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

Using timezone: Eastern Standard Time
* barmintor_ joins08:26
* awead joins08:34
* jcoyne joins08:53
* MohamedAR joins
* barmintor_ leaves09:01
<pivotal-bot>Jonathan Roby added comment: "Accordin g to the schema https://github.com/ModeShape/modeshape/blob/master/modeshape-jcr/src/main/resources�" https://www.pivotaltracker.com/story/show/8148290409:02
* jonroby joins09:12
* ksclarke joins09:43
<pivotal-bot>Andrew Woods added "Create CND nodetypes for OrderedList Ontology" https://www.pivotaltracker.com/story/show/8176467209:54
Andrew Woods added "Draft implementation of a WebAccessControls authz module" https://www.pivotaltracker.com/story/show/8176477009:55
* acoburn joins09:59
* osmandin joins10:03
afk10:05
<pivotal-bot>Andrew Woods added comment: "@jonathanroby, The idea would be to have the oaiprovider code look for specific properties on the root node o�" https://www.pivotaltracker.com/story/show/8148290410:14
* ajs6f joins10:17
<pivotal-bot>Andrew Woods edited "Redirect to federated content" https://www.pivotaltracker.com/story/show/8171406010:27
* github-ff joins10:32
[fcrepo4] ajs6f force-pushed PreferComboAction from 673fbc2 to 1722c88: http://git.io/55RrXg
fcrepo4/PreferComboAction ffd592e ajs6f: Adding MultiPrefer type to handle multiple Prefer headers
fcrepo4/PreferComboAction 1722c88 ajs6f: Adding MultiPrefer types and action
* github-ff leaves
<f4jenkins>Project fcrepo4 build #2522: UNSTABLE in 11 min: http://jenkins.fcrepo.org/job/fcrepo4/2522/10:44
* github-ff joins10:49
[fcrepo4] ajs6f pushed 1 new commit to PreferComboAction: http://git.io/aN-nMg
fcrepo4/PreferComboAction 25c12b7 ajs6f: Adding unit test
* github-ff leaves
* scossu joins10:51
* barmintor joins
<pivotal-bot>Stefano Cossu added comment: "@ajs6f : So far the out-of-the-box CND functionality provided by Fedora has fulfilled almost all of our stru�" https://www.pivotaltracker.com/story/show/8169853810:53
* scossu1 joins
* scossu1 leaves10:55
* scossu1 joins
* scossu leaves
<pivotal-bot>A. "Offratonix" Soroka added comment: "@stefanoc : The key point to understand here is that Fedora _doesn't_ supply that CND functionality�" https://www.pivotaltracker.com/story/show/8169853810:57
<barmintor>awoods: I’m squashing and rearranging that branch
* ajs6f leaves
<awoods>barmintor: great
* longshou joins
<pivotal-bot>Chris Beer added comment: "Why do they need to be pre-defined? " https://www.pivotaltracker.com/story/show/8176467210:59
<awoods>https://wiki.duraspace.org/display/FF/2014-10-30+-+Fedora+Committer+Meeting11:00
<cbeer>awoods: i'm on.11:01
<barmintor>escowles++11:03
* nikhiltri joins11:04
<f4jenkins>Yippee, build fixed!11:05
Project fcrepo4 build #2523: FIXED in 16 min: http://jenkins.fcrepo.org/job/fcrepo4/2523/
Project fcrepo4-client build #293: FAILURE in 1 min 42 sec: http://jenkins.fcrepo.org/job/fcrepo4-client/293/11:07
<pivotal-bot>Chris Beer added "Rename core models and services" https://www.pivotaltracker.com/story/show/8177062411:09
* Yinlin joins
<jcoyne>cbeer: is this a problem with the Link header:11:10
<http://www.w3.org/ns/ldp#Resource>;rel="type", <http://www.w3.org/ns/ldp#NonRDFSource>;rel="type", <http://localhost:8983/fedora/rest/test/7c/2f/da/3f/7c2fda3f-d0ac-4187-b6ec-e98c6acfe052/descMetadata/fcr:metadata>; rel="describedby"
see the space between ; and rel (for describedby)
?
<barmintor>Hmm, adding @since is breaking the build in fs-connector11:11
<pivotal-bot>Chris Beer added "Add external resource non-rdf source type" https://www.pivotaltracker.com/story/show/81770804
Chris Beer added comment: "Here's my proposal for dealing with that kind of object: ""11:12
https://www.pivotaltracker.com/story/show/81770804
M�" https://www.pivotaltracker.com/story/show/81714060
Chris Beer added "Add support for LDP interaction models" https://www.pivotaltracker.com/story/show/8177100211:13
* ajs6f joins11:14
<cbeer>jcoyne: no. it's just a difference between our hand-crafted, artisanal Link headers, and those we're generated through the JAX-RS framework.
<ajs6f>Is the committers' meeting on hangout?
<cbeer>ajs6f: no. on the dreadful readytalk.
<ajs6f>YAY!
omw11:15
<jcoyne>cbeer: so this? /^<([^>]+)>;\s?rel="([^"]+)"/
or \s+ ?11:16
<cbeer>jcoyne: you're looking for a regex to match link headers?
<jcoyne>well, for each line
yeah
<cbeer>\s*11:17
\s*;\s*
rather.
<jcoyne>cbeer: I see you have {0,1} here: https://github.com/cbeer/ldp/blob/d60389000503b44b98a8ce8b6dccc5ce0f4a02f4/lib/ldp/response.rb#L18
<cbeer>jcoyne: and you're not guaranteed order of the link params either.
jcoyne: make it someone elses problem? https://github.com/asplake/link_header/blob/master/lib/link_header.rb11:18
<pivotal-bot>Chris Beer edited "Rename core models and services" https://www.pivotaltracker.com/story/show/8177062411:20
Chris Beer edited "Replace hasContent/isContentOf in RDF with describes/isDescribedBy" https://www.pivotaltracker.com/story/show/8092139011:23
* github-ff joins11:24
[fcrepo4] ajs6f pushed 1 new commit to PreferComboAction: http://git.io/4f_Z6w
fcrepo4/PreferComboAction 83773bc ajs6f: License and checkstyle cleanup
* github-ff leaves
<pivotal-bot>Chris Beer edited "Add external resource non-rdf source type" https://www.pivotaltracker.com/story/show/8177080411:25
Chris Beer edited "Replace hasContent/isContentOf in RDF with describes/isDescribedBy" https://www.pivotaltracker.com/story/show/8092139011:26
Chris Beer edited "Rename core models and services" https://www.pivotaltracker.com/story/show/81770624
<cbeer>ajs6f: you're on the call, right?11:28
<ajs6f>y
* nikhiltri leaves
<ajs6f>Just letting @awwods talk first.
JCR is impl. Relying on impl classes breaks encapsulation, and that's always a dangerous move.11:30
(In this case, impl functionality, not classes, but the idea is the same.)
As @scossu discovered it's already tricky.11:31
* github-ff joins11:35
[fcrepo4] ajs6f pushed 1 new commit to PreferComboAction: http://git.io/vsdBJg
fcrepo4/PreferComboAction 7026911 ajs6f: Cleanup
* github-ff leaves
* github-ff joins
[fcrepo4] ajs6f opened pull request #608: Prefer combo action (master...PreferComboAction) http://git.io/NV8jGg
* github-ff leaves
<pivotal-bot>A. "Offratonix" Soroka added comment: "https://github.com/fcrepo4/fcrepo4/pull/608" https://www.pivotaltracker.com/story/show/81242476
A. "Offratonix" Soroka finished "Handle multiple Prefer: headers" https://www.pivotaltracker.com/story/show/8124247611:36
<cbeer>ajs6f: and, for 4.0, i'd argue it shouldn't bother with baked-in content modeling.11:38
<ajs6f>cbeer: it meaning Fedora?
<cbeer>ajs6f: and, i don't think scossu1 is talking about content modeling. He's really talking about validation.
ajs6f: you could model content in an OWL document somewhere, but that's not actionable.11:39
<barmintor>(in some sense, we are back to our discussion about our types as mixins, when scossu wanted to use primaryType, from last January)
<f4jenkins>Yippee, build fixed!
Project fcrepo4-client build #294: FIXED in 2 min 5 sec: http://jenkins.fcrepo.org/job/fcrepo4-client/294/
* nikhiltri joins11:40
<cbeer>data rots slower in a refrigerator.11:41
<f4jenkins>Project fcrepo-message-consumer build #843: UNSTABLE in 4 min 55 sec: http://jenkins.fcrepo.org/job/fcrepo-message-consumer/843/11:42
<barmintor>https://groups.google.com/forum/#!searchin/fedora-tech/Fedora$204$20primary$20types/fedora-tech/WuY3fqu_JeE/OjygC7ZD-kYJ11:43
<nikhiltri>I hear a *ba-dam tssss* whenever someone says "kosher". #fcrefrigerator11:47
<cbeer>awoods: in order to support fedora's needs, the CND has to be open enough to be useless for validation.11:52
<osmandin>afk11:56
<ajs6f>Fedora 4: What's in your 'fridge?12:09
<barmintor>awoods: I’m held up by inconsistent test failures in fs-connector on build12:11
<awoods>barmintor: is there any hope of bringing in some consistency?
