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

Corrections by Spyros: Scipt cleanup, corrections and re-arrangement

This commit is contained in:
Spyros Zoupanos 2020-06-13 13:35:53 +03:00
parent 477fa6234d
commit 657a40536b
19 changed files with 350 additions and 447 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/config-default.xml
View File

@ -1,12 +1,11 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
<value>${jobTracker}</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
<value>${nameNode}</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
@ -16,16 +15,6 @@
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>hive_db_name</name>
<!-- <value>stats_wf_db_galexiou_oozie</value> -->
<value>stats_wf_db_galexiou_oozie_beta</value>
</property>
<property>
<name>hive_source_db_name</name>
<!-- <value>openaire_prod</value> -->
<value>openaire2</value>
</property>
<property>
<name>hive_metastore_uris</name>
<value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
@ -34,5 +23,8 @@
<name>hive_jdbc_url</name>
<value>jdbc:hive2://iis-cdh5-test-m3.ocean.icm.edu.pl:10000</value>
</property>
</configuration>
<property>
<name>oozie.wf.workflow.notification.url</name>
<value>{serviceUrl}/v1/oozieNotification/jobUpdate?jobId=$jobId%26status=$status</value>
</property>
</configuration>

12
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/impala-shell.sh
View File

@ -1,4 +1,4 @@
export PYTHON_EGG_CACHE=/home/giorgos.alexiou/.python-eggs
export PYTHON_EGG_CACHE=/home/$(whoami)/.python-eggs
export link_folder=/tmp/impala-shell-python-egg-cache-$(whoami)
if ! [ -L $link_folder ]
then
@ -6,11 +6,13 @@ then
ln -sfn ${PYTHON_EGG_CACHE}${link_folder} ${link_folder}
fi
echo "getting file from " $3
echo "Getting file from " $3
hdfs dfs -copyToLocal $3
echo "running impala shell now"
echo "Running impala shell make the new database visible"
impala-shell -q "INVALIDATE METADATA;"
echo "Running impala shell to compute new table stats"
impala-shell -d $1 -f $2
echo "impala shell finished"
echo "Impala shell finished"
rm $2

77
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/impala-stats.sql
View File

@ -1,77 +0,0 @@
INVALIDATE METADATA;
-- CREATE TABLE chart_country_year AS SELECT org.country AS country, r.year AS year, count(distinct r.id) AS publications FROM result r, result_datasources rd, datasource d, datasource_organizations dor, organization org WHERE r.id=rd.id AND rd.datasource=d.id AND rd.datasource=dor.id AND dor.organization=org.id AND r.type='publication' AND r.bestlicence='Open Access' and r.year>='1990' AND r.year<=CAST(date_part('year',now()) AS STRING) group by org.country, r.year ORDER BY org.country, r.year;
-- CREATE TABLE chart_country_datasources AS SELECT org.country AS country, d.name AS datasource, count(distinct r.id) AS publications FROM result r, result_datasources rd, datasource d, datasource_organizations dor, organization org WHERE r.id=rd.id AND rd.datasource=d.id AND d.id=dor.id AND dor.organization=org.id AND r.type='publication' AND r.bestlicence='Open Access' GROUP BY org.country, d.name ORDER BY org.country, publications DESC;
-- CREATE TABLE chart_country_type AS SELECT org.country AS country, rc.type AS type, count(distinct r.id) AS publications FROM result r, result_datasources rd, datasource d, result_classifications rc, datasource_organizations dor, organization org WHERE r.id=rd.id AND r.id=rc.id AND rd.datasource=d.id AND rd.datasource=dor.id AND dor.organization=org.id AND r.type='publication' AND r.bestlicence='Open Access' GROUP BY org.country, rc.type;
-- CREATE TABLE chart_country_fp7 AS SELECT org.country AS country, r.year AS year, count(distinct r.id) AS publications FROM result r, result_datasources rd, datasource d, datasource_organizations dor, organization org, project_results pr, project p WHERE r.id=rd.id AND rd.datasource=d.id AND rd.datasource=dor.id AND dor.organization=org.id AND r.id=pr.result AND pr.id=p.id AND p.funding_lvl0='FP7' AND r.type='publication' AND r.year>='1990' AND r.year<=CAST(date_part('year',now()) AS STRING) GROUP BY org.country, r.year ORDER BY org.country, r.year;
-- CREATE TABLE chart_datasource_type AS SELECT rd.datasource, rc.type, count(distinct rd.id) FROM result_datasources rd, result_classifications rc WHERE rd.id=rc.id GROUP BY rd.datasource, rc.type;
-- CREATE TABLE chart_datasource_year AS SELECT rd.datasource, r.year, count(distinct rd.id) FROM result r, result_datasources rd WHERE rd.id=r.id GROUP By rd.datasource, r.year;
-- CREATE TABLE chart_datasource_funders AS SELECT rd.datasource, p.funder, count(distinct rd.id) FROM result_datasources rd, project p, project_results pr WHERE p.id=pr.id AND pr.result=rd.id GROUP BY rd.datasource, p.funder;
-- CREATE TABLE chart_datasource_projects_pubs AS SELECT rd.datasource, p.title, count(distinct rd.id) FROM result_datasources rd, project p, project_results pr, result r WHERE p.id=pr.id AND pr.result=rd.id AND pr.result=r.id AND r.type='publication' GROUP BY rd.datasource, p.title;
-- CREATE TABLE chart_datasource_projects_data AS SELECT rd.datasource, p.title, count(distinct rd.id) FROM result_datasources rd, project p, project_results pr, result r WHERE p.id=pr.id AND pr.result=rd.id AND pr.result=r.id and r.type='dataset' GROUP BY rd.datasource, p.title;
-- CREATE TABLE chart_project_year AS SELECT p.id, r.year, count( distinct r.id) FROM result r, project_results pr, project p WHERE r.id=pr.result AND p.id=pr.id AND r.year>=p.start_year GROUP BY p.id, r.year;
-- CREATE TABLE chart_project_license AS SELECT pr.id, r.bestlicence, count(distinct r.id) FROM result r, project_results pr WHERE r.id=pr.result AND r.type='publication' GROUP BY pr.id, r.bestlicence;
-- CREATE TABLE chart_project_repos AS SELECT pr.id, d.name, count (distinct r.id) FROM result r, project_results pr, datasource d, datasource_results dr WHERE r.id=dr.result AND d.id=dr.id AND r.id=pr.result AND r.type='publication' GROUP BY pr.id, d.name;
-- CREATE TABLE rd_distinct AS SELECT DISTINCT * FROM result_datasources;
-- DROP VIEW result_datasources;
-- ALTER TABLE rd_distinct RENAME TO result_datasources;
-- COMPUTE STATS chart_country_datasources;
-- COMPUTE STATS chart_country_fp7;
-- COMPUTE STATS chart_country_type;
-- COMPUTE STATS chart_country_year;
-- COMPUTE STATS chart_datasource_funders;
-- COMPUTE STATS chart_datasource_projects_data;
-- COMPUTE STATS chart_datasource_projects_pubs;
-- COMPUTE STATS chart_datasource_type;
-- COMPUTE STATS chart_datasource_year;
-- COMPUTE STATS chart_project_license;
-- COMPUTE STATS chart_project_repos;
-- COMPUTE STATS chart_project_year;
COMPUTE STATS dataset;
COMPUTE STATS dataset_citations;
COMPUTE STATS dataset_classifications;
COMPUTE STATS dataset_concepts;
COMPUTE STATS dataset_datasources;
COMPUTE STATS dataset_languages;
COMPUTE STATS dataset_oids;
COMPUTE STATS dataset_pids;
COMPUTE STATS dataset_topics;
COMPUTE STATS datasource;
COMPUTE STATS datasource_languages;
COMPUTE STATS datasource_oids;
COMPUTE STATS datasource_organizations;
COMPUTE STATS numbers_country;
COMPUTE STATS organization;
COMPUTE STATS otherresearchproduct;
COMPUTE STATS otherresearchproduct_citations;
COMPUTE STATS otherresearchproduct_classifications;
COMPUTE STATS otherresearchproduct_concepts;
COMPUTE STATS otherresearchproduct_datasources;
COMPUTE STATS otherresearchproduct_languages;
COMPUTE STATS otherresearchproduct_oids;
COMPUTE STATS otherresearchproduct_pids;
COMPUTE STATS otherresearchproduct_topics;
COMPUTE STATS project;
COMPUTE STATS project_oids;
COMPUTE STATS project_organizations;
COMPUTE STATS project_results;
COMPUTE STATS publication;
COMPUTE STATS publication_citations;
COMPUTE STATS publication_classifications;
COMPUTE STATS publication_concepts;
COMPUTE STATS publication_datasources;
COMPUTE STATS publication_languages;
COMPUTE STATS publication_oids;
COMPUTE STATS publication_pids;
COMPUTE STATS publication_topics;
COMPUTE STATS result_organization;
COMPUTE STATS result_projects;
COMPUTE STATS software;
COMPUTE STATS software_citations;
COMPUTE STATS software_classifications;
COMPUTE STATS software_concepts;
COMPUTE STATS software_datasources;
COMPUTE STATS software_languages;
COMPUTE STATS software_oids;
COMPUTE STATS software_pids;
COMPUTE STATS software_topic;

6
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step1.sql
View File

@ -1,11 +1,11 @@
-- DROP database if EXISTS ${hive_db_name} cascade;
-- CREATE database ${hive_db_name};
--
--
-- CREATE TABLE ${hive_db_name}.Persons (
-- PersonID int,
-- LastName varchar(255));
--
-- INSERT INTO ${hive_db_name}.Persons VALUES (1, "test_db_spyros_rec_111");
drop database if exists ${hive_db_name} cascade;
create database ${hive_db_name};
DROP database IF EXISTS ${stats_db_name} CASCADE;
CREATE database ${stats_db_name};

