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

Using timezone: Eastern Standard Time
* kaarefc joins03:00
* kaarefc leaves03:02
* kaarefc joins
* kaarefc leaves03:05
* kaarefc joins03:06
* kaarefc leaves03:47
* kaarefc joins04:21
* kaarefc leaves04:22
* kaarefc joins04:24
* anusha joins05:54
* eddies leaves06:38
* eddies joins06:47
* eddies leaves
* eddies joins
* eddies leaves06:51
* eddies joins06:58
* eddies leaves
* eddies joins
<pivotal-bot>Edwin Shin accepted "Move from fixed hierarchy under /objects to arbitrary hierarchy" https://www.pivotaltracker.com/story/show/4857822307:01
* eddies leaves07:43
* ppound joins08:01
* eddies joins08:27
* eddies leaves
* eddies joins
<pivotal-bot>Edwin Shin accepted "Parameterize the OpenWire port used by ActiveMQ" https://www.pivotaltracker.com/story/show/4877532708:37
Edwin Shin accepted "Wire the fcrepo-legacy-api into the kitchen sink, jenkins, etc" https://www.pivotaltracker.com/story/show/48835133
Edwin Shin accepted "Make the fcrepo-kernel path-based rather than pid/dsid based" https://www.pivotaltracker.com/story/show/4883499508:38
Edwin Shin accepted "fix the rdf generator with the globbing REST api" https://www.pivotaltracker.com/story/show/4880826308:39
Edwin Shin accepted "Create an alternative API proposal" https://www.pivotaltracker.com/story/show/48908691
Edwin Shin accepted "Fix kitchen sink on fcrepo4.fcrepo.org" https://www.pivotaltracker.com/story/show/48950797
Edwin Shin delivered "Extract the old fcrepo-http-api into a new fcrepo-legacy-api repository (and, kill the old one)" https://www.pivotaltracker.com/story/show/48835089
Edwin Shin accepted "Extract the old fcrepo-http-api into a new fcrepo-legacy-api repository (and, kill the old one)" https://www.pivotaltracker.com/story/show/48835089
Edwin Shin delivered "Databank filesystem connector" https://www.pivotaltracker.com/story/show/48257307
Edwin Shin accepted "Databank filesystem connector" https://www.pivotaltracker.com/story/show/48257307
Edwin Shin delivered "Run the FCRepo APIs directly on projected nodes" https://www.pivotaltracker.com/story/show/48657119
Edwin Shin accepted "Run the FCRepo APIs directly on projected nodes" https://www.pivotaltracker.com/story/show/48657119
Edwin Shin accepted "Create a fcr:fixity resource in the globbing branch" https://www.pivotaltracker.com/story/show/48799307
* kaarefc leaves09:11
* kaarefc joins09:34
* fasseg joins09:56
<cbeer>eddies: what's the plan for the call today?
<eddies>cbeer: same as last time. a sprint wrap-up at 12 Eastern09:57
<cbeer>ok. that's permanent then? we should adjust the calendar?09:58
* kaarefc leaves09:59
<eddies>i thought i did…weird. i guess my changes didn't save and now when i change it, it keeps reverting back10:01
sec
* barmintor joins
<fasseg>eddies: arg, I can't do 12.00h eastern :/10:03
<eddies>cbeer or barmintor: can you give me an example of posting a new tiff and fetching its profile?
<cbeer>nope. it doesn't work as advertised.10:04
<eddies>hrm. i'd rather it was just me10:05
<cbeer>NPE?
<eddies>javax.jcr.nodetype.ConstraintViolationException: The values for the '{http://www.jcp.org/jcr/1.0}content' property on node '/objects/a063b9b7-96e9-4297-910c-d6ca4c647384/TIFF' no longer satisfy the type and/or constraints on the 'jcr:content' property definition on the 'nt:file' node type definition10:07
this is trying to POST a tiff to /objects/a063b9b7-96e9-4297-910c-d6ca4c647384/TIFF/fcr:content
<cbeer>hm. i had no problem posting.10:08
<eddies>given a new object, e.g. "/objects/foo" how are you POSTing the new tiff datatream?10:09
POST to /objects/foo/fcr:new/fcr:content?10:10
<cbeer>https://www.pivotaltracker.com/story/show/48578951
<pivotal-bot>feature: Policy-driven storage strawman using mime types (delivered) / owner: Chris Beer
<eddies>i'm pretty sure that was one of the variations i tried
that was the first thing i tried
<cbeer>it worked for me, and continues to work.10:11
i can't understand the semantics behind fcr:new though.
