Hi Santiago
If you replace the file:
/data/collectory/bootstrap/meta.xml
with this one:
https://github.com/gbif/ala-install/blob/master/ansible/roles/collectory/fil...
this should fix the problem. This file should match the content of the DWCA meta.xml downloaded from GBIF. We need to remove the use of this file (as the format of the GBIF archives may change in future releases of their API), but updating it is a quick fix for this current problem.
Cheers
Dave
________________________________________ From: Santiago Martinez de la Riva [sama@gbif.es] Sent: 25 June 2014 23:10 To: Martin, Dave (CES, Black Mountain) Cc: ala-portal@lists.gbif.org Subject: Ingest problem :(
Hi Dave,
I'm checking out the ansible script, and I have a problem with the field mapping in the indexing process.
The question is that when I upload this resource http://api.gbif.org/v0.9/occurrence/download/request/0000874-140616093749225... ( http://www.gbif.org/dataset/835177da-f762-11e1-a439-00145eb45e9a) and I try to ingest the records. The mapping that the process applies for the indexing, it's wrong. (Look attach image please)
It's like if one column is moved one position left. The begining of this problem appears in the column:
<field index="54" term="http://purl.org/dc/terms/type"/>
The next columns are moved one position left. It's weird.
For this reason, you can see that the field "Associated taxa" is "Basis of record", and the same with the rest of the fields
Could be a error in the internal map? In the meta.xml the order of the fields is right. I was thinking that it could be a error in the internal process of indexing
Do you know what could be the problem?? Does anyone have the same problem as me??
Thanks so much. Kind Regards. --------------------------------------------------------------------------------------- Santiago Martínez de la Riva GBIF.ES, Unidad de Coordinación Tel. +34 91 4203017 x 273 Real Jardín Botánico - CSIC Fax +34 91 429 2405 Plaza de Murillo, 2 sama@gbif.es 28014 Madrid, Spain www.gbif.es