[API-users] Why is the downloads interface different than other occurrence endpoints?

Markus Döring mdoering at gbif.org
Mon Sep 29 21:36:48 CEST 2014


Hi Scott,
the download filter basics are explained here:
http://www.gbif.org/developer/occurrence#predicates

We seem to miss a link to the full Java docs for the Predicates which can be found here:
http://gbif.github.io/gbif-api/apidocs/org/gbif/api/model/occurrence/predicate/package-summary.html

As you can see there expressions make use of the OccurrenceSearchParameter:
http://gbif.github.io/gbif-api/apidocs/org/gbif/api/model/occurrence/search/OccurrenceSearchParameter.html

This is exactly the same enumeration as is used for the search API. The only difference is that the search API is case insensitive and allows for camelCase while the predicates must use the exact upper case spelling of the enumeration values.
Hope this helps a little.

Markus



On 29 Sep 2014, at 18:44, Scott Chamberlain <myrmecocystus at gmail.com> wrote:

> Hi Markus, 
> 
> Thanks for the explanation!  I asked Tim on twitter, but in case he misses the tweet: Is there a page that lists variable names to use for the occurrence/download/ endpoint? They seem to differ from occurrence/search endpoint. e.g. basisOfRecord in occurrence/search is BASIS_OF_RECORD in occurrence/download. I found this page https://github.com/gbif/occurrence/blob/master/occurrence-ws/src/test/java/org/gbif/occurrence/download/service/HiveQueryVisitorTest.java with tests that lists some variable names. 
> 
> Cheers, Scott
> 
> On Mon, Sep 29, 2014 at 9:34 AM, Markus Döring <mdoering at gbif.org> wrote:
> Hi Scott,
> 
> the download interface enables more complex queries with any kind of logical nesting similar to SQL clauses. Currently we do not actually make use of the possible complexity of download filters, but the initial design considered it.
> Personally I agree it would be nicer if the searhc and download interface would use the same filter definitions. We actually recognized this over a year ago but never had resources to address the discrepancy. Feel free to add comments:
> http://dev.gbif.org/issues/browse/POR-173
> 
> 
> Markus
> 
> 
> 
> 
> On 29 Sep 2014, at 18:11, Scott Chamberlain <myrmecocystus at gmail.com> wrote:
> 
>> Hi!
>> 
>> I'm curious why the occurrence/download/ endpoint has a different interface than the other occurrence endpoints (occurrence/search).  I ask because I work on the R client rgbif, and it would be nice to provide the same exact interface to users whether they are getting JSON data or spinning up a download.  However, the query interfaces are quite different.  This isn't a big deal, as I can make both occurrence/download and occurrence/search endpoints  as similar as possible for users despite them being different internally. 
>> 
>> Thanks! 
>> Scott Chamberlain
>> _______________________________________________
>> API-users mailing list
>> API-users at lists.gbif.org
>> http://lists.gbif.org/mailman/listinfo/api-users
> 
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.gbif.org/pipermail/api-users/attachments/20140929/e42c0e68/attachment.html 


More information about the API-users mailing list