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

Using timezone: Eastern Standard Time
* ksclarke leaves00:52
* pivotal-bot_____ leaves01:01
* pivotal-bot_____ joins
* kaarefc joins01:18
* kaarefc leaves01:24
* Matthias_ leaves03:07
* kaarefc joins06:40
* nbanks joins07:07
* dunix joins08:47
* ajs6f joins
<pivotal-bot_____>A. "Falubrious" Soroka started "Document runtime wiring options for F4 modules" https://www.pivotaltracker.com/story/show/5689494608:54
* mikeAtUVa joins09:05
* ajs6f leaves09:11
* ajs6f joins09:13
* kaarefc leaves
* awoods joins09:29
* kaarefc joins09:32
* kaarefc leaves09:37
* kaarefc joins09:39
* kaarefc leaves09:42
* kaarefc joins
* ksclarke joins09:43
<pivotal-bot_____>Andrew Woods edited "Document runtime wiring options for F4 modules" https://www.pivotaltracker.com/story/show/5689494610:13
Andrew Woods added "Create follow-on Solr tickets" https://www.pivotaltracker.com/story/show/5834133410:16
Andrew Woods edited "Create follow-on Solr tickets" https://www.pivotaltracker.com/story/show/58341334
* kaarefc leaves10:18
<pivotal-bot_____>Andrew Woods edited "Search" https://www.pivotaltracker.com/story/show/5715317610:19
* gregjansen joins10:31
* gregjansen leaves
* github-ff joins10:49
[fcrepo-jms-indexer-pluggable] yecao opened pull request #5: Merge mpdl (awoods...merge_mpdl) http://git.io/lIPsEQ
* github-ff leaves
* barmintor_ joins10:52
* osmandin joins10:54
<ajs6f>Sprint planning metting in Google Hangout, folks.11:01
<pivotal-bot_____>Nigel Banks edited "Add Mulgara support to JMS Indexer" https://www.pivotaltracker.com/story/show/5819379011:18
* gregjansen joins11:21
<ajs6f>2013-10-07 10:42:34.305 java[1546:60f] *** NSInvocation: warning: object 0x110c15390 of class 'ThreadUtilities' does not implement methodSignatureForSelector: -- trouble ahead
2013-10-07 10:42:34.306 java[1546:60f] *** NSInvocation: warning: object 0x110c15390 of class 'ThreadUtilities' does not implement doesNotRecognizeSelector: -- abort
Trace/BPT trap
* nbanks leaves11:32
* nbanks joins
* gregjansen leaves11:42
<nbanks>I got these for failed tests on the HTTP-API on the MAC12:00
FedoraNodesIT.testDescribeSize:588 No increment in size occurred when we expected one! FedoraNodesIT.testGetDatastream:225 expected:<404> but was:<200> FedoraNodesIT.testIngest:92 Got wrong Location header for ingest! expected:<...7/FedoraObjectsTest1[]> but was:<...7/FedoraObjectsTest1[/91fa595a-b700-4234-b953-6458ba4eacfc]>
FedoraNodesIT.testngestWithSlug:178 Didn't get a CREATED response! Got coGntent:
is an existing resource! expected:<201> but was:<409> FedoraTransactionsIT.testCreateDoStuffAndCommitTransaction:190 Expected to not find our object within the scope of the transaction expected:<404> but was:<200> FedoraWorkspacesIT.shouldDemonstratePathsAndWorkspaces:66 expected:<201> but was:<500>
I'm owrking on FedoraNodesIT.testGetDatastream now.12:01
<ajs6f>I'm getting more or less the asme failures, when I get failures. (And not arcane low-level errors.)
Okay, I fixed FedoraNodesIT.testGetDatastream, but I'm not happy, because I don't see how it ever worked.12:03
BUt whatever.
Here's a thought.
In itests, instead of using string constants like "Object3" and "ds2", let's use UUIDs. Then we won't have collisions.12:04
Here's a patch: https://gist.github.com/ajs6f/687039912:05
<nbanks>UUID's seems reasonable12:06
<ajs6f>It's exactly the semantics they support. "Don't collide things."
<pivotal-bot_____>Nigel Banks added comment: "https://wiki.duraspace.org/display/FF/Design+-+Indexing" https://www.pivotaltracker.com/story/show/5819379012:10
Nigel Banks added comment: "https://github.com/futures/fcrepo-jms-indexer-pluggable" https://www.pivotaltracker.com/story/show/5819379012:11
<pivotal-bot_____>Andrew Woods edited "Refactor business logic from HTTP-API" https://www.pivotaltracker.com/story/show/5706237012:17
A. "Falubrious" Soroka edited "Refactor business logic from HTTP-API" https://www.pivotaltracker.com/story/show/5706237012:18
Ye Cao added comment: "please check https://github.com/futures/fcrepo-jms-indexer-pluggable/pull/5" https://www.pivotaltracker.com/story/show/57453236
A. "Falubrious" Soroka added "Make build work on Macs" https://www.pivotaltracker.com/story/show/5835435212:24
A. "Falubrious" Soroka started "Make build work on Macs" https://www.pivotaltracker.com/story/show/58354352
Nigel Banks added "Flesh out what needs to be done to complete the rebasing of the Islandora Fedora 4 on top the most recent Islandora c..." https://www.pivotaltracker.com/story/show/5835485212:28
Nigel Banks edited "Flesh out what needs to be done to complete the rebasing of the Islandora Fedora 4 on top the most recent Islandora code." https://www.pivotaltracker.com/story/show/58354852
<nbanks>brb gotta get my charger
<pivotal-bot_____>Osman Din deleted "Update interface org.fcrepo.kernel.services.policy.StoragePolicyDecisionPoint to use appropriate data structure to ho..." https://www.pivotaltracker.com/story/show/5529665012:33
* ajs6f leaves12:55
* nbanks leaves12:56
* ajs6f joins12:58
* ajs6f leaves12:59
* ajs6f joins
<awoods>ajs6f: hi
<ajs6f>Getting confused about the semantics of our APIā€¦ maybe they've changed. Getting
FedoraNodesIT.testIngest:92 Got wrong Location header for ingest! expected:<...2/FedoraObjectsTest1[]> but was:<...2/FedoraObjectsTest1[/81639cfa-a9ff-4dec-a947-2e014e5546dc]>13:00
Can you look at that test?
