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

Using timezone: Eastern Standard Time
* ermadmix leaves08:35
* ermadmix joins09:02
* kaarefc leaves09:09
* kaarefc joins09:32
* osmandin joins09:36
* ksclarke joins09:42
* kaarefc leaves09:46
<awoods>fasseg: I have not seen the error. Could you have stale data, meaning, data from an earlier version of F4?
* ajs6f joins09:48
* github-ff joins10:00
[fcrepo4] ajs6f pushed 1 new commit to RDFIteration: http://git.io/eETRPQ
fcrepo4/RDFIteration 2c48a72 ajs6f: Stripping dead code
* github-ff leaves
* ajs6f leaves10:08
* ermadmix leaves10:09
* fasseg joins
* kaarefc joins10:11
* kaarefc leaves10:15
<bljenkins>Yippie, build fixed!10:18
Project fcrepo4 build #1206: FIXED in 17 min: http://ci.fcrepo.org/jenkins/job/fcrepo4/1206/
Project fcrepo-fixity-corrupter build #290: SUCCESS in 1 min 9 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/290/10:19
* ajs6f joins10:20
* gregjansen joins
<bljenkins>Project fcrepo-jms-indexer-pluggable build #69: UNSTABLE in 4 min 31 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-jms-indexer-pluggable/69/10:22
Project fcrepo-kitchen-sink build #544: SUCCESS in 4 min 58 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-kitchen-sink/544/10:26
* barmintor joins
* gregjansen leaves
* gregjansen joins10:27
* ermadmix joins10:47
* tecoripa joins10:53
* gregjansen leaves10:54
* github-ff joins10:57
[fcrepo-jms-indexer-pluggable] yecao opened pull request #9: Solr36 (master...solr36) http://git.io/V0uGsA
* github-ff leaves
* github-ff joins
[fcrepo4] ajs6f created AddingUnitTests (+1 new commit): http://git.io/lxQKqA
fcrepo4/AddingUnitTests e81c314 ajs6f: Cleaner EventType enum and better test coverage
* github-ff leaves
<ajs6f>awoods: Running a bit late, will be there in a few minutes10:58
<awoods>ajs6f: thanks
<fasseg>I hope you guys can understand me, i went to the dentist today and am still a bit high on painkillers and have a hard time speaking ;)11:00
* ajs6f leaves
* tecoripa1 joins11:01
<awoods>ermadmix: coming?11:02
* tecoripa leaves
* ajs6f joins11:04
* gregjansen joins11:05
<bljenkins>Project fcrepo4 build #1207: UNSTABLE in 17 min: http://ci.fcrepo.org/jenkins/job/fcrepo4/1207/11:15
<pivotal-bot_____>Gregory Jansen added "Add metadata reader role to basic roles PEP" https://www.pivotaltracker.com/story/show/5933596011:22
Gregory Jansen estimated "Add metadata reader role to basic roles PEP" as 2 points https://www.pivotaltracker.com/story/show/59335960
Gregory Jansen edited "Add metadata reader role to basic roles PEP" https://www.pivotaltracker.com/story/show/59335960
Gregory Jansen edited "Add metadata reader role to basic roles PEP" https://www.pivotaltracker.com/story/show/5933596011:23
awoods: The contract at the Java layer is as we left it when you left: the RDF source will produce triples. If it throws an exception, it's up to the client to decide what to do about that. The iterator will continue to work and produce triples until it is genuinely exhausted. We didn't finalize what to do at the HTTP layer, because Frank, Scott, and Greg discovered some subtleties and different possibilities for useful behavior there. 11:38
<awoods>ajs6f: thanks11:39
<ajs6f>No, awoods: thank _you_.
<tecoripa1>ajs6f: I see you have this on the object model wiki page (https://wiki.duraspace.org/display/FF/The+Fedora+4+object+model):11:48
Properties may be single- or multi-valued, and are typed. Some properties are reserved for internal-use under mutation only (e.g. those in the fedora-internal namespace), but can otherwise be used freely by implementors. By default, properties are multivalued string types.
<gregjansen>I am pretty sure that we will *not* want to throw a 403 if one child or property is restricted from the current user. But perhaps there is a way for the client to dictate that behavior
<tecoripa1>gregjansen: if the exception indicates that the error was an access control one, then the client could conceivably throw the 403, right?11:49
<ajs6f>tecoripa1: Sure. That sounds reasonable to me. But then, there are a lot of words in there, so what do I know?