9
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step10.sql
View File

@ -3,8 +3,9 @@
-- Organization table/view and Organization related tables/views
----------------------------------------------------------------
----------------------------------------------------------------
DROP TABLE IF EXISTS ${hive_db_name}.organization;
CREATE TABLE ${hive_db_name}.organization AS SELECT substr(o.id, 4) as id, o.legalname.value as name, o.country.classid as country from ${hive_source_db_name}.organization o WHERE o.datainfo.deletedbyinference=false;
DROP TABLE IF EXISTS ${stats_db_name}.organization;
CREATE TABLE ${stats_db_name}.organization AS SELECT substr(o.id, 4) as id, o.legalname.value AS name, o.country.classid AS country FROM ${openaire_db_name}.organization o WHERE o.datainfo.deletedbyinference=false;
CREATE OR REPLACE VIEW ${hive_db_name}.organization_datasources AS SELECT organization AS id, id AS datasource FROM ${hive_db_name}.datasource_organizations;
CREATE OR REPLACE VIEW ${hive_db_name}.organization_projects AS SELECT id AS project, organization as id FROM ${hive_db_name}.project_organizations;
CREATE OR REPLACE VIEW ${stats_db_name}.organization_datasources AS SELECT organization AS id, id AS datasource FROM ${stats_db_name}.datasource_organizations;
CREATE OR REPLACE VIEW ${stats_db_name}.organization_projects AS SELECT id AS project, organization as id FROM ${stats_db_name}.project_organizations;

10
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step11.sql
View File

@ -3,8 +3,8 @@
-- Tables/views from external tables/views (Fundref, Country, CountyGDP, roarmap, rndexpediture)
------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------
CREATE OR REPLACE VIEW ${hive_db_name}.fundref AS SELECT * FROM stats_ext.fundref;
CREATE OR REPLACE VIEW ${hive_db_name}.country AS SELECT * FROM stats_ext.country;
CREATE OR REPLACE VIEW ${hive_db_name}.countrygdp AS SELECT * FROM stats_ext.countrygdp;
CREATE OR REPLACE VIEW ${hive_db_name}.roarmap AS SELECT * FROM stats_ext.roarmap;
CREATE OR REPLACE VIEW ${hive_db_name}.rndexpediture AS SELECT * FROM stats_ext.rndexpediture;
CREATE OR REPLACE VIEW ${stats_db_name}.fundref AS SELECT * FROM ${external_stats_db_name}.fundref;
CREATE OR REPLACE VIEW ${stats_db_name}.country AS SELECT * FROM ${external_stats_db_name}.country;
CREATE OR REPLACE VIEW ${stats_db_name}.countrygdp AS SELECT * FROM ${external_stats_db_name}.countrygdp;
CREATE OR REPLACE VIEW ${stats_db_name}.roarmap AS SELECT * FROM ${external_stats_db_name}.roarmap;
CREATE OR REPLACE VIEW ${stats_db_name}.rndexpediture AS SELECT * FROM ${external_stats_db_name}.rndexpediture;

46
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step12.sql
View File

@ -3,42 +3,40 @@
----------------------------
--Datasource temporary table updates
UPDATE ${hive_db_name}.datasource_tmp set harvested ='true' WHERE datasource_tmp.id IN (SELECT DISTINCT d.id FROM ${hive_db_name}.datasource_tmp d, ${hive_db_name}.result_datasources rd where d.id=rd.datasource);
UPDATE ${stats_db_name}.datasource_tmp SET harvested='true' WHERE datasource_tmp.id IN (SELECT DISTINCT d.id FROM ${stats_db_name}.datasource_tmp d, ${stats_db_name}.result_datasources rd WHERE d.id=rd.datasource);
-- Project temporary table update and final project table creation with final updates that can not be applied to ORC tables
-- UPDATE ${hive_db_name}.project_tmp SET haspubs='yes' WHERE project_tmp.id IN (SELECT pr.id FROM ${hive_db_name}.project_results pr, ${hive_db_name}.result r WHERE pr.result=r.id AND r.type='publication');
UPDATE ${stats_db_name}.project_tmp SET haspubs='yes' WHERE project_tmp.id IN (SELECT pr.id FROM ${stats_db_name}.project_results pr, ${stats_db_name}.result r WHERE pr.result=r.id AND r.type='publication');
CREATE TABLE ${hive_db_name}.project stored as parquet as
SELECT p.id , p.acronym, p.title, p.funder, p.funding_lvl0, p.funding_lvl1, p.funding_lvl2, p.ec39, p.type, p.startdate, p.enddate, p.start_year, p.end_year, p.duration,
CASE WHEN prr1.id IS NULL THEN 'no' ELSE 'yes' END as haspubs,
CASE WHEN prr1.id IS NULL THEN 0 ELSE prr1.np END as numpubs,
CASE WHEN prr2.id IS NULL THEN 0 ELSE prr2.daysForlastPub END as daysforlastpub,
CASE WHEN prr2.id IS NULL THEN 0 ELSE prr2.dp END as delayedpubs,
DROP TABLE IF EXISTS ${stats_db_name}.project;
CREATE TABLE ${stats_db_name}.project stored as parquet as
SELECT p.id , p.acronym, p.title, p.funder, p.funding_lvl0, p.funding_lvl1, p.funding_lvl2, p.ec39, p.type, p.startdate, p.enddate, p.start_year, p.end_year, p.duration,
CASE WHEN prr1.id IS NULL THEN 'no' ELSE 'yes' END AS haspubs,
CASE WHEN prr1.id IS NULL THEN 0 ELSE prr1.np END AS numpubs,
CASE WHEN prr2.id IS NULL THEN 0 ELSE prr2.daysForlastPub END AS daysforlastpub,
CASE WHEN prr2.id IS NULL THEN 0 ELSE prr2.dp END AS delayedpubs,
p.callidentifier, p.code
FROM ${hive_db_name}.project_tmp p
FROM ${stats_db_name}.project_tmp p
LEFT JOIN (SELECT pr.id, count(distinct pr.result) AS np
FROM ${hive_db_name}.project_results pr INNER JOIN ${hive_db_name}.result r ON pr.result=r.id
WHERE r.type='publication'
FROM ${stats_db_name}.project_results pr INNER JOIN ${stats_db_name}.result r ON pr.result=r.id
WHERE r.type='publication'
GROUP BY pr.id) AS prr1 on prr1.id = p.id
LEFT JOIN (SELECT pp.id, max(datediff(to_date(r.date), to_date(pp.enddate)) ) as daysForlastPub , count(distinct r.id) as dp
FROM ${hive_db_name}.project_tmp pp, ${hive_db_name}.project_results pr, ${hive_db_name}.result r
WHERE pp.id=pr.id AND pr.result=r.id AND r.type='publication' AND datediff(to_date(r.date), to_date(pp.enddate)) > 0
LEFT JOIN (SELECT pp.id, max(datediff(to_date(r.date), to_date(pp.enddate)) ) AS daysForlastPub , count(distinct r.id) AS dp
FROM ${stats_db_name}.project_tmp pp, ${stats_db_name}.project_results pr, ${stats_db_name}.result r
WHERE pp.id=pr.id AND pr.result=r.id AND r.type='publication' AND datediff(to_date(r.date), to_date(pp.enddate)) > 0
GROUP BY pp.id) AS prr2
on prr2.id = p.id;
ON prr2.id = p.id;
-- Publication temporary table updates
UPDATE ${hive_db_name}.publication_tmp SET delayed = 'yes' WHERE publication_tmp.id IN (SELECT distinct r.id FROM ${hive_db_name}.result r, ${hive_db_name}.project_results pr, ${hive_db_name}.project_tmp p WHERE r.id=pr.result AND pr.id=p.id AND to_date(r.date)-to_date(p.enddate) > 0);
UPDATE ${stats_db_name}.publication_tmp SET delayed = 'yes' WHERE publication_tmp.id IN (SELECT distinct r.id FROM stats_wf_db_obs.result r, ${stats_db_name}.project_results pr, ${stats_db_name}.project_tmp p WHERE r.id=pr.result AND pr.id=p.id AND to_date(r.date)-to_date(p.enddate) > 0);
-- Dataset temporary table updates
UPDATE ${hive_db_name}.dataset_tmp SET delayed = 'yes' WHERE dataset_tmp.id IN (SELECT distinct r.id FROM ${hive_db_name}.result r, ${hive_db_name}.project_results pr, ${hive_db_name}.project_tmp p WHERE r.id=pr.result AND pr.id=p.id AND to_date(r.date)-to_date(p.enddate) > 0);
UPDATE ${stats_db_name}.dataset_tmp SET delayed = 'yes' WHERE dataset_tmp.id IN (SELECT distinct r.id FROM stats_wf_db_obs.result r, ${stats_db_name}.project_results pr, ${stats_db_name}.project_tmp p WHERE r.id=pr.result AND pr.id=p.id AND to_date(r.date)-to_date(p.enddate) > 0);
-- Software temporary table updates
UPDATE ${hive_db_name}.software_tmp SET delayed = 'yes' WHERE software_tmp.id IN (SELECT distinct r.id FROM ${hive_db_name}.result r, ${hive_db_name}.project_results pr, ${hive_db_name}.project_tmp p WHERE r.id=pr.result AND pr.id=p.id AND to_date(r.date)-to_date(p.enddate) > 0);
UPDATE ${stats_db_name}.software_tmp SET delayed = 'yes' WHERE software_tmp.id IN (SELECT distinct r.id FROM ${stats_db_name}.result r, ${stats_db_name}.project_results pr, ${stats_db_name}.project_tmp p WHERE r.id=pr.result AND pr.id=p.id AND to_date(r.date)-to_date(p.enddate) > 0);
-- Oherresearchproduct temporary table updates
UPDATE ${hive_db_name}.otherresearchproduct_tmp SET delayed = 'yes' WHERE otherresearchproduct_tmp.id IN (SELECT distinct r.id FROM ${hive_db_name}.result r, ${hive_db_name}.project_results pr, ${hive_db_name}.project_tmp p WHERE r.id=pr.result AND pr.id=p.id AND to_date(r.date)-to_date(p.enddate) > 0);
UPDATE ${stats_db_name}.otherresearchproduct_tmp SET delayed = 'yes' WHERE otherresearchproduct_tmp.id IN (SELECT distinct r.id FROM ${stats_db_name}.result r, ${stats_db_name}.project_results pr, ${stats_db_name}.project_tmp p WHERE r.id=pr.result AND pr.id=p.id AND to_date(r.date)-to_date(p.enddate) > 0);
CREATE OR REPLACE VIEW ${hive_db_name}.project_results_publication AS SELECT result_projects.id AS result, result_projects.project AS project_results, result.date as resultdate, project.enddate as projectenddate, result_projects.daysfromend as daysfromend FROM ${hive_db_name}.result_projects, ${hive_db_name}.result, ${hive_db_name}.project WHERE result_projects.id=result.id and result.type='publication' and project.id=result_projects.project;
CREATE OR REPLACE VIEW ${stats_db_name}.project_results_publication AS SELECT result_projects.id AS result, result_projects.project AS project_results, result.date as resultdate, project.enddate as projectenddate, result_projects.daysfromend AS daysfromend FROM ${stats_db_name}.result_projects, ${stats_db_name}.result, ${stats_db_name}.project WHERE result_projects.id=result.id AND result.type='publication' AND project.id=result_projects.project;

