D-Net
/
dnet-hadoop
14
5
Fork 11
Code Issues 14 Pull Requests 13 Projects 2 Releases 2 Wiki Activity

revised query implementation for indi_pub_gold_oa

This commit is contained in:
Claudio Atzori 2024-04-17 15:06:50 +02:00
parent 0486cea4c4
commit 96bddcc921
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql
View File

@ -237,16 +237,16 @@ create table if not exists ${stats_db_name}.indi_pub_gold_oa stored as parquet a
UNION ALL UNION ALL
select id, issn_online as issn from ${stats_db_name}.datasource d where d.id like '%doajarticles%' select id, issn_online as issn from ${stats_db_name}.datasource d where d.id like '%doajarticles%'
UNION ALL UNION ALL
select id, issn_printed as issn from ${stats_db_name}.datasource d join gold_oa on gold_oa.issn=d.issn_printed select id, issn_printed as issn from ${stats_db_name}.datasource d left semi join gold_oa on gold_oa.issn=d.issn_printed
UNION ALL UNION ALL
select id, issn_online as issn from ${stats_db_name}.datasource d join gold_oa on gold_oa.issn=d.issn_online) foo select id, issn_online as issn from ${stats_db_name}.datasource d left semi join gold_oa on gold_oa.issn=d.issn_online) foo
) )
SELECT DISTINCT pd.id, coalesce(is_gold, 0) as is_gold SELECT DISTINCT pd.id, coalesce(is_gold, 0) as is_gold
FROM ${stats_db_name}.publication_datasources pd FROM ${stats_db_name}.publication_datasources pd
left outer join ( left outer join (
select pd.id, 1 as is_gold select pd.id, 1 as is_gold
FROM ${stats_db_name}.publication_datasources pd FROM ${stats_db_name}.publication_datasources pd
join dd on dd.id=pd.datasource left semi join dd on dd.id=pd.datasource
left outer join ${stats_db_name}.result_accessroute ra on ra.id = pd.id where ra.accessroute = 'gold') tmp on tmp.id=pd.id; /*EOS*/ left outer join ${stats_db_name}.result_accessroute ra on ra.id = pd.id where ra.accessroute = 'gold') tmp on tmp.id=pd.id; /*EOS*/
drop table if exists ${stats_db_name}.indi_pub_hybrid_oa_with_cc purge; /*EOS*/ drop table if exists ${stats_db_name}.indi_pub_hybrid_oa_with_cc purge; /*EOS*/