<barmintor>awoods: working on it now- I think there’s some fishiness about how those tests were changed after the fs-connector went read-only
<awoods>barmintor: I will stay tuned12:12
<jonroby>So, anyone have any suggestions for OAI information predicates:prperty names for storage on the root node please?12:13
* longshou leaves
<ajs6f>What kind of information are you storing?12:14
<jonroby>reponame, repo description, admin email, and repo version12:15
<awoods>ajs6f: jonroby's question relates to: https://www.pivotaltracker.com/story/show/81482904
<pivotal-bot>feature: Allow admins to manage descriptive content about the repository and make the content available (started) / owner: Jonathan Roby
<awoods>- Admin email addresses
- Repository name
- Repository descrption
<ajs6f>Does OAI not define an RDF form in which this stuff will eventually be published?12:16
If not, how _will_ it be published?
<jonroby>ajs6f: the OAI provider information was canned, it was/is static text12:17
<awoods>ajs6f: we are talking about OAI-PMH
<jonroby>its the OAI-PMH Identify response
<awoods>ajs6f: The OAI details (mentioned above) will be stored as properties on the root node, then translated by the oaiprovider extension that fasseg just built into OAI-PMH responses.12:18
<ajs6f>Right. I don't understand why the vocabulary in use for persistence isn't just the same vocabulary as fasseg will use to create the repsonse.12:19
<awoods>ajs6f: is there an OAI vocabulary?
<ajs6f>http://www.openarchives.org/OAI/openarchivesprotocol.html#Identify
It's not RDF, and who cares?12:20
There's a namespace, there are local names.
<awoods>ajs6f: we would want those predicates to be de-referencable...
ajs6f: maybe the namespace would be: http://www.openarchives.org/OAI/openarchivesprotocol.html ?12:21
<ajs6f>awoods: I think dereferenc-ability here is not the most important concern.
the clients going against this endpoint are not HATEOS.12:22
<awoods>ajs6f: agreed, but there needs to be a namespace.
ajs6f: suggestions?12:23
<ajs6f>http://www.openarchives.org/OAI/2.0/
It doesn't dereference, and I don't care.
<awoods>ajs6f: ok. So use that namespace with the local names found in: http://www.openarchives.org/OAI/openarchivesprotocol.html#Identify12:24
<ajs6f>We can start a conversation with OAI if we really care.
awoods: yeah.
<awoods>jonroby: yeah
<jonroby>awoods: ok12:25
<awoods>ajs6f: ok
<ajs6f>all: ok
ok: ok
* acoburn leaves
<jonroby>ajs6f/awoods:thank you12:26
<awoods>cheers
<ajs6f>ok: jonroby
<barmintor>ok, for some reason the read-only fs-connector is consistently throwing exception on remove property, but not on write property12:28
* acoburn joins12:29
<jonroby>afk12:31
<cbeer>barmintor: this is on your branch?12:38
or master?
<barmintor>my branch12:40
<cbeer>barmintor: what's pushed? it just built fine for me12:43
<barmintor>cbeer: it’s the same branch, with a javadoc added. I don;t get it, but it was happening at home, too12:52
<cbeer>barmintor: can you push it too? sounds fishy.12:54
barmintor: or, maybe, did the read-only change leave something in a bad state for you, maybe?12:55
<barmintor>cbeer: maybe. I’ll try a reset12:56
<pivotal-bot>Chris Beer edited "Rename core models and services" https://www.pivotaltracker.com/story/show/8177062412:58
* longshou joins13:10
* github-ff joins13:16
[fcrepo4] barmintor force-pushed 80959252 from 17a767c to a442c92: http://git.io/nxQKtw
fcrepo4/80959252 7e2df86 Chris Beer: Use methods on FedoraResource instead of getting the JCR Node
fcrepo4/80959252 4c080d3 Benjamin Armintor: refactor getClusterExecutor into Function subclass
fcrepo4/80959252 9dc9305 Benjamin Armintor: push rarely-used repositoryService injection down inheritance tree
* github-ff leaves
* github-ff joins13:18
[fcrepo4] barmintor force-pushed 80959252 from a442c92 to 2e3e927: http://git.io/nxQKtw
fcrepo4/80959252 b84c957 Benjamin Armintor: refactor getClusterExecutor into Function subclass
fcrepo4/80959252 17da403 Benjamin Armintor: push rarely-used repositoryService injection down inheritance tree
fcrepo4/80959252 f8c655a Benjamin Armintor: refactor Service into a genric interface...
* github-ff leaves
<barmintor>that’s better
<cbeer>awoods: can we get that merged soon?13:19
<awoods>cbeer/barmintor: sure
<cbeer>awoods: same with 600 and 607 too. they're begging for merge conflicts with my next piece of work.13:20
<barmintor>cbeer: you were right re: inconsistent state, git reset and mvn clean fixed it
<ajs6f>barmintor: Don't forget to wipe your screen with a clean, damp cloth.
<awoods>cbeer: It would be helpful if you could look into: https://www.pivotaltracker.com/story/show/8171406013:21
<pivotal-bot>bug: Redirect to federated content (unstarted) / owner: Esme Cowles
<cbeer>awoods: indeed. i've made a proposal and linked it from that ticket.
* github-ff joins13:23
[fcrepo4-release-tests] lsitu opened pull request #3: Enabled configurable ports for cargo and fuseki. (master...feature/configuration) http://git.io/hUc1mA
* github-ff leaves
<awoods>cbeer: great, let get the series pushed through.13:24
cbeer: great, let's get the series pushed through.
<f4jenkins>Project fcrepo4-client build #296: FAILURE in 1 min 32 sec: http://jenkins.fcrepo.org/job/fcrepo4-client/296/13:29
<pivotal-bot>Andrew Woods added comment: "Part 1: https://github.com/fcrepo4/fcrepo4/pull/606" https://www.pivotaltracker.com/story/show/8095925213:31
<barmintor>so these size-related ServiceHelper methods...13:35
Not diminishing the value, but they appear to not be called anywhere?13:36
does this mean that estimates of subtree size are another area of LDP-expansion in FCR?13:37
Or should I delete these methods?
<f4jenkins>Project fcrepo4 build #2527: UNSTABLE in 9 min 55 sec: http://jenkins.fcrepo.org/job/fcrepo4/2527/
<escowles>barmintor: i think they were being called and the output added to the root node's RDF output, but that was disabled b/c it made mode load everything from a federation in order to calculate the size13:38
<barmintor>escowles: so… where does that leave us?
<escowles>barmintor: i'm not sure -- i think there's a ticket for fixing that, let me see if i can find it13:39
<barmintor>org.fcrepo.kernel.impl.rdf.impl huh?13:41
<escowles>barmintor: it's really impl-y13:42
<barmintor>org.fcrepo.kernel.impl.rdf.impl.mappings is pretty impl-y, too
* github-ff joins13:43
[fcrepo4] awoods pushed 1 new commit to master: http://git.io/yT8LZA
fcrepo4/master d62c1fc Andrew Woods: Merge pull request #606 from fcrepo4/80959252...
* github-ff leaves
<pivotal-bot>Longshou Situ added comment: "PR https://github.com/fcrepo4-labs/fcrepo4-release-tests/pull/3
We can run it with the following command now�" https://www.pivotaltracker.com/story/show/81640410
Longshou Situ finished "Enable random ports on fcrepo4-release-tests" https://www.pivotaltracker.com/story/show/81640410
<escowles>we're headed towards rdf-style striping org.fcrepo.kernel.impl.ref.impl.mappings.impl...
<barmintor>escowles: those packages should probably all be renamed with only a last-part impl, no?
we don’t get any access benefit from sub-packages, iirc13:44
<awoods>escowles/barmintor: Kai is supposedly working: https://www.pivotaltracker.com/story/show/78647248
<pivotal-bot>feature: Improve object/byte count performance, or move to separate REST endpoint (started) / owner: Kai Sternad
Longshou Situ added comment: "@escowles What do you suggest me to work for this ticket?" https://www.pivotaltracker.com/story/show/80653522
<awoods>barmintor: access benefit?13:45
<escowles>barmintor: there's also a org.fcrepo.kernel.impl.utils.impl -- i think that's the impl-est of them all
<barmintor>awoods: default access, protected access, etc
ajs6f: is IdentifierConverter a dead interface?13:46
<cbeer>barmintor: no.
<awoods>barmintor: in terms of package names, we do want each module to be in a separate package.
