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

Using timezone: Eastern Standard Time
* dwilcox joins02:14
* dwilcox leaves07:21
* dwilcox joins07:40
* dhlamb joins08:20
* peichman joins08:56
* peichman leaves08:57
* dwilcox leaves08:59
* whikloj joins09:01
* benpennell joins09:09
* bseeger joins09:18
* yamil_ joins09:23
* harringj joins10:02
* osmandin joins10:40
* dbernstein joins10:46
* github-ff joins10:52
[fcrepo-specification] awoods closed pull request #221: Fix typo: add missing hyphen (master...missing-hyphen) https://git.io/v5HZy
* github-ff leaves
* harringj leaves11:31
* bseeger leaves12:02
* bseeger joins12:09
* peichman joins13:04
* osmandin leaves13:14
* peichman leaves13:29
* bseeger leaves14:14
* bseeger joins15:20
* github-ff joins15:36
[fcrepo-specification] awoods pushed 1 new commit to master: https://git.io/vdIOY
fcrepo-specification/master a64f994 Esmé Cowles: Adding note about using Cache-Control: no-cache with fixity checks (#227)...
* github-ff leaves
<benpennell>is there a way to make it so that the import-export tool will re-export a resource if it has been modified since it was previously exported?15:50
or intention to add that
<bseeger>I think there is a way to start an export at a certain resource in the repo (as in it will export a tree starting from that resource), but I don't think it supports single resource export. I could very well be wrong about that.15:52
<benpennell>bseeger: no, i think you are right, but i gives me a message like "(Exporter) Already exported http://localhost:8080/rest/a/b" rather than exporting b, which has had a triple added to it15:53
also, if i export /a/b after adding c to it, it won't export c unless i directly point the export at /a/b/c15:54
although it is cool that it will add c to the existing export at the right path when directly targeted15:55
<bseeger>does it update /a/b as well in that scenario?
adding /a/b/c to it?15:56
<benpennell>no, /a/b is still missing its triples and the new contains statement
<bseeger>oh, that's an interesting test then - wonder what import would do in that scenario
<benpennell>i think it trims all the ldp:contains statements out of the import before submitting it since they are server managed15:57
i would probably lose the dc:title i added to it though
<bseeger>so then it goes by the file system directory structure? (actually, I haven't followed the bags stuff - it probably goes via some file in there maybe)15:58
<benpennell>yeah i think so, it just relies on fcrepo to generate those ldp:contains again when you import the children15:59
* peichman joins
<benpennell>might do something similar with other containment relations, can't remember for sure though
at the moment i'm trying to work out options for ongoing backs up of the structure of our repo, and was trying to figure out if the import-export tool was a reasonable way to do that16:01
seems like some deleting of old copies before exporting would be in order currently
i do wonder if anyone else is doing this, or if there's an expectation that normal backups of the fcrepo postgres/mysql database are sufficient16:02
oh, bseeger it looks like you can export single objects if you provide an empty --predicates parameter, since it won't follow ldp:contains anymore16:06
<bseeger>benpennell - thanks, that's good to know.16:10
* github-ff joins16:24
[fcrepo4] dbernstein opened pull request #1243: Sets proper actor values for emitted messages. (master...fcrepo-2603) https://git.io/vdIcW
* github-ff leaves
* yamil_ leaves16:30
* peichman leaves16:56
* benpennell leaves17:26
* bseeger leaves
* dwilcox joins17:55
* whikloj leaves17:59
* dhlamb leaves18:03
* peichman joins18:19
* peichman leaves18:25
* dwilcox leaves18:30
* dwilcox joins18:49
* dwilcox leaves19:41
* dwilcox joins
* dwilcox leaves19:51
* awoods leaves23:42