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

Using timezone: Eastern Standard Time
* kaarefc joins02:06
<pivotal-bot>Anusha Ranganathan started "Databank filesystem connector" https://www.pivotaltracker.com/story/show/4825730702:29
* kaarefc leaves02:39
* kaarefc joins02:42
* eddies leaves02:51
* anusha joins05:37
* nbanks joins06:22
* fasseg joins06:56
fasseg, nbanks: a noob question regarding running the mvn tests for the bagit modeshape connector.07:12
when I run mvn clean compile, I get "Non-resolvable import POM: Failure to find org.modeshape.bom"
<fasseg>hmm not usre i can help, didn't handle the bagit connector up until now
<anusha>Any idea how to solve this07:13
I am guessing this is my lack of java knowledge here
<fasseg>can you gist the the complete error message?
<nbanks>anusha: sorry I'm new to maven as well
<fasseg>because i had some problems with the arduno repository since yesterday, and this might be related07:14
<anusha>fasseg: sure, give me a sec
nbanks: glad I have someone in the same boat :)
fasseg: https://gist.github.com/anusharanganathan/6916b306b23a55a8a9f207:16
I've copied the error and pasted the link to the pom in the comments07:17
<fasseg>oh ok it seems just the <version> tags are missing from the cf dependencies...
let me check the pom....
try adding the CXF version to the pom https://github.com/futures/fcrepo-databank-modeshape-federation-connector/blob/master/pom.xml#L56 for the CXF dependencies07:19
normally this should be set by the parent pom of the project, but cxf will be dropped in the near future so the versions might have already been removed from the fcrepo4 paretn project07:20
or you will have to clean install the fcrepo4 parent project, if you did not do so already07:21
<anusha>fasseg: Hmm, I didn't clean install fcrepo4.
Does that just mean I cd to fcrepo4 and do mvn clean install07:22
<fasseg>try that before since the pom.xml from fcrepo4 is the connector's parent pom and should hold the version infos
might be worth a try
<anusha>thanks, will do that now
<fasseg>so versions can be set in the parent pom using "dependencymanagement" and all the child projects will use the same version. but if the parent version can not be looked up somehow this might be the cause of the problem07:23
<anusha>fasseg: many many Thanks. that seemed to have worked. Atleast, maven is putting out many line now07:28
<fasseg>nice, always glad to help07:29
* ppound joins07:43
* kaarefc leaves07:52
* VincentNG joins07:58
* VincentNG leaves08:02
* VincentNG joins
* ajs6f joins08:30
anusha: fasseg was definitely right. When you install a project that has a parent, you want to install the parent _first_, or make its metadata available to Maven in some other way. That's not obvious for us, and we have more refactoring to do in the fcrepo4 codebase to make building easier, but I'm waiting on fasseg's epic journey into JAX-RS to be complete before I began that.08:56
<fasseg>almost done last submodule atm.....im soo happy
and it would be nice if you could take a look at it later...08:57
<anusha>ajs64: Thanks. I started thinking along those lines and cloned fcrepo4 but then didn't go as far as installing, that is until frank helped me out
* ap2972 joins
<anusha>ajs64: I have a question for you about the bagit connector. The directoryPath seems to be a string. Where is the connection between directory path and object Id made09:00
ajs6f: Oops sorry typo in your name in the prev message09:05
<VincentNG>anusha: the directoryPath and where the object is projected is defined here: https://github.com/futures/fcrepo-databank-modeshape-federation-connector/blob/master/src/test/resources/test_repository.json#L1509:07
not sure if that's what you were looking for
<ajs6f>anusha: https://github.com/futures/fcrepo-databank-modeshape-federation-connector/blob/master/src/main/java/org/fcrepo/federation/bagit/BagItConnector.java#L14609:09
is where an id is received as part
of a request to MODE and translated into action against the filesystem.
Is that it?
<anusha>ajs6f: looks like it. Thanks09:11
<ajs6f>rock on.
<VincentNG>barmintor: you there?09:12
<ppound>I'm having trouble doing a build09:13
[INFO] fcrepo dc ......................................... SUCCESS [7.225s]
[INFO] fcrepo-generator-rdf .............................. FAILURE [0.588s]
<ajs6f>Not your fault.
A Maven repo elswhere in the world disappeared from the Net.
<ppound>k is the powermock09:14
<ajs6f> It's breaking stuff in fcrepo3, fcrepo4, and a couple of Apache projects I know.
is there a workaround like adding the dependancies manually?09:15
<ajs6f>yea, but are you actaully working on fcrepo-generator-rdf?
<ajs6f>if not, i would jut skip that part of the builfd09:16
nothing is dependant on it?
<nbanks>I can send you the files to throw into .m2/org/powermock ? Would that help?09:17
.m2/repository/org/powermock I mean
<ppound>if it's the only one not sure if there are others
I'll see if I can skip the rdf-builder as Adam suggested09:18
* github-ff joins09:26
[fcrepo4] fasseg pushed 1 new commit to jersey-grizzly: http://git.io/9Kirnw
fcrepo4/jersey-grizzly be4d127 fasseg: changed webapp, generators-rdf and webhooks to jersey impl and switched integration tests to use grizzly
* github-ff leaves
<ajs6f>ppound: No, nothing depends on fcrepo-generator-rdf.09:27
<fasseg>ajsf6: It seems im quite finished, you can check out the jersey-grizzly branch and give it a scrutinizing look...:
quite finished in every sense, i'll take a break now ;)09:28
<ajs6f>fasseg: Will do. IIWY I'd run a few cool-down laps. Maybe code up a simple text-editor to shake out your muscles. :)09:33
<fasseg>hehe, i was thiking about creating a JaxRsContainer for Resteasy that is spring compatible ;)09:36
* eddies joins09:38
* eddies leaves
* eddies joins
<pivotal-bot>Frank Asseg edited "Issues stopping Jetty via the jetty-maven-plugin in between integration tests" https://www.pivotaltracker.com/story/show/4764708109:40
Frank Asseg accepted "Issues stopping Jetty via the jetty-maven-plugin in between integration tests" https://www.pivotaltracker.com/story/show/4764708109:41
Frank Asseg edited "Use Jersey instead of CXF in fcrepo4" https://www.pivotaltracker.com/story/show/4618366709:44
<fasseg>hmm accepted my own ticket there...hope noone notices ;)
<ajs6f>fasseg: one thing I'm noticing— the old CXF Maven declarations seems all to be still there, and I don't see the new Jersey dependencies being managed in fcrepo4/pom.xml. That's leading to a lot of duplication of versions, etc. Did you hold off on that for some particular reason? (Like it was too early to figure out what versions were going to get used?)09:48
<fasseg>true. that should go in there...