<ajs6f>barmintor: Aren't ther eimpls?
<cbeer>https://github.com/fcrepo4/fcrepo4/blob/master/fcrepo-http-commons/src/main/java/org/fcrepo/http/commons/api/rdf/HttpResourceConverter.java#L71
<ajs6f>There y'go.13:47
<barmintor>cbeer: ok, it must be an injection thing. the rdfcontext constructors withthat param are showing up as unused
<cbeer>(although i wonder if we can kill it after we kill fcr:sparql... a lot of the non-Converter<> work supports that.)
<barmintor>awoods: I’m only arguing that ‘impl’ should probably only be in one place?13:48
<awoods>barmintor: no disagreement from me there.
cbeer: you haven't killed fcr:sparql yet!? ;)13:49
* github-ff joins13:50
[fcrepo4] cbeer force-pushed remove-html-datastream from 7880667 to b18ae21: http://git.io/DsnvwQ
fcrepo4/remove-html-datastream b18ae21 Chris Beer: Remove datastream references from HTML UI controls
* github-ff leaves
<awoods>cbeer!
I was just building/reviewing 600!
cbeer: should I wait?
<cbeer>awoods: barmintor's commit caused a conflict. i've rebased it.
<barmintor>I leave a trail of conflict and squashed dreams in my wake13:51
s/dreams/pull requests/13:52
<ajs6f>afk for lunch, bbl13:54
* ajs6f leaves
<f4jenkins>Yippee, build fixed!
Project fcrepo4 build #2528: FIXED in 11 min: http://jenkins.fcrepo.org/job/fcrepo4/2528/
* armintor: refactor getClusterExecutor into Function subclass
* armintor: push rarely-used repositoryService injection down inheritance tree
* armintor: refactor Service into a genric interface
* armintor: Checkstyle updates
* armintor: remove unused RDF type constants and regularize produced serializations
Project fcrepo-message-consumer build #846: FAILURE in 1 min 16 sec: http://jenkins.fcrepo.org/job/fcrepo-message-consumer/846/13:55
Yippee, build fixed!13:56
Project fcrepo4-client build #297: FIXED in 1 min 50 sec: http://jenkins.fcrepo.org/job/fcrepo4-client/297/
<awoods>escowles: Is there anything in UCSD pilot effort that longshou can begin working?13:59
<escowles>awoods: i don't think so -- the ingest is chugging along right now14:01
* github-ff joins14:02
[fcrepo4] awoods pushed 2 new commits to master: http://git.io/2mTIWA
fcrepo4/master b18ae21 Chris Beer: Remove datastream references from HTML UI controls
fcrepo4/master f974ea1 Andrew Woods: Merge pull request #600 from fcrepo4/remove-html-datastream...
* github-ff leaves
<pivotal-bot>Andrew Woods delivered "Remove fedora:object/fedora:datastream references from the HTML UI "Create object" panel (and JAX-RS endpoint?)" https://www.pivotaltracker.com/story/show/81601312
<f4jenkins>Project fcrepo4 build #2529: UNSTABLE in 9 min 54 sec: http://jenkins.fcrepo.org/job/fcrepo4/2529/14:04
<pivotal-bot>Andrew Woods edited "Cleanup Modeshape config properties exposed at root of repository" https://www.pivotaltracker.com/story/show/75047684
Andrew Woods added "Put Root Node properties in a jquery accordion" https://www.pivotaltracker.com/story/show/8178664014:06
Andrew Woods edited "Put Root Node properties in a jquery accordion" https://www.pivotaltracker.com/story/show/81786640
<awoods>cbeer: it looks like 607 needs a rebase
<cbeer>awoods: does it?14:07
awoods: maybe my github ui is just slow to update, but it looks fine to me (and i'd be surprised if it conflicted with the last couple..)
<awoods>cbeer: I got a compilation error in kernel-impl14:08
cbeer: I will try again
<barmintor>I’m showing green for #607
<cbeer>awoods: oh, i thought you meant there was a conflict, not that it was broken.14:09
<awoods>cbeer: yes, it is broken
* github-ff joins14:13
[fcrepo4] cbeer force-pushed no-jcr-node from f0158f4 to a7c933e: http://git.io/Qp85BQ
fcrepo4/no-jcr-node 1c7f2e0 Chris Beer: Use FedoraResource#getProperty/hasProperty
fcrepo4/no-jcr-node a7c933e Chris Beer: Push versioned ancestor logic up to FedoraResource
* github-ff leaves
<f4jenkins>Yippee, build fixed!14:15
Project fcrepo4 build #2530: FIXED in 11 min: http://jenkins.fcrepo.org/job/fcrepo4/2530/
Chris Beer: Remove datastream references from HTML UI controls
<cbeer>awoods: psuhed.14:16
<awoods>cbeer: gto it
<f4jenkins>Project fcrepo4 build #2531: FAILURE in 1 min 51 sec: http://jenkins.fcrepo.org/job/fcrepo4/2531/14:17
<barmintor>escowles: HttpPidMinter is only called from tests - sounds like an injection false negative?14:20
<f4jenkins>Project fcrepo-message-consumer build #847: NOW UNSTABLE in 5 min 37 sec: http://jenkins.fcrepo.org/job/fcrepo-message-consumer/847/14:21
<escowles>barmintor: i think the only way it would be used is if someone updated their spring config to use it instead of the UUID pid minter
barmintor: is that the same as 'injection false negative'?14:22
<barmintor>escowles: more or less
this makePid factory method in the PidMinter interface, though14:23
that looks like it needs to go14:24
which also means the BasePidMinter class is a do-nothing layabout
* ajs6f joins14:26
<pivotal-bot>Esme Cowles added comment: "@longshous The current ingest is running on lib-devsandbox1 with modeshape storage and the RDF serializer. Yo�" https://www.pivotaltracker.com/story/show/8065352214:28
* travis-ci joins
fcrepo4/fcrepo4#3065 (no-jcr-node - a7c933e : Chris Beer): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/f0158f47c3ac...a7c933ead7d5
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39521046
* travis-ci leaves
<barmintor>ajs6f: org.fcrepo.kernel.impl.rdf.impl.mappings.ZippingIterator14:29
ajs6f: convince me not to delete it14:30
<ajs6f>Do we not use it for something?
<barmintor>exactly
<ajs6f>?
<barmintor>we do not appear to use it for something
Yes, we don’t14:31
<ajs6f>Oh, in that case, axe it, by all means.
<barmintor>ajs6f: I mean, I’l be sorry to see that glorious FQCN go...
<ajs6f>barmintor: Think we'll ever see another like it?14:32
* barmintor gets misty eyed
<ajs6f>You're weird.
I wonder what it was for.
<barmintor>if only we could ask the author
<ajs6f>Who wrote it?
<barmintor>Somebody who’s you14:33
<ajs6f>Why did I write it? What was the purpose of it?
<barmintor>How am I not myself?
* jonroby leaves
<barmintor>These are questions not answered by javadocs
<ajs6f>Those are NP-hard questions.14:34
<pivotal-bot>Andrew Woods delivered "Where possible, don't use the JCR node directly." https://www.pivotaltracker.com/story/show/81689998
* github-ff joins
[fcrepo4] awoods pushed 1 new commit to master: http://git.io/jnRkZA
fcrepo4/master 3432611 Andrew Woods: Merge pull request #607 from fcrepo4/no-jcr-node...
* github-ff leaves
<cbeer>ajs6f; it had something to do with RDF serialization.
<ajs6f>cbeer: Thanks! Oh, well, if it's not in use any more, the heck with it.
I wonder if there's _any_ other code that isn't doing anything… naaaahhhh….14:35
<barmintor>awoods was right that component-scanning really complicates finding actually-dead code14:36
<pivotal-bot>Longshou Situ added comment: "@escowles We just disabled the FedoraFileSystemConnector to be writable and make it to be always read-only (�" https://www.pivotaltracker.com/story/show/80653522
<ajs6f>barmintor: Yes, it's a question of static vs. dynamic analysis. There's a reason some of us like powerful type systems.
<awoods>ajs6f: your PR to support multiple Prefer headers touched 111 files... I would have thought our support for Prefer headers was less pervasive.14:37
https://github.com/fcrepo4/fcrepo4/pull/608
<ajs6f>awoods: Looks like some carelessness on my part pulling weird files into Git. Let me see if I can clean it up quick.14:38
<barmintor>ajs6f: mebbe also some reorganization of import statements
<ajs6f>barmintor: Yeah, but that works for me, because I have to change the imports anyway.14:39
<awoods>it is important that everyone have a chance to organize import statements just right.
<ajs6f>awoods++
<barmintor>ajs6f: looks like we might want to .gitignore test-output
<ajs6f>barmintor: Yep, looks like that was it.14:40
fcrepo-integration-ldp/report-basic/test-output/navigator-bullet.png
I think all this stuff is coming from fcrepo-integration-ldp, but maybe not...
cbeer: Did you see anything weird like this when you did fcrepo-integration-ldp?
