From 40b98d8182957618e2c0024ea2c3abe4c4a2d165 Mon Sep 17 00:00:00 2001 From: dimitrispie Date: Fri, 22 Dec 2023 10:29:20 +0200 Subject: [PATCH 01/22] Changes to indicators and funders definition - Changes result_refereed definition - Added result_country indicator - Added indi_pub_green_with_license indicator - Added country from jurisdiction to funders --- .../graph/stats/oozie_app/scripts/step15.sql | 96 ++++++++++++++++--- .../scripts/step16-createIndicatorsTables.sql | 19 +++- .../scripts/step20-createMonitorDB.sql | 4 +- .../scripts/step20-createMonitorDBAll.sql | 7 +- .../step20-createMonitorDB_institutions.sql | 3 +- .../graph/stats/oozie_app/scripts/step6.sql | 3 +- 6 files changed, 112 insertions(+), 20 deletions(-) diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15.sql index 066b197e6..343438cf4 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15.sql @@ -8,31 +8,103 @@ DROP TABLE IF EXISTS ${stats_db_name}.publication_refereed purge; +--CREATE TABLE IF NOT EXISTS ${stats_db_name}.publication_refereed STORED AS PARQUET as +--select substr(r.id, 4) as id, inst.refereed.classname as refereed +--from ${openaire_db_name}.publication r lateral view explode(r.instance) instances as inst +--where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE; CREATE TABLE IF NOT EXISTS ${stats_db_name}.publication_refereed STORED AS PARQUET as -select substr(r.id, 4) as id, inst.refereed.classname as refereed -from ${openaire_db_name}.publication r lateral view explode(r.instance) instances as inst -where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE; +with peer_reviewed as +(select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed +from ${openaire_db_name}.publication r lateral +view explode(r.instance) instances as inst +where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE +and inst.refereed.classname='peerReviewed'), +non_peer_reviewed as +(select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed +from ${openaire_db_name}.publication r lateral +view explode(r.instance) instances as inst +where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE +and inst.refereed.classname='nonPeerReviewed') +select peer_reviewed.* from peer_reviewed +union all +select non_peer_reviewed.* from non_peer_reviewed +left join peer_reviewed on peer_reviewed.id=non_peer_reviewed.id +where peer_reviewed.id is null; DROP TABLE IF EXISTS ${stats_db_name}.dataset_refereed purge; +--CREATE TABLE IF NOT EXISTS ${stats_db_name}.dataset_refereed STORED AS PARQUET as +--select substr(r.id, 4) as id, inst.refereed.classname as refereed +--from ${openaire_db_name}.dataset r lateral view explode(r.instance) instances as inst +--where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE; CREATE TABLE IF NOT EXISTS ${stats_db_name}.dataset_refereed STORED AS PARQUET as -select substr(r.id, 4) as id, inst.refereed.classname as refereed -from ${openaire_db_name}.dataset r lateral view explode(r.instance) instances as inst -where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE; +with peer_reviewed as +(select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed +from ${openaire_db_name}.dataset r lateral +view explode(r.instance) instances as inst +where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE +and inst.refereed.classname='peerReviewed'), +non_peer_reviewed as +(select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed +from ${openaire_db_name}.dataset r lateral +view explode(r.instance) instances as inst +where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE +and inst.refereed.classname='nonPeerReviewed') +select peer_reviewed.* from peer_reviewed +union all +select non_peer_reviewed.