45
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step13.sql
View File

@ -1,39 +1,38 @@
------------------------------------------------------------------------------------------------------
-- Creating parquet tables from the updated temporary tables and removing unnecessary temporary tables
------------------------------------------------------------------------------------------------------
DROP TABLE IF EXISTS ${hive_db_name}.datasource;
CREATE TABLE ${hive_db_name}.datasource stored as parquet as select * from ${hive_db_name}.datasource_tmp;
DROP TABLE IF EXISTS ${hive_db_name}.publication;
CREATE TABLE ${hive_db_name}.publication stored as parquet as select * from ${hive_db_name}.publication_tmp;
DROP TABLE IF EXISTS ${stats_db_name}.datasource;
CREATE TABLE ${stats_db_name}.datasource stored AS parquet AS SELECT * FROM ${stats_db_name}.datasource_tmp;
DROP TABLE IF EXISTS ${hive_db_name}.dataset;
CREATE TABLE ${hive_db_name}.dataset stored as parquet as select * from ${hive_db_name}.dataset_tmp;
DROP TABLE IF EXISTS ${stats_db_name}.publication;
CREATE TABLE ${stats_db_name}.publication stored AS parquet AS SELECT * FROM ${stats_db_name}.publication_tmp;
DROP TABLE IF EXISTS ${hive_db_name}.software;
CREATE TABLE ${hive_db_name}.software stored as parquet as select * from ${hive_db_name}.software_tmp;
DROP TABLE IF EXISTS ${stats_db_name}.dataset;
CREATE TABLE ${stats_db_name}.dataset stored AS parquet AS SELECT * FROM ${stats_db_name}.dataset_tmp;
DROP TABLE IF EXISTS ${hive_db_name}.otherresearchproduct;
CREATE TABLE ${hive_db_name}.otherresearchproduct stored as parquet as select * from ${hive_db_name}.otherresearchproduct_tmp;
DROP TABLE IF EXISTS ${stats_db_name}.software;
CREATE TABLE ${stats_db_name}.software stored AS parquet AS SELECT * FROM ${stats_db_name}.software_tmp;
DROP TABLE ${hive_db_name}.project_tmp;
DROP TABLE ${hive_db_name}.datasource_tmp;
DROP TABLE ${hive_db_name}.publication_tmp;
DROP TABLE ${hive_db_name}.dataset_tmp;
DROP TABLE ${hive_db_name}.software_tmp;
DROP TABLE ${hive_db_name}.otherresearchproduct_tmp;
DROP TABLE IF EXISTS ${stats_db_name}.otherresearchproduct;
CREATE TABLE ${stats_db_name}.otherresearchproduct stored AS parquet AS SELECT * FROM ${stats_db_name}.otherresearchproduct_tmp;
DROP TABLE ${stats_db_name}.project_tmp;
DROP TABLE ${stats_db_name}.datasource_tmp;
DROP TABLE ${stats_db_name}.publication_tmp;
DROP TABLE ${stats_db_name}.dataset_tmp;
DROP TABLE ${stats_db_name}.software_tmp;
DROP TABLE ${stats_db_name}.otherresearchproduct_tmp;
----------------------------------------------
-- Re-creating views from final parquet tables
---------------------------------------------
-- Result
CREATE OR REPLACE VIEW ${hive_db_name}.result as SELECT *, bestlicence as access_mode FROM ${hive_db_name}.publication UNION ALL SELECT *, bestlicence as access_mode FROM ${hive_db_name}.software UNION ALL SELECT *, bestlicence as access_mode FROM ${hive_db_name}.dataset UNION ALL SELECT *, bestlicence as access_mode FROM ${hive_db_name}.otherresearchproduct;
CREATE OR REPLACE VIEW ${stats_db_name}.result AS SELECT *, bestlicence AS access_mode FROM ${stats_db_name}.publication UNION ALL SELECT *, bestlicence as access_mode FROM ${stats_db_name}.software UNION ALL SELECT *, bestlicence AS access_mode FROM ${stats_db_name}.dataset UNION ALL SELECT *, bestlicence AS access_mode FROM ${stats_db_name}.otherresearchproduct;
-- cleanup
drop view if exists ${hive_db_name}.delayedpubs;
drop view if exists ${hive_db_name}.project_pub_count;
drop view if exists ${hive_db_name}.delayedpubs;
drop view if exists ${hive_db_name}.project_results_publication;
CREATE TABLE ${hive_db_name}.numbers_country AS SELECT org.country AS country, count(distinct rd.datasource) AS datasources, count(distinct r.id) AS publications FROM ${hive_db_name}.result r, ${hive_db_name}.result_datasources rd, ${hive_db_name}.datasource d, ${hive_db_name}.datasource_organizations dor, ${hive_db_name}.organization org WHERE r.id=rd.id AND rd.datasource=d.id AND d.id=dor.id AND dor.organization=org.id AND r.type='publication' AND r.bestlicence='Open Access' GROUP BY org.country;
-------------------------------------------------------------------------------
-- To see with Antonis if the following is needed and where it should be placed
-------------------------------------------------------------------------------
CREATE TABLE ${stats_db_name}.numbers_country AS SELECT org.country AS country, count(distinct rd.datasource) AS datasources, count(distinct r.id) AS publications FROM ${stats_db_name}.result r, ${stats_db_name}.result_datasources rd, ${stats_db_name}.datasource d, ${stats_db_name}.datasource_organizations dor, ${stats_db_name}.organization org WHERE r.id=rd.id AND rd.datasource=d.id AND d.id=dor.id AND dor.organization=org.id AND r.type='publication' AND r.bestlicence='Open Access' GROUP BY org.country;

47
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step14.sql Normal file
View File

@ -0,0 +1,47 @@
COMPUTE STATS dataset;
COMPUTE STATS dataset_citations;
COMPUTE STATS dataset_classifications;
COMPUTE STATS dataset_concepts;
COMPUTE STATS dataset_datasources;
COMPUTE STATS dataset_languages;
COMPUTE STATS dataset_oids;
COMPUTE STATS dataset_pids;
COMPUTE STATS dataset_topics;
COMPUTE STATS datasource;
COMPUTE STATS datasource_languages;
COMPUTE STATS datasource_oids;
COMPUTE STATS datasource_organizations;
COMPUTE STATS numbers_country;
COMPUTE STATS organization;
COMPUTE STATS otherresearchproduct;
COMPUTE STATS otherresearchproduct_citations;
COMPUTE STATS otherresearchproduct_classifications;
COMPUTE STATS otherresearchproduct_concepts;
COMPUTE STATS otherresearchproduct_datasources;
COMPUTE STATS otherresearchproduct_languages;
COMPUTE STATS otherresearchproduct_oids;
COMPUTE STATS otherresearchproduct_pids;
COMPUTE STATS otherresearchproduct_topics;
COMPUTE STATS project;
COMPUTE STATS project_oids;
COMPUTE STATS project_organizations;
COMPUTE STATS project_results;
COMPUTE STATS publication;
COMPUTE STATS publication_citations;
COMPUTE STATS publication_classifications;
COMPUTE STATS publication_concepts;
COMPUTE STATS publication_datasources;
COMPUTE STATS publication_languages;
COMPUTE STATS publication_oids;
COMPUTE STATS publication_pids;
COMPUTE STATS publication_topics;
COMPUTE STATS result_organization;
COMPUTE STATS result_projects;
COMPUTE STATS software;
COMPUTE STATS software_citations;
COMPUTE STATS software_classifications;
COMPUTE STATS software_concepts;
COMPUTE STATS software_datasources;
COMPUTE STATS software_languages;
COMPUTE STATS software_oids;
COMPUTE STATS software_pids;