<tecoripa1>ajs6f: do we have a document anywhere (rdf schema, owl ontology, etc.) that lists the internal properties and their use?
is it the fedora 3 rdf schema?11:50
<gregjansen>tecoripa: when you say the client could throw the 403, you imply that fedora would not throw one, correct?
<ajs6f>tecoripa1: We have a full fledged ontology in Github: https://github.com/futures/ontology/11:51
<tecoripa1>gregjansen: define "fedora". The internal http client built into fedora (fedora webapp?) may throw it, or may not, or...?
ajs6f: cool, thanks. Would it make sense to link that in the wiki page, or is it a voltaile, moving target that will confuse more than it elucidates?11:52
<ajs6f>That's up to us. I think it has still some wiggle in its hips.11:54
<gregjansen>tecoripa: I mean that the fedora REST API would not throw a 403 response code. In other words the http client would get all the information that they can access. It would need to throw the 403 if they have no read permission on the object level though.
an alternative is to return a triple that indicates a forbidden resource, but here I defer to that LDP inlining spec. We should see where that takes us.11:55
<tecoripa1>gregjansen: ah, I see. yeah, that's a tough question. How do you indicate in the REST response that one item out of many was closed off to you? Do you even want it to be invisible (i.e., not expose anything you that is closed off from you)?
my gut reaction is to silently return everything you can, ignore what you can't, and log server-side what was discared because of an access-exception.11:56
but my gut is notoriously loud, and often wrong.11:57
<gregjansen>tecoripa: this impacts the authz mechanisms I think, b/c currently the mode session only has the perms of the connected user.
tecoripa: I have a feeling that "list" is granted if you can "read" the parent object. We will need to verify that.11:58
<tecoripa1>gregjansen: right. the PEP will be the one that decides to throw an access exception, correct? When Adam's iterator asks the PEP for what it can get?
<gregjansen>yup, on a pretty granular basis.
<ajs6f>Oh, phooey. I'm not doing those kinds of checks at all… I'm just pulling from the JCR as per the old impl.
Not yet, anyway.11:59
<tecoripa1>gregjansen: that may be a good starting assumption. And document when we implement it that it may be wrong, when more use cases show up.
<ajs6f>awoods: We may need to delay merge until I catch up with what gregjansen and tecoripa1 have done to provide guarded access to the repo.
<awoods>ajs6f: on a call
<tecoripa1>I need to stop chatting and start coding. You guys are slave drivers.12:00
<ajs6f>awoods: I could have guessed that.12:01
* osmandin leaves
<gregjansen>ajs6f: should enforce perms as long as you don't use an internal session
<ajs6f>gregjansen: I know, I just didn't know that I could. Where should I go to start learning about how to do that? I'm just using the Session as passed into me from the client...12:02
<tecoripa1>ajs6f: re -- the ontology. I'll add the link to the github link to the object model page, and we can remove it or change it when something more stable shows up somewhere else.
gregjansen, ajs6f: perfomance consideration -- is an internal session much faster than an external one?
<gregjansen>ajs6f: if you are using an injected session provider, then you are using the auth stuff12:03
* github-ff joins
[fcrepo4] yulgit1 opened pull request #121: update modeshape version to 3.5.0.Final (master...updateModeshapeVer) http://git.io/zwJt2A
* github-ff leaves
<gregjansen>tecoripa: they bypass the PEP entirely. internal sessions are anonymous. they are used mostly for setup tasks, i.e. loading new CND files for a module12:04
<ajs6f>gregjansen: I'm not using a session provider at all— I accept the session as part of the contract. You can't get RDF back without giving me a session. So I guess that puts it on the client, which is pretty reasonable.
<gregjansen>ajs6f: yep
<gregjansen>ajs6f: reason acheived
<tecoripa1>++gregjansen, ajs6f
<tecoripa1>out of here. I'll be back shortly.12:06
* tecoripa1 leaves
* github-ff joins12:17
[fcrepo4] ajs6f pushed 2 new commits to AddingUnitTests: http://git.io/zoQisA
fcrepo4/AddingUnitTests 69a3db2 ajs6f: More cleanup for EventTypeTest
fcrepo4/AddingUnitTests 9977b4d ajs6f: Improved DefaultFilterTest
* github-ff leaves
<ajs6f>Can anyone see: https://github.com/futures/fcrepo4?12:19
Oh, wait, there it comes back again.