<eddies>you didn't add a mixin query parameter?
<cbeer>ah, so maybe we're missing an error code when you don't provide one10:13
curl -X POST "http://fcrepo4.fcrepo.org/fcrepo/v3/objects/new"
curl -X POST -H "Content-Type: image/tiff" -D @StatusOfResponse-1.png "http://fcrepo4.fcrepo.org/fcrepo/v3/objects/4fb299f0-740d-4c3a-ae67-409ba5a279dd/datastreams/TIFF"
also works for me10:14
still trying to understand the NPE i see on the fcr:describe though
<eddies>the NPE at http://fcrepo4.fcrepo.org/fcrepo/rest/objects/4fb299f0-740d-4c3a-ae67-409ba5a279dd/TIFF/fcr:describe ?10:15
<cbeer>yes
<eddies>now i remember. this is what i was seeing earlier so i thought i was creating the resource incorrectly and started fussing with other ways to create the resources10:16
<cbeer>hm. i wonder if modeshape never calls start() on these stores10:18
hm. not sure we have integration test coverage on fcr:describe on datastreams10:22
<eddies>yeah, i'm not seeing one in http-api10:26
<barmintor>I think I must have failed to git-add a couple of IT last week- I know I implemented POST fcr:new/fcr:content, and I'm pretty sure I added one for the ds describe10:28
<cbeer>ok, i can recreate the failure locally with the CompositeCacheStore.10:30
seems like i should answer the question: "seems like we have to start it, not sure why."
* github-ff joins10:32
[fcrepo4] cbeer pushed 2 new commits to master: http://git.io/DuGfkA
fcrepo4/master 0db91fe Chris Beer: add IT for fcr:describe
fcrepo4/master 46bd4de Chris Beer: explicitly start our compositeStore, still for reasons unknown.
* github-ff leaves
<cbeer>ah, right. the point i tap into modeshape i get a different BinaryStore back.10:35
because they lock the real BinaryStore into a private field10:38
or protected method, i guess.. i guess this is where i should use reflection?10:39
<barmintor>if it's protectdc, you can move your code into their package to work around it. If it's private, you need to use reflection.10:40
<cbeer>ok, so... i could move the GetBinaryStore function into their package and it should "just work"?10:41
<barmintor>cbeer: I *think* so.10:43
* travis-ci joins10:44
[travis-ci] futures/fcrepo4#447 (master - 46bd4de : Chris Beer): The build has errored.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/66cd2184814d...46bd4de0dea9
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6759388
* travis-ci leaves
<fasseg>eddies: I always look at this page, this isn't accurate anymore than?10:49
<barmintor>cbeer: grabbing a coffee; let me know if I can help in a few minutes
<fasseg>oops forgot the link: same as last time. a sprint wrap-up at 12 Eastern10:50
argl: https://wiki.duraspace.org/display/FF/TT+Meetings
<cbeer>i think i've got it.10:51
<eddies>fasseg: it's up to date now10:53
thanks, ahem, for the reminder
<cbeer>powermockito+-11:16
<barmintor>PowerMockito: Making you feel awkward about your tests since 200811:17
<cbeer>final final everywhere.11:21
* kaarefc joins11:24
<cbeer>that's it. i give up. Integration test it is.11:25
<barmintor>Nooooooooooooo11:26
* github-ff joins11:28
[fcrepo4] cbeer pushed 1 new commit to master: http://git.io/vK3ruw
fcrepo4/master 50f2a5b Chris Beer: rewrite GetBinaryStore to reach into the Modeshape runningState to pull out the BinaryStore instance
* github-ff leaves
<cbeer>assertTrue(true);
is the other option.
it's final members all the way down.