30
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step2.sql
View File

@ -5,28 +5,28 @@
--------------------------------------------------------------
-- Publication temporary table
DROP TABLE IF EXISTS ${hive_db_name}.publication_tmp;
CREATE TABLE ${hive_db_name}.publication_tmp (id STRING, title STRING, publisher STRING, journal STRING, date STRING, year STRING, bestlicence STRING, embargo_end_date STRING, delayed BOOLEAN, authors INT, source STRING, abstract BOOLEAN, type STRING ) clustered by (id) into 100 buckets stored as orc tblproperties('transactional'='true');
DROP TABLE IF EXISTS ${stats_db_name}.publication_tmp;
-- The following fails
--
INSERT INTO ${hive_db_name}.publication_tmp SELECT substr(p.id, 4) as id, p.title[0].value as title, p.publisher.value as publisher, p.journal.name as journal,
CREATE TABLE ${stats_db_name}.publication_tmp (id STRING, title STRING, publisher STRING, journal STRING, date STRING, year STRING, bestlicence STRING, embargo_end_date STRING, delayed BOOLEAN, authors INT, source STRING, abstract BOOLEAN, type STRING ) clustered by (id) into 100 buckets stored as orc tblproperties('transactional'='true');
INSERT INTO ${stats_db_name}.publication_tmp SELECT substr(p.id, 4) as id, p.title[0].value as title, p.publisher.value as publisher, p.journal.name as journal ,
p.dateofacceptance.value as date, date_format(p.dateofacceptance.value,'yyyy') as year, p.bestaccessright.classname as bestlicence,
p.embargoenddate.value as embargo_end_date, false as delayed, size(p.author) as authors , concat_ws('\u003B',p.source.value) as source,
case when size(p.description) > 0 then true else false end as abstract,
'publication' as type
from ${hive_source_db_name}.publication p
from ${openaire_db_name}.publication p
where p.datainfo.deletedbyinference=false;
CREATE TABLE ${stats_db_name}.publication_classifications AS SELECT substr(p.id, 4) as id, instancetype.classname as type from ${openaire_db_name}.publication p LATERAL VIEW explode(p.instance.instancetype) instances as instancetype;
CREATE TABLE ${hive_db_name}.publication_classifications AS SELECT substr(p.id, 4) as id, instancetype.classname as type from ${hive_source_db_name}.publication p LATERAL VIEW explode(p.instance.instancetype) instances as instancetype;
CREATE TABLE ${hive_db_name}.publication_concepts AS SELECT substr(p.id, 4) as id, contexts.context.id as concept from ${hive_source_db_name}.publication p LATERAL VIEW explode(p.context) contexts as context;
CREATE TABLE ${stats_db_name}.publication_concepts AS SELECT substr(p.id, 4) as id, contexts.context.id as concept from ${openaire_db_name}.publication p LATERAL VIEW explode(p.context) contexts as context;
CREATE TABLE ${hive_db_name}.publication_datasources as SELECT p.id, case when d.id is null then 'other' else p.datasource end as datasource FROM (SELECT substr(p.id, 4) as id, substr(instances.instance.hostedby.key, 4) as datasource
from ${hive_source_db_name}.publication p lateral view explode(p.instance) instances as instance) p LEFT OUTER JOIN
(SELECT substr(d.id, 4) id from ${hive_source_db_name}.datasource d WHERE d.datainfo.deletedbyinference=false) d on p.datasource = d.id;
CREATE TABLE ${stats_db_name}.publication_datasources as SELECT p.id, case when d.id is null then 'other' else p.datasource end as datasource FROM (SELECT substr(p.id, 4) as id, substr(instances.instance.hostedby.key, 4) as datasource from ${openaire_db_name}.publication p lateral view explode(p.instance) instances as instance) p LEFT OUTER JOIN (SELECT substr(d.id, 4) id from ${openaire_db_name}.datasource d WHERE d.datainfo.deletedbyinference=false) d on p.datasource = d.id;
CREATE TABLE ${hive_db_name}.publication_languages AS select substr(p.id, 4) as id, p.language.classname as language from ${hive_source_db_name}.publication p;
CREATE TABLE ${hive_db_name}.publication_oids AS SELECT substr(p.id, 4) as id, oids.ids as oid from ${hive_source_db_name}.publication p LATERAL VIEW explode(p.originalid) oids as ids;
create table ${hive_db_name}.publication_pids as select substr(p.id, 4) as id, ppid.qualifier.classname as type, ppid.value as pid from ${hive_source_db_name}.publication p lateral view explode(p.pid) pids as ppid;
create table ${hive_db_name}.publication_topics as select substr(p.id, 4) as id, subjects.subject.qualifier.classname as type, subjects.subject.value as topic from ${hive_source_db_name}.publication p lateral view explode(p.subject) subjects as subject;
CREATE TABLE ${stats_db_name}.publication_languages AS select substr(p.id, 4) as id, p.language.classname as language FROM ${openaire_db_name}.publication p;
CREATE TABLE ${stats_db_name}.publication_oids AS SELECT substr(p.id, 4) AS id, oids.ids AS oid FROM ${openaire_db_name}.publication p LATERAL VIEW explode(p.originalid) oids AS ids;
CREATE TABLE ${stats_db_name}.publication_pids AS SELECT substr(p.id, 4) AS id, ppid.qualifier.classname AS type, ppid.value as pid FROM ${openaire_db_name}.publication p LATERAL VIEW explode(p.pid) pids AS ppid;
CREATE TABLE ${stats_db_name}.publication_topics as select substr(p.id, 4) AS id, subjects.subject.qualifier.classname AS TYPE, subjects.subject.value AS topic FROM ${openaire_db_name}.publication p LATERAL VIEW explode(p.subject) subjects AS subject;

4
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step3.sql
View File

@ -1,4 +1,2 @@
-- The following fails
--
-- 3. Publication_citations
CREATE TABLE ${hive_db_name}.publication_citations AS SELECT substr(p.id, 4) AS id, xpath_string(citation.value, "//citation/id[@type='openaire']/@value") AS result from ${hive_source_db_name}.publication p lateral view explode(p.extrainfo) citations AS citation WHERE xpath_string(citation.value, "//citation/id[@type='openaire']/@value") !="";
CREATE TABLE ${stats_db_name}.publication_citations AS SELECT substr(p.id, 4) AS id, xpath_string(citation.value, "//citation/id[@type='openaire']/@value") AS result FROM ${openaire_db_name}.publication p lateral view explode(p.extrainfo) citations AS citation WHERE xpath_string(citation.value, "//citation/id[@type='openaire']/@value") !="";

40
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step4.sql
View File

@ -5,22 +5,32 @@
------------------------------------------------------
-- Dataset temporary table supporting updates
DROP TABLE IF EXISTS ${hive_db_name}.dataset_tmp;
CREATE TABLE ${hive_db_name}.dataset_tmp (id STRING, title STRING, publisher STRING, journal STRING, date STRING, year STRING, bestlicence STRING, embargo_end_date STRING, delayed BOOLEAN, authors INT, source STRING, abstract BOOLEAN, type STRING ) clustered by (id) into 100 buckets stored as orc tblproperties('transactional'='true');
INSERT INTO ${hive_db_name}.dataset_tmp SELECT substr(d.id, 4) as id, d.title[0].value as title, d.publisher.value as publisher, cast(null as string) as journal, d.dateofacceptance.value as date,
date_format(d.dateofacceptance.value,'yyyy') as year, d.bestaccessright.classname as bestlicence, d.embargoenddate.value as embargo_end_date, false as delayed, size(d.author) as authors,
concat_ws ('\u003B',d.source.value) as source,
case when size(d.description) > 0 then true else false end as abstract, 'dataset' as type from ${hive_source_db_name}.dataset d where d.datainfo.deletedbyinference=false;
DROP TABLE IF EXISTS ${stats_db_name}.dataset_tmp;
CREATE TABLE ${stats_db_name}.dataset_tmp (id STRING, title STRING, publisher STRING, journal STRING, date STRING, year STRING, bestlicence STRING, embargo_end_date STRING, delayed BOOLEAN, authors INT, source STRING, abstract BOOLEAN, type STRING ) clustered by (id) into 100 buckets stored AS orc tblproperties('transactional'='true');
INSERT INTO ${stats_db_name}.dataset_tmp SELECT substr(d.id, 4) AS id, d.title[0].value AS title, d.publisher.value AS publisher, cast(null AS string) AS journal,
d.dateofacceptance.value as date, date_format(d.dateofacceptance.value,'yyyy') AS year, d.bestaccessright.classname AS bestlicence,
d.embargoenddate.value AS embargo_end_date, false AS delayed, size(d.author) AS authors , concat_ws('\u003B',d.source.value) AS source,
CASE WHEN SIZE(d.description) > 0 THEN TRUE ELSE FALSE end AS abstract,
'dataset' AS type
FROM ${openaire_db_name}.dataset d
WHERE d.datainfo.deletedbyinference=FALSE;
-- Dataset_citations
CREATE TABLE ${hive_db_name}.dataset_citations as select substr(d.id, 4) as id, xpath_string(citation.value, "//citation/id[@type='openaire']/@value") as result from ${hive_source_db_name}.dataset d lateral view explode(d.extrainfo) citations as citation where xpath_string(citation.value, "//citation/id[@type='openaire']/@value") !="";
CREATE TABLE ${stats_db_name}.dataset_citations AS SELECT substr(d.id, 4) AS id, xpath_string(citation.value, "//citation/id[@type='openaire']/@value") AS result FROM ${openaire_db_name}.dataset d LATERAL VIEW explode(d.extrainfo) citations AS citation WHERE xpath_string(citation.value, "//citation/id[@type='openaire']/@value") !="";
CREATE TABLE ${hive_db_name}.dataset_classifications AS SELECT substr(p.id, 4) as id, instancetype.classname as type from ${hive_source_db_name}.dataset p LATERAL VIEW explode(p.instance.instancetype) instances as instancetype;
CREATE TABLE ${hive_db_name}.dataset_concepts AS SELECT substr(p.id, 4) as id, contexts.context.id as concept from ${hive_source_db_name}.dataset p LATERAL VIEW explode(p.context) contexts as context;
CREATE TABLE ${hive_db_name}.dataset_datasources as SELECT p.id, case when d.id is null then 'other' else p.datasource end as datasource FROM (SELECT substr(p.id, 4) as id, substr(instances.instance.hostedby.key, 4) as datasource from ${hive_source_db_name}.dataset p lateral view explode(p.instance) instances as instance) p LEFT OUTER JOIN (SELECT substr(d.id, 4) id from ${hive_source_db_name}.datasource d WHERE d.datainfo.deletedbyinference=false) d on p.datasource = d.id;
CREATE TABLE ${hive_db_name}.dataset_languages AS SELECT substr(p.id, 4) as id, p.language.classname as language from ${hive_source_db_name}.dataset p;
CREATE TABLE ${hive_db_name}.dataset_oids AS SELECT substr(p.id, 4) as id, oids.ids as oid from ${hive_source_db_name}.dataset p LATERAL VIEW explode(p.originalid) oids as ids;
CREATE TABLE ${hive_db_name}.dataset_pids AS SELECT substr(p.id, 4) as id, ppid.qualifier.classname as type, ppid.value as pid from ${hive_source_db_name}.dataset p lateral view explode(p.pid) pids as ppid;
CREATE TABLE ${hive_db_name}.dataset_topics AS SELECT substr(p.id, 4) as id, subjects.subject.qualifier.classname as type, subjects.subject.value as topic from ${hive_source_db_name}.dataset p lateral view explode(p.subject) subjects as subject;
CREATE TABLE ${stats_db_name}.dataset_classifications AS SELECT substr(p.id, 4) AS id, instancetype.classname AS type FROM ${openaire_db_name}.dataset p LATERAL VIEW explode(p.instance.instancetype) instances AS instancetype;
CREATE TABLE ${stats_db_name}.dataset_concepts AS SELECT substr(p.id, 4) as id, contexts.context.id as concept from ${openaire_db_name}.dataset p LATERAL VIEW explode(p.context) contexts as context;
CREATE TABLE ${stats_db_name}.dataset_datasources AS SELECT p.id, case when d.id IS NULL THEN 'other' ELSE p.datasource END AS datasource FROM (SELECT substr(p.id, 4) as id, substr(instances.instance.hostedby.key, 4) AS datasource
FROM ${openaire_db_name}.dataset p LATERAL VIEW explode(p.instance) instances AS instance) p LEFT OUTER JOIN
(SELECT substr(d.id, 4) id FROM ${openaire_db_name}.datasource d WHERE d.datainfo.deletedbyinference=false) d ON p.datasource = d.id;
CREATE TABLE ${stats_db_name}.dataset_languages AS SELECT substr(p.id, 4) AS id, p.language.classname AS language FROM ${openaire_db_name}.dataset p;
CREATE TABLE ${stats_db_name}.dataset_oids AS SELECT substr(p.id, 4) AS id, oids.ids AS oid FROM ${openaire_db_name}.dataset p LATERAL VIEW explode(p.originalid) oids AS ids;
CREATE TABLE ${stats_db_name}.dataset_pids AS SELECT substr(p.id, 4) AS id, ppid.qualifier.classname AS type, ppid.value AS pid FROM ${openaire_db_name}.dataset p LATERAL VIEW explode(p.pid) pids AS ppid;
CREATE TABLE ${stats_db_name}.dataset_topics AS SELECT substr(p.id, 4) AS id, subjects.subject.qualifier.classname AS type, subjects.subject.value AS topic FROM ${openaire_db_name}.dataset p LATERAL VIEW explode(p.subject) subjects AS subject;