<ajs6f>Maybe make a new one-point ticket?
<ajs6f>"Cause maybe someone else should do that, to force us to be familiar with your new code.09:49
<pivotal-bot>Frank Asseg added "Configure Jersey/Grizzly dependency management and remove CXF leftovers" https://www.pivotaltracker.com/story/show/4832426909:50
Frank Asseg edited "Configure Jersey/Grizzly dependency management and remove CXF leftovers" https://www.pivotaltracker.com/story/show/48324269
Frank Asseg edited "Configure Jersey/Grizzly dependency management and remove CXF leftovers" https://www.pivotaltracker.com/story/show/4832426909:51
<fasseg>owner: none then ;)
<ajs6f>Fedo Radmin
<pivotal-bot>Frank Asseg edited "Configure Jersey/Grizzly dependency management and remove CXF leftovers" https://www.pivotaltracker.com/story/show/48324269
<bljenkins>Project fcrepo-kitchen-sink build #179: SUCCESS in 5 min 2 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/179/
<fasseg>which is used for?09:52
or just instead of none?
<ajs6f>It's Fedo. Need we say any more?
<fasseg>Now, I need a pet so i can call him that.
<bljenkins>Project fcrepo-fixity build #170: SUCCESS in 4 min 21 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/170/09:53
<pivotal-bot>A. "Forbulo" Soroka started "Configure Jersey/Grizzly dependency management and remove CXF leftovers" https://www.pivotaltracker.com/story/show/4832426909:58
<ajs6f>fasseg: Running the i-tests, I get an error from fcrepo-webapp's sanity check test:10:10
2013-04-18 10:08:45.488:WARN:oejs.Holder:
java.lang.ClassNotFoundException: com.sun.jersey.spi.spring.container.servlet.SpringServlet
at org.codehaus.plexus.classworlds.strategy.SelfFirstStrategy.loadClass(SelfFirstStrategy.java:50)
Any thoughts?
<fasseg>eddies, ajs6f, cbeer: If we can switch to sesame 2.6.10 we can use the maven central repos: http://search.maven.org/#search%7Cgav%7C1%7Cg%3A%22org.openrdf.sesame%22%20AND%20a%3A%22sesame%22
spring-test is missing i think...lemme check10:11
<eddies>fasseg: i think any23 is the one that has the aduna dependencies, right?
<fasseg>jersey-spring 1.17.1 seems to be missing from the classpath somehow
i thought openrdf10:12
<eddies>so as long as any23 can deal with the newer versions
<fasseg>yeah aou're right it's a dpeendency on apache-any
<eddies>which project declares openrdf dependency?
<eddies>so, yes, i'd say it's worth a try10:13
i have no idea what happened to aduna
<fasseg>just include newer versions on top the?
<ajs6f>Aduna's disappearance is bugging a couple of different projs.
<fasseg>biy the domain and put you're resumee there ;)10:14
<ajs6f>fasseg: should I add jersey-spring as a dep for fcrepo-webapp, in test scope?
<eddies>i forget: if you declare the newer versions of those dependencies in the parent pom's dependency management…will that override the transitive deps?
i made the horrible experience just now10:15
<fasseg>since webhooks needs httpclient at compile time but parent declares it as test-scoped
<eddies>then i think that's where it should go
<ajs6f>What should go?
<fasseg>the dependey declaration for the openrdf version10:16
so that we can pull directly from maven central
ajs6f: did you get rid of the sanity check classnotfoundexception?10:19
or idi i forget to include somethig in the pom?
<ajs6f>No, I don't know. You said something ^^^ about jersey-spring?10:20
<fasseg>right, SpringServlet is part of the jersey-spring 1.17.1 dependency.
<ajs6f>So add jersey-spring as a test scope dep, or complile scope dep?10:21
<fasseg>test-scoped...but it's already in there at the bottom, last dependency
compile-scoped though ;)
<ajs6f>So that's not the prob.10:22
<fasseg>can you gist the whole output plx?
arg committer call in 8 mins?10:23
<ppound>still having some issues I'm now stuck on fcrepo web app.
get this message Could not transfer artifact org.aduna-software:aduna-opensource-root-pom:pom:5 from/to aduna-software-release-repo (http://repo.aduna-software.org/maven2/releases
i think it wants to down load openrdf sesame
<ajs6f>fasseg: yah, give me a sec to generate a clean log of an attempt.
<fasseg>yep same problem, since webapp imports generator-rdf
ppound ^^
<ajs6f>ppound: Right. I was wrong to say that nothing depdends on generator-rdf,10:24
because fcrepo-webapp and kitchen-sink do.
<ppound>Nigel sent me the open sesame files
how do I make maven aware of them
<fasseg>ppound: you can try overriding the sesame version in the dependency management of fcrepo4/pom.xml10:25
mvn install:install-file
<ajs6f>ppound: yeah, you should need to install the files.
