enrico.ottonello
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

integrating suggestion from #9699 to improve the result_country table construction

This commit is contained in:
Claudio Atzori 2024-04-18 11:23:43 +02:00
parent 43e123c624
commit 18fdaaf548
1 changed files with 12 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

19
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql
View File

@ -1000,13 +1000,18 @@ left outer join (
drop table if exists ${stats_db_name}.result_country purge; /*EOS*/
create table ${stats_db_name}.result_country stored as parquet as
select distinct ro.id, coalesce(o.country, f.country) as country
from ${stats_db_name}.result_organization ro
left outer join ${stats_db_name}.organization o on o.id=ro.organization
left outer join ${stats_db_name}.result_projects rp on rp.id=ro.id
left outer join ${stats_db_name}.project p on p.id=rp.project
left outer join ${stats_db_name}.funder f on f.name=p.funder
where coalesce(o.country, f.country) IS NOT NULL; /*EOS*/
select distinct *
from (
select ro.id, o.country
from ${stats_db_name}.result_organization ro
left outer join ${stats_db_name}.organization o on o.id=ro.organization
union all
select rp.id, f.country
from ${stats_db_name}.result_projects
left outer join ${stats_db_name}.project p on p.id=rp.project
left outer join ${stats_db_name}.funder f on f.name=p.funder
) rc
where rc.country is not null; /*EOS*/
drop table if exists ${stats_db_name}.indi_result_oa_with_license purge; /*EOS*/
create table ${stats_db_name}.indi_result_oa_with_license stored as parquet as