40
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step5.sql
View File

@ -5,28 +5,32 @@
--------------------------------------------------------
-- Software temporary table supporting updates
DROP TABLE IF EXISTS ${hive_db_name}.software_tmp;
CREATE TABLE ${hive_db_name}.software_tmp ( id STRING, title STRING, publisher STRING, journal STRING, date STRING, year STRING, bestlicence STRING, embargo_end_date STRING, delayed BOOLEAN, authors INT, source STRING, abstract BOOLEAN, type STRING ) clustered by (id) into 100 buckets stored as orc tblproperties('transactional'='true');
INSERT INTO ${hive_db_name}.software_tmp select substr(s.id, 4) as id, s.title[0].value as title, s.publisher.value as publisher, cast(null as string) as journal,
s.dateofacceptance.value as date, date_format(s.dateofacceptance.value,'yyyy') as year, s.bestaccessright.classname as bestlicence,
s.embargoenddate.value as embargo_end_date, false as delayed, size(s.author) as authors , concat_ws('\u003B',s.source.value) as source,
case when size(s.description) > 0 then true else false end as abstract,
DROP TABLE IF EXISTS ${stats_db_name}.software_tmp;
CREATE TABLE ${stats_db_name}.software_tmp (id STRING, title STRING, publisher STRING, journal STRING, date STRING, year STRING, bestlicence STRING, embargo_end_date STRING, delayed BOOLEAN, authors INT, source STRING, abstract BOOLEAN, type STRING ) clustered by (id) INTO 100 buckets stored AS orc tblproperties('transactional'='true');
INSERT INTO ${stats_db_name}.software_tmp SELECT substr(s.id, 4) as id, s.title[0].value AS title, s.publisher.value AS publisher, CAST(NULL AS string) AS journal,
s.dateofacceptance.value AS DATE, date_format(s.dateofacceptance.value,'yyyy') AS YEAR, s.bestaccessright.classname AS bestlicence,
s.embargoenddate.value AS embargo_end_date, FALSE AS delayed, SIZE(s.author) AS authors , concat_ws('\u003B',s.source.value) AS source,
CASE WHEN SIZE(s.description) > 0 THEN TRUE ELSE FALSE END AS abstract,
'software' as type
from ${hive_source_db_name}.software s
from ${openaire_db_name}.software s
where s.datainfo.deletedbyinference=false;
-- Software_citations
Create table ${hive_db_name}.software_citations as select substr(s.id, 4) as id, xpath_string(citation.value, "//citation/id[@type='openaire']/@value") as result from ${hive_source_db_name}.software s lateral view explode(s.extrainfo) citations as citation where xpath_string(citation.value, "//citation/id[@type='openaire']/@value") !="";
CREATE TABLE ${stats_db_name}.software_citations AS SELECT substr(s.id, 4) as id, xpath_string(citation.value, "//citation/id[@type='openaire']/@value") AS RESULT FROM ${openaire_db_name}.software s LATERAL VIEW explode(s.extrainfo) citations as citation where xpath_string(citation.value, "//citation/id[@type='openaire']/@value") !="";
CREATE TABLE ${hive_db_name}.software_classifications AS SELECT substr(p.id, 4) as id, instancetype.classname as type from ${hive_source_db_name}.software p LATERAL VIEW explode(p.instance.instancetype) instances as instancetype;
CREATE TABLE ${hive_db_name}.software_concepts AS SELECT substr(p.id, 4) as id, contexts.context.id as concept from ${hive_source_db_name}.software p LATERAL VIEW explode(p.context) contexts as context;
CREATE TABLE ${stats_db_name}.software_classifications AS SELECT substr(p.id, 4) AS id, instancetype.classname AS type FROM ${openaire_db_name}.software p LATERAL VIEW explode(p.instance.instancetype) instances AS instancetype;
CREATE TABLE ${hive_db_name}.software_datasources as SELECT p.id, case when d.id is null then 'other' else p.datasource end as datasource FROM (SELECT substr(p.id, 4) as id, substr(instances.instance.hostedby.key, 4) as datasource
from ${hive_source_db_name}.software p lateral view explode(p.instance) instances as instance) p LEFT OUTER JOIN
(SELECT substr(d.id, 4) id from ${hive_source_db_name}.datasource d WHERE d.datainfo.deletedbyinference=false) d on p.datasource = d.id;
CREATE TABLE ${stats_db_name}.software_concepts AS SELECT substr(p.id, 4) AS id, contexts.context.id AS concept FROM ${openaire_db_name}.software p LATERAL VIEW explode(p.context) contexts AS context;
CREATE TABLE ${hive_db_name}.software_languages AS select substr(p.id, 4) as id, p.language.classname as language from ${hive_source_db_name}.software p;
CREATE TABLE ${hive_db_name}.software_oids AS SELECT substr(p.id, 4) as id, oids.ids as oid from ${hive_source_db_name}.software p LATERAL VIEW explode(p.originalid) oids as ids;
create table ${hive_db_name}.software_pids as select substr(p.id, 4) as id, ppid.qualifier.classname as type, ppid.value as pid from ${hive_source_db_name}.software p lateral view explode(p.pid) pids as ppid;
create table ${hive_db_name}.software_topics as select substr(p.id, 4) as id, subjects.subject.qualifier.classname as type, subjects.subject.value as topic from ${hive_source_db_name}.software p lateral view explode(p.subject) subjects as subject;
CREATE TABLE ${stats_db_name}.software_datasources AS SELECT p.id, CASE WHEN d.id IS NULL THEN 'other' ELSE p.datasource end as datasource FROM (SELECT substr(p.id, 4) AS id, substr(instances.instance.hostedby.key, 4) AS datasource
FROM ${openaire_db_name}.software p LATERAL VIEW explode(p.instance) instances AS instance) p LEFT OUTER JOIN
(SELECT substr(d.id, 4) id FROM ${openaire_db_name}.datasource d WHERE d.datainfo.deletedbyinference=false) d ON p.datasource = d.id;
CREATE TABLE ${stats_db_name}.software_languages AS select substr(p.id, 4) AS id, p.language.classname AS language FROM ${openaire_db_name}.software p;
CREATE TABLE ${stats_db_name}.software_oids AS SELECT substr(p.id, 4) AS id, oids.ids AS oid FROM ${openaire_db_name}.software p LATERAL VIEW explode(p.originalid) oids AS ids;
CREATE TABLE ${stats_db_name}.software_pids AS SELECT substr(p.id, 4) AS id, ppid.qualifier.classname AS type, ppid.value AS pid FROM ${openaire_db_name}.software p LATERAL VIEW explode(p.pid) pids AS ppid;
CREATE TABLE ${stats_db_name}.software_topics AS SELECT substr(p.id, 4) AS id, subjects.subject.qualifier.classname AS type, subjects.subject.value AS topic FROM ${openaire_db_name}.software p LATERAL VIEW explode(p.subject) subjects AS subject;

