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

Using timezone: Eastern Standard Time
* robyj joins01:43
anyone on?01:45
* jonroby leaves01:46
<robyj>I Guess not. Ok, What do I use instead of JQLConverter now to send the core sparql queries? the oai provider makes/made use of JQLConverter and now as its removed, it'll have to be fixed. thanks.01:52
* fcrepo-bot joins02:13
* fcrepo-bot leaves03:17
* escowles leaves03:21
* dwilcox joins05:45
* nikhiltri joins08:00
* nikhiltri leaves08:04
<pivotal-bot>Jonathan Roby added comment: "removal of JQLConverter has somewhat broken the OAI functionality until I/someone figures out an alternative�" https://www.pivotaltracker.com/story/show/8148290408:24
* MohamedAR joins08:25
* ksclarke joins08:26
* jgpawletko joins08:30
* awead joins08:48
<pivotal-bot>Jonathan Roby added comment: "After some testing of what exactly produces the 400 HTTP code, it seems that everything is fine until Node.g�" https://www.pivotaltracker.com/story/show/8185050408:49
* acoburn joins09:00
* robyj leaves09:03
<pivotal-bot>Andrew Woods added comment: "Will you have some time today to give this a look, @ajs6f?" https://www.pivotaltracker.com/story/show/8198503809:52
* ksclarke leaves09:53
<pivotal-bot>Andrew Woods added "Fix oaiprovider from impacts of fcrepo4 updates" https://www.pivotaltracker.com/story/show/8212372209:58
Andrew Woods started "Fix oaiprovider from impacts of fcrepo4 updates" https://www.pivotaltracker.com/story/show/82123722
* dwilcox leaves10:01
* awead leaves10:03
* awead joins10:05
* ksclarke joins10:09
* ksclarke leaves
* ksclarke joins
* ruebot_ joins10:23
* ruebot leaves10:24
* jonroby joins
* longshou joins10:34
<jonroby>What do I use instead of JQLConverter now to send the core sparql queries? the oai provider makes/made use of JQLConverter and now as its removed, it'll have to be fixed. thanks.10:51
<f4jenkins>Yippee, build fixed!10:56
Project fcrepo4-T2 build #44: FIXED in 8 min 15 sec: http://jenkins.fcrepo.org/job/fcrepo4-T2/44/
* Chris Beer: Remove fcr:sparql endpoint
* Chris Beer: Always store in-domain URIs are reference properties, and serialize them in the current HTTP context
* dwilcox joins10:57
<awoods>jonroby: I am working: https://www.pivotaltracker.com/story/show/8212372210:58
<pivotal-bot>feature: Fix oaiprovider from impacts of fcrepo4 updates (started) / owner: Andrew Woods
* scossu joins11:02
* ruebot_ leaves11:06
* ruebot joins
* ruebot leaves
* ruebot joins
<pivotal-bot>Andrew Woods added "Create initial Grinder test" https://www.pivotaltracker.com/story/show/8213226211:26
Andrew Woods edited "Create initial Grinder test" https://www.pivotaltracker.com/story/show/82132262
Andrew Woods edited "Configure fcrepo-test-grinder for fcrepo4 work" https://www.pivotaltracker.com/story/show/76247430
* dwilcox leaves11:32
<ksclarke>awoods: you mentioned you had a graphite question/comment too?11:42
<awoods>ksclarke: yes, I was just wondering what possibilities you saw for: http://grinder-to-graphite.readthedocs.org/en/latest/
<ksclarke>yeah, I have that bookmarked as well11:43
definitely worth configuring, imho, once both are spin-up-able
<awoods>ksclarke: it seems like an interesting "next step"11:44
<ksclarke>yes, indeed
<awoods>cbeer: ping
* longshou leaves12:08
* scossu leaves12:14
* scossu joins12:16
<pivotal-bot>Andrew Woods edited "Create initial Grinder test" https://www.pivotaltracker.com/story/show/8213226212:20
* dwilcox joins