* from non_peer_reviewed +left join peer_reviewed on peer_reviewed.id=non_peer_reviewed.id +where peer_reviewed.id is null; DROP TABLE IF EXISTS ${stats_db_name}.software_refereed purge; +--CREATE TABLE IF NOT EXISTS ${stats_db_name}.software_refereed STORED AS PARQUET as +--select substr(r.id, 4) as id, inst.refereed.classname as refereed +--from ${openaire_db_name}.software r lateral view explode(r.instance) instances as inst +--where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE; CREATE TABLE IF NOT EXISTS ${stats_db_name}.software_refereed STORED AS PARQUET as -select substr(r.id, 4) as id, inst.refereed.classname as refereed -from ${openaire_db_name}.software r lateral view explode(r.instance) instances as inst -where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE; +with peer_reviewed as +(select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed +from ${openaire_db_name}.software r lateral +view explode(r.instance) instances as inst +where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE +and inst.refereed.classname='peerReviewed'), +non_peer_reviewed as +(select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed +from ${openaire_db_name}.software r lateral +view explode(r.instance) instances as inst +where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE +and inst.refereed.classname='nonPeerReviewed') +select peer_reviewed.* from peer_reviewed +union all +select non_peer_reviewed.* from non_peer_reviewed +left join peer_reviewed on peer_reviewed.id=non_peer_reviewed.id +where peer_reviewed.id is null; DROP TABLE IF EXISTS ${stats_db_name}.otherresearchproduct_refereed purge; +--CREATE TABLE IF NOT EXISTS ${stats_db_name}.otherresearchproduct_refereed STORED AS PARQUET as +--select substr(r.id, 4) as id, inst.refereed.classname as refereed +--from ${openaire_db_name}.otherresearchproduct r lateral view explode(r.instance) instances as inst +--where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE; CREATE TABLE IF NOT EXISTS ${stats_db_name}.otherresearchproduct_refereed STORED AS PARQUET as -select substr(r.id, 4) as id, inst.refereed.classname as refereed -from ${openaire_db_name}.otherresearchproduct r lateral view explode(r.instance) instances as inst -where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE; +with peer_reviewed as +(select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed +from ${openaire_db_name}.otherresearchproduct r lateral +view explode(r.instance) instances as inst +where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE +and inst.refereed.classname='peerReviewed'), +non_peer_reviewed as +(select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed +from ${openaire_db_name}.otherresearchproduct r lateral +view explode(r.instance) instances as inst +where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE +and inst.refereed.classname='nonPeerReviewed') +select peer_reviewed.* from peer_reviewed +union all +select non_peer_reviewed.* from non_peer_reviewed +left join peer_reviewed on peer_reviewed.id=non_peer_reviewed.id +where peer_reviewed.id is null; CREATE VIEW IF NOT EXISTS ${stats_db_name}.result_refereed as select * from ${stats_db_name}.publication_refereed diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql index 5aa14e2c2..f8213030e 100755 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql @@ -1195,6 +1195,21 @@ and pf.publicly_funded='yes') foo) select distinct p.id, coalesce(publicly_funded, 0) as publicly_funded from ${stats_db_name}.publication p left outer join ( -select distinct ro.id, 1 as publicly_funded from result_organization ro +select distinct ro.id, 1 as publicly_funded from ${stats_db_name}.result_organization ro join ${stats_db_name}.organization o on o.id=ro.organization -join publicly_funded_orgs pfo on o.name=pfo.name) tmp on p.id=tmp.id; \ No newline at end of file +join publicly_funded_orgs pfo on o.name=pfo.name) tmp on p.id=tmp.id; + +create table ${stats_db_name}.indi_pub_green_with_license stored as parquet as +select distinct p.id, coalesce(green_with_license, 0) as green_with_license +from ${stats_db_name}.publication p +left outer join ( +select distinct p.id, 1 as green_with_license from ${stats_db_name}.publication p +join ${stats_db_name}.result_instance ri on ri.id = p.id +join ${stats_db_name}.datasource on datasource.id = ri.hostedby +where ri.license is not null and datasource.type like '%Repository%' and datasource.name!