instance here at UVaLib, and once I get fcrepo4 on it, we'll work on getting the committers access and connecting it to whatever external services we need to. It _should_ run faster than what we have now, and it will remain a supported service for a good while to come.
And then maybe we can retire the Columbia part.16:28
<awoods>ajs6f: That is great, thanks. Would you be able to support the other pieces that are also currently hosted at Columbia? sonar, irclogger, and maven-repo16:31
<ajs6f>awoods: to quote a great man: c'est possible.16:32
We have a Maven repo now.
We've looked breifly at Sonar.
Let's start with Jenkins and bite off other pieces once we swallow it16:33
<bljenkins>Project fcrepo-fixity-corrupter build #430: SUCCESS in 1 min 20 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/430/16:36
* nbanks leaves16:39
* github-ff joins16:41
[fcrepo4] ajs6f pushed 1 new commit to KillWarnings: http://git.io/Gu9LKA
fcrepo4/KillWarnings 4ed0f7b ajs6f: Using non-deprecated HTTP client machinery
* github-ff leaves
* jonathangee leaves16:42
<bljenkins>Yippie, build fixed!16:43
Project fcrepo-jms-indexer-pluggable build #213: FIXED in 7 min 45 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-jms-indexer-pluggable/213/
<awoods>sounds like a plan, ajs6f16:44
bbi ~30min
* github-ff joins16:48
[fcrepo4] ajs6f pushed 1 new commit to KillWarnings: http://git.io/XYlwYQ
fcrepo4/KillWarnings 53762ba ajs6f: Using non-deprecated HTTP multipart machinery
* github-ff leaves
<pivotal-bot>A. "Pholeotic" Soroka added "Killing compiler warnings" https://www.pivotaltracker.com/story/show/6081025016:49
A. "Pholeotic" Soroka started "Killing compiler warnings" https://www.pivotaltracker.com/story/show/60810250
A. "Pholeotic" Soroka edited "Killing compiler warnings" https://www.pivotaltracker.com/story/show/60810250
A. "Pholeotic" Soroka added comment: "https://github.com/futures/fcrepo4/pull/152" https://www.pivotaltracker.com/story/show/6081025016:50
* osmandin leaves16:51
* travis-ci joins16:56
[travis-ci] futures/fcrepo4#1185 (KillWarnings - 4ed0f7b : ajs6f): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/51558ee93ae3...4ed0f7b229fb
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/13993924
* travis-ci leaves
* ajs6f leaves16:59
<bljenkins>Project fcrepo-fixity-corrupter build #431: SUCCESS in 1 min 55 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/431/
* ajs6f joins17:01
* ajs6f leaves
* travis-ci joins17:05
[travis-ci] futures/fcrepo4#1186 (KillWarnings - 53762ba : ajs6f): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/4ed0f7b229fb...53762ba6e962
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/13994205
* travis-ci leaves
<barmintor>"There are 58 checkstyle errors." dammit17:30
* barmintor has no one to blame but himself
<awoods>barmintor: unless you do not like the rules
barmintor: you are eclipse, no?
<barmintor>No, this is just me being expeditious in the short-term
<awoods>I like the forcefulness
I like ajs6f's suggestion to move the checksum rules into the baseline. Objections to pulling that module in?17:32
<pivotal-bot>Andrew Woods accepted "Improve the HTML rendering of the fedora 4 rest API to make it more welcoming to new users" https://www.pivotaltracker.com/story/show/5905659617:33
<bljenkins>Project fcrepo-fixity-corrupter build #432: SUCCESS in 2 min 38 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/432/
Yippie, build fixed!17:42
Project fcrepo-fedora3-federation-connector build #210: FIXED in 10 min: http://ci.fcrepo.org/jenkins/job/fcrepo-fedora3-federation-connector/210/
Project fcrepo-kitchen-sink build #634: STILL FAILING in 9 min 17 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-kitchen-sink/634/17:43
* barmintor <3 intermittent JMS IT failures17:59
they actually have a pretty nice js api
and at some point, i'd like to reorg our REST API page to be more like https://www.pivotaltracker.com/help/api/rest/v5#Iterations18:04
<barmintor>that is pretty nice18:06
* edInCo leaves
* github-ff joins18:27
[fcrepo4] barmintor pushed 1 new commit to jersey2: http://git.io/RHssSQ
fcrepo4/jersey2 3840478 Benjamin Armintor: lengthen the timeout in the TransactionIT to prevent premature reaping; use long instead of date for expiry time; ditch MultiExceptionMapper
* github-ff leaves
* ermadmix leaves18:35
* github-ff joins18:39
[fcrepo4] awoods pushed 1 new commit to master: http://git.io/DRTDbg
fcrepo4/master 1c94c0c osmandin: Add package-info, mostly to top level packages....
* github-ff leaves
<pivotal-bot>Andrew Woods added comment: "Resolved with: https://github.com/futures/fcrepo4/commit/1c94c0c4747438dff38693cfeafe1dc07f4d11da" https://www.pivotaltracker.com/story/show/5495224418:40
Andrew Woods delivered "Javadoc Package Documentation" https://www.pivotaltracker.com/story/show/54952244
Andrew Woods edited "Simple Administrative Search" https://www.pivotaltracker.com/story/show/6071379618:41
Andrew Woods edited "Create design document for simple admin search" https://www.pivotaltracker.com/story/show/60713838
* travis-ci joins
[travis-ci] futures/fcrepo4#1187 (jersey2 - 3840478 : Benjamin Armintor): The build failed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/4a26e39876f0...38404783e3de
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/13999190
* travis-ci leaves
<pivotal-bot>Andrew Woods edited "Killing compiler warnings" https://www.pivotaltracker.com/story/show/60810250
<bljenkins>Project fcrepo-fixity-corrupter build #433: SUCCESS in 1 min 11 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/433/18:43
Project fcrepo-fedora3-federation-connector build #211: UNSTABLE in 5 min 23 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fedora3-federation-connector/211/18:47
Project fcrepo-fixity-corrupter build #434: SUCCESS in 1 min 21 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/434/19:01
Yippie, build fixed!19:05
Project fcrepo-fedora3-federation-connector build #212: FIXED in 5 min 24 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fedora3-federation-connector/212/
Project fcrepo-kitchen-sink build #635: STILL FAILING in 6 min 5 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-kitchen-sink/635/19:07
* nbanks joins19:11
* nbanks leaves19:24
* mikeAtUVa leaves19:45
* ksclarke leaves20:40
* jonathangee joins20:50
* jonathangee_ joins
* jonathangee leaves
* ksclarke joins21:49
* fasseg leaves