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

Using timezone: Eastern Standard Time
* fcrepo-bot joins03:55
* fcrepo-bot leaves05:01
* dwilcox joins07:36
* github-ff joins07:38
[fcrepo-jms-indexer-pluggable] escowles pushed 1 new commit to jms-baseurl: http://git.io/PlneAQ
fcrepo-jms-indexer-pluggable/jms-baseurl 25bd357 Esmé Cowles: Adding VisibleForTesting annotation, replacing magic numbers with variables for likely webapp/REST API path components
* github-ff leaves
* mohamed joins09:30
* jcoyne joins09:37
<pivotal-bot>Esme Cowles started "Get baseURL for JMS events from JAX-RS instead of requiring it be configured separately" https://www.pivotaltracker.com/story/show/7235191809:40
Esme Cowles finished "Get baseURL for JMS events from JAX-RS instead of requiring it be configured separately" https://www.pivotaltracker.com/story/show/72351918
<jcoyne>awoods: could we have the POST a new object request return the RDF Graph?09:42
awoods: I'm always following up a POST with a GET to retrieve the created/modified date properties.
<awoods>jcoyne: That is a possibility. Can you add a pivotal ticket, and the team can discuss.09:43
jcoyne: I gather that you would like the same response from create object (POST) as you get from GET object?09:44
* gregjansen joins
<pivotal-bot>Justin Coyne added "POST new object should return the RDF graph" https://www.pivotaltracker.com/story/show/7589032009:45
<jcoyne>awoods: there you have it.09:46
awoods: The same response would be ideal unless there is some reason not to.
<awoods>jcoyne: thanks for the request.09:47
<pivotal-bot>Andrew Woods edited "POST new object should return the RDF graph" https://www.pivotaltracker.com/story/show/75890320
<jcoyne>awoods: I'm not sure who is in charge of the website, but it would be helpful if this page (http://www.fedora-commons.org/community/developers) had a link to API documentation. It currently links to the Wiki, but it doesn't make it clear that the wiki contains the documentation.09:50
<awoods>jcoyne: The static website (fedora-commons.org) is in a state of limbo, there is a mix of F3 and F4 and it is a bit difficult to make the status and differences clear to people coming new to the project.09:52
<jcoyne>awoods: also if http://fcrepo.org/ and http://fedorarepository.org/ all redirected to fedora-commons.org, instead of hosting the same site, it would improve SEO
awoods: I agree, it's a bit of a jumble.
<awoods>jcoyne: but yes, I will add a link to the F4 REST API
jcoyne: thanks for the SEO tip
<jcoyne>awoods: currently I get all three of those sites in my top 5 results, and I'm confused as to which is the right one.09:53
<awoods>jcoyne: what is an example search?09:54
jcoyne: I do not get fcrepo.org
<jcoyne>"fedora futures"
<awoods>got it09:55
jcoyne: http://www.fedora-commons.org/community/developers10:01
* github-ff joins10:11
[fcrepo4] escowles pushed 1 new commit to jaxrs-baseurl: http://git.io/FU9JaA
fcrepo4/jaxrs-baseurl a9fda13 Esmé Cowles: Updating logging and access, adding comment to explain fail-fast strategy
* github-ff leaves
* ksclarke joins10:12
<pivotal-bot>Andrew Woods added comment: "@longshous, the description of the ticket is misleading. You can see the issue that this ticket is trying to �" https://www.pivotaltracker.com/story/show/7132661810:17
Andrew Woods unstarted "Add API support for root fedora object" https://www.pivotaltracker.com/story/show/71326618
<awoods>jcoyne: This section makes me think that a POST response should not contain the RDF graph: http://www.w3.org/TR/ldp/#ldpc-post-created20110:28
* travis-ci joins
[travis-ci] fcrepo4/fcrepo4#2191 (jaxrs-baseurl - a9fda13 : Esmé Cowles): The build passed.
[travis-ci] Change view : https://github.com/fcrepo4/fcrepo4/compare/629cf3bcca3e...a9fda13a421f
[travis-ci] Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/31139908
* travis-ci leaves
<jcoyne>awoods: this is ambiguous: "Clients shall not expect"10:30
awoods: certainly it's not part of the LDP that we must return a response, but perhaps we may?
awoods: It certainly seems like an inefficiency in the protocol if it doesn't allow that.10:33
* scossu joins10:47
* mohamed1 joins11:03
* mohamed leaves11:05
* mohamed1 leaves11:07
* mohamed1 joins11:45
<jcoyne>awoods: not sure if this is a bug or not. If I make a node and set versionable and then try to post to the /fcr:content subnode I get a:11:47
STATUS: 500 javax.jcr.nodetype.ConstraintViolationException: Unable to determine a valid node definition for the node "/test/de/10/5e/b6/de105eb6-72a3-4b9b-9d0d-39ef2f6d6467/DS1/jcr:content" in workspace "default" of "repo"
awoods: If I switch the order: post the content first, then update the parent node to have versionable, it works fine.11:48
<awoods>on a call
<jcoyne>Seems like it should never show me a 500 though.