='Other') tmp +on p.id= tmp.id; + +create table ${stats_db_name}.result_country stored as parquet as +select distinct ro.id, o.country from ${stats_db_name}.result_organization ro +join ${stats_db_name}.organization o on o.id=ro.organization +join ${stats_db_name}.funder f on f.country=o.country; \ No newline at end of file diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB.sql index cc8348f26..a99a75798 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB.sql @@ -180,4 +180,6 @@ create view TARGET.indi_funder_openess as select * from SOURCE.indi_funder_opene create view TARGET.indi_funder_findable as select * from SOURCE.indi_funder_findable; create view TARGET.indi_ris_fairness as select * from SOURCE.indi_ris_fairness; create view TARGET.indi_ris_openess as select * from SOURCE.indi_ris_openess; -create view TARGET.indi_ris_findable as select * from SOURCE.indi_ris_findable; \ No newline at end of file +create view TARGET.indi_ris_findable as select * from SOURCE.indi_ris_findable; +create table TARGET.indi_pub_green_with_license stored as parquet as select * from SOURCE.indi_pub_green_with_license orig where exists (select 1 from TARGET.result r where r.id=orig.id); +create table TARGET.result_country stored as parquet as select * from SOURCE.result_country orig where exists (select 1 from TARGET.result r where r.id=orig.id); diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDBAll.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDBAll.sql index 42812d159..671a9ea9e 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDBAll.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDBAll.sql @@ -80,8 +80,8 @@ create table TARGET.result stored as parquet as 'openorgs____::0fccc7640f0cb44d5cd1b06b312a06b9', -- Cardiff University 'openorgs____::8839b55dae0c84d56fd533f52d5d483a', -- Leibniz Institute of Ecological Urban and Regional Development 'openorgs____::526468206bca24c1c90da6a312295cf4', -- Cyprus University of Technology - 'openorgs____::b5ca9d4340e26454e367e2908ef3872f' -- Alma Mater Studiorum University of Bologna - + 'openorgs____::b5ca9d4340e26454e367e2908ef3872f', -- Alma Mater Studiorum University of Bologna + 'openorgs____::a6340e6ecf60f6bba163659df985b0f2' -- TU Dresden ) )) foo; create view if not exists TARGET.category as select * from SOURCE.category; @@ -263,5 +263,6 @@ create view TARGET.indi_funder_findable as select * from SOURCE.indi_funder_find create view TARGET.indi_ris_fairness as select * from SOURCE.indi_ris_fairness; create view TARGET.indi_ris_openess as select * from SOURCE.indi_ris_openess; create view TARGET.indi_ris_findable as select * from SOURCE.indi_ris_findable; - +create table TARGET.indi_pub_green_with_license stored as parquet as select * from SOURCE.indi_pub_green_with_license orig where exists (select 1 from TARGET.result r where r.id=orig.id); +create table TARGET.result_country stored as parquet as select * from SOURCE.result_country orig where exists (select 1 from TARGET.result r where r.id=orig.id); diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB_institutions.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB_institutions.sql index 2c0ac337c..5c347eb6b 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB_institutions.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB_institutions.sql @@ -60,5 +60,6 @@ create table TARGET.result stored as parquet as 'openorgs____::0fccc7640f0cb44d5cd1b06b312a06b9', -- Cardiff University 'openorgs____::8839b55dae0c84d56fd533f52d5d483a', -- Leibniz Institute of Ecological Urban and Regional Development 'openorgs____::526468206bca24c1c90da6a312295cf4', -- Cyprus University of Technology - 'openorgs____::b5ca9d4340e26454e367e2908ef3872f' -- Alma Mater Studiorum University of Bologna + 'openorgs____::b5ca9d4340e26454e367e2908ef3872f', -- Alma Mater Studiorum University of Bologna + 'openorgs____::a6340e6ecf60f6bba163659df985b0f2' -- TU Dresden ))) foo; \ No newline at end of file diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step6.