* ajs6f joins11:31
<cbeer>ajs6f: what do you think about moving the fcrepo-object-serialization machinery (but probably not the serializations) into the http-api/kernel/whatever? turns out rubydora has a dependency on being able to export and import objects, and making it part of the default package would be nice11:34
<pivotal-bot>Chris Beer edited "standardize on a common fcrepo4 prefix for things" https://www.pivotaltracker.com/story/show/48922513
<ajs6f>If we consider it core functionality, sounds good to me. Keep in mind that the default serialization (which is the only one included directly in that module) is the JCR XML export.
Is anyone else having problems building master? I'm getting failures in http-api.11:35
<pivotal-bot>Chris Beer edited "standardize on a common fcrepo4 prefix for things" https://www.pivotaltracker.com/story/show/48922513
<cbeer>ajs6f: let me try. i may have put things in a bad state.
<pivotal-bot>Chris Beer edited "Promote fcrepo-object-serialization into the core fcrepo4 API" https://www.pivotaltracker.com/story/show/4891115311:37
<cbeer>ajs6f: yup, i broke it11:38
<ajs6f>Well, there are no really bad states, just misunderstood ones.
* travis-ci joins11:39
[travis-ci] futures/fcrepo4#448 (master - 50f2a5b : Chris Beer): The build has errored.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/46bd4de0dea9...50f2a5b732bc
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6761023
* travis-ci leaves
<pivotal-bot>Chris Beer estimated "Update Rubydora to use the latest fcrepo4 http APIs" as 8 points https://www.pivotaltracker.com/story/show/4882472111:43
* github-ff joins11:44
[fcrepo4] cbeer pushed 1 new commit to master: http://git.io/wGUTRQ
fcrepo4/master f1e5a41 Chris Beer: fix FedoraDescribeIT.java
* github-ff leaves
* fasseg leaves
* fasseg joins
<ajs6f>Looks like that last commit fixed it for me.11:46
<eddies>time to hangout12:00
* travis-ci joins
[travis-ci] futures/fcrepo4#449 (master - f1e5a41 : Chris Beer): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/50f2a5b732bc...f1e5a414c4c6
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6761391
* travis-ci leaves
<cbeer>barmintor anusha fasseg jonathangee?12:01
<bljenkins>Yippie, build fixed!
Project fcrepo4 build #492: FIXED in 17 min: http://ci.projectblacklight.org/jenkins/job/fcrepo4/492/
Chris Beer: fix FedoraDescribeIT.java
* fasseg leaves12:06
<bljenkins>Project fcrepo-kitchen-sink build #224: SUCCESS in 5 min 22 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/224/12:07
<cbeer>fixed: curl http://fcrepo4.fcrepo.org/fcrepo/rest/4624fa20-87a0-4026-af41-548b58fb1754/fcr:describe12:08
i'm not sure where the dsstores are coming from... seems like it's pulling all the stores, not just the one our object is in12:11
<bljenkins>Project fcrepo-kitchen-sink build #225: SUCCESS in 4 min 15 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/225/12:12
<cbeer>oh, duh. i've continued to push the same file with different mime types12:13
<barmintor>sorry, meeting w/ Stephen ran over12:21
there in a sec
<cbeer>please, fire me!!12:27
<jonathangee>eddies++12:38
<barmintor>Who is LEKing?12:39
<cbeer>afk12:42
back12:45
<ajs6f>How do you LEK?12:48
<barmintor>it's someone asking for membership in the FF group12:49
<ajs6f>Oh...
LE King?
<anusha>I'm off now. bye12:51
<ajs6f>http://docs.codehaus.org/display/SONAR/Using+Sonar+in+Eclipse12:52
<barmintor>thanks ajs6f!
<ajs6f>http://docs.codehaus.org/display/SONAR/Toxicity+Chart+Plugin12:54
Toxicity!