* mohamed1 leaves11:56
* mohamed1 joins12:01
<awoods>jcoyne: that seems odd. let me give it a look.12:04
* mohamed1 leaves12:08
* mohamed1 joins12:09
* dwilcox leaves12:14
<pivotal-bot>Andrew Woods added comment: "Discussion: https://community.jboss.org/thread/24319112:15
Modeshape ticket: https://issues.jboss.org/browse/MODE-�" https://www.pivotaltracker.com/story/show/75459836
* dwilcox joins12:16
<pivotal-bot>Andrew Woods added comment: "Related: http://www.w3.org/TR/ldp/#ldpc-post-created201" https://www.pivotaltracker.com/story/show/7589032012:17
* ksclarke leaves12:24
* ksclarke joins12:30
* edInCo joins
* edInCo leaves12:39
<awoods>jcoyne: I am not able to reproduce your issue (re: versionable and fcr:content)12:48
<pivotal-bot>Mohamed Mohideen Abdul Rasheed added comment: "Attached profiling results highlights JcrSeesion.save and JcrNode.getPath a�" https://www.pivotaltracker.com/story/show/7545983612:50
<awoods>jcoyne: POST /rest/12:51
> http://localhost:8080/rest/11/03/f2/e2/1103f2e2-4df3-4bec-876d-18184f48119f (resource)
PATCH /rest/resource/ (with versionable)
PUT /rest/resource/DS1/fcr:content (with binary payload)
...no issue
* danromano joins12:52
* danromano leaves12:55
* mohamed1 leaves13:25
<pivotal-bot>Andrew Woods added "Not all versions displayed in UI" https://www.pivotaltracker.com/story/show/7591020413:32
<awoods>jcoyne: I got a kick out of the jmsBorker as well13:33
<jcoyne>awoods: Here's a transcript of that 500 error: https://gist.github.com/jcoyne/ed9f3596454ffd289176
* mohamed joins13:34
<awoods>jcoyne: I may see the issue13:36
jcoyne: it looks like you are first creating an object: /resource/DS1
then you are subsequently trying to turn that object into a datastream13:37
<jcoyne>yeah, I'm creating /resource/DS1 in order to apply the versionable property
<awoods>instead, do not create "/resource/DS1", but rather go straight to creating "/resource/DS1/fcr:content"
<jcoyne>awoods: which is what I did to fix the problem.
awoods: but I Imagine it should be a 4xx error, not a 50013:38
<awoods>jcoyne: agreed. I will put in the ticket for the scenario to respond with a 4xx13:39
<jcoyne>awoods: after doing "/resource/DS1/fcr:content", I'm able to come back and set the versionable property on ""/resource/DS1", so it's funny that it works fine that way.
<awoods>jcoyne: it has to do with the way we create/manage/mask the underlying nodes13:40
jcoyne: and it is a bit "funny"13:41
jcoyne: you should be able to get around the issue...13:44
<jcoyne>yes, I have.
<awoods>by explicitly saying that you are creating a "datastream": curl -i -X POST -H "Content-Type:text/plain" "http://localhost:8080/rest?mixin=fedora:datastream"
<jcoyne>Ah, okay.
<jcoyne>I was wondering if there wasn't something like that.13:45
<awoods>jcoyne: I would be interested to get your feedback on that
<jcoyne>It's sort of a pain to do that with a URL in a generic LDP client
it would be nicer to assert RDF.type
Is that possible?13:46
<awoods>jcoyne: Here is the logic: https://github.com/fcrepo4/fcrepo4/blob/master/fcrepo-http-api/src/main/java/org/fcrepo/http/api/FedoraNodes.java#L66513:47
<pivotal-bot>Andrew Woods edited "Not all versions displayed in UI" https://www.pivotaltracker.com/story/show/7591020414:15
* mohamed leaves14:17
* mohamed joins14:20
<jcoyne>Does fc4 have fixity checking?14:22
Oh, just found this https://wiki.duraspace.org/display/FF/Durability14:24
* github-ff joins
[fcrepo4] awoods pushed 1 new commit to master: http://git.io/NCU8iw
fcrepo4/master ae12d03 Esmé Cowles: Getting baseURL for JMS messages from event user data field, which is set using by HTTP API from UriInfo baseURL...