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step6.sql index e5b3f504e..165f77946 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step6.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step6.sql @@ -95,7 +95,8 @@ DROP TABLE IF EXISTS ${stats_db_name}.funder purge; create table ${stats_db_name}.funder STORED AS PARQUET as select distinct xpath_string(fund, '//funder/id') as id, xpath_string(fund, '//funder/name') as name, - xpath_string(fund, '//funder/shortname') as shortname + xpath_string(fund, '//funder/shortname') as shortname, + xpath_string(fundingtree[0].value, '//funder/jurisdiction') as country from ${openaire_db_name}.project p lateral view explode(p.fundingtree.value) fundingtree as fund; DROP TABLE IF EXISTS ${stats_db_name}.project_organization_contribution purge; -- 2.17.1 From ffdd03d2f4aed561dd8753a2595e729e0fb64ffe Mon Sep 17 00:00:00 2001 From: dimitrispie Date: Fri, 22 Dec 2023 11:05:24 +0200 Subject: [PATCH 02/22] Monitor Irish Stats WF Parameters (with examples): stats_db_name=openaire_beta_stats_20231208 monitor_irish_db_name=openaire_beta_stats_monitor_ie_20231208b monitor_irish_db_prod_name=openaire_beta_stats_monitor_ie graph_db_name=openaire_beta_20231208 monitor_irish_db_shadow_name=openaire_beta_stats_monitor_ie_shadow hive_timeout=150000 hadoop_user_name=dnet.beta resumeFrom=Step1-buildIrishMonitorDB --- dhp-workflows/dhp-stats-monitor-irish/pom.xml | 32 +++ .../oozie_app/config-default.xml | 30 +++ .../oozie_app/copyDataToImpalaCluster.sh | 70 ++++++ .../oozie_app/finalizeImpalaCluster.sh | 23 ++ .../oozie_app/monitor_irish.sh | 28 +++ .../oozie_app/scripts/buildIrishMonitorDB.sql | 224 ++++++++++++++++++ .../oozie_app/workflow.xml | 118 +++++++++ 7 files changed, 525 insertions(+) create mode 100644 dhp-workflows/dhp-stats-monitor-irish/pom.xml create mode 100644 dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/config-default.xml create mode 100644 dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/copyDataToImpalaCluster.sh create mode 100644 dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/finalizeImpalaCluster.sh create mode 100644 dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/monitor_irish.sh create mode 100644 dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/scripts/buildIrishMonitorDB.sql create mode 100644 dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/workflow.xml diff --git a/dhp-workflows/dhp-stats-monitor-irish/pom.xml b/dhp-workflows/dhp-stats-monitor-irish/pom.xml new file mode 100644 index 000000000..9d850a4bb --- /dev/null +++ b/dhp-workflows/dhp-stats-monitor-irish/pom.xml @@ -0,0 +1,32 @@ + + + + dhp-workflows + eu.dnetlib.dhp + 1.2.4-SNAPSHOT + + 4.0.0 + dhp-stats-monitor-irish + + + org.apache.spark + spark-core_2.11 + + + org.apache.spark + spark-sql_2.11 + + + + + + pl.project13.maven + git-commit-id-plugin + 2.1.11 + + false + + + + + diff --git a/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/config-default.xml b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/config-default.xml new file mode 100644 index 000000000..b2a1322e6 --- /dev/null +++ b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/config-default.xml @@ -0,0 +1,30 @@ + + + jobTracker + ${jobTracker} + + + nameNode + ${nameNode} + + + oozie.use.system.libpath + true + + + oozie.action.sharelib.for.spark + spark2 + + + hive_metastore_uris + thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083 + + + hive_jdbc_url + jdbc:hive2://iis-cdh5-test-m3.ocean.icm.edu.pl:10000/;UseNativeQuery=1;?spark.executor.memory=22166291558;spark.yarn.executor.memoryOverhead=3225;spark.driver.memory=15596411699;spark.yarn.driver.memoryOverhead=1228 + + + oozie.wf.workflow.notification.url + {serviceUrl}/v1/oozieNotification/jobUpdate?jobId=$jobId%26status=$status + + \ No newline at end of file diff --git a/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/copyDataToImpalaCluster.sh b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/copyDataToImpalaCluster.sh new file mode 100644 index 000000000..78c44ed20 --- /dev/null +++ b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/copyDataToImpalaCluster.sh @@ -0,0 +1,70 @@ +export PYTHON_EGG_CACHE=/home/$(whoami)/.python-eggs +export link_folder=/tmp/impala-shell-python-egg-cache-$(whoami) +if ! [ -L $link_folder ] +then + rm -Rf "$link_folder" + ln -sfn ${PYTHON_EGG_CACHE}${link_folder} ${link_folder} +fi + +export HADOOP_USER_NAME=$2 + +function copydb() { + + db=$1 + FILE=("hive_wf_tmp_"$RANDOM) + hdfs dfs -mkdir hdfs://impala-cluster-mn1.openaire.eu:8020/tmp/$FILE/ + + # change ownership to impala +# hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db + hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/ + + + # copy the databases from ocean to impala + echo "copying $db" + hadoop distcp -Dmapreduce.map.memory.mb=6144 -pb hdfs://nameservice1/user/hive/warehouse/${db}.db hdfs://impala-cluster-mn1.openaire.eu:8020/tmp/$FILE/ + + hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db + + # drop tables from db + for i in `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} --delimited -q "show tables"`; + do + `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} -q "drop table $i;"`; + done + + # drop views from db + for i in `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} --delimited -q "show tables"`; + do + `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} -q "drop view $i;"`; + done + + # delete the database + impala-shell -i impala-cluster-dn1.openaire.eu -q "drop database if exists ${db} cascade"; + + # create the databases + impala-shell -i impala-cluster-dn1.openaire.eu -q "create database ${db}"; + + impala-shell -q "INVALIDATE METADATA" + echo "creating schema for ${db}" + for (( k = 0; k < 5; k ++ )); do + for i in `impala-shell -d ${db} --delimited -q "show tables"`; + do + impala-shell -d ${db} --delimited -q "show create table $i"; + done | sed 's/"$/;/' | sed 's/^"//' | sed 's/[[:space:]]\date[[:space:]]/`date`/g' | impala-shell --user $HADOOP_USER_NAME -i impala-cluster-dn1.openaire.eu -c -f - + done + + # load the data from /tmp in the respective tables + echo "copying data in tables and computing stats" + for i in `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} --delimited -q "show tables"`; + do + impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} -q "load data inpath '/tmp/$FILE/${db}.db/$i' into table $i"; + impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} -q "compute stats $i"; + done + + # deleting the remaining directory from hdfs +hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -rm -R /tmp/$FILE/${db}.db +} + +MONITOR_DB=$1 +#HADOOP_USER_NAME=$2 +copydb $MONITOR_DB + diff --git a/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/finalizeImpalaCluster.sh b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/finalizeImpalaCluster.sh new file mode 100644 index 000000000..38a2f61bc --- /dev/null +++ b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/finalizeImpalaCluster.sh @@ -0,0 +1,23 @@ +export PYTHON_EGG_CACHE=/home/$(whoami)/.python-eggs +export link_folder=/tmp/impala-shell-python-egg-cache-$(whoami) +if ! [ -L $link_folder ] +then + rm -Rf "$link_folder" + ln -sfn ${PYTHON_EGG_CACHE}${link_folder} ${link_folder} +fi + +SOURCE=$1 +PRODUCTION=$2 +echo ${SOURCE} +echo ${PRODUCTION} + +#echo "Updating ${PRODUCTION} monitor database old cluster" +#impala-shell -q "create database if not exists ${PRODUCTION}" +#impala-shell -d ${PRODUCTION} -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}./" | sed "s/$/;/" | impala-shell -c -f - +#impala-shell -d ${SOURCE} -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}.\1 as select * from ${SOURCE}.\1;/" | impala-shell -c -f - + +echo "Updating ${PRODUCTION} monitor database" +impala-shell -i impala-cluster-dn1.openaire.eu -q "create database if not exists ${PRODUCTION}" +impala-shell -i impala-cluster-dn1.openaire.eu -d ${PRODUCTION} -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}./" | sed "s/$/;/" | impala-shell -i impala-cluster-dn1.openaire.eu -c -f - +impala-shell -i impala-cluster-dn1.openaire.eu -d ${SOURCE} -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}.