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

Using timezone: Eastern Standard Time
* mohamedar joins07:59
* jcoyne joins08:14
* jcoyne leaves08:18
* dhlamb joins08:28
* mohamedar leaves08:38
* jcoyne joins
* diegopino joins09:02
* bseeger joins09:24
* mikeAtUVa joins09:25
* mohamedar joins09:32
* github-ff joins09:53
[fcrepo-audit] escowles pushed 2 new commits to master: https://git.io/vgaP7
fcrepo-audit/master 8f71865 Andrew Woods: Remove method: FedoraResource.setURIProperty()...
fcrepo-audit/master cc4e64c Esmé Cowles: Merge pull request #28 from awoods/fcrepo-1894...
* github-ff leaves
* ajs6f joins09:54
* github-ff joins09:56
[fcrepo4] awoods pushed 2 new commits to master: https://git.io/vgaXE
fcrepo4/master eba66c2 Aaron Coburn: remove setURIProperty and getProperty from the kernel-api
fcrepo4/master d9840a3 Andrew Woods: Merge pull request #977 from acoburn/fcrepo-1894...
* github-ff leaves
* travis-ci joins10:00
fcrepo4-exts/fcrepo-audit#73 (master - cc4e64c : Esmé Cowles): The build passed.
Change view : https://github.com/fcrepo4-exts/fcrepo-audit/compare/64118e196891...cc4e64c5b869
Build details : https://travis-ci.org/fcrepo4-exts/fcrepo-audit/builds/108292837
* travis-ci leaves
<diegopino>Hi, any people working with Hydra? I have an Islandora compat question. What are you using for example for UUID's in fedora4 resources? Which ontology, predicate?10:08
Basically how are you dealing with identity matching between rdf/ldp storage and front end.10:09
<awoods>diegopino: see #projecthydra10:11
<diegopino>Thanks awoods: sorry for the wrong channel question10:12
* travis-ci joins10:19
fcrepo4/fcrepo4#4256 (master - d9840a3 : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/7cfd64c4348f...d9840a30d4e0
Build details : https://travis-ci.org/fcrepo4/fcrepo4/builds/108293452
* travis-ci leaves
* whikloj joins10:35
* bseeger leaves10:37
* github-ff joins10:38
[fcrepo4] escowles pushed 2 new commits to master: https://git.io/vgadR
fcrepo4/master f805f67 Andrew Woods: Add configuration for a MySQL backend, instead of LevelDB...
fcrepo4/master a8b1273 Esmé Cowles: Merge branch 'fcrepo-1833'
* github-ff leaves
* github-ff joins10:39
[fcrepo4] escowles closed pull request #978: Add configuration for a MySQL backend, instead of LevelDB (master...fcrepo-1833) https://git.io/vgEvr
* github-ff leaves
* peichman joins10:43
* barmintor joins10:51
* jcoyne leaves11:14
* bseeger joins
* jcoyne joins11:15
* osmandin joins11:19
* diegopino leaves11:22
* bseeger leaves11:23
* bseeger joins11:24
* dwilcox joins11:31
* bseeger leaves11:35
* bseeger joins11:38
* github-ff joins11:50
[fcrepo-camel-toolbox] awoods pushed 2 new commits to master: https://git.io/vgVOE
fcrepo-camel-toolbox/master d9af7db Aaron Coburn: update camel/fcrepo dependencies
fcrepo-camel-toolbox/master c92eeff Andrew Woods: Merge pull request #70 from acoburn/fcrepo-1896...
* github-ff leaves
* escowles joins11:53
* ajs6f leaves12:02
* travis-ci joins12:04
fcrepo4-exts/fcrepo-camel-toolbox#200 (master - c92eeff : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4-exts/fcrepo-camel-toolbox/compare/4198ccc52306...c92eeffa095f
Build details : https://travis-ci.org/fcrepo4-exts/fcrepo-camel-toolbox/builds/108323785
* travis-ci leaves
<whikloj_AFK>Generic version question: What property is the Slug stored as? When I try to get the PropertiesRdfContext of a version and loop through the available triples I get12:16
com.hp.hpl.jena.shared.InvalidPropertyURIException: v0
* dwilcox leaves
<escowles>whikloj: i thought the slug was stored as the version label, not as a property
<whikloj>escowles: possibly it is and it is being picked up incorrectly by this context?12:17
escowles: Is a label not a property on the node?