<barmintor>http://marketplace.eclipse.org/content/ecobertura#.UX_3kyuG0qs12:55
<ajs6f>Sprint 9: Navelgazing!12:56
* anusha leaves
* ajs6f leaves12:58
<eddies>barmintor: that plugin looks promising except for the fact that i think it might not work against current versions of eclipse13:00
<barmintor>Grr
<eddies>maybe it does work: http://stackoverflow.com/questions/6112561/clearing-coverage-highlighting-in-eclipse13:02
seems like folks are running it against juno
this is the first time i've seen someone say "Moved from http://github.com/ to http://bitbucket.org/ because I like http://mercurial.selenic.com/ better…"13:04
from the ecobertura dev (http://ecobertura.johoop.de/)
<barmintor>yeah
I'm going to have a look at the Emma plugin, and see how closely they match13:05
I think Emma uses JaCoCo, and reports very different numbers13:09
<cbeer>now i'm totally confused how policy driven storage even appears to work on futures613:22
:/
<eddies>barmintor: as i understand, sonar uses cobertura for unit test coverage and supports jacoco for integration test coverage13:26
(for the latter, we need to do some gymnastics which i don't fully grok: http://www.sonarsource.org/measure-coverage-by-integration-tests-with-sonar-updated/)
<barmintor>I've got the cobertura eclipse plugin running on Indigo
<eddies>cbeer: call?13:42
<cbeer>eddies: sure.
on the hangout?13:43
* ajs6f joins13:45
<pivotal-bot>Chris Beer added "Batch retrieval API" https://www.pivotaltracker.com/story/show/4900848113:56
Chris Beer edited "Batch retrieval API" https://www.pivotaltracker.com/story/show/4900848113:57
Chris Beer deleted "Create an transactions-based REST API which allows multiple requests happening under a single "transaction"" https://www.pivotaltracker.com/story/show/4476550114:01
<ajs6f>cbeer: Are you assembling issues for a transaction theme/epic?14:02
<cbeer>ajs6f: yes.
<pivotal-bot>Chris Beer added "Transactions-enabled API" https://www.pivotaltracker.com/story/show/49009209
<barmintor>well hell14:03
<ajs6f>Rocking.
<cbeer>wiki.apache.org/jackrabbit/DavidsModel14:05
<barmintor>eddies: I'm seeing a lot of stuff online that cobertura doesn't parse Java 7
and sonar appears to say "use JaCoCo"
<cbeer>http://www.day.com/maven/jsr170/javadocs/jcr-2.0/javax/jcr/version/VersionManager.html#merge(java.lang.String, java.lang.String, boolean)14:07
<pivotal-bot>Paul Pound added comment: "code that worked with fcrepo4 just prior to the fcrepo4 globbing updates lives in fcrepo-experimental branche..." https://www.pivotaltracker.com/story/show/48351401
<barmintor>eddies: Can we try configuring Sonar to work with JaCoCo or Emma? I think some of our missing coverage is that it can't parse the classes.14:10
<ajs6f>barmintor: shouldn't that throw some kind of message?
<cbeer>http://fcrepo4.fcrepo.org/fcrepo/v3/describe/modeshape
<barmintor>ajs6f: It does in the Eclipse plugin- I expect that they get swallowed somewhere
<ajs6f>oh.14:11
<barmintor>"java.lang.VerifyError: Expecting a stackmap frame at branch target 207 in org.fcrepo…"14:12
<pivotal-bot>Chris Beer edited "Transactions-enabled API" https://www.pivotaltracker.com/story/show/49009209
Edwin Shin edited "Transactions-enabled API" https://www.pivotaltracker.com/story/show/4900920914:13
<barmintor>Hmm… I might also be able to add some command line thing to change the JDK class veirifcation...14:14
<pivotal-bot>Chris Beer edited "Transactions-enabled API" https://www.pivotaltracker.com/story/show/49009209
A. "Gorbuncular" Soroka edited "Transactions-enabled API" https://www.pivotaltracker.com/story/show/4900920914:15
Edwin Shin edited "Transactions-enabled API" https://www.pivotaltracker.com/story/show/49009209
Chris Beer edited "Transactions-enabled API" https://www.pivotaltracker.com/story/show/49009209
A. "Gorbuncular" Soroka started "Transactions-enabled API" https://www.pivotaltracker.com/story/show/49009209
<eddies>ajs6f: please pause editing that ticket14:16
<ajs6f>k
<eddies>you keep stomping over my edits :P
<ajs6f>Sorry.