44
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step6.sql
View File

@ -5,27 +5,33 @@
--------------------------------------------------------------------------------
-- Otherresearchproduct temporary table supporting updates
DROP TABLE IF EXISTS ${hive_db_name}.otherresearchproduct_tmp;
CREATE TABLE ${hive_db_name}.otherresearchproduct_tmp ( id STRING, title STRING, publisher STRING, journal STRING, date STRING, year STRING, bestlicence STRING, embargo_end_date STRING, delayed BOOLEAN, authors INT, source STRING, abstract BOOLEAN, type STRING ) clustered by (id) into 100 buckets stored as orc tblproperties('transactional'='true');
INSERT INTO ${hive_db_name}.otherresearchproduct_tmp select substr(o.id, 4) as id, o.title[0].value as title, o.publisher.value as publisher, cast(null as string) as journal,
o.dateofacceptance.value as date, date_format(o.dateofacceptance.value,'yyyy') as year, o.bestaccessright.classname as bestlicence,
o.embargoenddate.value as embargo_end_date, false as delayed, size(o.author) as authors , concat_ws('\u003B',o.source.value) as source,
case when size(o.description) > 0 then true else false end as abstract,
'other' as type
from ${hive_source_db_name}.otherresearchproduct o
where o.datainfo.deletedbyinference=false;
DROP TABLE IF EXISTS ${stats_db_name}.otherresearchproduct_tmp;
CREATE TABLE ${stats_db_name}.otherresearchproduct_tmp ( id STRING, title STRING, publisher STRING, journal STRING, date STRING, year STRING, bestlicence STRING, embargo_end_date STRING, delayed BOOLEAN, authors INT, source STRING, abstract BOOLEAN, type STRING ) CLUSTERED BY (id) INTO 100 buckets stored AS orc tblproperties('transactional'='true');
INSERT INTO ${stats_db_name}.otherresearchproduct_tmp SELECT substr(o.id, 4) AS id, o.title[0].value AS title, o.publisher.value AS publisher, CAST(NULL AS string) AS journal,
o.dateofacceptance.value AS DATE, date_format(o.dateofacceptance.value,'yyyy') AS year, o.bestaccessright.classname AS bestlicence,
o.embargoenddate.value as embargo_end_date, FALSE AS delayed, SIZE(o.author) AS authors , concat_ws('\u003B',o.source.value) AS source,
CASE WHEN SIZE(o.description) > 0 THEN TRUE ELSE FALSE END AS abstract,
'other' AS type
FROM ${openaire_db_name}.otherresearchproduct o
WHERE o.datainfo.deletedbyinference=FALSE;
-- Otherresearchproduct_citations
Create table ${hive_db_name}.otherresearchproduct_citations as select substr(o.id, 4) as id, xpath_string(citation.value, "//citation/id[@type='openaire']/@value") as result from ${hive_source_db_name}.otherresearchproduct o lateral view explode(o.extrainfo) citations as citation where xpath_string(citation.value, "//citation/id[@type='openaire']/@value") !="";
CREATE TABLE ${stats_db_name}.otherresearchproduct_citations AS SELECT substr(o.id, 4) AS id, xpath_string(citation.value, "//citation/id[@type='openaire']/@value") AS RESULT FROM ${openaire_db_name}.otherresearchproduct o LATERAL VIEW explode(o.extrainfo) citations AS citation WHERE xpath_string(citation.value, "//citation/id[@type='openaire']/@value") !="";
CREATE TABLE ${hive_db_name}.otherresearchproduct_classifications AS SELECT substr(p.id, 4) as id, instancetype.classname as type from ${hive_source_db_name}.otherresearchproduct p LATERAL VIEW explode(p.instance.instancetype) instances as instancetype;
CREATE TABLE ${hive_db_name}.otherresearchproduct_concepts AS SELECT substr(p.id, 4) as id, contexts.context.id as concept from ${hive_source_db_name}.otherresearchproduct p LATERAL VIEW explode(p.context) contexts as context;
CREATE TABLE ${stats_db_name}.otherresearchproduct_classifications AS SELECT substr(p.id, 4) AS id, instancetype.classname AS type FROM ${openaire_db_name}.otherresearchproduct p LATERAL VIEW explode(p.instance.instancetype) instances AS instancetype;
CREATE TABLE ${hive_db_name}.otherresearchproduct_datasources as SELECT p.id, case when d.id is null then 'other' else p.datasource end as datasource FROM (SELECT substr(p.id, 4) as id, substr(instances.instance.hostedby.key, 4) as datasource
from ${hive_source_db_name}.otherresearchproduct p lateral view explode(p.instance) instances as instance) p LEFT OUTER JOIN
(SELECT substr(d.id, 4) id from ${hive_source_db_name}.datasource d WHERE d.datainfo.deletedbyinference=false) d on p.datasource = d.id;
CREATE TABLE ${stats_db_name}.otherresearchproduct_concepts AS SELECT substr(p.id, 4) AS id, contexts.context.id AS concept FROM ${openaire_db_name}.otherresearchproduct p LATERAL VIEW explode(p.context) contexts AS context;
CREATE TABLE ${hive_db_name}.otherresearchproduct_languages AS select substr(p.id, 4) as id, p.language.classname as language from ${hive_source_db_name}.otherresearchproduct p;
CREATE TABLE ${hive_db_name}.otherresearchproduct_oids AS SELECT substr(p.id, 4) as id, oids.ids as oid from ${hive_source_db_name}.otherresearchproduct p LATERAL VIEW explode(p.originalid) oids as ids;
create table ${hive_db_name}.otherresearchproduct_pids as select substr(p.id, 4) as id, ppid.qualifier.classname as type, ppid.value as pid from ${hive_source_db_name}.otherresearchproduct p lateral view explode(p.pid) pids as ppid;
create table ${hive_db_name}.otherresearchproduct_topics as select substr(p.id, 4) as id, subjects.subject.qualifier.classname as type, subjects.subject.value as topic from ${hive_source_db_name}.otherresearchproduct p lateral view explode(p.subject) subjects as subject;
CREATE TABLE ${stats_db_name}.otherresearchproduct_datasources AS SELECT p.id, CASE WHEN d.id IS NULL THEN 'other' ELSE p.datasource END AS datasource FROM (SELECT substr(p.id, 4) AS id, substr(instances.instance.hostedby.key, 4) AS datasource
from ${openaire_db_name}.otherresearchproduct p lateral view explode(p.instance) instances as instance) p LEFT OUTER JOIN
(SELECT substr(d.id, 4) id from ${openaire_db_name}.datasource d WHERE d.datainfo.deletedbyinference=false) d on p.datasource = d.id;
CREATE TABLE ${stats_db_name}.otherresearchproduct_languages AS SELECT substr(p.id, 4) AS id, p.language.classname AS language FROM ${openaire_db_name}.otherresearchproduct p;
CREATE TABLE ${stats_db_name}.otherresearchproduct_oids AS SELECT substr(p.id, 4) AS id, oids.ids AS oid FROM ${openaire_db_name}.otherresearchproduct p LATERAL VIEW explode(p.originalid) oids AS ids;
CREATE TABLE ${stats_db_name}.otherresearchproduct_pids AS SELECT substr(p.id, 4) AS id, ppid.qualifier.classname AS type, ppid.value AS pid FROM ${openaire_db_name}.otherresearchproduct p LATERAL VIEW explode(p.pid) pids AS ppid;
CREATE TABLE ${stats_db_name}.otherresearchproduct_topics AS SELECT substr(p.id, 4) AS id, subjects.subject.qualifier.classname AS type, subjects.subject.value AS topic FROM ${openaire_db_name}.otherresearchproduct p LATERAL VIEW explode(p.subject) subjects AS subject;

21
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step7.sql
View File

@ -4,18 +4,21 @@
------------------------------------------------------
------------------------------------------------------
-- Project_oids Table
DROP TABLE IF EXISTS ${hive_db_name}.project_oids;
CREATE TABLE ${hive_db_name}.project_oids AS SELECT substr(p.id, 4) as id, oids.ids as oid from ${hive_source_db_name}.project p LATERAL VIEW explode(p.originalid) oids as ids;
DROP TABLE IF EXISTS ${stats_db_name}.project_oids;
CREATE TABLE ${stats_db_name}.project_oids AS SELECT substr(p.id, 4) AS id, oids.ids AS oid FROM ${openaire_db_name}.project p LATERAL VIEW explode(p.originalid) oids AS ids;
-- Project_organizations Table
DROP TABLE IF EXISTS ${hive_db_name}.project_organizations;
CREATE TABLE ${hive_db_name}.project_organizations AS SELECT substr(r.source, 4) as id, substr(r.target, 4) AS organization from ${hive_source_db_name}.relation r WHERE r.reltype='projectOrganization';
DROP TABLE IF EXISTS ${stats_db_name}.project_organizations;
CREATE TABLE ${stats_db_name}.project_organizations AS SELECT substr(r.source, 4) AS id, substr(r.target, 4) AS organization from ${openaire_db_name}.relation r WHERE r.reltype='projectOrganization';
-- Project_results Table
DROP TABLE IF EXISTS ${hive_db_name}.project_results;
CREATE TABLE ${hive_db_name}.project_results AS SELECT substr(r.target, 4) as id, substr(r.source, 4) AS result from ${hive_source_db_name}.relation r WHERE r.reltype='resultProject';
DROP TABLE IF EXISTS ${stats_db_name}.project_results;
CREATE TABLE ${stats_db_name}.project_results AS SELECT substr(r.target, 4) AS id, substr(r.source, 4) AS result FROM ${openaire_db_name}.relation r WHERE r.reltype='resultProject';
-- Project table
----------------
-- Creating and populating temporary Project table
DROP TABLE IF EXISTS ${hive_db_name}.project_tmp;
CREATE TABLE ${hive_db_name}.project_tmp (id STRING, acronym STRING, title STRING, funder STRING, funding_lvl0 STRING, funding_lvl1 STRING, funding_lvl2 STRING, ec39 STRING, type STRING, startdate STRING, enddate STRING, start_year STRING, end_year STRING, duration INT, haspubs STRING, numpubs INT, daysforlastpub INT, delayedpubs INT, callidentifier STRING, code STRING) clustered by (id) into 100 buckets stored as orc tblproperties('transactional'='true');
INSERT INTO ${hive_db_name}.project_tmp SELECT substr(p.id, 4) as id, p.acronym.value as acronym, p.title.value as title, xpath_string(p.fundingtree[0].value, '//funder/name') as funder, xpath_string(p.fundingtree[0].value, '//funding_level_0/name') as funding_lvl0, xpath_string(p.fundingtree[0].value, '//funding_level_1/name') as funding_lvl1, xpath_string(p.fundingtree[0].value, '//funding_level_2/name') as funding_lvl2, p.ecsc39.value as ec39, p.contracttype.classname as type, p.startdate.value as startdate, p.enddate.value as enddate, date_format(p.startdate.value, 'yyyy') as start_year, date_format(p.enddate.value, 'yyyy') as end_year, 0 as duration, 'no' as haspubs, 0 as numpubs, 0 as daysforlastpub, 0 as delayedpubs, p.callidentifier.value as callidentifier, p.code.value as code from ${hive_source_db_name}.project p WHERE p.datainfo.deletedbyinference=false;
DROP TABLE IF EXISTS ${stats_db_name}.project_tmp;
CREATE TABLE ${stats_db_name}.project_tmp (id STRING, acronym STRING, title STRING, funder STRING, funding_lvl0 STRING, funding_lvl1 STRING, funding_lvl2 STRING, ec39 STRING, type STRING, startdate STRING, enddate STRING, start_year STRING, end_year STRING, duration INT, haspubs STRING, numpubs INT, daysforlastpub INT, delayedpubs INT, callidentifier STRING, code STRING) CLUSTERED BY (id) INTO 100 buckets stored AS orc tblproperties('transactional'='true');
INSERT INTO ${stats_db_name}.project_tmp SELECT substr(p.id, 4) AS id, p.acronym.value AS acronym, p.title.value AS title, xpath_string(p.fundingtree[0].value, '//funder/name') AS funder, xpath_string(p.fundingtree[0].value, '//funding_level_0/name') AS funding_lvl0, xpath_string(p.fundingtree[0].value, '//funding_level_1/name') AS funding_lvl1, xpath_string(p.fundingtree[0].value, '//funding_level_2/name') AS funding_lvl2, p.ecsc39.value AS ec39, p.contracttype.classname AS type, p.startdate.value AS startdate, p.enddate.value AS enddate, date_format(p.startdate.value, 'yyyy') AS start_year, date_format(p.enddate.value, 'yyyy') AS end_year, 0 AS duration, 'no' AS haspubs, 0 AS numpubs, 0 AS daysforlastpub, 0 AS delayedpubs, p.callidentifier.value AS callidentifier, p.code.value AS code FROM ${openaire_db_name}.project p WHERE p.datainfo.deletedbyinference=false;