<fasseg>lots of examples on the web
<ppound>i'll take a look
<fasseg>and the man page is nice too
<ajs6f>fasseg: https://gist.github.com/ajs6f/541313410:26
<fasseg>this on is the culprit i think: ClassNotFoundException: org.springframework.web.context.ContextLoaderListener10:29
spring-web maybe ...I'll check
<ajs6f>fasseg: thanks!10:30
* github-ff joins10:34
[fcrepo4] fasseg pushed 1 new commit to jersey-grizzly: http://git.io/qSLBSA
fcrepo4/jersey-grizzly 55422e1 fasseg: added spring-web as a test-scoped dep
* github-ff leaves
<fasseg>ajs6f: can you pull my pom-xml change and try to build again?
<ajs6f>Yeah, I'll try.
<fasseg>i just added spring-web as a test-dep
<pivotal-bot>feature: Copy node from one Infinispan cache to another (unstarted) / owner: Chris Beer
feature: Policy-aware BinaryStoreValueFactory implementation (unstarted) / owner: Chris Beer
<ajs6f>fasseg: Not quite there:
org.springframework.beans.factory.BeanDefinitionStoreException: Failed to read candidate component class: URL [jar:file:/Users/ajs6f/.m2/repository/org/modeshape/modeshape-web-jcr-rest/3.2-SNAPSHOT/modeshape-web-jcr-rest-3.2-SNAPSHOT.jar!/org/modeshape/web/jcr/rest/form/FileUploadForm.class]; nested exception is java.lang.IncompatibleClassChangeError: class org.springframework.core.type.classreading.ClassMetadataReadingVisitor has int
looks like our new stuff is breaking the MODE HTTP API that we try to load into fcrepo-webapp.10:40
and https://github.com/ModeShape/modeshape/blob/master/modeshape-jcr/src/main/java/org/modeshape/jcr/JcrValue.java#L178
<fasseg>hmm i don't get it...lemme try to reproduce...10:50
<ajs6f>fasseg: Here is the full error:10:52
* github-ff joins10:53
[fcrepo4] fasseg pushed 1 new commit to jersey-grizzly: http://git.io/PCVdoA
fcrepo4/jersey-grizzly 134551c fasseg: removed test-scope and moved dependencies
* github-ff leaves
<fasseg>ajs6f: can you try again pls?10:54
although I think I had a diff exception before, but now it runs again for me
<ajs6f>Urg. I'm getting conflicts10:56
<pivotal-bot>Edwin Shin added "Copy node from federated filesystem to Infinispan cache in a cluster" https://www.pivotaltracker.com/story/show/4833159311:03
Edwin Shin edited "Copy node from federated filesystem to Infinispan cache in a cluster" https://www.pivotaltracker.com/story/show/48331593
A. "Forbulo" Soroka finished "Copy node from federated filesystem to Infinispan cache" https://www.pivotaltracker.com/story/show/4825242311:05
A. "Forbulo" Soroka added comment: "https://github.com/ajs6f/node-goes-from-federated-to-ispn" https://www.pivotaltracker.com/story/show/48252423
Failed tests:11:08
BagItConnectorTest.testIdFor:116 expected:</foo/[data/]bar> but was:</foo/[]ba
Tests in error:
BagItConnectorTest.testFileFor:106 ╗ StringIndexOutOfBounds String index out o
BagItConnectorTest.testGetBagInfo:82 ╗ StringIndexOutOfBounds String index out
BagItConnectorTest.getDocumentById:94 ╗ StringIndexOutOfBounds String index ou
barmintor: nope, clean test no good11:09
let me make sure i have the latest
<pivotal-bot>Edwin Shin added "Update fcrepo4 parent pom to use Maven Central for aduna dependencies" https://www.pivotaltracker.com/story/show/4833221511:10
Edwin Shin edited "Update fcrepo4 parent pom to use Maven Central for aduna dependencies" https://www.pivotaltracker.com/story/show/48332215
* nbanks leaves11:11
<VincentNG>barmintor: i have the latest from https://github.com/futures/fcrepo-bagit-modeshape-federation-connector.git
master branch11:12
<barmintor>the last commit is "owner property is required for FedoraObject"
<ajs6f>fasseg: Still failing, but now with:
2013-04-18 11:12:03.414:WARN:oejuc.AbstractLifeCycle:FAILED jersey-serlvet: javax.servlet.UnavailableException: com.sun.jersey.spi.spring.container.servlet.SpringServlet
javax.servlet.UnavailableException: com.sun.jersey.spi.spring.container.servlet.SpringServlet
<VincentNG>ya, i even tried wiping it out and clone it all over again11:13
* github-ff joins11:17
[fcrepo4] ajs6f pushed 1 new commit to jersey-grizzly: http://git.io/-u7tEg
fcrepo4/jersey-grizzly d49a167 ajs6f: Cleaned out CXF stuff and moved version control to fcrepo/pom.xml
* github-ff leaves
<ajs6f>fasseg: I just pushed back the code I'm working with, which differs from your by having the CXF stuff removed.
<ajs6f>fasseg: I don't think that could be relevant, but you never know...
<bljenkins>Project fcrepo-kitchen-sink build #180: SUCCESS in 4 min 31 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/180/
<fasseg>ill try..
<ajs6f>At least we can try and duplicate the error.11:18
<bljenkins>Project fcrepo-fixity build #171: SUCCESS in 4 min 15 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/171/11:19
<fasseg>ajs6f: after merging the changes i get this one: java.lang.ClassNotFoundException: com.sun.jersey.spi.spring.container.servlet.SpringServlet11:20
ajs6f: build-breaker :)11:21
<ajs6f>fasseg: So… what's missing?11:22
<fasseg>no idea checking atm...yeah seems like jersey-spring to me
<anusha>eddies: correct11:23
<VincentNG>barmintor: the filesystem connector works fine for me with simply files
<barmintor>with what test?
<VincentNG>if i remove the json properties
let me run your test against my filesystem11:24
<barmintor>I found that it looked like it worked when you called federateFrom, but it didn't actually work when you tried to get the DS
* ap2972 leaves11:27
<pivotal-bot>Edwin Shin edited "Do MODE-1710" https://www.pivotaltracker.com/story/show/48184899
<fasseg>ajs6f: i can reproduce now: UnavailableException: com.sun.jersey.spi.spring.container.servlet.SpringServlet11:29
<ajs6f>fasseg: cool.