<gregjansen>lunch, back in a bit12:20
* github-ff joins12:22
[fcrepo4] ajs6f opened pull request #122: Adding unit tests (master...AddingUnitTests) http://git.io/WKrhXQ
* github-ff leaves
* github-ff joins12:32
[fcrepo4] ajs6f pushed 1 new commit to RDFIteration: http://git.io/F6EsoQ
fcrepo4/RDFIteration b836596 ajs6f: Tiny code cleanup in RdfStream
* github-ff leaves
<bljenkins>Yippie, build fixed!12:37
Project fcrepo4 build #1208: FIXED in 19 min: http://ci.fcrepo.org/jenkins/job/fcrepo4/1208/
* tecoripa joins
<bljenkins>Project fcrepo-fixity-corrupter build #291: SUCCESS in 1 min 7 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/291/12:38
* ermadmix leaves12:42
<bljenkins>Yippie, build fixed!12:45
Project fcrepo-jms-indexer-pluggable build #70: FIXED in 8 min 35 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-jms-indexer-pluggable/70/
Project fcrepo4 build #1209: UNSTABLE in 17 min: http://ci.fcrepo.org/jenkins/job/fcrepo4/1209/12:54
<pivotal-bot_____>Ye Cao added comment: "Latest https://github.com/futures/fcrepo-jms-indexer-pluggable/pull/912:55
1.Clean SolrIndexer dependences: only modi..." https://www.pivotaltracker.com/story/show/57453236
* ermadmix joins13:07
* github-ff joins13:17
[fcrepo4] ajs6f pushed 1 new commit to RDFIteration: http://git.io/Ivda5Q
fcrepo4/RDFIteration 409bf45 ajs6f: Testing behavior decided upon for clients
* github-ff leaves
<pivotal-bot_____>Scott Prater added comment: "Ben committed fixes for the two issues mentioned above." https://www.pivotaltracker.com/story/show/5923906813:19
<ajs6f>awoods: Going to have time to merge that RDfIterators PR today? Perhaps gregjansen or tecoripa can do it instead?13:45
<awoods>ajs6f: It is next on my list... after this call.
<ajs6f>k, i'll wait
<awoods>ajs6f: 15min
<cbeer>ajs6f: do you know if LDPath has any ontology-aware features when selecting properties?14:01
like, if X is a type of Y, and I ask for Y, will it give me X?14:02
maybe it's back-end specific.
<ajs6f>Sometimes but not always, it depends entirely on the backend for that. If the backend supports that, then yes.
e.g. The Solr backends that I often use with Stanbol, no. If I were to use a Jena backend, yes.14:03
I hope you're thoroughly destroying the RDF.
<cbeer>we're destroying something, all right.14:04
<awoods>cbeer: What might that be?14:05
<cbeer>maybe my brain.
<awoods>cbeer: Hydrate.
<ajs6f>A good cask ale should do it.14:06
<cbeer>ajs6f: did that last night.
<ajs6f>That explains the destroyed brain.
* ermadmix leaves14:07
<pivotal-bot_____>Andrew Woods added comment: "I saw a pull request. Are you "Finished" with this ticket, @ericjames?" https://www.pivotaltracker.com/story/show/5928508014:20
* github-ff joins
[fcrepo4] ajs6f pushed 1 new commit to AddingUnitTests: http://git.io/PQhmkQ
fcrepo4/AddingUnitTests d42d998 ajs6f: Small tweaks to use Guava predicates instead of homebrew
* github-ff leaves
<awoods>ajs6f: With which ticket is this "AddingUnitTests" branch associated?14:23
<ajs6f>None. It's general cleanup, while I wait to move on with the RDF stuff.
We can make a placeholder ticket "Make unit tests better".
<awoods>ajs6f: I use tickets to know when to review and push branches.14:24
<ajs6f>I'll make a placehlder, and you can watch that.
<awoods>ajs6f: I will watch it intently
<pivotal-bot_____>A. "Brulestion" Soroka added "Make unit tests better." https://www.pivotaltracker.com/story/show/5935203414:25
<awoods>ajs6f: after reviewing/pushing your latest iterator updates
<ajs6f>thnx— nothing there should be provocative. I just put in a unit test for the behavior we agreed on this morning.
* ermadmix joins14:26
* osmandin joins14:30
<awoods>ajs6f: Your whitespace update commit is causing merge conflicts with the iteration work, arghh. https://github.com/futures/fcrepo4/commit/5a8c48dec520de427b11f31b588695504a61a28914:39
<ajs6f>awoods: That is why I argue for separate commits inside one PR. We could even decide that all non-functional commits in a PR must precede the first functional commit, which would make review no more difficult than with separate PRs.14:45
Otherwise, there _will_ be merging.