<cbeer>ajs6f: yes, but it should be git-ignored.14:41
https://github.com/fcrepo4/fcrepo4/blob/master/.gitignore#L17
<ajs6f>cbeer: Hm, maybe I need to do what barmintor did to fix whatever was going wrong with whatever he was doing that other time.
<pivotal-bot>Esme Cowles added comment: "@longshous The next ingest I was planning was to copy the files to the federated filesystem, and then create o�" https://www.pivotaltracker.com/story/show/8065352214:42
<f4jenkins>Yippee, build fixed!14:46
Project fcrepo4 build #2532: FIXED in 11 min: http://jenkins.fcrepo.org/job/fcrepo4/2532/
* Chris Beer: Use FedoraResource#getProperty/hasProperty
* Chris Beer: Push versioned ancestor logic up to FedoraResource
* github-ff joins14:47
[fcrepo4] ajs6f pushed 1 new commit to PreferComboAction: http://git.io/8DOtWA
fcrepo4/PreferComboAction 66e056f ajs6f: Removing LDP test cruft
* github-ff leaves
<ajs6f>awoods: ^^^ 12 files changed. Much better.
<f4jenkins>Project fcrepo4 build #2533: FAILURE in 41 sec: http://jenkins.fcrepo.org/job/fcrepo4/2533/
<ajs6f>awoods: Hang on, I see a place where I could clean up something.14:49
<pivotal-bot>Esme Cowles added "Attempting to create a node that already exists should return 409 Conflict, not 415 Unsupported Media Type" https://www.pivotaltracker.com/story/show/81790304
Esme Cowles started "Attempting to create a node that already exists should return 409 Conflict, not 415 Unsupported Media Type" https://www.pivotaltracker.com/story/show/81790304
A. "Offratonix" Soroka started "Handle multiple Prefer: headers" https://www.pivotaltracker.com/story/show/81242476
<cbeer>escowles: i disagree about 81790304.14:51
escowles: in the first request, we create a LDP-RS14:52
escowles: once it is an LDP-RS, you must give us RDF.
<escowles>cbeer: sure -- and if there's a request body that isn't RDF, then 415 is fine
but if we're doing a PUT with no body, it's not the content type that's wrong, it's the lack of a body14:53
<cbeer>escowles: unless what i'm trying to do is create a 0 byte LDP-NR.14:54
escowles: or, totally erase everything in that object.
in which case, an empty body is totally valid, it's just not typed.14:55
<escowles>cbeer: if we have no request content, and no content-type specified...
<pivotal-bot>Andrew Woods edited "Draft implementation of a WebAccessControls authz module" https://www.pivotaltracker.com/story/show/8176477014:57
A. "Offratonix" Soroka finished "Handle multiple Prefer: headers" https://www.pivotaltracker.com/story/show/8124247614:58
* github-ff joins
[fcrepo4] ajs6f pushed 1 new commit to PreferComboAction: http://git.io/-UN1gA
fcrepo4/PreferComboAction b851e43 ajs6f: Small cleanup
* github-ff leaves
<escowles>cbeer: more to the point, there's already code to do what i'm talking about, it's just not working: https://github.com/fcrepo4/fcrepo4/blob/master/fcrepo-http-api/src/main/java/org/fcrepo/http/api/FedoraLdp.java#L26615:03
<cbeer>escowles: hm. then i'm not sure /that/ logic is right either. you should be able to wipe things out.15:04
<escowles>cbeer: how about: 409 if resource exists and body and content-type are both null, 415 if the content-type is non-null and non-RDF15:06
* travis-ci joins
fcrepo4/fcrepo4#3068 (PreferComboAction - 66e056f : ajs6f): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/7026911ff842...66e056fa03b9
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39524474
* travis-ci leaves
<cbeer>escowles: but only for LDP-RS.
<escowles>cbeer: sure, you can do anything you want to a non-RDF resource15:07
<pivotal-bot>Andrew Woods added comment: "So that: ""
1. So that the structural aspects of the OLO would be supported by Fedora
2. So that Fedora will be �" https://www.pivotaltracker.com/story/show/81764672
<cbeer>escowles: sounds right.
<pivotal-bot>Chris Beer added comment: "Ok, so we're not just talking about pre-defining predicate, but actually using them for same-name-siblings?" https://www.pivotaltracker.com/story/show/8176467215:08
<cbeer>awoods: no, wait, i'm still confused. why does the CND need those predicates?
<awoods>cbeer: The expectation would be that a user would apply the mixin:OrderedList or mixin:Slot15:11
cbeer: rather, apply those RDF types
which may include SNS implications15:12
<f4jenkins>Project fcrepo-module-auth-rbacl build #325: ABORTED in 1 day 4 hr: http://jenkins.fcrepo.org/job/fcrepo-module-auth-rbacl/325/15:13
<cbeer>awoods: i'm still confused what applying that in the CND gets us beyond just making those RDF assertions.
<f4jenkins>Project fcrepo-module-auth-rbacl build #326: FAILURE in 57 sec: http://jenkins.fcrepo.org/job/fcrepo-module-auth-rbacl/326/
<ajs6f>Just how far as we going to go to model RDF/RDFS/etc. structures in JCR?15:14
I'm not being snarky— it's a serious question.
<awoods>cbeer: have you looked at: http://smiy.sourceforge.net/olo/spec/orderedlistontology.html15:15
<cbeer>awoods: yes, i'm aware of the ontology.
awoods: and i agree we should promote it as a better alternative to lists.
awoods: but my question is, why do we need to teach JCR about it?
<barmintor>than rdf lists? *anything* is better than rdf lists
<awoods>cbeer: SNS is one example...15:16
* travis-ci joins
fcrepo4/fcrepo4#3069 (PreferComboAction - b851e43 : ajs6f): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/66e056fa03b9...b851e432f13a
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39525618
* travis-ci leaves
<awoods>cbeer: escowles? mikeAtUVa? scossu1? re:OLO15:17
<ajs6f>"why do we need to teach JCR about it?"
Yes, this.
<escowles>cbeer: i think the feature request is: serialize OLO lists as JCR ordered children
<ajs6f>If someone wants to use it, what is stopping them?
<escowles>ajs6f: nothing is stopping them, they are asking for pushing the order down into the modeshape/JCR layer
<cbeer>escowles: ok. that's not what i understand from that ticket, and i don't think it requires teaching JCR about OLO. we'll need code to translate that on our end, not JCR.15:18
<barmintor>escowles: everything in olo seems pretty straightforward to model as JCR nodes
<escowles>cbeer: i think the ultimate functionality desired is things like returing the child nodes in the OLO order, when paging through large sets of child nodes, etc.
<ajs6f>If it's a matter of translating into JCR, I'm vaguely against that.
afk bbs15:19
<escowles>ajs6f: for the record, i'm vaguely against it too -- but if there is a good use case that just using OLO in fedora nodes doesn't satisfy
<awoods>stepping away for lunch15:20
* mikeAtUVa joins15:21
* Yinlin leaves
<mikeAtUVa>As for ordered lists... there was a general idea that taking advantage of JCR's native ordering and mapping that to some standard ordered list ontology would allow performance/storage benefits and let people do things in JCR and have them work in RDF. That said it's clear that we're moving away from the approach of using JCR similar/analogous JCR constructs to represent linked data concepts.15:25
<ajs6f>In an ideal world, it would be possible to offer extension modules to the HTTP API for this kind of thing, but we're a long way away from that.15:26
<mikeAtUVa>Some of us (me, grenjansen) would benefit from and use special fedora rest API calls that did things like "insert after" or "get ordered children"... for large lists it beats loading the whole set and sorting.
<cbeer>mikeAtUVa: sure. that doesn't explain why we need JCR node types for OLO objects https://www.pivotaltracker.com/n/projects/684825/stories/8176467215:29
<ajs6f>mikeAtUva: One way here would be for you and gregjansen to get together, cut that module, and contriubute it out, not to the main codebase, but just a s a handy utility or those who like to use it.
<pivotal-bot>Chris Beer estimated "Rename core models and services" as 3 points https://www.pivotaltracker.com/story/show/8177062415:36
Chris Beer started "Rename core models and services" https://www.pivotaltracker.com/story/show/81770624
<scossu1>Having the OLO CND does not mean that repo users have to know about it. They would just assign Slot and OrderedList mixins to nodes and they have to be aware of the properties necessary for them to work.15:41
<ajs6f>afk bbs
* ajs6f leaves
<scossu1>But having CND constraints is important to ensure that the OLO slots and OrderedLists are valid.15:43
Unless someone wants to provide that validation on the Fedora layer.15:44
* ajs6f1 joins15:45
scossu: What's the advtange there over just using rdf:type?15:46
<scossu1>ajsf1: How would you make sure that a Slot doesn't have an index of −1 or has a non-existing reference otherwise?15:48
<ajs6f1>scossu: You don't. This, in fact, is the crucial distinction between content modeling and validation.