<fasseg>class incompability: jersey-spring imports spring-aop 3.0.0 :/11:31
<pivotal-bot>feature: Policy-aware BinaryStoreValueFactory implementation (unstarted) / owner: Chris Beer
feature: Copy node from one Infinispan cache to another (unstarted) / owner: Chris Beer
* ap2972 joins11:37
<ajs6f>fasseg: is that a problem? Is that a version conflict for us?11:41
<fasseg>yeah it seems some internals have changed, but im on the way to fix it, ill come back to you....
<ajs6f>fasseg: cool.11:42
<eddies>ffs. there must be 2 dozen sesame dependencies
<ajs6f>Seseme is a lot more than an RDF API. There's all kinds of stuff fin there.
persistnce and inferencing inter al.
<eddies>where is the aduna maven repo referenced?11:43
<cbeer>gotta store all those AbstractFactoryBeanProxyFactoriesBeanFactory somewhere
<eddies>is it in the any23 pom?
<ajs6f>via apache any23
<eddies>i don't see it in any of ours…ok
<cbeer>ajs6f: any chance the any23 people will solve the problem for us?
<ajs6f>They're probably having the same problem.
I don't know what they're doing.
(about it)
We could try to find out.11:44
<cbeer>ok. i'm trying to find their mailing list on their website and am failing
<ajs6f>you have to find it via an RDF link. :)
it's all part of THE GRAPH.
* kaarefc joins11:45
<cbeer>oh, right. i remember why i gave up yesterday.11:46
the JcrValue doesn't contain a reference to the node either
<bljenkins>Project fcrepo-kitchen-sink build #181: SUCCESS in 3 min 48 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/181/
<ajs6f>cbeer: yea, Values are supposed to be "atomic".
directly comparable for identity w/o respect to their "owners"11:47
* anusha leaves
<cbeer>oh, right. but it doesn't matter. we pre-wrap our Binary properties
<bljenkins>Project fcrepo-fixity build #172: SUCCESS in 4 min 29 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/172/11:49
<cbeer>right. it's at the property level where we actually care.11:50
<barmintor>eddies: stumptown has a roaster in midtown. acceptable?
<barmintor>ajs6f: it's a town made of stumps.
<ajs6f>That is powerful architecture.
<eddies>barmintor: ack. really? had i known that i wouldn't have ordered http://buy.stumptowncoffee.com/kenya-gaturiri-2.html11:51
<eddies>which was a) pretty expensive on its own and b) shipping was a bit mad
oh well
<ajs6f>why don't we roast our won?
<fasseg>ajs6f: Is it possible you pushed some changes to FedoraIdentifiers? there is a "close" call in there now which i dont have in my sources...
<eddies>actually, let me check. usps claims some of the coffee i ordered should arrive in salem by tomorrow
<fasseg>so is jersey-grizzly now half merged with master?
<eddies>but given that it's still in washington state, i'm a bit doubtful11:53
<ajs6f>Yes, I did, because Guave version bump deprecated some of our code. But I don't see how it affects anything else, and all of the tests pass for that.
<eddies>barmintor: so yes, anything from stumptown would be fine
<ajs6f>fasseg: no, not merged at all.
can you roast coffee beans in a bbq grill?
<cbeer>ajs6f: ok, modeshape is wacky: https://github.com/ModeShape/modeshape/blob/master/modeshape-jcr/src/main/java/org/modeshape/jcr/JcrSingleValueProperty.java#L216
<eddies>just don't get the exact same one i ordered, i guess
ajs6f: you can roast in a wok or air popper
<cbeer>if you ask for the property's stream, it'll stick it through your binary store in order to deliver it to you?
<ajs6f>well, doesn't it _have_ to go to the store?11:54
eddies: we need a wok anyway, for the amazing stir fry that we're going to eat.
coffee bean stir fry!
<cbeer>ajs6f: i don't understand where property values get stored. i thought they got serialized into the MODE/ISPN "schematic"11:55
with references for binary content
<ajs6f>I thought so too.
<cbeer>and just the value for other things
Binaries are special.
<cbeer>so, if I have a string property.. and i ask for getStream(), it's going to round-trip that through ISPN
as I read that code
(answer, don't do that.)
<ajs6f>Yeah, because asking for a Stream from a String value is non-idiomatic.
Just ask for the String.11:56
If your String is really that big,
you ought to store it as a Binary.
That's my understanding, anyway.
<cbeer>right. so, do stupid stuff, get a stupid side-effect.
That's why there are all those different dignatures for getting values from properties.11:57
* nbanks joins
<ajs6f>As barmintor knows, Scala would make that look a lot cleaner. )
Urg. CLustering seems to need JNDI.11:58
<fasseg>ajs6f: it seems guava 13.0.1 doesn't have a Range.close method so i cant compile FedoraIdentifiers...can you give me a hint on how to fix that?
<ajs6f>fasseg: are you using the code I pushed? Because I bumped the Guava version (cause we were well behind).11:59
<fasseg>i thought so...
<ajs6f>Well, i cahnged the version int he same commit as changing the code...
What's your Guava version in fcrepo/pom.xxml
<fasseg>commit d49a1679bbf41821f26fab7a0d4a395d3acdfc84 guava 13.0.112:00
in fcrepo-kernel
<version>13.0.1</version> in parent pom
<ajs6f>That's not the same as mine.12:01
commit d49a1679bbf41821f26fab7a0d4a395d3acdfc84
Hm. That's the right commit.
<fasseg>"Cleaned out CXF stuff and moved version control to fcrepo/pom.xml"
But I have (in parent):
* ap2972 leaves
<eddies>so far as i can tell, none of the aduna-* dependencies (of any version) are in maven central :(12:02
<ajs6f>Some kind of incomplete merge?
<eddies>i have all the sesame-* dependencies managed and working
i seriously wonder if aduna went out of business or something
that's going to suck for openrdf12:03
<ajs6f>And a lot of other people who use it. Like us.
