Issue 17408

Country page publishing summary chart has bad BoR Observation values

17408
Reporter: omeyn
Assignee: bko
Type: Bug
Summary: Country page publishing summary chart has bad BoR Observation values
Priority: Major
Resolution: Fixed
Status: Resolved
Created: 2015-03-10 12:07:23.363
Updated: 2015-03-17 10:52:30.103
Resolved: 2015-03-17 10:52:30.078
        
Description: The chart generation script appears to be asking the cube for Basis of Record Observation and gets that back, but that does not include HUMAN_OBSERVATION or MACHINE_OBSERVATION, which it should. Suggest the query sums these 3 as 3 calls and writes that value in the chart.

http://www.gbif.org/country/LU/publishing

http://api.gbif.org/v1/occurrence/count?country=LU&basisOfRecord=HUMAN_OBSERVATION

http://api.gbif.org/v1/occurrence/count?country=LU&basisOfRecord=OBSERVATION]]>
    


Author: mdoering@gbif.org
Created: 2015-03-10 12:53:27.211
Updated: 2015-03-10 12:53:52.197
        
I am not convinced this is very helpful if it only gets fixed in those tables. Links to occurrence searches will show other counts then and the metrics are likely to pop up elsewhere too. Dataset stats pages (struts), both types checklist (with occ) and occurrences have also that table.

To do it properly we should consider the BoR vocabulary hierarchical and do the magic inside the metrics cube
    


Author: mdoering@gbif.org
Created: 2015-03-10 13:37:10.609
Updated: 2015-03-10 13:37:10.609
        
There are also pie charts of BoR breakdowns next to struts2 metrics tables:
http://www.gbif.org/dataset/93814afc-6a43-435c-bc18-93ed1f2192a8/stats
http://www.gbif.org/dataset/bb6113e0-9d52-4828-b9a8-4a37c2d6e95d/stats
    


Author: bko@gbif.org
Created: 2015-03-12 15:50:07.502
Updated: 2015-03-12 15:50:07.502
        
Observation count now includes human and machine ones.
https://github.com/gbif/gbif-drupal/blob/master/sites/all/modules/custom/gbifs/gbif_participant/js/loadMetrics.js

See also: http://www.gbif.org/country/LU/about

    


Author: bko@gbif.org
Comment: It seems the total count is not exact adding-up of all subcategories. There is usually just small difference. Guess it's because there are a few records not indexed in the shown categories?
Created: 2015-03-12 15:53:42.589
Updated: 2015-03-12 15:53:42.589


Author: bko@gbif.org
Created: 2015-03-17 10:52:30.101
Updated: 2015-03-17 10:52:30.101
        
The observation count now includes human and machine observation on the JavaWeb, too.

https://github.com/gbif/portal-web/commit/1edf031f85c05809dfb1972686862f0177c9b16f