<scossu1>ajs6f1: But we need validation here, don't we?15:49
<barmintor>I’m not sure that CND wouldn’t make OLO management worse
since the slots point to predecessors and successors
<ajs6f1>scossu1: I don't see that. You can make assertions about ordering, and you can make guarantees about ordering. Doing the first doesn't demand doing the second.15:50
<barmintor>relying on CND would leave the list invalid on deletes
* ajs6f joins15:51
<scossu1>barmintor: how would that happen?
<ajs6f>barmintor: That's nearly as bad as RDF Collections.barmintor: That's nearly as bad as RDF Collections.
<barmintor>S1 <-> S2 <-> S3, delete S2. CND will not be able to leave the list in a valid state
<ajs6f>barmintor: Or there will have to be Fedora-macinery to clean up.15:52
Which is not great, to say the least.
<jcoyne>But OLO is only one of many datastructures
<barmintor>ajs6f: in which case, why are we taking the bad of it for no good
<jcoyne>Are we going to have machinery for each?
<ajs6f>jcoyne: Exactly.
barmintor: That was gnomic, even for this chat room.15:53
<barmintor>I ride a fox to work
<ajs6f>Maybe we should all take in Game of Thrones speeches.
<scossu1>barmintor: I don't see anything in the example CND that would invalidate that list.15:54
<ajs6f>The evil of Fedora is as cold as the ice on a DSpace 1.0 instance.
* ajs6f1 leaves
<jcoyne>http://pydoc.net/Python/game-of-thrones/0.1.7/game_of_thrones/15:55
<barmintor>scossu1: do you make the next and previous properties both optional? Something has to know to build S1 <-> S3
<ajs6f>Not invalid per JCR, invalid per OLO.
<escowles>barmintor: that's a problem for rdf:List, but not for OLO
<ajs6f>awoods: We need a demo vid narrated by the guy who plays the dwarf. His voice is fantastic.15:56
<scossu1>barmintor: olo:previous and olo:next are inferred by Fedora as per proposal.
<ajs6f>scossu: Inferred by Fedora? Sounds like specialized machinery for a specific data structure.
<scossu1>That's the Fedora "machinery".
<ajs6f>scossu: That's what we're trying to avoid, until we have extremely compelling uses.15:57
It's not obvious to me that we yet do.
But I welcome correction.
<scossu1>ajs6f: https://wiki.duraspace.org/display/FF/Design+-+Ordered+Lists15:58
<ajs6f>scossu: Looks like we can rely on https://wiki.duraspace.org/display/~mdurbin, https://wiki.duraspace.org/display/~escowles@ucsd.edu, https://wiki.duraspace.org/display/~scossu to bring forward a plan for the construction of the aforementioned Fedora machinery.
scossu: I'm not trying to be snarky.
<scossu1>the proposal was to implement a data structure at first, then in a second phase, the automation to maintain ordering and presenting the ordered nodes.15:59
<ajs6f>scossu: I'm being (I think) realistic about what we should be doing in the weeks before release.
<escowles>ajs6f: i agree this is 4.1 functionality we're talking about here
<ajs6f>scossu: Well, if we take it at that level, I'm still completely opposed because we're talking about intriducing datastructures via CND.
escowles/scossu: Maybe we can leave this for then?
* osmandin leaves16:00
<scossu1>ajs6f, escowles: absolutely. I was talking about post-4.0.0 plans.
<ajs6f>scossu: Sorry— I didn't understand that.
<jcoyne>Are there any plans for another "released" beta in the near future?
<ajs6f>Fedora 4.Thanksgiving.
Fedora 4.Thanksgiving.Turkey?16:01
<scossu1>ajs6f: my proposal was to just nail down one implementation for ordered lists at first.
So one can start building lists in a repo even without functionality attached.16:02
<ajs6f>scossu: Spec'ing is good, at least as an exercise. I'm yet to understand that having a canonical data structure of any kind (except the adminstrative hierarchy) is within Fedora's remit, but let that go. I'm only objecting to committing any resources to any impl plans before release.
<scossu1>After that, Fedora can provide some convenience automation.
<ajs6f>scossu: If that is something Fedora ends up doing as part of the core. I'm more inclined to thikn of it as an opportunity for a community-curated add-on, like OAI.16:03
<barmintor>This is definitely an add-on16:04
<scossu1>ajs6f: It can be an add-on, my concern is that there is a group of users who agree on a valid implementation in order to join efforts.16:05
<ajs6f>That's how I'm feeling it. I'd love to see a "Data Structures Toolbox" filled with handy mechanisms for lists, etc.
scossu1: +1
That's how it should happen.
Gathering that group of interested parties is nothing but good.
<barmintor>… but I’m wary of thinking that it should be baked into the FCR layer
<ajs6f>barmintor: I don't disagree, but I don't think we have ground to rule it out.16:06
barmintor: If we have folks adding validation of any kind to the repo, this seems like a natural outgrpwth.
<barmintor>If we have default data structure impls in the core distro, you can go ahead and start counting the days before it becomes a pain point
<ajs6f>barmintor: NOT in the core. In some add-on.16:07
For which a known agency is responsible, as scossu suggests.
Validation, to me, is an add-on.
<scossu1>I'm talking from a layman's point of view with little knowledge of Fedora's code architecture, but a case to solve.16:08
^^which can be the worst kind of trouble makers, I know...
<ajs6f>scossu: and gathering folks who share the case and raising a unified voice is a-ok with me!
scossu: I'm sure the committers would rather hear from everyone at once, so to speak.16:09
everyone sharing the case, that is.
<scossu1>ajs6f: I agree with building validation externally, but wouldn't you distinguish validation for user content and validation for core or add-on structures (such as OLO)?16:10
<ajs6f>scossU: Not until you break down for me what you mean by "validation for user content". is that data value validation? Structural validation? Something else?16:11
<scossu1>My point is: letting repo managers build their own CND to validate user-generated content may be unsupported, but why not use CND to validate structures that the end user cannot change, such as the OLO definitions?16:12
s/cannot/shouldn't16:13
<ajs6f>scossu: For one thing, it's not clear to me that there is agreement that the OLO structures "rank" any differently thatn the user-generated content you mention.
They're just user-imported structures with more users.
* github-ff joins16:16
[fcrepo4] escowles created repeated-put-conflict (+1 new commit): http://git.io/WjFFvQ
fcrepo4/repeated-put-conflict c335ba2 Esmé Cowles: Updating logic to successfully detect empty request body
* github-ff leaves
<pivotal-bot>Esme Cowles added comment: "https://github.com/fcrepo4/fcrepo4/pull/609" https://www.pivotaltracker.com/story/show/81790304
* github-ff joins16:17
[fcrepo4] escowles opened pull request #609: Updating logic to successfully detect empty request body (master...repeated-put-conflict) http://git.io/f6R8aQ
* github-ff leaves
<pivotal-bot>Esme Cowles finished "Attempting to create a node that already exists should return 409 Conflict, not 415 Unsupported Media Type" https://www.pivotaltracker.com/story/show/81790304
<scossu1>ajs6f: The OLO definitions would be provided by the developers. Why not use CND to validate those structures? Unless you foresee conflicts between CND validation and the Fedora layer — which would be why you are against any CND validation?
<ajs6f>scossu:: I am not against any CND validation. I am against any CND validation that is visible to the end user / repo dev without intermediation by Fedora.16:18
scossu: If Fedora takes on OLO as a feature, then forward we go!
scossu: But it's not clear to me yet that Fedora should take on OLO as a feature.16:19
<cbeer>ajs6f: i'd raise the bar.. s/should/needs to/
<ajs6f>cbeer: Fari Engouh, as we say.16:20
A small core is a strong core.
<scossu1>ajs6f, re. end user visibility: That is the point. The user doesn't ever need to see the CND. They just know that they have to provide an index for a Slot that is > 1 and other constraints.16:21
<barmintor>LDP. Binaries. Versions. TX.
<scossu1>And agreed that OLO should be an add-on.
* github-ff joins
[fcrepo4] cbeer created rename-all-the-models (+1 new commit): http://git.io/PWbuYQ
fcrepo4/rename-all-the-models 6883b68 Chris Beer: Rename fedora model objects to reflect usage, not legacy baggage
* github-ff leaves
<ajs6f>scossu: If we're talking about an add-on, then this question about CND and Fedora machinery and so forth is an impl decision for the group taking responsbility for that add-on (advised by the committers, yes).16:22
<barmintor>developing it as an add-on in the repository layer implies committing to versioned Java APIs as well, presumably.
<ajs6f>barmintor: Are you saying those things should have been add=ons? Isn't that the direction we're trying to take by partitioning the HTTP API?