<fasseg>eddies: http://www.openrdf.org/download.jsp you can get openrdf at least from maven central from 2.6.10 onward
<eddies>there are no openrdf dependencies in generator-rdf that i see
<fasseg>from apache-any23
<ajs6f>Let's write our own RDF parser in Perl. That will fix this.12:05
<fasseg>in rdf not dc
<eddies>i see sesame-* and aduna-* dependencies
<eddies>i don't see any openrdf dependencies
<pivotal-bot>Highest karma: ajs6f (12), cbeer (8), fasseg (4)
Lowest karma: ---------------------------- (-2), ------------------------------6271f127ca48 (-2), metrics (-4)
<Scala4Win4Eva>ajs6f: s/perl/Scala/G
<pivotal-bot>Highest karma: ajs6f (12), cbeer (8), fasseg (4)
Lowest karma: ------------------------------6271f127ca48 (-2), ---------------------------- (-2), metrics (-4)
* github-ff joins
[fcrepo4] fasseg pushed 1 new commit to jersey-grizzly: http://git.io/KK_23A
fcrepo4/jersey-grizzly 01dfdf0 fasseg: fixed some versions in dependency management, bumped guava version to 14
* github-ff leaves
<fasseg>ajs6f: can you try to build the latest commit?
<eddies>fasseg: can you point me to a particular openrdf dependency?
<ajs6f>fasseg: in our branch?
<fasseg>eddies: fcrepo-generator-rdf/apache-any23-core/sesame-model 2.6.1 [compile]12:07
ajs6f: the jersey-grizzly branch..yeah12:08
* github-ff joins
[fcrepo-bagit-modeshape-federation-connector] barmintor pushed 1 new commit to master: http://git.io/TJMOzQ
fcrepo-bagit-modeshape-federation-connector/master 53832cd Benjamin Armintor: change path substitutions to not assume Unix FS
* github-ff leaves
<Scala4Win4Eva>VincentNG: ^^
<ajs6f>fasseg: k, doing it now
<eddies>fasseg: nm. re-read what i said above. there are sesame-* and aduna-* dependencies. i wasn't talking about things in the org.openrdf *group*, i was saying i don't see things like openrdf-util or openrdf-model referenced12:10
<ajs6f>fasseg: blew out with:
2013-04-18 12:10:24.584:WARN:oejw.WebAppContext:Failed startup of context o.m.j.p.JettyWebAppContext{/,file:/Users/ajs6f/Documents/fcrepo4/fcrepo4/fcrepo-webapp/src/main/webapp/},file:/Users/ajs6f/Documents/fcrepo4/fcrepo4/fcrepo-webapp/src/main/webapp/
javax.servlet.UnavailableException: com.sun.jersey.spi.spring.container.servlet.SpringServlet
<fasseg>eddies: oh i misunderstood sry12:11
ajs6f: oh no you don't!
<bljenkins>Project fcrepo-bagit-modeshape-federation-connector build #3: SUCCESS in 3 min 30 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-bagit-modeshape-federation-connector/3/
<ajs6f>fasseg: i'd rather not.
<fasseg>ajs6f: http://stackoverflow.com/questions/6375675/jersey-rest-issues but i guess you did a complete clean/install before you tries it...12:13
<ajs6f>mvn clean install12:14
<fasseg>on the parent level...
<fasseg>with java7
<ajs6f>:) Yes.
<fasseg>i need a break....brb12:15
* github-ff joins12:24
[fcrepo4] fasseg pushed 1 new commit to jersey-grizzly: http://git.io/Gz7-Yg
fcrepo4/jersey-grizzly 661bb98 fasseg: changed spring-aop version on the parent level and hoping for the best
* github-ff leaves
<fasseg>ajs6f: ...and once more pls
i hope i just forgot to include the aop version in the dep management12:25
<ajs6f>k, trying
cbeer: did you ever run into JNDI problems while getting clustering configured?12:26
<cbeer>no, never saw anything about JNDI
fasseg: success!12:27
<ajs6f>fasseg: enjoy rebasing :)12:28
<fasseg>hehe only 2 behind ;)
i rewrote on top startinf from last friday
<ajs6f>nice move.
Shouldn't be a prob then.
I wonder if we can do some kind of automatic check to make sure that CXF is gone.12:29
Maybe Sonar tells us that kind of thing?
<fasseg>no idea
but we could use the banning mechanis we use for commons-logging
<ajs6f>Well, we could set a rule for Maven enforcer...
* ap2972 joins
eddies: you set that up, right?
<eddies>ajs6f: there should already be enforcer rules for commons-logging12:30
<ajs6f>ww want to exclude cxf.
<fasseg>eddies: can you take a look at the jersey-grizzly branch too, and if it builds for you, I'll rebase on master tommorrow morning..
<eddies>fasseg: is it still not working for ajs6f?
<fasseg>eddies: it is now
<ajs6f>eddies: hold back for a minute— I'm going to add a cxf excluder rule.
<fasseg>but I wouldn't want to break the travis/jenkins build when everyone is developing, and in the morning i have almost all the infrastructure for me12:32
* VincentNG leaves
<fasseg>perks of being in europe i guess
* VincentNG joins12:33
<ajs6f>eddies: okay, please try pulling jersey-grizzly and building12:36
* github-ff joins
[fcrepo4] ajs6f pushed 1 new commit to jersey-grizzly: http://git.io/Os625A
fcrepo4/jersey-grizzly d58545d ajs6f: Added CXF excluder rule
* github-ff leaves
<VincentNG>Scala4Win4Eva: most of the tests are passing now. only BagItConnectorTest.testFileFor:111 is failing
you prob just missed the path on that test tho. no biggie12:37
<Scala4Win4Eva>VincentNG: oh, yeah12:38
wrong kind of slashes in the failure message?