<pivotal-bot>Andrew Woods added "JMS Message property list" https://www.pivotaltracker.com/story/show/8213517612:22
Andrew Woods edited "JMS Message property list" https://www.pivotaltracker.com/story/show/82135176
Andrew Woods edited "JMS Message property list" https://www.pivotaltracker.com/story/show/82135176
A. "Offratonix" Soroka added comment: "Sorry, probably not today, but very likely tomorrow." https://www.pivotaltracker.com/story/show/8198503812:25
Mohamed Mohideen Abdul Rasheed added comment: "AccessDeniedException stacktrace while creating datastream/childnode " https://www.pivotaltracker.com/story/show/7298294812:39
* longshou joins12:54
* MohamedAR leaves13:02
* awead_ joins
<pivotal-bot>Longshou Situ started "Create initial Grinder test" https://www.pivotaltracker.com/story/show/8213226213:03
* awead leaves13:05
* acoburn leaves13:06
<pivotal-bot>Andrew Woods added "Fix oaiprovider ListSetsIT" https://www.pivotaltracker.com/story/show/8214139413:08
* MohamedAR joins13:11
<pivotal-bot>Andrew Woods added comment: "Resolved with: https://github.com/fcrepo4-labs/fcrepo4-oaiprovider/commit/1f6c82aa7e95b81927674c370154633b998�" https://www.pivotaltracker.com/story/show/8212372213:13
Andrew Woods delivered "Fix oaiprovider from impacts of fcrepo4 updates" https://www.pivotaltracker.com/story/show/82123722
<awoods>jonroby: oaiprovider should be ready for you now13:14
<jonroby>awoods: thank you, i'll check it out
<awoods>MohamedAR: Where do things stand on the AuthZ issue?13:16
<MohamedAR>awoods: FedoraTypeUtils:getClosestExistingAncestor is where the problem occurs.13:22
<awoods>MohamedAR: I was seeing that as well. I have a patch for that... but then the same issue shows up lower in the stack.13:23
MohamedAR: One moment and I will share the patch.13:24
* awoods currently building another branch
<MohamedAR>awoods: ok
<jonroby>awoods: oaiprovider works out of the box, now to modify. thanks for the work13:25
<awoods>jonroby: glad to hear it
* github-ff joins13:29
[fcrepo4] awoods pushed 2 new commits to master: http://git.io/n1ennw
fcrepo4/master 0f97167 lsitu: Added the toString() method for FedoraResource.
fcrepo4/master abc3fb1 Andrew Woods: Merge pull request #622 from lsitu/feature/resource2string...
* github-ff leaves
<pivotal-bot>Andrew Woods edited "Fix oaiprovider ListSetsIT" https://www.pivotaltracker.com/story/show/8214139413:31
Andrew Woods edited "Fix oaiprovider ListSetsIT" https://www.pivotaltracker.com/story/show/82141394
Andrew Woods edited "Fix oaiprovider ListSetsIT" https://www.pivotaltracker.com/story/show/82141394
Andrew Woods delivered "FedoraResource.toString()" https://www.pivotaltracker.com/story/show/8187229813:32
<awoods>MohamedAR: https://github.com/awoods/fcrepo4/compare/create-gchild13:41
MohamedAR: That is just a draft, so the tests have not been updated.
afk lunch13:45
* nikhiltri joins13:48
* travis-ci joins13:50
fcrepo4/fcrepo4#3135 (master - abc3fb1 : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/4a3630763b25...abc3fb1a584b
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/40093085
* travis-ci leaves
<jonroby>awoods: question14:35
<awoods>jonroby: the beauty of IRC is that it is asynchronous. You do not need to wait for acknowledgement to proceed.14:38
<jonroby>awoods: well, I didn't want it to get lost in the chatter14:39
awoods: Can you think of a reason why root.hasProperty("fedora:writable") would work (200 HTTP code, test is OK) and root.hasProperty("oai:adminEmail") would produce 400, test fails??14:40
its the oai namespace causing the 400 error?