<barmintor>ajs6f: I’m saying that’s all core should be, IMO16:23
<ajs6f>barmintor: Oh, right, cool.
barmintor: I think the plan of partition now, with only CRUD as true core, is even fiercer.
barmintor: In that language, your core is the "Workflow Package" of modules.16:24
<barmintor>ajs6f: excellent. smaller the better.
<ajs6f>barmintor: Once we cut it down to nothing, your job of maintaining Fedora 4 forever will be much easier.
<barmintor>I can see my future diminishing as we speak.16:25
<ajs6f>barmintor: It's not impossible, it seems to me, to develop such an add-on using HTTP as the layer between it and the repo, although I admit that would be horribly inefficient.
* barmintor shrugs
in which case, you’ve just decided to develop in Java what you might have developed in Downstream Technology X16:26
<ajs6f>I think we can leave the whole question until scossu assembles his horde of OLOites.
barmintor:Isn't part of the point of spec'ing the API exactly to make it easier to make the decision to develop in tech X, instead of being forced to learn the innards of Fedora?16:27
<barmintor>but in my recent experience, versioned network APIs + versioned backend + versioned Java API = annual minor releases
<ajs6f>I'm not arguing for versioning Java API, I'm suggesting we stop the madness at HTTP.16:28
<scossu1>ajs6f: OLOites sounds quite scary. Do they use axes or spears?
<ajs6f>(And the ontologies, without which HTTP is just syntax)
scossu: I think they wield sharp-edged notions of order and collection.
<barmintor>ajs6f: yes. In fact, I think there should be a lightweight reference impl of the REST API that is not the MODE-backed FCR416:29
but I am a radical in this
<ajs6f>barmintor: Yes. awoods and I were talking about Marmotta earlier today, and it came up that therre are some micro-LDP-impls already.
Yes meaning I agree, not that you are radical. Although you both radical and tubular.16:30
<barmintor>\m/
<ajs6f>I've said it before, and I'll say it again, that as far as a community-sourced Fedora impl that is only of interest to the Fedora community, I would like for this to be the _last_ one.
<barmintor>cbeer ajs6f escowles when y’all were talking about dead code, was there a major area y’all had in mind? I feel a little like I’m chipping away little bits that weren’t even well-known enough to merit a ticket16:31
<ajs6f>barmintor: Not being snarky, but if there were well-known bits, I hope there would be a ticket.
Right?16:32
This is another reason that having clean module boundaries makes everything better.
<barmintor>I mean, these are ZippingIterator-style, 20 line jobs that you wouldn’t even have known were dead
<ajs6f>I didn't. I didn't even know that it was alive.16:33
barmintor: In the absence of static analysis, the only thing that seems easy is to rip out what you suspect and run the ITs.
* travis-ci joins16:34
fcrepo4/fcrepo4#3072 (rename-all-the-models - 6883b68 : Chris Beer): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/commit/6883b68eabbe
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39533954
* travis-ci leaves
<ajs6f>barminttor: If you can't find anything to kill, want to work on the HTTP API outcomes for TWG?16:37
* github-ff joins16:39
[fcrepo-webapp-plus] mohideen opened pull request #3: Xacml fix (master...xacml-fix) http://git.io/5lu7XA
* github-ff leaves
<pivotal-bot>Mohamed Mohideen Abdul Rasheed added comment: "xacml profile fix PR: https://github.com/fcrepo4-labs/fcrepo-webapp-plus/pull/3" https://www.pivotaltracker.com/story/show/8143362816:40
<ajs6f>out for the day
* ajs6f leaves
* github-ff joins
[fcrepo4] cbeer force-pushed rename-all-the-models from 6883b68 to 48bea42: http://git.io/NdtLMQ
fcrepo4/rename-all-the-models 48bea42 Chris Beer: Rename fedora model objects to reflect usage, not legacy baggage
* github-ff leaves
* travis-ci joins16:42
fcrepo4/fcrepo4#3070 (repeated-put-conflict - c335ba2 : Esmé Cowles): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/commit/c335ba20d91b
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39533621
* travis-ci leaves
* github-ff joins16:45
[fcrepo4] cbeer force-pushed rename-all-the-models from 48bea42 to 9b6bd6f: http://git.io/NdtLMQ
fcrepo4/rename-all-the-models 9b6bd6f Chris Beer: Rename fedora model objects to reflect usage, not legacy baggage
* github-ff leaves
<pivotal-bot>Chris Beer finished "Rename core models and services" https://www.pivotaltracker.com/story/show/81770624
Chris Beer added comment: "https://github.com/fcrepo4/fcrepo4/pull/610" https://www.pivotaltracker.com/story/show/81770624
* github-ff joins16:46
[fcrepo4] cbeer opened pull request #610: Rename fedora model objects to reflect usage, not legacy baggage (master...rename-all-the-models) http://git.io/13H0gg
* github-ff leaves
<pivotal-bot>Chris Beer edited "Rename core models and services" https://www.pivotaltracker.com/story/show/81770624
<cbeer>awoods: can you try to get to that PR soon? 100% guaranteed to make merge conflicts for everyone.
* travis-ci joins16:47
fcrepo4/fcrepo4#3073 (rename-all-the-models - 48bea42 : Chris Beer): The build has errored.
Change view : https://github.com/fcrepo4/fcrepo4/compare/6883b68eabbe...48bea4238f36
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39535785
* travis-ci leaves
<cbeer>awoods: it's no new functionality, just renaming stuff
* mikeAtUVa leaves16:50
* travis-ci joins16:57
fcrepo4/fcrepo4#3074 (rename-all-the-models - 9b6bd6f : Chris Beer): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/48bea4238f36...9b6bd6f5bbcf
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39536220
* travis-ci leaves
* github-ff joins16:59
[fcrepo4] barmintor created 80959252 (+5 new commits): http://git.io/L2EI7Q
fcrepo4/80959252 5c57e1f Benjamin Armintor: remove duplicate jcrRdfTools and unused session accessor in RDF iterators
fcrepo4/80959252 9801aa8 Benjamin Armintor: remove unused pid-function-factory method from minters
fcrepo4/80959252 ff3e7aa Benjamin Armintor: remove unused function and input iterator-traversing ZippingIterator
* github-ff leaves
* scossu1 leaves
<awoods>jcoyne: I would like to get a beta-04 out at the end of this sprint... if the dust would settle, even a little.17:18
<pivotal-bot>Chris Beer started "Replace hasContent/isContentOf in RDF with describes/isDescribedBy" https://www.pivotaltracker.com/story/show/80921390
* travis-ci joins
fcrepo4/fcrepo4#3076 (80959252 - 1a229cd : Benjamin Armintor): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/5c57e1f1631a^...1a229cd09f4a
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39537569
* travis-ci leaves
* github-ff joins17:26
[fcrepo4] cbeer created describes-predicate (+1 new commit): http://git.io/xOBfVQ
fcrepo4/describes-predicate 203d3ff Chris Beer: Use IANA describes/describedby predicates to link a non-rdf source and its description
* github-ff leaves
<pivotal-bot>Chris Beer added comment: "https://github.com/fcrepo4/fcrepo4/pull/611" https://www.pivotaltracker.com/story/show/8092139017:27
Chris Beer finished "Replace hasContent/isContentOf in RDF with describes/isDescribedBy" https://www.pivotaltracker.com/story/show/80921390
* github-ff joins
[fcrepo4] cbeer opened pull request #611: Use IANA describes/describedby predicates to link a non-rdf source and its description (master...describes-predicate) http://git.io/MPg55w
* github-ff leaves
* github-ff joins
[fcrepo4] awoods closed pull request #610: Rename fedora model objects to reflect usage, not legacy baggage (master...rename-all-the-models) http://git.io/13H0gg
* github-ff leaves
<pivotal-bot>Andrew Woods delivered "Rename core models and services" https://www.pivotaltracker.com/story/show/81770624
* github-ff joins17:28
[fcrepo4] cbeer force-pushed describes-predicate from 203d3ff to 3ad9b11: http://git.io/Okeodg
fcrepo4/describes-predicate 3ad9b11 Chris Beer: Use IANA describes/describedby predicates to link a non-rdf source and its description
* github-ff leaves
<pivotal-bot>Andrew Woods accepted "fcr:fixity in HTML view" https://www.pivotaltracker.com/story/show/8171302217:30
Andrew Woods accepted "Where possible, don't use the JCR node directly." https://www.pivotaltracker.com/story/show/81689998
Andrew Woods accepted "Remove fedora:object/fedora:datastream references from the HTML UI "Create object" panel (and JAX-RS endpoint?)" https://www.pivotaltracker.com/story/show/81601312
Andrew Woods accepted "getResourceTriples should avoid switching subjects too much." https://www.pivotaltracker.com/story/show/81620916
Andrew Woods accepted "Restoring a version of non-RDF resources results in error "This operation requires that the node be versionable"" https://www.pivotaltracker.com/story/show/81615202
Andrew Woods accepted "Only include inbound references when explicitly included" https://www.pivotaltracker.com/story/show/81611896
Andrew Woods accepted "fcr:versions should default to text/turtle like other RDF resources" https://www.pivotaltracker.com/story/show/81583556
Andrew Woods accepted "Support the full range of LDP containers" https://www.pivotaltracker.com/story/show/81411416
Andrew Woods accepted "Make FedoraFileSystemConnector Read-Only" https://www.pivotaltracker.com/story/show/80775226
Andrew Woods accepted "The sparql recipes no longer pass for 4.0.0-beta-04-SNAPSHOT" https://www.pivotaltracker.com/story/show/79143930
Andrew Woods added comment: "Please rebase on master... and ideally revert the import statement modifications." https://www.pivotaltracker.com/story/show/8124247617:31
Andrew Woods rejected "Handle multiple Prefer: headers" https://www.pivotaltracker.com/story/show/81242476
* travis-ci joins17:32
fcrepo4/fcrepo4#3077 (describes-predicate - 203d3ff : Chris Beer): The build has errored.