* github-ff joins12:42
[fcrepo-bagit-modeshape-federation-connector] barmintor pushed 1 new commit to master: http://git.io/gfregA
fcrepo-bagit-modeshape-federation-connector/master 7ac04a4 Benjamin Armintor: typo in test (still had Unix path seps)
* github-ff leaves
* github-ff joins12:43
[fcrepo4] eddies pushed 1 new commit to master: http://git.io/rQDC5w
fcrepo4/master 0181f32 Edwin Shin: Because of the disappearance of the repo-aduno Maven repository, adds...
* github-ff leaves
<pivotal-bot>Edwin Shin added comment: "Fixed in https://github.com/futures/fcrepo4/commit/0181f324f353d3b345027b8b3f490cea7a78cc1f" https://www.pivotaltracker.com/story/show/4833221512:44
Edwin Shin accepted "Update fcrepo4 parent pom to use Maven Central for aduna dependencies" https://www.pivotaltracker.com/story/show/48332215
<eddies>ppound: you should be good to go
<ppound>ok I'll give it a try, thanks12:45
<bljenkins>Project fcrepo-bagit-modeshape-federation-connector build #4: SUCCESS in 3 min 11 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-bagit-modeshape-federation-connector/4/
<eddies>fasseg/ajs6f: am i just running mvn verify on the jersey-grizzly branch?12:46
<ajs6f>eddies: I did mvn clean install, but yeah, just making sure it builds
<eddies>ajs6f: mvn clean install succeeds on jersey-grizzly for me12:49
fasseg: merge away
<fasseg>cool...i will tommorrow, but now i'll end my day...i need some food and im getting the angry looks again ;)12:51
<ajs6f>eddies: after fasseg's merge, I'm going to open a ticket for splitting out the modules in fcrepo into separate Git repos.12:54
<pivotal-bot>A. "Forbulo" Soroka estimated "Move SessionFactory out of fcrepo-http-commons into a new fcrepo-authn module with default impl for HTTP...." https://www.pivotaltracker.com/story/show/47895901
<eddies>fasseg: np. of course, the longer you wait, the more to rebase ;-)12:55
<pivotal-bot>A. "Forbulo" Soroka finished "Configure Jersey/Grizzly dependency management and remove CXF leftovers" https://www.pivotaltracker.com/story/show/48324269
A. "Forbulo" Soroka delivered "Configure Jersey/Grizzly dependency management and remove CXF leftovers" https://www.pivotaltracker.com/story/show/48324269
<eddies>ajs6f: great. although let's treat that as a low-priority ticket/chore12:56
<pivotal-bot>A. "Forbulo" Soroka added comment: "Depends on https://www.pivotaltracker.com/story/show/48184899" https://www.pivotaltracker.com/story/show/47380619
<ajs6f>eddies: Sure. But the longer we wait, the more of a pain it'll be.
<fasseg>hehe true, so hold back on the commits today!
<ajs6f>eddies: so I do want to get it done.
<fasseg>have fun guys!
* fasseg leaves
<ajs6f>vincentNG/Scala4Win4Eva: is this:12:58
<pivotal-bot>feature: Create ModeShape BagIt federation connector (started) / owner: A. "Forbulo" Soroka
<ajs6f>effectively done?
<pivotal-bot>A. "Forbulo" Soroka started "Copy node from federated filesystem to Infinispan cache in a cluster" https://www.pivotaltracker.com/story/show/4833159312:59
<Scala4Win4Eva>I would say it's not done
I'm not really working on. SHould I transfer it to you or VNG?13:00
<pivotal-bot>A. "Forbulo" Soroka edited "Create ModeShape BagIt federation connector" https://www.pivotaltracker.com/story/show/47089843
<eddies>ajs6f: i don't disagree at all. but our priority *has* to be on storage features for this sprint. i'm willing to incur the increased debt of postponing breaking those modules out. if we get everything else sorted, ahem, early…we'll have time at the end of the sprint, right? =)13:01
<ajs6f>eddies: all I'm doing is making a ticket. for Fedo.13:02
* travis-ci joins13:05
[travis-ci] futures/fcrepo4#373 (master - 0181f32 : Edwin Shin): The build has errored.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/efc3e7709c18...0181f324f353
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6449360
* travis-ci leaves
<eddies>ppound: i guess i missed org.aduna-software:aduna-opensource-root-pom13:14
just added that to our nexus repo and restarted travis
<ppound>eddies: ok I'll try it just need one second13:16
<cbeer>design docs about binary stores: https://community.jboss.org/message/72786613:17
interestingly, rhauch proposes an API where the user can provide a key
and multiple binary stores!13:18
all hope is not lost13:20
<eddies>cbeer: that's a great find. and the very first questions raised are exactly what we were asking ourselves13:25
<cbeer>yes, it's a pretty good read
on page 2 are some questions about connector synchronization
and large video binaries
<eddies>it's almost like…we're not alone in our universe of needs13:26
<ppound>eddies: I'm failing at the same point with a mvn clean install13:27
eddies: the error is here https://gist.github.com/ppound/5414595
<cbeer>yeah. i should probably see if Jonathan Fields has any other forum posts of interest13:28
<eddies>ppound: i'm not sure why it's explicitly going after that resource from repo-aduna
travis is doing the same
(even though that artifact is now available on maven.fcrepo.org)
<ppound>eddies: that makes two of us13:29
<ajs6f>eddies/ppound: It may be the ordering of repository elements in the calculated POM.
We can't really do much with that.
They claim to be able to support non SHA keys for binaries now:13:30
<cbeer>time to fork any23?
<ajs6f>FORK EVERYTHING.13:31
<cbeer>ajs6f: yes, the BinaryKey contract doesn't enforce SHA-1s.
<ajs6f>cbeer: oh, cool. I didn't realize that.
