15011
Reporter: mdoering
Assignee: mdoering
Type: Bug
Summary: dc:modified not interpreted
Priority: Major
Resolution: CantReproduce
Status: Resolved
Created: 2014-02-07 15:27:02.248
Updated: 2014-02-13 12:56:31.517
Resolved: 2014-02-12 17:20:38.693
Description: http://appdev.gbif.org/occurrence/30945/verbatim has dc:modified
but not in the interpreted version: http://appdev.gbif.org/occurrence/30945]]>
Author: kbraak@gbif.org
Created: 2014-02-07 16:55:16.674
Updated: 2014-02-07 16:55:16.674
Since Occurrence.modified is supposed to interpret dc:modified, something must be wrong with the interpretation. The verbatim term is obviously being stored properly, since it's showing on the verbatim page.
Occurrence model object, see:
https://github.com/gbif/gbif-api/blob/master/src/main/java/org/gbif/api/model/occurrence/Occurrence.java
Author: mdoering@gbif.org
Created: 2014-02-12 17:20:38.715
Updated: 2014-02-12 17:20:38.715
The code is looking good and there are numerous examples that do have dc:modified interpreted correctly. For example cat number AVES-105348
http://appdev.gbif.org/occurrence/695
"It was last updated according to the publisher on 27.02.2011."
From verbatim
As we wipe and recrawl occurrence ids change all the time and its impossible to find the exact same record again without any kind if stable property. After hours of searching for bad records I give up.
Dataset pages also do not show any invalid dc_modified:
http://appdev.gbif.org/dataset/b8ed9b7c-5b73-4e6d-a6e1-b0911a09f947/stats
http://appdev.gbif.org/dataset/af3bce08-0599-45a6-9bfc-08188bcd868e/stats
http://appdev.gbif.org/dataset/2f46fe6f-b9fd-4dba-8ca2-2cb1368ceed8/stats
Author: mdoering@gbif.org
Comment: A bug in persistence cause the date to be wrongly handled
Created: 2014-02-13 12:56:31.517
Updated: 2014-02-13 12:56:31.517