Change view : https://github.com/fcrepo4/fcrepo4/commit/203d3ff3904d
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39540243
* travis-ci leaves
<pivotal-bot>Andrew Woods added comment: "Pending response to code review comment." https://www.pivotaltracker.com/story/show/8164041017:36
Andrew Woods rejected "Enable random ports on fcrepo4-release-tests" https://www.pivotaltracker.com/story/show/81640410
Andrew Woods added comment: "Please rebase on master." https://www.pivotaltracker.com/story/show/8179030417:38
Andrew Woods rejected "Attempting to create a node that already exists should return 409 Conflict, not 415 Unsupported Media Type" https://www.pivotaltracker.com/story/show/81790304
* MohamedAR leaves17:40
* github-ff joins
[fcrepo4] barmintor force-pushed 80959252 from 1a229cd to ea63ee0: http://git.io/nxQKtw
fcrepo4/80959252 f573139 Benjamin Armintor: remove duplicate jcrRdfTools and unused session accessor in RDF iterators
fcrepo4/80959252 3331551 Benjamin Armintor: remove unused pid-function-factory method from minters
fcrepo4/80959252 b1cf57a Benjamin Armintor: remove unused function and input iterator-traversing ZippingIterator
* github-ff leaves
<pivotal-bot>Benjamin Armintor finished "Remove dead code" https://www.pivotaltracker.com/story/show/8095925217:42
* github-ff joins17:43
[fcrepo4] barmintor opened pull request #612: Remove dead code (master...80959252) http://git.io/pmBlyw
* github-ff leaves
<pivotal-bot>Benjamin Armintor added comment: "Part 2: https://github.com/fcrepo4/fcrepo4/pull/612" https://www.pivotaltracker.com/story/show/80959252
* travis-ci joins17:46
fcrepo4/fcrepo4#3080 (describes-predicate - 3ad9b11 : Chris Beer): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/203d3ff3904d...3ad9b11595ef
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39540351
* travis-ci leaves
* travis-ci joins17:48
fcrepo4/fcrepo4#3079 (master - bc44e2d : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/34326117ce31...bc44e2da38a3
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39540262
* travis-ci leaves
* github-ff joins17:50
[fcrepo4] lsitu opened pull request #613: Added support to the web UI to trigger the file picker when no file attached for datastream creation. (master...feature/datastream-upload) http://git.io/L3jVBw
* github-ff leaves
<pivotal-bot>Longshou Situ added comment: "https://github.com/fcrepo4/fcrepo4/pull/613" https://www.pivotaltracker.com/story/show/78847348
Longshou Situ finished "Adding datastream without file in HTML UI causes ugly error" https://www.pivotaltracker.com/story/show/78847348
Longshou Situ started "Enable random ports on fcrepo4-release-tests" https://www.pivotaltracker.com/story/show/81640410
* acoburn leaves17:53
<f4jenkins>Project fcrepo-message-consumer build #848: ABORTED in 3 hr 15 min: http://jenkins.fcrepo.org/job/fcrepo-message-consumer/848/18:01
Project fcrepo4 build #2534: ABORTED in 3 hr 4 min: http://jenkins.fcrepo.org/job/fcrepo4/2534/18:02
* travis-ci joins18:05
fcrepo4/fcrepo4#3082 (80959252 - ea63ee0 : Benjamin Armintor): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/1a229cd09f4a...ea63ee03b82c
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39541855
* travis-ci leaves
<pivotal-bot>Andrew Woods added comment: "Are there actual IANA local names: describedby and describes?
Or are these pure fabrications?" https://www.pivotaltracker.com/story/show/80921390
Benjamin Armintor added comment: "http://www.iana.org/assignments/link-relations/link-relations.xml" https://www.pivotaltracker.com/story/show/8092139018:07
<cbeer>awoods: they're real.
<pivotal-bot>Andrew Woods added comment: "Does PR-3 above finish this ticket?" https://www.pivotaltracker.com/story/show/8143362818:08
<awoods>cbeer: where can I look them up?
<barmintor>awoods: check that link I posted
<awoods>barmintor: where did you do your link posting?
<barmintor>in the issue, but it’s http://www.iana.org/assignments/link-relations/link-relations.xml18:09
<cbeer>awoods: and, somewhere, an RFC says how to turn those names into predicates.
* jcoyne leaves
<awoods>cbeer/barmintor: you two are ok with the fact that these do not resolve, nor point to anything that gives any indication what the terms mean? http://www.iana.org/assignments/relation/describedby18:10
http://www.iana.org/assignments/relation/describes
<barmintor>cbeer: http://www.w3.org/TR/2009/REC-powder-dr-20090901/#semlink
<cbeer>awoods: yes. being resolvable is not important.
<barmintor>although not being resolvable makes me a little sad18:11
<awoods>barmintor: agreed
<barmintor>but that POWDER stuff is a really real thing
<awoods>what about the tenents of linked data?18:12
<cbeer>awoods: resolvability is only a nice-to-have, and when it comes to predicates really doesn't matter.18:13
so, we can be sad. but it's still a thing.
<barmintor>awoods: also, the unresolvable form under iana exists for both describes and describedby, but w3c’s resolvable form only for describedby18:14
akaik, anyway
<cbeer>barmintor: powder doesn't define describes.
(and, we don't actually want POWDER's describedby either, iirc... they do something funky with the domain or range of the predicate)18:15
<barmintor>cbeer: right, it’s an extension
oh, I see: POWDER subtypes IANA18:16
<cbeer>barmintor: right. and we just want to say "there's this link between them.. we included it in the HTTP header, but here it is again"
<f4jenkins>Yippee, build fixed!
Project fcrepo4 build #2535: FIXED in 13 min: http://jenkins.fcrepo.org/job/fcrepo4/2535/
Chris Beer: Rename fedora model objects to reflect usage, not legacy baggage
* github-ff joins
[fcrepo4] awoods deleted describes-predicate at 3ad9b11: http://git.io/AfC0uA
* github-ff leaves
<barmintor>yeah. I also see the Range funkiness.
<pivotal-bot>Andrew Woods delivered "Replace hasContent/isContentOf in RDF with describes/isDescribedBy" https://www.pivotaltracker.com/story/show/8092139018:17
<f4jenkins>Project fcrepo4-client build #300: FAILURE in 2 min 8 sec: http://jenkins.fcrepo.org/job/fcrepo4-client/300/18:19
* github-ff joins
[fcrepo4] awoods deleted 80959252 at ea63ee0: http://git.io/UVaVkA
* github-ff leaves
<awoods>barmintor: I assume you are still working the "Remove dead code" ticket?
<barmintor>awoods: I think I’m as done with it as I can get for now.18:20
<pivotal-bot>Andrew Woods delivered "Remove dead code" https://www.pivotaltracker.com/story/show/80959252
<awoods>barmintor: what are you thinking about tackling next?18:22
<pivotal-bot>Chris Beer estimated "Add external resource non-rdf source type" as 2 points https://www.pivotaltracker.com/story/show/8177080418:23
Chris Beer started "Add external resource non-rdf source type" https://www.pivotaltracker.com/story/show/81770804
<barmintor>cbeer, awoods worth starting to work on “Update Nodes API Docs”? I need to learn the LDP stuff better to know whether the FCR3 backport is possible, anyway18:24
<awoods>barmintor: FCR3 backport--18:25
<barmintor>what I do in the privacy of my own home is my business!