<cbeer>although there are vestigal SHA-1 elements scattered around
and the InfinispanBinaryStore does enforce SHA-1
and in practice every BinaryStore is using SHA-1
<eddies>ppound/ajs6f: i'm going to try just excluding that dependency. we don't need it—in fact all it mostly does is add the dead repository13:34
<ppound>eddies: I see reference to aduna in openrdf-root-pom-5.pom.lastupdated13:35
in my .m2 dir
<ajs6f>which dependency/
<ppound>pound is the file
<ajs6f>Okay. I don't know why we would need that.
<eddies>ajs6f: we don't13:37
<ajs6f>Well, there ya go.
* travis-ci joins13:38
[travis-ci] futures/fcrepo4#373 (master - 0181f32 : Edwin Shin): The build has errored.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/efc3e7709c18...0181f324f353
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6449360
* travis-ci leaves
<eddies>hmm. it's not even showing up as a transitive dependency13:44
and enforcer doesn't catch it either
<ajs6f>I'm going into the office. Back online in a little while.13:51
* ajs6f leaves
* ap2972 leaves13:55
<cbeer>cbeer: next time you can't compile modeshape with this error:13:57
Could not transfer artifact oswego-concurrent:concurrent:pom:1.3.4-jboss from/to repository.jboss.org (http://repository.jboss.org/maven2/): Access denied to: http://repository.jboss.org/maven2/oswego-concurrent/concurrent/1.3.4-jboss/concurrent-1.3.4-jboss.pom
you forgot to read the README
compile with mvn clean install -s settings.xml !
* nbanks leaves
* nbanks joins
<eddies>ppound: you still there?14:12
eddies: I'm trying on a fresh box14:13
<eddies>can you cd to fcrepo-generator-rdf and run "mvn -U test"
no, fresh box won't help
one sec
<eddies>ppound: just stepping away for a min. maven always seems to drive me to drink :P14:14
<ppound>eddies: it's running14:15
eddies: says success14:16
<eddies>well hot diggity14:17
<ppound>eddies: seems to have built on the new vm image now as well
<eddies>travis is still failing, but it's on a latest aduna dependency i uploaded… so i think after some expiry period, it should work if it's now working for you paul14:19
*the latest
<ppound>eddies: yeah i'm building the whole thing now on my mac I'll see how it goes14:20
<eddies>ok. i'm really going to fix myself that drink now
<ppound>eddies: make it a double
eddies: and have one for me too14:21
<eddies>scala4win4eva: do you have your arrival/departure schedule finalized? if so, can you add it to https://wiki.duraspace.org/display/FF/2013-04+Hackfest ?14:23
<Scala4Win4Eva>my arrival is the same as ajs6f
<eddies>ok. b/c i need to make arrangements for the key pickup for whoever is first to the house14:24
we didn't exactly sort out if/how i was going to pick you two up did we?
i need to provide an estimated arrival if i'm going to be the first to arrive
<Scala4Win4Eva>not really, no
afaik we will get to south station ~ 7:00pm14:25
<eddies>how about you two just split a cab? =)14:26
or at least a cab to the point you can catch the direct commuter rail to salem
it looks like a cab is ~$65 total14:27
<Scala4Win4Eva>we'll work something out, but I will prefer it not to be a $65 cab ride14:28
<pivotal-bot>Chris Beer added "Crank up the MINIMUM_BINARY_SIZE_IN_BYTES as appropriate" https://www.pivotaltracker.com/story/show/4834859114:31
Chris Beer added "Use separate named caches for metadata and binary stores" https://www.pivotaltracker.com/story/show/4834877114:32
* ajs6f joins
eddies: I had planned to take commuter rail up to Salem for a few dollars.14:38
cbeer: The only thing that matters that MODE's settings.xml is adding is their maven repo. You can add that to your own personal settings.xml (e.g. ~/.m2/settings.xml) and forget about it.14:39
* ap2972 joins
<ajs6f>Scala4Win4Eva: The line we want to catch (Newburyport/Rockport) leaves North Station at 8:30, which should give us plenty of time. I think.14:40
And get us to Salem about 9. Then a few minutes walk to the house.14:41
* travis-ci joins14:42
[travis-ci] futures/fcrepo4#375 (master - 0181f32 : Edwin Shin): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/ae48eb4bf086...0181f324f353
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6451495
* travis-ci leaves
<eddies>YES! it's alive!14:44
travis-ci++ ;)
* travis-ci joins14:50
[travis-ci] futures/fcrepo4#377 (master - 0181f32 : Edwin Shin): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/ae48eb4bf086...0181f324f353
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6451792
* travis-ci leaves
* travis-ci joins
[travis-ci] futures/fcrepo4#374 (master - 0181f32 : Edwin Shin): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/ae48eb4bf086...0181f324f353
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6450361
* travis-ci leaves
<eddies>fyi, i've been adding version markers to Sonar to roughly coincide w/ each sprint: http://sonar.fcrepo.org/dashboard/index/1?did=414:51
our violations are up, comment and coverage are down :(14:52
* BHSPiMonkey leaves14:55
* BHSPiMonkey joins
<ajs6f>eddie: for a lot of "style" type violations, (not using brackets for for loops, that kind of thing) Eclipse has automatic clean up facilites.15:03
<pivotal-bot>Paul Pound added "make islandora work with latest built of fcrepo4. " https://www.pivotaltracker.com/story/show/4835140115:04
Paul Pound started "make islandora work with latest built of fcrepo4. " https://www.pivotaltracker.com/story/show/48351401
<cbeer>i'll assign some of those new style violations to me. i think i introduced some of them killing off the old bookkeeping code15:21
<ajs6f>cbeer: do Source -> Clean up and take a look at some of the custom options. There's a lot in there and it can save a lot of time.15:23
* github-ff joins15:27
[fcrepo4] cbeer pushed 1 new commit to master: http://git.io/TLI-rw
fcrepo4/master 05cad00 Chris Beer: clean up style errors after removal of bookkeeping synchronous code
* github-ff leaves
<cbeer>eddies: i think comments are down because we junked all that bookkeeping code, that was prolifically commented.15:29
and we're getting 3x dinged on the style things i just fixed.