33
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step8.sql
View File

@ -5,20 +5,27 @@
----------------------------------------------------
-- Views on temporary tables that should be re-created in the end
CREATE OR REPLACE VIEW ${hive_db_name}.result as SELECT *, bestlicence as access_mode FROM ${hive_db_name}.publication_tmp UNION ALL SELECT *,bestlicence as access_mode FROM ${hive_db_name}.software_tmp UNION ALL SELECT *,bestlicence as access_mode FROM ${hive_db_name}.dataset_tmp UNION ALL SELECT *,bestlicence as access_mode FROM ${hive_db_name}.otherresearchproduct_tmp;
CREATE OR REPLACE VIEW ${stats_db_name}.result as SELECT *, bestlicence AS access_mode FROM ${stats_db_name}.publication_tmp UNION ALL SELECT *,bestlicence AS access_mode FROM ${stats_db_name}.software_tmp UNION ALL SELECT *,bestlicence AS access_mode FROM ${stats_db_name}.dataset_tmp UNION ALL SELECT *,bestlicence AS access_mode FROM ${stats_db_name}.otherresearchproduct_tmp;
-- Views on final tables
CREATE OR REPLACE VIEW ${hive_db_name}.result_datasources as SELECT * FROM ${hive_db_name}.publication_datasources UNION ALL SELECT * FROM ${hive_db_name}.software_datasources UNION ALL SELECT * FROM ${hive_db_name}.dataset_datasources UNION ALL SELECT * FROM ${hive_db_name}.otherresearchproduct_datasources;
CREATE OR REPLACE VIEW ${hive_db_name}.result_citations as SELECT * FROM ${hive_db_name}.publication_citations UNION ALL SELECT * FROM ${hive_db_name}.software_citations UNION ALL SELECT * FROM ${hive_db_name}.dataset_citations UNION ALL SELECT * FROM ${hive_db_name}.otherresearchproduct_citations;
CREATE OR REPLACE VIEW ${hive_db_name}.result_classifications as SELECT * FROM ${hive_db_name}.publication_classifications UNION ALL SELECT * FROM ${hive_db_name}.software_classifications UNION ALL SELECT * FROM ${hive_db_name}.dataset_classifications UNION ALL SELECT * FROM ${hive_db_name}.otherresearchproduct_classifications;
CREATE OR REPLACE VIEW ${hive_db_name}.result_concepts as SELECT * FROM ${hive_db_name}.publication_concepts UNION ALL SELECT * FROM ${hive_db_name}.software_concepts UNION ALL SELECT * FROM ${hive_db_name}.dataset_concepts UNION ALL SELECT * FROM ${hive_db_name}.otherresearchproduct_concepts;
CREATE OR REPLACE VIEW ${hive_db_name}.result_languages as SELECT * FROM ${hive_db_name}.publication_languages UNION ALL SELECT * FROM ${hive_db_name}.software_languages UNION ALL SELECT * FROM ${hive_db_name}.dataset_languages UNION ALL SELECT * FROM ${hive_db_name}.otherresearchproduct_languages;
CREATE OR REPLACE VIEW ${hive_db_name}.result_oids as SELECT * FROM ${hive_db_name}.publication_oids UNION ALL SELECT * FROM ${hive_db_name}.software_oids UNION ALL SELECT * FROM ${hive_db_name}.dataset_oids UNION ALL SELECT * FROM ${hive_db_name}.otherresearchproduct_oids;
CREATE OR REPLACE VIEW ${hive_db_name}.result_pids as SELECT * FROM ${hive_db_name}.publication_pids UNION ALL SELECT * FROM ${hive_db_name}.software_pids UNION ALL SELECT * FROM ${hive_db_name}.dataset_pids UNION ALL SELECT * FROM ${hive_db_name}.otherresearchproduct_pids;
CREATE OR REPLACE VIEW ${hive_db_name}.result_topics as SELECT * FROM ${hive_db_name}.publication_topics UNION ALL SELECT * FROM ${hive_db_name}.software_topics UNION ALL SELECT * FROM ${hive_db_name}.dataset_topics UNION ALL SELECT * FROM ${hive_db_name}.otherresearchproduct_topics;
CREATE OR REPLACE VIEW ${stats_db_name}.result_datasources AS SELECT * FROM ${stats_db_name}.publication_datasources UNION ALL SELECT * FROM ${stats_db_name}.software_datasources UNION ALL SELECT * FROM ${stats_db_name}.dataset_datasources UNION ALL SELECT * FROM ${stats_db_name}.otherresearchproduct_datasources;
DROP TABLE IF EXISTS ${hive_db_name}.result_organization;
CREATE TABLE ${hive_db_name}.result_organization AS SELECT substr(r.target, 4) as id, substr(r.source, 4) as organization from ${hive_source_db_name}.relation r where r.reltype='resultOrganization';
CREATE OR REPLACE VIEW ${stats_db_name}.result_citations AS SELECT * FROM ${stats_db_name}.publication_citations UNION ALL SELECT * FROM ${stats_db_name}.software_citations UNION ALL SELECT * FROM ${stats_db_name}.dataset_citations UNION ALL SELECT * FROM ${stats_db_name}.otherresearchproduct_citations;
DROP TABLE IF EXISTS ${hive_db_name}.result_projects;
CREATE TABLE ${hive_db_name}.result_projects AS select pr.result as id, pr.id as project, datediff(p.enddate, p.startdate) as daysfromend from ${hive_db_name}.result r join ${hive_db_name}.project_results pr on r.id=pr.result join ${hive_db_name}.project_tmp p on p.id=pr.id;
CREATE OR REPLACE VIEW ${stats_db_name}.result_classifications AS SELECT * FROM ${stats_db_name}.publication_classifications UNION ALL SELECT * FROM ${stats_db_name}.software_classifications UNION ALL SELECT * FROM ${stats_db_name}.dataset_classifications UNION ALL SELECT * FROM ${stats_db_name}.otherresearchproduct_classifications;
CREATE OR REPLACE VIEW ${stats_db_name}.result_concepts AS SELECT * FROM ${stats_db_name}.publication_concepts UNION ALL SELECT * FROM ${stats_db_name}.software_concepts UNION ALL SELECT * FROM ${stats_db_name}.dataset_concepts UNION ALL SELECT * FROM ${stats_db_name}.otherresearchproduct_concepts;
CREATE OR REPLACE VIEW ${stats_db_name}.result_languages AS SELECT * FROM ${stats_db_name}.publication_languages UNION ALL SELECT * FROM ${stats_db_name}.software_languages UNION ALL SELECT * FROM ${stats_db_name}.dataset_languages UNION ALL SELECT * FROM ${stats_db_name}.otherresearchproduct_languages;
CREATE OR REPLACE VIEW ${stats_db_name}.result_oids AS SELECT * FROM ${stats_db_name}.publication_oids UNION ALL SELECT * FROM ${stats_db_name}.software_oids UNION ALL SELECT * FROM ${stats_db_name}.dataset_oids UNION ALL SELECT * FROM ${stats_db_name}.otherresearchproduct_oids;
CREATE OR REPLACE VIEW ${stats_db_name}.result_pids AS SELECT * FROM ${stats_db_name}.publication_pids UNION ALL SELECT * FROM ${stats_db_name}.software_pids UNION ALL SELECT * FROM ${stats_db_name}.dataset_pids UNION ALL SELECT * FROM ${stats_db_name}.otherresearchproduct_pids;
CREATE OR REPLACE VIEW ${stats_db_name}.result_topics AS SELECT * FROM ${stats_db_name}.publication_topics UNION ALL SELECT * FROM ${stats_db_name}.software_topics UNION ALL SELECT * FROM ${stats_db_name}.dataset_topics UNION ALL SELECT * FROM ${stats_db_name}.otherresearchproduct_topics;
DROP TABLE IF EXISTS ${stats_db_name}.result_organization;
CREATE TABLE ${stats_db_name}.result_organization AS SELECT substr(r.target, 4) AS id, substr(r.source, 4) AS organization FROM ${openaire_db_name}.relation r WHERE r.reltype='resultOrganization';
DROP TABLE IF EXISTS ${stats_db_name}.result_projects;
CREATE TABLE ${stats_db_name}.result_projects AS select pr.result AS id, pr.id AS project, datediff(p.enddate, p.startdate) AS daysfromend FROM ${stats_db_name}.result r JOIN ${stats_db_name}.project_results pr ON r.id=pr.result JOIN ${stats_db_name}.project_tmp p ON p.id=pr.id;