:P
<awoods>barmintor: cleaning up the docs would be great... cbeer has already done some of it... and it continues to be a moving target. But go for it: https://www.pivotaltracker.com/story/show/7960049218:26
<pivotal-bot>feature: Update Nodes API documentation (unstarted) / owner: Chris Beer
<barmintor>I mean, it looks like it’s either that or WAC, and WAC looks terrifying18:27
* travis-ci joins18:28
fcrepo4/fcrepo4#3086 (master - e619ab4 : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/4c5a5ed6bd62...e619ab49f2db
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39545002
* travis-ci leaves
<awoods>barmintor: There is not a ticket for it, but just having you walk through the codebase, identify points of clean-up/refactoring would also be helpful.
<barmintor>ok18:29
cbeer: did that generic Service interface do what you were looking for?
<awoods>barmintor: If you go that way, just make a ticket for it.
<barmintor>awoods: sure thing
<cbeer>barmintor: yep. i'm wondering if NodeService is actually a Service<>, but i'll take a look later18:30
<awoods>barmintor: also, cbeer's recent massive renaming will undoubtedly break the external modules (message-consumer, module-auth-*, fcrepo4-client, etc). Help fixing those would also be great.18:31
<barmintor>cbeer: minimally, it’s weird to keep calling it Node* when it’s all really FedoraResource*
<awoods>barmintor: http://jenkins.fcrepo.org/
<barmintor>yep, lots of clouds18:32
<awoods>it's gloomy out there
* travis-ci joins18:33
fcrepo4/fcrepo4#3085 (master - 4c5a5ed : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/bc44e2da38a3...4c5a5ed6bd62
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39544836
* travis-ci leaves
<awoods>the jenkins machine only has 6GB of memory, but Mohamed is looking into increasing that.
some of the jenkins breakage is related to machine resources.
<pivotal-bot>Andrew Woods added comment: "Please rebase on master." https://www.pivotaltracker.com/story/show/7884734818:34
Andrew Woods rejected "Adding datastream without file in HTML UI causes ugly error" https://www.pivotaltracker.com/story/show/78847348
* awead leaves18:37
* ksclarke leaves18:40
<pivotal-bot>Andrew Woods finished "Adding datastream without file in HTML UI causes ugly error" https://www.pivotaltracker.com/story/show/7884734818:44
* nikhiltri leaves18:46
* github-ff joins18:47
[fcrepo4] cbeer created external-body (+1 new commit): http://git.io/FET7mw
fcrepo4/external-body c9a435b Chris Beer: Use message/external-body MIME type for registering 'redirect' binaries
* github-ff leaves
* github-ff joins18:50
[fcrepo4] cbeer opened pull request #614: Use message/external-body MIME type for registering 'redirect' binaries (master...external-body) http://git.io/GknOIw
* github-ff leaves
<pivotal-bot>Longshou Situ added comment: "@awoods I've rebased it locally. Do you want me to push it for the PR?" https://www.pivotaltracker.com/story/show/7884734818:53
Andrew Woods added comment: "No, actually the initial PR was clean. You are fine." https://www.pivotaltracker.com/story/show/7884734818:54
* awead joins19:00
* travis-ci joins19:01
fcrepo4/fcrepo4#3087 (external-body - c9a435b : Chris Beer): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/commit/c9a435bf232f
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39547437
* travis-ci leaves
* awead leaves19:05
* github-ff joins19:06
[fcrepo4] cbeer force-pushed external-body from c9a435b to d048bb1: http://git.io/RXY5Ag
fcrepo4/external-body d048bb1 Chris Beer: Use message/external-body MIME type for registering 'redirect' binaries
* github-ff leaves
* github-ff joins19:10
[fcrepo4] cbeer force-pushed external-body from d048bb1 to 6b73765: http://git.io/RXY5Ag
fcrepo4/external-body 6b73765 Chris Beer: Use message/external-body MIME type for registering 'redirect' binaries
* github-ff leaves
<pivotal-bot>Longshou Situ added comment: "But I see it failed for the element names change in the form: ""19:13
mixin => new_mixin, fcr:datastream => binary.
�" https://www.pivotaltracker.com/story/show/78847348
* travis-ci joins19:18
fcrepo4/fcrepo4#3089 (external-body - d048bb1 : Chris Beer): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/c9a435bf232f...d048bb12b8cd
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39548817
* travis-ci leaves
* github-ff joins19:22
[fcrepo4] cbeer force-pushed external-body from 6b73765 to 0e1a10e: http://git.io/RXY5Ag
fcrepo4/external-body 0e1a10e Chris Beer: Use message/external-body MIME type for registering 'redirect' binaries
* github-ff leaves
* travis-ci joins19:23
fcrepo4/fcrepo4#3090 (external-body - 6b73765 : Chris Beer): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/d048bb12b8cd...6b737657273c
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39549131
* travis-ci leaves
<pivotal-bot>Chris Beer edited "Add support for resources that redirect to other HTTP resource" https://www.pivotaltracker.com/story/show/81770804
Chris Beer added comment: "https://github.com/fcrepo4/fcrepo4/pull/614" https://www.pivotaltracker.com/story/show/81770804
Chris Beer finished "Add support for resources that redirect to other HTTP resource" https://www.pivotaltracker.com/story/show/81770804
Chris Beer edited "Add support for resources that redirect to other HTTP resource" https://www.pivotaltracker.com/story/show/81770804
* travis-ci joins19:31
fcrepo4/fcrepo4#3092 (external-body - 0e1a10e : Chris Beer): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/6b737657273c...0e1a10e99eb1
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39550011
* travis-ci leaves
<pivotal-bot>Andrew Woods added comment: "Agreed." https://www.pivotaltracker.com/story/show/78847348
Andrew Woods rejected "Adding datastream without file in HTML UI causes ugly error" https://www.pivotaltracker.com/story/show/7884734819:32
Longshou Situ started "Adding datastream without file in HTML UI causes ugly error" https://www.pivotaltracker.com/story/show/7884734819:40
Longshou Situ added comment: "PR https://github.com/fcrepo4-labs/fcrepo4-release-tests/pull/3, commit https://github.com/lsitu/fcrepo4-rel�" https://www.pivotaltracker.com/story/show/8164041019:41
Longshou Situ finished "Enable random ports on fcrepo4-release-tests" https://www.pivotaltracker.com/story/show/81640410
Longshou Situ started "Put Root Node properties in a jquery accordion" https://www.pivotaltracker.com/story/show/8178664019:42
<f4jenkins>Project fcrepo4 build #2536: ABORTED in 1 hr 26 min: http://jenkins.fcrepo.org/job/fcrepo4/2536/19:43
Chris Beer: Use IANA describes/describedby predicates to link a non-rdf source and its description
<pivotal-bot>Esme Cowles started "Attempting to create a node that already exists should return 409 Conflict, not 415 Unsupported Media Type" https://www.pivotaltracker.com/story/show/8179030419:49
* github-ff joins
[fcrepo4] escowles force-pushed repeated-put-conflict from c335ba2 to c90db60: http://git.io/4qkesQ
fcrepo4/repeated-put-conflict c90db60 Esmé Cowles: Updating logic to successfully detect empty request body
* github-ff leaves
<pivotal-bot>Esme Cowles added comment: "https://github.com/fcrepo4/fcrepo4/pull/609 has been rebased and force-pushed." https://www.pivotaltracker.com/story/show/81790304
Esme Cowles finished "Attempting to create a node that already exists should return 409 Conflict, not 415 Unsupported Media Type" https://www.pivotaltracker.com/story/show/81790304
* travis-ci joins19:59
fcrepo4/fcrepo4#3093 (repeated-put-conflict - c90db60 : Esmé Cowles): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/c335ba20d91b...c90db60fb955
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/39551657
* travis-ci leaves
* jcoyne joins20:15
* ruebot leaves20:33
* awead joins20:59
<pivotal-bot>Longshou Situ added comment: "PR https://github.com/fcrepo4/fcrepo4/pull/613 again, commit21:20
https://github.com/lsitu/fcrepo4/commit/69bd82�" https://www.pivotaltracker.com/story/show/78847348
Longshou Situ finished "Adding datastream without file in HTML UI causes ugly error" https://www.pivotaltracker.com/story/show/78847348
* longshou leaves21:21
* ruebot joins21:32
* awead leaves21:38
* awead joins21:40
* ksclarke joins21:43
* ksclarke leaves
* ksclarke joins
<pivotal-bot>Andrew Woods added comment: "When I run: mvn clean install, I get the following error: ""22:35
Failed tests:
SparqlRecipesIT.testSparqlQueries1�" https://www.pivotaltracker.com/story/show/81640410
Andrew Woods rejected "Enable random ports on fcrepo4-release-tests" https://www.pivotaltracker.com/story/show/81640410
<f4jenkins>Project fcrepo4 build #2538: FAILURE in 39 sec: http://jenkins.fcrepo.org/job/fcrepo4/2538/22:36
* jcoyne leaves22:45
* jcoyne joins22:54
* awead leaves23:16
* awead joins23:20
* scossu joins23:22

Generated by Sualtam