I seem to have started it, too. Ugr.
<eddies>s'ok =) if only we had transaction support on our tickets
<pivotal-bot>Edwin Shin unstarted "Transactions-enabled API" https://www.pivotaltracker.com/story/show/49009209
<barmintor>ok, telling the JDK not to use the Java7 class format verification vaults one class's test coverage from 13.5% to 80%
<ajs6f>But our classes _are_ j7, right?14:17
<barmintor>yes, but for some reason running w/ cobertura jacks that all up
<ajs6f>Weeeird.
<barmintor>well, cobertura does some weird stuff.
<ajs6f>Why would Cob affect the output, tho'? Oughtn't it do its thing without side-efffects except on its own artifiacts14:18
?
<pivotal-bot>Chris Beer added "Make existing REST API transactions-aware" https://www.pivotaltracker.com/story/show/49010571
<barmintor>b/c it chokes when it tries to execute the test
<pivotal-bot>Chris Beer edited "Make existing REST API transactions-aware" https://www.pivotaltracker.com/story/show/49010571
Edwin Shin edited "Transactions-enabled API" https://www.pivotaltracker.com/story/show/49009209
<ajs6f>Oh, right.
<pivotal-bot>Chris Beer edited "Make existing REST API transactions-aware" https://www.pivotaltracker.com/story/show/4901057114:19
Chris Beer added "Update JMeter test to demonstrate starting a transaction, make changes, and commit the changes" https://www.pivotaltracker.com/story/show/4901075714:20
Chris Beer edited "Update JMeter test to demonstrate starting a transaction, make changes, and commit the changes" https://www.pivotaltracker.com/story/show/49010757
Chris Beer added "Add Transactions support to Rubydora" https://www.pivotaltracker.com/story/show/49010781
Chris Beer edited "Add Transactions support to Rubydora" https://www.pivotaltracker.com/story/show/49010781
<ajs6f>cbeer: If we use JCR Sessions for lightweight transactions, we may need to be able to pass a Session into what are now JAX-RS endpoints to satisfy: 49010571
<pivotal-bot>Chris Beer added "Add transactions support to Tuque" https://www.pivotaltracker.com/story/show/4901079714:21
Chris Beer edited "Add transactions support to Tuque" https://www.pivotaltracker.com/story/show/49010797
Chris Beer edited "Add transactions support to Tuque" https://www.pivotaltracker.com/story/show/49010797
Chris Beer edited "Add Transactions support to Rubydora" https://www.pivotaltracker.com/story/show/49010781
Chris Beer edited "Update JMeter test to demonstrate starting a transaction, make changes, and commit the changes" https://www.pivotaltracker.com/story/show/49010757
<cbeer>ajs6f: i'll add this to the ticket, but we were concerned about how sessions may or may not handle conflicts14:22
<pivotal-bot>Edwin Shin edited "Batch retrieval API" https://www.pivotaltracker.com/story/show/4900848114:23
<ajs6f>Last one to save() blows up on a conflict.
There is no merge semantic defined by JCR.
Although I guess impls might offer one.
<cbeer>right. so we might want some locking ability14:24
<ajs6f>Seems like the next step beyond Session is XA xactions.
<pivotal-bot>Chris Beer added "Add batch retrieval API support to JMeter test suite" https://www.pivotaltracker.com/story/show/49011083
<ajs6f>We _could_ bring the JCR Locking semantics out into HTTP API.