43
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step9.sql
View File

@ -6,28 +6,31 @@
-- Datasource table creation & update
-------------------------------------
-- Creating and populating temporary datasource table
DROP TABLE IF EXISTS ${hive_db_name}.datasource_tmp;
create table ${hive_db_name}.datasource_tmp(`id` string, `name` string, `type` string, `dateofvalidation` string, `yearofvalidation` string, `harvested` boolean, `piwik_id` int, `latitude` string, `longitude` string, `websiteurl` string, `compatibility` string) clustered by (id) into 100 buckets stored as orc tblproperties('transactional'='true');
insert into ${hive_db_name}.datasource_tmp SELECT substr(d.id, 4) as id, officialname.value as name, datasourcetype.classname as type, dateofvalidation.value as dateofvalidation, date_format(d.dateofvalidation.value,'yyyy') as yearofvalidation, false as harvested, 0 as piwik_id, d.latitude.value as latitude, d.longitude.value as longitude, d.websiteurl.value as websiteurl, d.openairecompatibility.classid as compatibility
from ${hive_source_db_name}.datasource d
WHERE d.datainfo.deletedbyinference=false;
DROP TABLE IF EXISTS ${stats_db_name}.datasource_tmp;
CREATE TABLE ${stats_db_name}.datasource_tmp(`id` string, `name` STRING, `type` STRING, `dateofvalidation` STRING, `yearofvalidation` string, `harvested` BOOLEAN, `piwik_id` INT, `latitude` STRING, `longitude`STRING, `websiteurl` STRING, `compatibility` STRING) CLUSTERED BY (id) INTO 100 buckets stored AS orc tblproperties('transactional'='true');
-- Updating temporary table with everything that is not based on results -> This is done with the following "dual" table. To see if default values are there
INSERT INTO ${stats_db_name}.datasource_tmp SELECT substr(d.id, 4) AS id, officialname.value AS name, datasourcetype.classname AS type, dateofvalidation.value AS dateofvalidation, date_format(d.dateofvalidation.value,'yyyy') AS yearofvalidation, FALSE AS harvested, 0 AS piwik_id, d.latitude.value AS latitude, d.longitude.value AS longitude, d.websiteurl.value AS websiteurl, d.openairecompatibility.classid AS compatibility
FROM ${openaire_db_name}.datasource d
WHERE d.datainfo.deletedbyinference=FALSE;
-- Updating temporary table with everything that is not based on results -> This is done with the following "dual" table.
-- Creating a temporary dual table that will be removed after the following insert
CREATE TABLE ${hive_db_name}.dual(dummy char(1));
INSERT INTO ${hive_db_name}.dual values('X');
INSERT INTO ${hive_db_name}.datasource_tmp (`id`, `name`, `type`, `dateofvalidation`, `yearofvalidation`, `harvested`, `piwik_id`, `latitude`, `longitude`, `websiteurl`, `compatibility`)
SELECT 'other', 'Other', 'Repository', NULL, NULL, false, 0, NULL, NULL, NULL, 'unknown' FROM ${hive_db_name}.dual WHERE 'other' not in (SELECT id FROM ${hive_db_name}.datasource_tmp WHERE name='Unknown Repository');
DROP TABLE ${hive_db_name}.dual;
CREATE TABLE ${stats_db_name}.dual(dummy CHAR(1));
INSERT INTO ${stats_db_name}.dual VALUES('X');
INSERT INTO ${stats_db_name}.datasource_tmp (`id`, `name`, `type`, `dateofvalidation`, `yearofvalidation`, `harvested`, `piwik_id`, `latitude`, `longitude`, `websiteurl`, `compatibility`)
SELECT 'other', 'Other', 'Repository', NULL, NULL, false, 0, NULL, NULL, NULL, 'unknown' FROM ${stats_db_name}.dual WHERE 'other' not in (SELECT id FROM ${stats_db_name}.datasource_tmp WHERE name='Unknown Repository');
DROP TABLE ${stats_db_name}.dual;
UPDATE ${hive_db_name}.datasource_tmp SET name='Other' where name='Unknown Repository';
UPDATE ${hive_db_name}.datasource_tmp SET yearofvalidation=null WHERE yearofvalidation='-1';
UPDATE ${stats_db_name}.datasource_tmp SET name='Other' WHERE name='Unknown Repository';
UPDATE ${stats_db_name}.datasource_tmp SET yearofvalidation=null WHERE yearofvalidation='-1';
DROP TABLE IF EXISTS ${hive_db_name}.datasource_languages;
CREATE TABLE ${hive_db_name}.datasource_languages AS SELECT substr(d.id, 4) as id, langs.languages as language from ${hive_source_db_name}.datasource d LATERAL VIEW explode(d.odlanguages.value) langs as languages;
DROP TABLE IF EXISTS ${hive_db_name}.datasource_oids;
CREATE TABLE ${hive_db_name}.datasource_oids AS SELECT substr(d.id, 4) as id, oids.ids as oid from ${hive_source_db_name}.datasource d LATERAL VIEW explode(d.originalid) oids as ids;
DROP TABLE IF EXISTS ${hive_db_name}.datasource_organizations;
CREATE TABLE ${hive_db_name}.datasource_organizations AS select substr(r.target, 4) as id, substr(r.source, 4) as organization from ${hive_source_db_name}.relation r where r.reltype='datasourceOrganization';
DROP TABLE IF EXISTS ${stats_db_name}.datasource_languages;
CREATE TABLE ${stats_db_name}.datasource_languages AS SELECT substr(d.id, 4) AS id, langs.languages AS language FROM ${openaire_db_name}.datasource d LATERAL VIEW explode(d.odlanguages.value) langs AS languages;
CREATE OR REPLACE VIEW ${hive_db_name}.datasource_results AS SELECT datasource AS id, id AS result FROM ${hive_db_name}.result_datasources;
DROP TABLE IF EXISTS ${stats_db_name}.datasource_oids;
CREATE TABLE ${stats_db_name}.datasource_oids AS SELECT substr(d.id, 4) AS id, oids.ids AS oid FROM ${openaire_db_name}.datasource d LATERAL VIEW explode(d.originalid) oids AS ids;
DROP TABLE IF EXISTS ${stats_db_name}.datasource_organizations;
CREATE TABLE ${stats_db_name}.datasource_organizations AS SELECT substr(r.target, 4) AS id, substr(r.source, 4) AS organization FROM ${openaire_db_name}.relation r WHERE r.reltype='datasourceOrganization';
CREATE OR REPLACE VIEW ${stats_db_name}.datasource_results AS SELECT datasource AS id, id AS result FROM ${stats_db_name}.result_datasources;

2
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step9_6.sql
View File

@ -1,2 +0,0 @@
DROP TABLE IF EXISTS ${stats_db_name}.datasource_languages;
CREATE TABLE ${stats_db_name}.datasource_languages AS SELECT substr(d.id, 4) as id, langs.languages as language from ${openaire_db_name}.datasource d LATERAL VIEW explode(d.odlanguages.value) langs as languages;

266
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/workflow.xml
View File

@ -1,12 +1,25 @@
<workflow-app name="import_infospace_graph" xmlns="uri:oozie:workflow:0.5">
<workflow-app name="graph_stats" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>hive_db_name</name>
<description>the target hive database name</description>
<name>stats_db_name</name>
<description>the target stats database name</description>
</property>
<property>
<name>hive_source_db_name</name>
<description>the source hive database name</description>
<name>openaire_db_name</name>
<description>the original graph database name</description>
</property>
<property>
<name>external_stats_db_name</name>
<value>stats_ext</value>
<description>the external stats that should be added since they are not included in the graph database</description>
</property>
<property>
<name>hive_metastore_uris</name>
<description>hive server metastore URIs</description>
</property>
<property>
<name>hive_jdbc_url</name>
<description>hive server jdbc url</description>
</property>
<property>
<name>hive_timeout</name>
@ -20,7 +33,7 @@
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hiveMetastoreUris}</value>
<value>${hive_metastore_uris}</value>
</property>
<property>
<name>hive.txn.timeout</name>
@ -29,271 +42,170 @@
</configuration>
</global>
<start to="Step1-CreateDB"/>
<start to="Step1"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="Step1-CreateDB">
<action name="Step1">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
</configuration>
<jdbc-url>${hive_jdbc_url}/${hive_db_name}</jdbc-url>
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step1.sql</script>
<param>hive_db_name=${hive_db_name}</param>
<param>hive_source_db_name=${hive_source_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step2"/>
<error to="Kill"/>
</action>
<action name="Step2">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
</configuration>
<jdbc-url>${hive_jdbc_url}/${hive_db_name}</jdbc-url>
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step2.sql</script>
<param>hive_db_name=${hive_db_name}</param>
<param>hive_source_db_name=${hive_source_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step3"/>
<error to="Kill"/>
</action>
<action name="Step3">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
</configuration>
<jdbc-url>${hive_jdbc_url}/${hive_db_name}</jdbc-url>
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step3.sql</script>
<param>hive_db_name=${hive_db_name}</param>
<param>hive_source_db_name=${hive_source_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step4"/>
<!-- <ok to="End"/> -->
<error to="Kill"/>
</action>
<action name="Step4">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
<property>
<name>hive.variable.substitute</name>
<value>true</value>
</property>
</configuration>
<jdbc-url>${hive_jdbc_url}/${hive_db_name}</jdbc-url>
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step4.sql</script>
<param>hive_db_name=${hive_db_name}</param>
<param>hive_source_db_name=${hive_source_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step5"/>
<error to="Kill"/>
</action>
<action name="Step5">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
</configuration>
<jdbc-url>${hive_jdbc_url}/${hive_db_name}</jdbc-url>
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step5.sql</script>
<param>hive_db_name=${hive_db_name}</param>
<param>hive_source_db_name=${hive_source_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step6"/>
<error to="Kill"/>
</action>
<action name="Step6">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
</configuration>
<jdbc-url>${hive_jdbc_url}/${hive_db_name}</jdbc-url>
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step6.sql</script>
<param>hive_db_name=${hive_db_name}</param>
<param>hive_source_db_name=${hive_source_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step7"/>
<error to="Kill"/>
</action>
<action name="Step7">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
</configuration>
<jdbc-url>${hive_jdbc_url}/${hive_db_name}</jdbc-url>
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step7.sql</script>
<param>hive_db_name=${hive_db_name}</param>
<param>hive_source_db_name=${hive_source_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step8"/>
<error to="Kill"/>
</action>
<action name="Step8">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
</configuration>
<jdbc-url>${hive_jdbc_url}/${hive_db_name}</jdbc-url>
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step8.sql</script>
<param>hive_db_name=${hive_db_name}</param>
<param>hive_source_db_name=${hive_source_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step9"/>
<error to="Kill"/>
</action>
<action name="Step9">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
</configuration>
<jdbc-url>${hive_jdbc_url}/${hive_db_name}</jdbc-url>
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step9.sql</script>
<param>hive_db_name=${hive_db_name}</param>
<param>hive_source_db_name=${hive_source_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step10"/>
<error to="Kill"/>
</action>
<action name="Step10">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
</configuration>
<jdbc-url>${hive_jdbc_url}/${hive_db_name}</jdbc-url>
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step10.sql</script>
<param>hive_db_name=${hive_db_name}</param>
<param>hive_source_db_name=${hive_source_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step11"/>
<error to="Kill"/>
</action>
<action name="Step11">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
</configuration>
<jdbc-url>${hive_jdbc_url}/${hive_db_name}</jdbc-url>
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step11.sql</script>
<param>hive_db_name=${hive_db_name}</param>
<param>hive_source_db_name=${hive_source_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>external_stats_db_name=${external_stats_db_name}</param>
</hive2>
<ok to="Step12"/>
<error to="Kill"/>
</action>
</action>
<action name="Step12">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
</configuration>
<jdbc-url>${hive_jdbc_url}/${hive_db_name}</jdbc-url>
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step12.sql</script>
<param>hive_db_name=${hive_db_name}</param>
<param>hive_source_db_name=${hive_source_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>external_stats_db_name=${external_stats_db_name}</param>
</hive2>
<ok to="Step13"/>
<error to="Kill"/>
</action>
</action>
<action name="Step13">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
</configuration>
<jdbc-url>${hive_jdbc_url}/${hive_db_name}</jdbc-url>
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step13.sql</script>
<param>hive_db_name=${hive_db_name}</param>
<param>hive_source_db_name=${hive_source_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step14"/>
<error to="Kill"/>
</action>
<action name="Step14">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<!-- <configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
</configuration> -->
<exec>impala-shell.sh</exec>
<argument>${hive_db_name}</argument>
<argument>impala-stats.sql</argument>
<argument>/user/giorgos.alexiou/graph/oozie_app/impala-stats.sql</argument>
<argument>${stats_db_name}</argument>
<argument>step14.sql</argument>
<argument>/user/${wf:user()}/oa/graph/stats/oozie_app/scripts/step14.sql</argument>
<file>impala-shell.sh</file>
<!-- <param>hive_db_name=${hive_db_name}</param> -->
</shell>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>