* github-ff leaves
<jcoyne>anyone have a link to "Fedora Futures developed a fixity queueing and reporting service (that is only loosely coupled to Fedora)."?14:25
* github-ff joins
[fcrepo4] awoods closed pull request #420: Getting baseURL for JMS messages from event user data field, which is set using by HTTP API from UriInfo baseURL (master...jaxrs-baseurl) http://git.io/YycndQ
* github-ff leaves
<pivotal-bot>Andrew Woods added comment: "Resolved with: https://github.com/fcrepo4/fcrepo4/commit/ae12d0308a721e51e3e3a1b050978b901eb3fd3c" https://www.pivotaltracker.com/story/show/72351918
* github-ff joins
[fcrepo4] awoods deleted jaxrs-baseurl at a9fda13: http://git.io/_f7f1Q
* github-ff leaves
<pivotal-bot>Andrew Woods delivered "Get baseURL for JMS events from JAX-RS instead of requiring it be configured separately" https://www.pivotaltracker.com/story/show/72351918
<jcoyne>awoods: is that this: https://github.com/fcrepo4-labs/fcrepo-fixity14:27
<awoods>jcoyne: That project went silent before my time... but it does resonate with the "Durability" quote you found.14:28
jcoyne: there is also: https://wiki.duraspace.org/display/FF/RESTful+HTTP+API+-+Fixity14:29
<pivotal-bot>Andrew Woods added "Update object with /fcr:content should be 4xx" https://www.pivotaltracker.com/story/show/7591535014:31
Andrew Woods edited "Update object with /fcr:content should be 4xx" https://www.pivotaltracker.com/story/show/75915350
<jcoyne>awoods: is fcr:content supposed to be versionable?14:33
Ah, nm, I think I just forgot to create a version.
* scossu leaves
<awoods>jcoyne: you are hitting f4 pretty hard. Do you have a short-term goal?14:34
<jcoyne>awoods: make Sufia work on F4
<awoods>jcoyne: do you have a sense of timeline? or how close that goal is?
<jcoyne>awoods: I've got almost everything going except for version recall, audit (fixity checking) and I still have to deal with translating IDs into the hierarchical URL structure.14:35
<awoods>jcoyne: btw, I will be offline starting in the next few minutes until next Wednesday14:36
<jcoyne>Okay, sorry to drive you away. ;)
* travis-ci joins14:49
[travis-ci] fcrepo4/fcrepo4#2192 (master - ae12d03 : Esmé Cowles): The build passed.
[travis-ci] Change view : https://github.com/fcrepo4/fcrepo4/compare/384c0593a919...ae12d0308a72
[travis-ci] Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/31167521
* travis-ci leaves
* scossu joins15:03
* dwilcox leaves15:33
<pivotal-bot>Esme Cowles estimated "Define F4 client interfaces" as 2 points https://www.pivotaltracker.com/story/show/7505228416:12
Esme Cowles started "Define F4 client interfaces" https://www.pivotaltracker.com/story/show/75052284
Esme Cowles added comment: "The first thing I notice upon looking at the kernel API is that most methods require a JCR Session object, and�" https://www.pivotaltracker.com/story/show/7505228416:15
* awoods leaves16:47
* bljenkins joins16:57
* bljenkins leaves16:58
* scossu leaves17:05
* bljenkins joins17:06
* gregjansen leaves17:12
* bljenkins leaves17:17
* bljenkins joins17:18
* bljenkins leaves17:20
* bljenkins joins17:27
* scossu joins17:35
* scossu leaves17:40
* bljenkins leaves17:45
* bljenkins joins
* scossu joins17:49
* bljenkins leaves17:53
* bljenkins joins17:54
* jcoyne leaves17:59
* jcoyne joins18:01
* jcoyne leaves18:03
* jcoyne joins18:05
* jcoyne leaves18:13
* dwilcox joins18:17
* scossu leaves18:19
* mohamed leaves18:23
* dwilcox leaves18:26
* dwilcox joins19:26
* mohamed1 joins19:54
* dwilcox leaves
* jcoyne joins20:12
* jcoyne leaves20:13
* mohamed1 leaves20:16
* jcoyne joins20:35
* jcoyne leaves21:40
* ksclarke leaves22:37
* ksclarke joins22:40
* jcoyne joins23:27

Generated by Sualtam