Merge pull request 'Added fields: totalcost, fundedamount, currency, in project table' (#257) from antonis.lempesis/dnet-hadoop:beta into beta

Reviewed-on: #257
This commit is contained in:
Claudio Atzori 2022-10-31 13:49:27 +01:00
commit 22873c9172
4 changed files with 14 additions and 8 deletions

View File

@ -21,7 +21,7 @@
</property>
<property>
<name>hive_jdbc_url</name>
<value>jdbc:hive2://iis-cdh5-test-m3.ocean.icm.edu.pl:10000</value>
<value>jdbc:hive2://iis-cdh5-test-m3.ocean.icm.edu.pl:10000/;UseNativeQuery=1;?spark.executor.memory=19166291558;spark.yarn.executor.memoryOverhead=3225;spark.driver.memory=11596411699;spark.yarn.driver.memoryOverhead=1228</value>
</property>
<property>
<name>oozie.wf.workflow.notification.url</name>

View File

@ -42,7 +42,9 @@ SELECT p.id,
CASE WHEN prr2.id IS NULL THEN 0 ELSE prr2.dp END AS delayedpubs,
p.callidentifier,
p.code,
p.totalcost
p.totalcost,
p.fundedamount,
p.currency
FROM ${stats_db_name}.project_tmp p
LEFT JOIN (SELECT pr.id, count(distinct pr.result) AS np
FROM ${stats_db_name}.project_results pr

View File

@ -59,7 +59,7 @@ UNION ALL
SELECT * FROM ${stats_db_name}.otherresearchproduct_sources;
create table ${stats_db_name}.result_orcid STORED AS PARQUET as
CREATE TABLE IF NOT EXISTS ${stats_db_name}.result_orcid STORED AS PARQUET as
select distinct res.id, regexp_replace(res.orcid, 'http://orcid.org/' ,'') as orcid
from (
SELECT substr(res.id, 4) as id, auth_pid.value as orcid
@ -69,7 +69,7 @@ from (
LATERAL VIEW explode(auth.pid.qualifier.classid) apt as author_pid_type
WHERE res.datainfo.deletedbyinference = FALSE and res.datainfo.invisible = FALSE and author_pid_type = 'orcid') as res;
create table ${stats_db_name}.result_result stored as parquet as
CREATE TABLE IF NOT EXISTS ${stats_db_name}.result_result stored as parquet as
select substr(rel.source, 4) as source, substr(rel.target, 4) as target, relclass, subreltype
from ${openaire_db_name}.relation rel
join ${openaire_db_name}.result r1 on rel.source=r1.id
@ -82,7 +82,7 @@ where reltype='resultResult'
and r2.resulttype.classname != 'other'
and rel.datainfo.deletedbyinference=false and rel.datainfo.invisible = FALSE;
create table ${stats_db_name}.result_citations_oc stored as parquet as
CREATE TABLE IF NOT EXISTS ${stats_db_name}.result_citations_oc stored as parquet as
select substr(target, 4) as id, count(distinct substr(source, 4)) as citations
from ${openaire_db_name}.relation rel
join ${openaire_db_name}.result r1 on rel.source=r1.id
@ -97,7 +97,7 @@ where relClass='Cites' and rel.datainfo.provenanceaction.classid = 'sysimport:cr
and rel.datainfo.deletedbyinference=false and rel.datainfo.invisible = FALSE
group by substr(target, 4);
create table ${stats_db_name}.result_references_oc stored as parquet as
CREATE TABLE IF NOT EXISTS ${stats_db_name}.result_references_oc stored as parquet as
select substr(source, 4) as id, count(distinct substr(target, 4)) as references
from ${openaire_db_name}.relation rel
join ${openaire_db_name}.result r1 on rel.source=r1.id

View File

@ -48,7 +48,9 @@ CREATE TABLE ${stats_db_name}.project_tmp
delayedpubs INT,
callidentifier STRING,
code STRING,
totalcost FLOAT
totalcost FLOAT,
fundedamount FLOAT,
currency STRING
) CLUSTERED BY (id) INTO 100 buckets stored AS orc tblproperties ('transactional' = 'true');
INSERT INTO ${stats_db_name}.project_tmp
@ -72,7 +74,9 @@ SELECT substr(p.id, 4) AS id,
0 AS delayedpubs,
p.callidentifier.value AS callidentifier,
p.code.value AS code,
p.totalcost AS totalcost
p.totalcost AS totalcost,
p.fundedamount AS fundedamount,
p.currency.value AS currency
FROM ${openaire_db_name}.project p
WHERE p.datainfo.deletedbyinference = false and p.datainfo.invisible=false;