<awoods>jonroby: likely the 400 is due to the undefined namespace, yes.14:41
<jonroby>awoods: so solutions would be to either add the namespace to root or use the properties under another namespace?14:42
<awoods>jonroby: or to use the fully qualified namespace.
<jonroby>awoods: I wasn't sure about that as all the examples i've found seem to only use the compact namespace, even in the test harnesses14:43
<awoods>jonroby: give it a try
<jonroby>i'll try the FQNS first
awoods: cheers
* awead_lunch leaves14:44
<awoods>MohamedAR: With the patch, the AccessDenied exception is now coming from: https://github.com/fcrepo4/fcrepo4/blob/master/fcrepo-kernel-impl/src/main/java/org/fcrepo/kernel/impl/services/AbstractService.java#L4914:50
* awead joins14:51
<awoods>MohamedAR: so the issue is coming from: https://github.com/ModeShape/modeshape/blob/master/modeshape-jcr-api/src/main/java/org/modeshape/jcr/api/JcrTools.java#L41514:54
* dwilcox leaves14:56
<MohamedAR>awoods: should we create a patch for modeshape?14:57
<awoods>MohamedAR: It seems like there are two options: create an integration test against ModeShape that demonstrates the issue...14:58
MohamedAR: and/or investigate how we handle this in fcrepo-module-auth-rbacl
MohamedAR: it does not look like the block of code in ModeShape has changed in a few years.14:59
MohamedAR: RBACL uses an "internal session" - https://github.com/fcrepo4/fcrepo-module-auth-rbacl/blob/master/fcrepo-auth-roles-common/src/main/java/org/fcrepo/auth/roles/common/AbstractRolesAuthorizationDelegate.java#L10115:01
<jonroby>awoods: FQNS does not work with the oaiprovider15:02
<awoods>jonroby: same error?
<jonroby>awoods: yes
<awoods>jonroby: can you put your debugger on it to see the exact issue? we are speculating it to be the namespace.
<jonroby>awoods: ok15:03
<awoods>jonroby: did you try the FQNS for the "fedora" namespace?
<jonroby>awoods: no, fedora was just fedora:writable. want me to try the FQNS?
<awoods>jonroby: it would be good to know that the FQNS works15:04
<jonroby>awoods: for a property named fedora:adminEmail, using http://fedora.info/definitions/v4/rest-api#adminEmail produces a 400 error15:06
<awoods>jonroby: sounds like you need to put the debugger on it15:07
<jonroby>awoods: ok15:08
<awoods>MohamedAR: I am not seeing how it could be reasonable to pass in an "internal session" to AbstractService.findOrCreate().15:09
MohamedAR: Would you be interested in creating a ModeShape integration test, and initiating the conversation with them?15:11
<MohamedAR>awoods: sure
<awoods>MohamedAR: thanks
MohamedAR: It may save time to feel the waters with the modeshape team on IRC, first.15:17
<MohamedAR>awoods: ok15:18
<pivotal-bot>Andrew Woods accepted "FedoraResource.toString()" https://www.pivotaltracker.com/story/show/8187229815:20
Andrew Woods accepted "Fix oaiprovider from impacts of fcrepo4 updates" https://www.pivotaltracker.com/story/show/82123722
* dwilcox joins15:21
<pivotal-bot>Andrew Woods started "Configure fcrepo-test-grinder for fcrepo4 work" https://www.pivotaltracker.com/story/show/76247430
Andrew Woods edited "Configure fcrepo-test-grinder for fcrepo4 work" https://www.pivotaltracker.com/story/show/76247430
<awoods>jonroby: any news?15:50
jonroby: did you get to the bottom of the 400 error?