<pivotal-bot>Chris Beer edited "Add batch retrieval API support to JMeter test suite" https://www.pivotaltracker.com/story/show/49011083
Chris Beer edited "Add batch retrieval API support to JMeter test suite" https://www.pivotaltracker.com/story/show/49011083
Chris Beer edited "Batch retrieval API" https://www.pivotaltracker.com/story/show/49008481
Edwin Shin added "Document class design to support more testable code" https://www.pivotaltracker.com/story/show/4901126114:26
Chris Beer added "Create a pleasing text/html representation for /fcr:describe with HATEAOS links to actions" https://www.pivotaltracker.com/story/show/4901135314:27
Chris Beer edited "Add POST /fcr:new request" https://www.pivotaltracker.com/story/show/4895418114:28
Chris Beer edited "Add POST /fcr:new request" https://www.pivotaltracker.com/story/show/48954181
Chris Beer estimated "Add POST /fcr:new request" as 1 point https://www.pivotaltracker.com/story/show/48954181
Chris Beer edited "RESTful error codes" https://www.pivotaltracker.com/story/show/48918595
Chris Beer estimated "RESTful error codes" as 5 points https://www.pivotaltracker.com/story/show/48918595
Edwin Shin edited "Create a pleasing text/html representation for /fcr:describe with HATEAOS links to actions" https://www.pivotaltracker.com/story/show/4901135314:29
Chris Beer edited "RESTful error codes" https://www.pivotaltracker.com/story/show/4891859514:30
<cbeer>http://www.w3.org/TR/sparql11-update/14:32
ajs6f: ^
is that crazy?
(we're talking relationships API now.. at least for the limited amount rubydora/tuque might use it)14:33
so this might be overkill
<ajs6f>Not crazy.
Hard to impl.
You'd kind of want a RDF store behind it.14:34
Which we don't want.
Are there other options being thrown around?14:35
<pivotal-bot>Chris Beer deleted "[TODO] relationships REST endpoint" https://www.pivotaltracker.com/story/show/48911229
Chris Beer estimated "Promote fcrepo-object-serialization into the core fcrepo4 API" as 3 points https://www.pivotaltracker.com/story/show/4891115314:36
Chris Beer edited "Promote fcrepo-object-serialization into the core fcrepo4 API" https://www.pivotaltracker.com/story/show/48911153
Chris Beer estimated "Create a pleasing text/html representation for /fcr:describe with HATEAOS links to actions" as 1 point https://www.pivotaltracker.com/story/show/49011353
<ajs6f>eddies/all: If we're going to do HATEOS HTML, can I do the MessageWriter thing first?
https://www.pivotaltracker.com/story/show/4853737914:37
<pivotal-bot>feature: Move HTML templating to a JAX-RS entity provider (unscheduled) / owner:
Chris Beer edited "make islandora work with latest built of fcrepo4. " https://www.pivotaltracker.com/story/show/4835140114:38
Chris Beer edited "make islandora work with latest built of fcrepo4. " https://www.pivotaltracker.com/story/show/48351401
Chris Beer added "Ensure REST API supports workspace identifier prefixes by adding tests" https://www.pivotaltracker.com/story/show/4901238914:39
Chris Beer estimated "Ensure REST API supports workspace identifier prefixes by adding tests" as 2 points https://www.pivotaltracker.com/story/show/49012389
Chris Beer edited "Ensure REST API supports workspace identifier prefixes by adding tests" https://www.pivotaltracker.com/story/show/49012389
A. "Reticulated" Soroka estimated "Move HTML templating to a JAX-RS entity provider" as 3 points https://www.pivotaltracker.com/story/show/4853737914:40
Chris Beer edited "Document class design to support more testable code" https://www.pivotaltracker.com/story/show/49011261
A. "Reticulated" Soroka started "Move HTML templating to a JAX-RS entity provider" https://www.pivotaltracker.com/story/show/48537379
<barmintor>eddies: http://stackoverflow.com/a/9583305
Do you want me to monkey with the java args in mvn, or switch to JaCoCo/Emma/etc. ?