<pivotal-bot_____>Eric James added comment: "https://github.com/futures/fcrepo4/pull/121" https://www.pivotaltracker.com/story/show/5928508014:48
Eric James finished "Upgrade Modeshape Version to: 3.5.0.Final" https://www.pivotaltracker.com/story/show/5928508014:49
<awoods>ajs6f: I still believe reformatting updates should be in their separate PRs. But with any in-progress branches, before submitting the merge request, the implementor should bring the branch up-to-date with master... which is what I am doing now for the RDFIteration branch.14:55
<ajs6f>The PR _was_ up-to-date when I finished the ticket. Between now and then I completed several other PRs.14:56
afk bbi 10
* ajs6f leaves14:57
* ajs6f joins15:04
* github-ff joins15:07
[fcrepo4] awoods pushed 1 new commit to master: http://git.io/_cBu9Q
fcrepo4/master b225254 ajs6f: Refactor RDF production to leverage iterators...
* github-ff leaves
<pivotal-bot_____>Andrew Woods added comment: "Resolved with: https://github.com/futures/fcrepo4/commit/b225254f08783111f457632e9207e19c2ee7f386" https://www.pivotaltracker.com/story/show/5730442615:08
Andrew Woods delivered "Refactor RDF generation in kernel" https://www.pivotaltracker.com/story/show/57304426
* github-ff joins15:09
[fcrepo4] awoods closed pull request #117: Rdf iteration (master...RDFIteration) http://git.io/rGkvoQ
* github-ff leaves
* github-ff joins15:11
[fcrepo4] ajs6f force-pushed AddingUnitTests from d42d998 to 3f82598: http://git.io/QonwRw
fcrepo4/AddingUnitTests 93f470f ajs6f: Cleaner EventType enum and better test coverage
fcrepo4/AddingUnitTests 82b188b ajs6f: More cleanup for EventTypeTest
fcrepo4/AddingUnitTests 478fdf5 ajs6f: Improved DefaultFilterTest
* github-ff leaves
* github-ff joins15:14
[fcrepo4] ajs6f created ExposeRdfStream from master (+0 new commits): http://git.io/dPEWUA
* github-ff leaves
* travis-ci joins15:23
[travis-ci] futures/fcrepo4#1041 (master - b225254 : ajs6f): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/4efdca81bf07...b225254f0878
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/12895814
* travis-ci leaves
<bljenkins>Yippie, build fixed!15:27
Project fcrepo4 build #1212: FIXED in 19 min: http://ci.fcrepo.org/jenkins/job/fcrepo4/1212/
awoods: Refactor RDF production to leverage iterators
Project fcrepo-fixity-corrupter build #292: SUCCESS in 1 min 12 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/292/15:28
Project fcrepo-legacy-api build #103: FAILURE in 59 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-legacy-api/103/15:29
* github-ff joins15:41
[fcrepo4] awoods pushed 1 new commit to master: http://git.io/NduI6g
fcrepo4/master 99564d9 Eric James: update modeshape version to 3.5.0.Final
* github-ff leaves
<pivotal-bot_____>Andrew Woods added comment: "Resolved with: https://github.com/futures/fcrepo4/commit/99564d96fdea924618342443168cc12621be4633" https://www.pivotaltracker.com/story/show/5928508015:43
Andrew Woods delivered "Upgrade Modeshape Version to: 3.5.0.Final" https://www.pivotaltracker.com/story/show/59285080
Andrew Woods accepted "Refactor RDF generation in kernel" https://www.pivotaltracker.com/story/show/57304426
* github-ff joins
[fcrepo4] awoods closed pull request #121: update modeshape version to 3.5.0.Final (master...updateModeshapeVer) http://git.io/zwJt2A
* github-ff leaves
<bljenkins>Project fcrepo4 build #1213: UNSTABLE in 16 min: http://ci.fcrepo.org/jenkins/job/fcrepo4/1213/15:44
* ajs6f leaves15:58
* travis-ci joins
[travis-ci] futures/fcrepo4#1043 (master - 99564d9 : Eric James): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/b225254f0878...99564d96fdea
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/12897352
* travis-ci leaves
<pivotal-bot_____>Andrew Woods accepted "Upgrade Modeshape Version to: 3.5.0.Final" https://www.pivotaltracker.com/story/show/5928508015:59
Andrew Woods added "Move Auth PEP project into Futures" https://www.pivotaltracker.com/story/show/5935958416:01
Andrew Woods edited "Move Auth PEP project into Futures" https://www.pivotaltracker.com/story/show/59359584
Andrew Woods added "Document performance numbers in wiki" https://www.pivotaltracker.com/story/show/5935984016:04
Andrew Woods edited "Document performance numbers in wiki" https://www.pivotaltracker.com/story/show/59359840
* ermadmix leaves16:08
* gregjansen leaves16:17
* gregjansen joins16:21
* ermadmix joins16:22
<pivotal-bot_____>Scott Prater finished "Create wiki page with matrix of basic roles, permissions to test" https://www.pivotaltracker.com/story/show/5928733816:27
<tecoripa>awoods: you deliver the finished tickets, correct?