* acoburn joins16:11
* dwilcox leaves16:23
<MohamedAR>awoods: Modeshape forum response on the JcrTools issue "The very simple JcrTools utility will not likely work with such advanced permissions schemas. Feel free to suggest a fix, or simply use a custom utility." https://developer.jboss.org/thread/24996216:38
<awoods>MohamedAR: A fix may be in order...16:40
MohamedAR: The fix that occurs to me is building up the path from the full path towards the root, versus the way ModeShape is doing it from the root to the full path.16:47
<MohamedAR>awoods: similar to your patch earlier?16:49
<awoods>MohamedAR: similar, but likely without the Guava.Joiner and with a means of collecting the path segments that will need to be created while walking up the full path towards the root.16:50
<MohamedAR>awoods: ok, I will put together a patch16:51
<awoods>MohamedAR: that would be great.
<jonroby>awoods: I got pulled into a meeting about a storage problem. I've looked at the code but not run a debugger on it. I've found that the fcrepo NS also produces a 400 code. I'm not sure why fedora: is so special that it works, other than its the REST-API??16:54
<awoods>jonroby: it would be nice to understand the issue you are hitting16:55
* longshou leaves16:56
<jonroby>awoods: I am still looking
* longshou joins
* awead leaves16:59
* awead joins17:03
<scossu>cbeer: are you planning to capitalize indexing:indexable node type for fcrepo-message-consumer?17:17
* dwilcox joins17:18
* jgpawletko_away leaves17:19
* jonroby leaves17:21
* awead leaves17:47
* MohamedAR leaves17:57
* dwilcox leaves18:06
* scossu1 joins18:37
* scossu1 leaves
* scossu leaves18:39
* acoburn leaves18:47
* ksclarke leaves18:50
* nikhiltri leaves19:28
<pivotal-bot>Andrew Woods edited "Fix oaiprovider ListSetsIT" https://www.pivotaltracker.com/story/show/8214139420:32
Andrew Woods added comment: "Namespace expansion will need to happen here: ""20:36
https://github.com/fcrepo4/fcrepo4/blob/master/fcrepo-kernel-im�" https://www.pivotaltracker.com/story/show/82135176
* jonroby joins20:49
awoods? believe 400 error is related to github.com/fcrepo4/issues/30720:51
<awoods>jonroby: that would make sense... since you are presumably passing in an invalid namespace prefix.20:52
<jonroby>awoods: sigh...so now figuring out how to add oai as a preferred namespace via sparql update20:53
awoods: as shown at the bottom of that ticket...but with oai20:54
<awoods>jonroby: ...or you can add it to your CND file.
jonroby: like "dc" https://github.com/fcrepo4/fcrepo4/blob/master/fcrepo-kernel-impl/src/main/resources/fedora-node-types.cnd#L1920:55
<jonroby>awoods: wow...ok, i'll give that a try
awoods: thats a confiug file, no recompilation needed?20:56
<awoods>jonroby: you probably need to compile since the CND is baked into the fcrepo-kernel-impl jar.20:57
<jonroby>awoods: Ah ok, good to know. thanks for the info, i'll let you know how it goes
<awoods>jonroby: you could also update your "repository.json" (which specifies where the CND can be found), then provide a SystemProperty to your servlet container on startup to point to where your repository.json file can be found.20:59
jonroby: https://wiki.duraspace.org/display/FF/Training+-+Administrator+Introduction#Training-AdministratorIntroduction-ContentModeling
afk - dinner21:00
<jonroby>awoods: sorry for disturbing you, cheers
* longshou leaves21:34
* jonroby leaves21:37
<pivotal-bot>Aaron Coburn added comment: "@awoods thanks for the pointer -- I had been looking through the code but wasn't sure where the change needed�" https://www.pivotaltracker.com/story/show/8213517621:54
Andrew Woods added comment: "@aaroncoburn, the trick is getting the NamespaceRegistry... which requires a Session. Since the messaging mac�" https://www.pivotaltracker.com/story/show/8213517622:07
* jonroby joins22:45
awoods: adding oai to fedora-node-types.cnd worked22:46
* scossu joins23:14
* scossu leaves23:51
* jonroby leaves00:46
* jonroby joins00:47
* scossu joins01:01
* scossu leaves01:06
* jonroby leaves01:25

Generated by Sualtam