<escowles>whikloj: IIRC, the version id and version label are properties of the version object, not properties of the node (which contains only the properties on the resource)
<whikloj>escowles: right, what I have is a version. I am trying to get the baseVersion, which seems also to be a property.12:19
<awoods>whikloj: https://github.com/fcrepo4/fcrepo4/blob/master/fcrepo-kernel-modeshape/src/main/java/org/fcrepo/kernel/modeshape/services/VersionServiceImpl.java#L181
<whikloj>awoods: yes, if I have a regular versionable resource. I have the version which WebAC needs to authenticate against to allow the "read" permission to fulfill the creation of it12:20
<awoods>whikloj: is there a ticket you are working? It is not clear to me what you are trying to tackle.12:21
<whikloj>awoods: FCREPO-183012:22
* dwilcox joins12:24
<awoods>whikloj: is the present challenge trying to get the "baseVersion" from a handle to a specific version of a resource?12:27
<whikloj>awoods: kind of the opposite, I get a resource that is in the system version space. My thought process (however bent) was to locate the baseVersion of that version. If you have permissions on it, then you are good.12:28
awoods: So maybe not the opposite
* bseeger leaves
<awoods>whikloj: when you say "baseVersion", do you mean the real resource itself?12:29
* dwilcox_ joins12:30
<whikloj>awoods: yes
<awoods>whikloj: that sounds like good logic...
<whikloj>awoods: I need to give it something with a path to match against the ACLs accessTo statements, so I can get the triple baseVersion which is like12:31
sorry lost the line12:32
* dwilcox leaves
<whikloj>awoods: info:fedora/versionResourceUri @http://fedora.info/definitions/v4/repository#baseVersion info:fedora/jcr:system/jcr:versionStorage/08/6c/80/086c80ef06f43c234fb6d7d1df14ae550f1524a5/1.012:33
<awoods>whikloj: I agree, it is not immediately apparent how to get to the "real resource" from a version of that resource... although it must be possible.12:34
whikloj: have you talked with mikeAtUVa?
<whikloj>awoods: yes, this seems to be one of the limitations of our implementation of versioning.
<awoods>whikloj: a limitation of understanding?12:35
<whikloj>awoods: more that the versions are stored in system space, so you are presenting JCR information. Instead of a representation of the LDP version
* dwilcox_ leaves
<whikloj>or that was how I understood it
mikeAtUVa: ^^ I am capturing that correctly
<awoods>whikloj: have you tried using your IDE's debugger and inspecting the underlying properties of a version resource?12:36
<whikloj>awoods: debugging using Eclipse has been hit and miss for the last little while, I'll give it a try again12:37
<barmintor>escowles: did my reply re: the difference btwn external-message and file fed make sense?12:38
<escowles>barmintor: sure -- there could be a different impl (and that could vary by the fedora impl too)
* dwilcox joins
<escowles>(but, imho, filesystem federation is less janky)12:39
<barmintor>escowles: in addition to this kind of impl difference, we should also note the places where we push JCR/MODE-isms out as places where a question is raised about (dun dun dun) specifiable API12:40
<escowles>barmintor: i do keep saying "federation", but i don't think that mode-isms are the key here: one approach is you have a whole tree that shows up in the repo, the other approach is you post something to create a reference to a particular file12:41
for external URLs, i think you have no choice, but for files on disk, the repo can easily know about them
<barmintor>escowles: yes, it is possible. but again, the mechanism.12:42
escowles: fwiw, I think there is a consderable authZ shortcut in handling it via federation
<escowles>barmintor: we could add some notion of serving files on disk to the spec, but i'm not sure what that would look like
barmintor: do you mean shortcut as in (solves a problem for us), or as in (circumvents authZ)?12:43
<barmintor>escowles: since iirc you don't federate via the REST api
escowles: the first
escowles: depending on admin intention, I guess
<escowles>barmintor: right now, i don't think you can assign access control to the federated content, because of the impl. details of the auth modules12:44
<escowles>barmintor: maybe it was only with rbacl, but i remember something about needing to create namespaced nodes, which isn't supported in the federated space12:45
maybe webac fixes it?12:46
<awoods>whikloj: you should be able to retrieve the "real resource" with: "jcr:baseVersion=87a0a8c317f1e7b8fb844e-dc0c-440c-a386-a127ed50d6ab"
escowles: I do not believe any testing has taken place on federation with webac12:47
<escowles>awoods: can you link from the webacl to the controlled resource? if you have to link from the resource to the webacl, then the read-only federation might be a problem for using it12:48
<awoods>escowles: you should be able to link from the authorization to the federated resource.12:49
<whikloj>awoods, escowles: you need the acl:accessControl link on the resource as well
<awoods>escowles: creating a link from the federated resource to the ACL would not work if the federation is read-only.
<escowles>awoods: whikloj: hmm -- seems worth testing to see what would happen if you linked from the webacl to a federated resource12:51
<whikloj>awoods, escowles: Best I can think of at this stage would be to allow for a modified root_authorization object which would have accessTo and accessToClass triples
escowles: My guess is it wouldn't be found as the acl:accessControl triple is used to locate the appropriate ACL to apply12:52
<escowles>whikloj: you might be able to mount your federation at /rest/foo/federation and attach webacl to /rest/foo/ that would be an effective federation-wide acl
<awoods>escowles: true
<whikloj>escowles: Yes, that would work
awoods: I don't understand how to use your "jcr:baseVersion" syntax to retrieve a resource.12:55
* github-ff joins
[fcrepo-camel] awoods pushed 1 new commit to master: https://git.io/vgVuG
fcrepo-camel/master 74259ac Aaron Coburn: Add pax-exam to integration tests...