<ajs6f>I know there's a google amount of the rdf generator that could be JavaDoc'ed with a 2x4.15:30
<eddies>we don't need to specifically deal with it now. but just to keep mind of it as we continue to do new work15:31
* ajs6f leaves15:34
* ajs6f joins15:37
* nbanks leaves15:42
<ajs6f>eddies/all: Does Sonar check every branch, or just master?15:46
<eddies>i assume only master
since the instrumenting is initiated from jenkins, which is only building master, afaik15:47
<ajs6f>Oh, well. I was going to take five minutes and run the Eclipse clean up routines over the whole fcrepo4 source and see if that cured some of what ails Sonar.
But I would only do that in a branch, becaus it's likely to make a big delta.
<eddies>ah, all of sonar's violations come from the various pmd, checkstyle and findbugs. so assuming you use the same config, you should be able to install those plugins in eclipse and see the same15:49
but you *really* oughtn't take the time right now =)
<ajs6f>Yeah, but we want to not just see them, but make them go away. WIthout spending a lot of time doing it. (I'm trying to automate my life.)15:50
<eddies>cluster & copy/move your nodes first ;-)
<ajs6f>eddies; If you know how to config those Eclipse plugins, you ought to add that info to the "Setting up Eclipse for fcrepo4" doc.15:54
* ppound leaves15:57
* ajs6f leaves16:03
<VincentNG>is anyone able to do a mvn clean install on this: https://github.com/futures/fcrepo-bagit-modeshape-federation-connector16:07
ben's able to build it but i'm not.
* ajs6f joins16:08
* nbanks joins16:13
<eddies>vincentng: works for me16:15
* travis-ci joins
[travis-ci] futures/fcrepo4#378 (master - 05cad00 : Chris Beer): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/0181f324f353...05cad008df47
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6453565
* travis-ci leaves
<VincentNG>eddies: thanks. my maven must be corrupted or something. i tried it on two computes
<Scala4Win4Eva>I do think there's a dependency issue with the LOC lib16:16
which I'm working on
<eddies>i thought i saw ben push something that was about platform-specific behavior...
vincentng: are you running windows?
<VincentNG>eddies: yes. i'm startin to feel like i'm the only one :-(16:17
<eddies>have you tried on anything else, e.g. linux?
<VincentNG>i don't have anything else. we're a windows shop at the CDC
i use windows at home at the moment
* nbanks leaves16:18
* github-ff joins16:23
[fcrepo-bagit-modeshape-federation-connector] eddies pushed 1 new commit to master: http://git.io/seI8JQ
fcrepo-bagit-modeshape-federation-connector/master db5c9a9 Edwin Shin: Removes Maven warnings about using the wildcards for the artifactId in dependency exclusions
* github-ff leaves
<eddies>vincentng: have you posted the error you're getting somewhere?16:24
<bljenkins>Project fcrepo-bagit-modeshape-federation-connector build #5: SUCCESS in 2 min 32 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-bagit-modeshape-federation-connector/5/16:25
<VincentNG>eddies: i manually deleted my .m2/repo/org/fcrepo.fcrepo-bagit.... and did a new mvn install and got this: https://gist.github.com/vnguyen745/541592916:26
<eddies>mvn -U verify16:27
and do a pull first
then mvn -U clean verify
and if it's still failing, enable debugging (mvn -X verify)16:28
<VincentNG>it's still failing. let me try mvn -X verify16:29
<eddies>well, if you read the stacktrace in the first gist you provided, i think it gives you the answer anyway16:30
tryProgrammaticAccess(org.fcrepo.federation.bagit.BagItConnectorIT) Time elapsed: 0 sec <<< FAILURE!16:31
java.lang.AssertionError: null
at org.fcrepo.federation.bagit.BagItConnector.fileFor(BagItConnector.java:310)
at org.fcrepo.federation.bagit.BagItConnector.getDocumentById(BagItConnector.java:150)
* ap2972 leaves
is assert id.startsWith(JCR_PATH_DELIMITER);
and JCR_PATH_DELIMITER = "/";16:32
* ajs6f leaves16:33
<eddies>i assume on windows when you get the absolute path, it starts with a "C:\" or somesuch
<Scala4Win4Eva>oh, that's no good
<eddies>well, i dunno, i didn't trace it back all the way so i may be wrong16:35
<Scala4Win4Eva>something else
w/e you're talkig about is probably fine
<eddies>but vincentng should be able to drop in a few printlns to verify what's going on there
or actually, you put in logging there, so just mvn -X should show16:36
<VincentNG>eddies: thanks. i'll do that
* ajs6f joins16:38
* Scala4Win4Eva leaves16:40
* github-ff joins16:41
[fcrepo-bagit-modeshape-federation-connector] barmintor created inlinebagit (+1 new commit): http://git.io/1JRidA
fcrepo-bagit-modeshape-federation-connector/inlinebagit 236d054 Benjamin Armintor: begin inlining bagit to fix lib incompat around commons-io
* github-ff leaves
<bljenkins>Project fcrepo-bagit-modeshape-federation-connector build #6: SUCCESS in 2 min 46 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-bagit-modeshape-federation-connector/6/16:43
* kaarefc leaves16:53
* ajs6f leaves16:59
* VincentNG leaves17:00
* nbanks joins17:24
* nbanks leaves17:29
* nbanks joins18:25
* nbanks leaves18:31
* BHSPiMonkey leaves18:57
* nbanks joins19:25
* nbanks leaves19:30
* ap2972 joins20:14
* ap2972 leaves20:20
* nbanks joins20:26
* nbanks leaves20:31
* github-ff joins21:12
[node-goes-from-federated-to-ispn] ajs6f created clustered (+1 new commit): http://git.io/ZhAziw
node-goes-from-federated-to-ispn/clustered 39fc9a7 ajs6f: First attempt at clustering, still failing
* github-ff leaves
* nbanks joins21:26
* nbanks leaves21:31
* VincentNG joins23:14
* VincentNG leaves23:15

Generated by Sualtam