<awoods>sure... after my current call.
<ajs6f>If I'm not wrong, the result (according to the API we advertise)
_shopuld_ be /FedoraObjectsTest1.
Not sure where the UUID is arising.
ITM, I'll move on to another broken test13:01
All: Can anyone tell me what FedoraNodesIT.testIngestWithSlug() is supposed to test? What is a slug, in this context?13:03
<cbeer>ajs6f: https://github.com/futures/fcrepo4/blob/master/fcrepo-http-api/src/test/java/org/fcrepo/integration/http/api/FedoraNodesIT.java#L171?13:04
Is it some HTTP thing?13:05
<cbeer>ajs6f: Slug is an HTTP header (defined.. somewhere. maybe the atom spec?.. and used by LDP)
<ajs6f>Oh, coolness.
<cbeer>and it's used as a hint to the thing-that-creates-objects about where you want to put it
<ajs6f>Okay. Thanks, cbeer! It's giving me a 409 and now I'm going to go find out why.
<cbeer>that i can't help with.13:06
<ajs6f>I must ride alone, with only my trusty debugger to keep me company. And a teddy bear.
* fcrepo-bot joins13:08
* gregjansen joins13:10
<ajs6f>Urg. It was another ID conflict. I'm about _this_ (holds fingers really very close) far from writing a ticket to replace all of the "FedoraObjectsTest3" constants in ITs with UUID strings.13:12
* nbanks joins13:16
<ajs6f>Looks like almost all of these Mac test failures are due to the same dang thing. Oh well, at last it's an easy fix. Maybe the Macs, for unknown reasons, are keeping files around longer and the test contexts aren't as fresh as they used to be.13:27
* nbanks leaves13:37
* github-ff joins13:40
[fcrepo4] ajs6f created MakeMacWork (+1 new commit): http://git.io/8_7zbA
fcrepo4/MakeMacWork 1b08474 ajs6f: Fixed fcrepo-http-api integration tests for Mac builds by using UUIDs instead of fixed names for test resources
* github-ff leaves
<ajs6f>Moving onto other failing tests tomorrow...
* ajs6f leaves
<pivotal-bot_____>Andrew Woods added comment: "Initial fixes: http://git.io/8_7zbA" https://www.pivotaltracker.com/story/show/5835435214:09
<awoods>ajs6f: Is it correct that there are more Mac build errors that are not addressed by your new branch (MakeMacWork)?14:10
* fcrepo-bot leaves14:11
<pivotal-bot_____>Scott Prater added "Investigate CDI framework" https://www.pivotaltracker.com/story/show/5836703414:37
Scott Prater started "Investigate CDI framework" https://www.pivotaltracker.com/story/show/5836703414:40
* gregjansen leaves14:57
* github-ff joins15:42
[fcrepo-jms-indexer-pluggable] yecao closed pull request #5: Merge mpdl (awoods...merge_mpdl) http://git.io/lIPsEQ
* github-ff leaves
<pivotal-bot_____>Mike Durbin edited "Investigate CDI framework" https://www.pivotaltracker.com/story/show/5836703415:59
* mikeAtUVa didn't... he just *looked* at it.16:01
* jonathangee leaves16:09
* github-ff joins16:25
[fcrepo4] ajs6f pushed 1 new commit to MakeMacWork: http://git.io/-9GlvA
fcrepo4/MakeMacWork 5f1f093 ajs6f: Fixed fcrepo-generator-dc integration tests for Mac builds by using UUIDs instead of fixed names for test resources
* github-ff leaves
* github-ff joins16:39
[fcrepo4] ajs6f opened pull request #115: Make mac work (master...MakeMacWork) http://git.io/83AYMw
* github-ff leaves
* osmandin leaves16:46
* dunix leaves16:56
* jonathangee joins17:17
* nbanks joins17:21
* jongibson_ joins17:27
* jongibson leaves17:34
* pivotal-bot_____ leaves
* mikeAtUVa leaves
* pivotal-bot_____ joins
* mikeAtUVa joins17:35
* ksclarke leaves18:20
* ksclarke joins20:31
* nbanks leaves21:55