<pivotal-bot>Chris Beer deleted "Add new people to the google docs" https://www.pivotaltracker.com/story/show/4772678714:42
Chris Beer deleted "Transparent serialization of object structure" https://www.pivotaltracker.com/story/show/4545859114:43
Chris Beer added "Introspect the bson output for modeshape schematics and report back how friendly the output is" https://www.pivotaltracker.com/story/show/49012799
* ajs6f leaves14:44
* ajs6f joins14:47
Procedure question:14:50
DO we have a way of linking issue IDs to Git branches?
<barmintor>Not afaik
<ajs6f>Good. Then I'm not doing it wrong.
<barmintor>but seems pretty trivial
<ajs6f>Just use the ID?
<barmintor>that'd be my first instinct14:51
<ajs6f>k
<barmintor>I think once you tell PowerMock to mockStatic(MyClass.class), all of that classes static methods are mocks in the scope of that test14:52
<pivotal-bot>Chris Beer added "Propose a straw-man way to support whole-object versioning" https://www.pivotaltracker.com/story/show/4901357714:53
<barmintor>So there's a motivation to put static methods in a separate helper class, and test independently
<ajs6f>Like for the *Service guys?14:54
Or are they all static already?
<barmintor>most of the Service's are objects
PathService isn't
ServiceHelpers
and...
<ajs6f>But we should factor out the static stuff?
<barmintor>if you want to test it. At minimum, the test for the static stuff need to be a sep test class14:55
Yeah, you need to break them out into a separate test class15:10
<ajs6f>Okay. Well, that's not a bad thing anyway (breaking out the static helpers).15:11
<barmintor>yeah- the sep helper class is going to be annoying sometimes. I think I just need to doc the test pattern15:12
esp. since it also effects trying to mock constructor calls
<ajs6f>Where are our mixins?!15:14
<barmintor>what mixins/
<ajs6f>In Scala, of course.
* github-ff joins15:28
[fcrepo4] barmintor pushed 1 new commit to master: http://git.io/-D-LYA
fcrepo4/master 894861d Benjamin Armintor: adding -XX:-UseSplitVerifier to surefire CLI args to try and coax cobertura into instrumenting Java7 better
* github-ff leaves
* ajs6f leaves15:29
* ajs6f joins15:33
<bljenkins>Project fcrepo4 build #493: UNSTABLE in 15 min: http://ci.projectblacklight.org/jenkins/job/fcrepo4/493/15:43
armintor: adding -XX:-UseSplitVerifier to surefire CLI args to try and coax cobertura into instrumenting Java7 better
* travis-ci joins15:45
[travis-ci] futures/fcrepo4#450 (master - 894861d : Benjamin Armintor): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/f1e5a414c4c6...894861dba9f4
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6767544
* travis-ci leaves
* ppound leaves15:46
<barmintor>Grr15:47
something is just not right
my cobetura eclipse plugin says org.fcrepo.jms.legacy.LegacyMethod has 80.15% terst coverage15:48
<ajs6f>Awesome!
<barmintor>sonar says it has 8.9%
<ajs6f>Rounding error.15:49
<barmintor>:|
<ajs6f>Is it supposed to be just unit tests?15:50
<barmintor>yeah.
<ajs6f>Is one of them counting i-tests?
<barmintor>I think part of the discrepancy is that I can't figure out how to exclude the IT from the cobertura test runner
^^
<ajs6f>Ah.
<barmintor>I still get 54.92% when I exclude the IT, and just run the unit test for that class15:56
Grr.