* github-ff leaves
* github-ff joins12:56
[fcrepo-camel] awoods closed pull request #102: Add pax-exam to integration tests (master...fcrepo-1734) https://git.io/vgR1z
* github-ff leaves
* travis-ci joins12:59
fcrepo4-exts/fcrepo-camel#245 (master - 74259ac : Aaron Coburn): The build passed.
Change view : https://github.com/fcrepo4-exts/fcrepo-camel/compare/abc0ccd9e18f...74259acc2a53
Build details : https://travis-ci.org/fcrepo4-exts/fcrepo-camel/builds/108338977
* travis-ci leaves
<awoods>whikloj: http://www.day.com/maven/javax.jcr/javadocs/jcr-2.0/javax/jcr/Session.html#getNodeByIdentifier(java.lang.String)13:04
* mcritchlow joins13:06
* jgpawletko joins13:08
* bseeger joins
<whikloj>awoods: thanks13:10
* github-ff joins13:40
[fcrepo4] awoods pushed 1 new commit to master: https://git.io/vgVMP
fcrepo4/master 99a30ff Yinlin Chen: Logging exceptions should normally be at level ERROR....
* github-ff leaves
* github-ff joins
[fcrepo4] awoods closed pull request #973: Logging for an exception that will be presented to the client should … (master...fcrepo-1874-update) https://git.io/vgIAq
* github-ff leaves
* travis-ci joins13:59
fcrepo4/fcrepo4#4258 (master - 99a30ff : Yinlin Chen): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/a8b1273b7ac3...99a30ffdb6c9
Build details : https://travis-ci.org/fcrepo4/fcrepo4/builds/108348864
* travis-ci leaves
* ajs6f joins14:52
awoods: As we rollinto the work of API specing, we're going to want to make that work visible in Jira, right? Beause that's how we know what we're doing.15:19
<awoods>ajs6f: are you referring to the "spec writing" tasks?15:20
<ajs6f>awoods: I think so— not sure what distinction you are making with that phrase...?
awoods: For example, I might include liasion woik with LDP as part of that API specing work.15:21
work, not woik
Woiking is like walking, but you wiggle around like Curley fro the Three Stooges.15:22
<awoods>ajs6f: sure, using the "api-spec" label and "API Specification" Roadmap Theme would work.15:23
ajs6f: with a "Fix Version" of 5.0.015:24
<ajs6f>awoods: Well, that's what I was going to raise. What do you think of having components for each spec piece (CRUD, fixity, etc.)
<awoods>ajs6f: that could be helpful: spec-crud, spec-fixity, etc15:25
ajs6f: to distinguish the spec components from software components
<ajs6f>awoods: Yeah, exactly. Some of those efforts could get involved enough to warrant some structuring.
* dwilcox leaves15:35
* dhlamb leaves15:54
* bseeger leaves16:13
* bseeger joins
* dwilcox joins16:17
* dwilcox_ joins16:39
* anarchivist_ joins16:40
* jgpawletko leaves16:43
* jcoyne1 joins16:44
* dwilcox leaves16:48
* jcoyne leaves
* anarchivist leaves
* bseeger leaves16:49
* dwilcox_ leaves16:56
* mikeAtUVa leaves16:59
* ajs6f leaves17:10
* petertr7z joins17:18
* anarchivist_ joins17:19
* dwilcox joins17:21
* anarchivist leaves17:22
* petertr7_away leaves
* mohamedar leaves17:34
* dwilcox leaves17:35
* peichman leaves17:37
* whikloj_AFK leaves18:02
* jcoyne1 leaves18:10
* jgpawletko joins18:56
* dwilcox joins20:01
* dwilcox leaves20:04
* dwilcox joins20:12
* dwilcox leaves20:13
* dwilcox joins20:14
* peichman joins20:24
* dwilcox leaves
* github-ff joins20:26
[fcrepo4] awoods opened pull request #979: Bump ModeShape version to 4.5.0.Final (master...fcrepo-1886) https://git.io/vgrKO
* github-ff leaves
* peichman leaves20:36
* dhlamb joins20:44
* peichman joins20:48
* peichman leaves21:20
* jcoyne joins21:21
* jcoyne leaves21:22
* jcoyne joins21:24
* dhlamb leaves22:11
* mcritchlow leaves22:16
* jgpawletko leaves22:19
* mcritchlow joins23:17
* mcritchlow leaves23:22
* f4jenkins leaves23:52
* f4jenkins joins23:55

Generated by Sualtam