<awoods>tecoripa: Yes, I do.16:28
<tecoripa>awoods: okay, thanks. I'll leave it to you, then. This ticket was wiki documentation, no pull request.
<awoods>tecoripa: When I see a ticket ready for "Delivery", that is my queue to review it and push it through the remaining states.
<pivotal-bot_____>Scott Prater added comment: "Greg and I also updated and fleshed out the Access Roles Module wiki page: https://wiki.duraspace.org/displ..." https://www.pivotaltracker.com/story/show/59287338
* tecoripa leaves16:52
* osmandin leaves16:55
<pivotal-bot_____>Scott Prater added comment: "Just a thought -- package-info.java is a good place to put package-level annotations to be used in the pack..." https://www.pivotaltracker.com/story/show/5495224416:59
A. "Brulestion" Soroka added comment: "Yes, it is unquestionably the correct place to do that, and you can see an example in our few JAX..." https://www.pivotaltracker.com/story/show/5495224417:11
* tecoripa joins17:17
* tecoripa leaves
* gregjansen leaves17:23
* ermadmix leaves18:00
* github-ff joins18:34
[fcrepo-jms-indexer-pluggable] awoods pushed 1 new commit to master: http://git.io/J8kdbg
fcrepo-jms-indexer-pluggable/master 8d8c307 Ye Cao: Add initial implementation of SolrIndexer...
* github-ff leaves
<pivotal-bot_____>Andrew Woods added comment: "Resolved with: https://github.com/futures/fcrepo-jms-indexer-pluggable/commit/8d8c307be6d327d63cd3af967338b..." https://www.pivotaltracker.com/story/show/57453236
Andrew Woods delivered "Create Solr implementation of Indexer interface" https://www.pivotaltracker.com/story/show/57453236
* github-ff joins18:35
[fcrepo-jms-indexer-pluggable] awoods closed pull request #9: Solr36 (master...solr36) http://git.io/V0uGsA
* github-ff leaves
* github-ff joins18:36
[fcrepo-jms-indexer-pluggable] awoods closed pull request #8: adopt constructor-arg for solrServerFactory and solrIndexer in indexer-c... (master...spring_injection_refactory) http://git.io/zyspNA
* github-ff leaves
<pivotal-bot_____>Andrew Woods added comment: "For this ticket to be effective, we should break it into specific unit tests for specific classes." https://www.pivotaltracker.com/story/show/5935203418:37
Andrew Woods edited "Make unit tests better." https://www.pivotaltracker.com/story/show/59352034
* ksclarke leaves18:38
* travis-ci joins18:45
[travis-ci] futures/fcrepo-jms-indexer-pluggable#25 (master - 8d8c307 : Ye Cao): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo-jms-indexer-pluggable/compare/3a9e0b4844a6...8d8c307be6d3
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo-jms-indexer-pluggable/builds/12905348
* travis-ci leaves
<pivotal-bot_____>Andrew Woods edited "Make unit tests better." https://www.pivotaltracker.com/story/show/5935203419:25
Andrew Woods accepted "Create Solr implementation of Indexer interface" https://www.pivotaltracker.com/story/show/5745323619:28
* fasseg leaves21:28
* ksclarke joins21:47
<pivotal-bot_____>Scott Prater added comment: "Ben and I finished reviewing his changes; Ben made a few more tweaks, and we now feel it is ready to release." https://www.pivotaltracker.com/story/show/5923906823:34
Scott Prater accepted "Review F3 Patch" https://www.pivotaltracker.com/story/show/59239068

Generated by Sualtam