<ajs6f>Is it counting i-tests for another resource?15:57
<barmintor>No, this is just running a single unit test
<ajs6f>Is it a difference in the algorithms for coverage? Do we know what those are?15:59
<barmintor>I don't. I don't see anything in the mvn plugin config16:00
* ajs6f leaves16:13
* ajs6f joins16:16
* ajs6f leaves16:33
<pivotal-bot>Chris Beer added "Update datastream and object responses to be path-oriented." https://www.pivotaltracker.com/story/show/4902382316:47
<barmintor>where the hell is sonar running from?17:02
<pivotal-bot>Chris Beer added "Either fix PUT /rest/{path} (FedoraObjects#modifyObject) or kill the idea of object/datastream labels entirely." https://www.pivotaltracker.com/story/show/4902583917:10
<cbeer>barmintor: the sonar that jenkins triggers?17:11
<barmintor>cbeer: yeah. I'm looking around gluck for something that ties sonar to fcrepo4
<cbeer>it's triggered by the CI build.. i think through a rest api
let me double check that17:12
yes, jenkins knows about sonar.fcrepo.org over HTTP and JDBC access to the underlying database17:13
<barmintor>ok
so jenkins sends a REST request to sonar, and sonar checks out the project and runs its plugins over it?17:17
* barmintor trundles off looking for sonar's java args17:18
<cbeer>sounds right. i think jenkins has some way to hand off java args.17:22
barmintor: look in the fcrepo4 build config => Post-build Actions => Sonar => MAVEN_OPTS and Additional properteis17:23
(maybe i'm wrong.. maybe jenkins shells out and runs maven comments)17:27
<barmintor>guh. cbeer, are you some kind of admin that can reset a password on jenkins?17:33
I don't know what on earth my creds are there
<cbeer>yeah, if i can remember how17:34
barmintor: you're using your gmail address
reset to 'changeme'
<barmintor>thank you17:35
<cbeer>barmintor: hm. looks like DELETE /objects/test:1 doesn't cascade to delete the children.17:39
i think it should, right?
<barmintor>cbeer: I would think so. I'm kind of fascinated it doesn't.17:40
<cbeer>yeah. i was banging my head with these rubydora tests
<barmintor>suggests something interesting about the tree- not really a tree.
<cbeer>GET /rest/objects/test%3A1/fcr:describe?format=xml => 404
GET /rest/objects/test%3A1 => 200
POST /rest/objects/test%3A1 => 409
hm, no. that's not the behavior17:41
hm. can't recreate it with curl..17:43
hm. i wonder if it's something to do with sessions..17:47
it looks like fcr:describe uses an authenticated session
and the regular path uses a read-only session
<barmintor>seems like they should both be using the same session17:48
* kaarefc leaves17:54
<cbeer>reading https://docs.jboss.org/author/display/MODE/Repository+and+Session i'm not sure our read-only session was a good idea17:55
<barmintor>probably better to fire off on request, eh?17:57
* bljenkins leaves18:19
* bljenkins joins
* github-ff joins18:55
[fcrepo4] barmintor pushed 1 new commit to master: http://git.io/GUFq8Q
fcrepo4/master 4b427eb Benjamin Armintor: cleaning up violations
* github-ff leaves
<bljenkins>Yippie, build fixed!19:08
Project fcrepo4 build #496: FIXED in 13 min: http://ci.projectblacklight.org/jenkins/job/fcrepo4/496/
armintor: cleaning up violations
<barmintor>g'night19:10
* travis-ci joins19:12
[travis-ci] futures/fcrepo4#451 (master - 4b427eb : Benjamin Armintor): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/894861dba9f4...4b427ebb73aa
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6772212
* travis-ci leaves
<bljenkins>Project fcrepo-kitchen-sink build #226: SUCCESS in 4 min 58 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/226/19:14
* eddies leaves22:01
* eddies joins22:03
* eddies leaves
* eddies joins
* eddies leaves22:08
* github-ff joins23:18
[fcrepo4] barmintor pushed 1 new commit to master: https://github.com/futures/fcrepo4/commit/8233045d6c1eb9a1828484c79fd9e6b9a2379b93
fcrepo4/master 8233045 Benjamin Armintor: unit testing fcrepo-kernel/org.fcrepo.observer
* github-ff leaves
* travis-ci joins23:41
[travis-ci] futures/fcrepo4#452 (master - 8233045 : Benjamin Armintor): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/4b427ebb73aa...8233045d6c1e
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6776540
* travis-ci leaves
* barmintor leaves00:34
<bljenkins>Project fcrepo-kitchen-sink build #227: FAILURE in 1 hr 26 min: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/227/01:15

Generated by Sualtam