\1 as select * from ${SOURCE}.\1;/" | impala-shell -i impala-cluster-dn1.openaire.eu -c -f - +echo "Production monitor db ready!" diff --git a/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/monitor_irish.sh b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/monitor_irish.sh new file mode 100644 index 000000000..27e399e98 --- /dev/null +++ b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/monitor_irish.sh @@ -0,0 +1,28 @@ +export PYTHON_EGG_CACHE=/home/$(whoami)/.python-eggs +export link_folder=/tmp/impala-shell-python-egg-cache-$(whoami) +if ! [ -L $link_folder ] +then + rm -Rf "$link_folder" + ln -sfn ${PYTHON_EGG_CACHE}${link_folder} ${link_folder} +fi + +export SOURCE=$1 +export TARGET=$2 +export SHADOW=$3 +export SCRIPT_PATH=$4 +export GRAPHDB=$5 + + +export HIVE_OPTS="-hiveconf mapred.job.queue.name=analytics -hiveconf hive.spark.client.connect.timeout=120000ms -hiveconf hive.spark.client.server.connect.timeout=300000ms -hiveconf spark.executor.memory=19166291558 -hiveconf spark.yarn.executor.memoryOverhead=3225 -hiveconf spark.driver.memory=11596411699 -hiveconf spark.yarn.driver.memoryOverhead=1228" +export HADOOP_USER_NAME="oozie" + +echo "Getting file from " $4 +hdfs dfs -copyToLocal $4 + +#update Monitor DB IRISH +#cat CreateDB.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2/g1" | sed "s/GRAPHDB/$3/g1" > foo +cat buildIrishMonitorDB.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2/g1" | sed "s/GRAPHDB/$5/g1" > foo +hive $HIVE_OPTS -f foo + +echo "Hive shell finished" + diff --git a/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/scripts/buildIrishMonitorDB.sql b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/scripts/buildIrishMonitorDB.sql new file mode 100644 index 000000000..db76fdab3 --- /dev/null +++ b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/scripts/buildIrishMonitorDB.sql @@ -0,0 +1,224 @@ +drop database if exists TARGET cascade; +create database if not exists TARGET; + +create view if not exists TARGET.category as select * from SOURCE.category; +create view if not exists TARGET.concept as select * from SOURCE.concept; +create view if not exists TARGET.context as select * from SOURCE.context; +create view if not exists TARGET.country as select * from SOURCE.country; +create view if not exists TARGET.countrygdp as select * from SOURCE.countrygdp; +create view if not exists TARGET.creation_date as select * from SOURCE.creation_date; +--create view if not exists TARGET.funder as select * from SOURCE.funder; +create view if not exists TARGET.fundref as select * from SOURCE.fundref; +create view if not exists TARGET.rndexpenditure as select * from SOURCE.rndexpediture; +create view if not exists TARGET.rndgdpexpenditure as select * from SOURCE.rndgdpexpenditure; +create view if not exists TARGET.doctoratestudents as select * from SOURCE.doctoratestudents; +create view if not exists TARGET.totalresearchers as select * from SOURCE.totalresearchers; +create view if not exists TARGET.totalresearchersft as select * from SOURCE.totalresearchersft; +create view if not exists TARGET.hrrst as select * from SOURCE.hrrst; + +drop table if exists TARGET.irish_funders; + +create TEMPORARY table TARGET.irish_funders as +select distinct xpath_string(fundingtree[0].value, '//funder/name') as funder from GRAPHDB.project + where xpath_string(fundingtree[0].value, '//funder/jurisdiction')='IE'; +--create TEMPORARY table TARGET.irish_funders as +--select distinct name as funder from SOURCE.fundref where country='IE'; + +drop table if exists TARGET.result; + +create table TARGET.result stored as parquet as + select distinct * from SOURCE.result r where r.id in (select ro.id from SOURCE.result_organization ro + left outer join SOURCE.organization o on o.id=ro.organization + left outer join SOURCE.result_projects rp on rp.id=ro.id + left outer join SOURCE.project p on p.id=rp.project + join TARGET.irish_funders irf + where o.country='IE' or p.funder=irf.funder); + +create view if not exists TARGET.category as select * from SOURCE.category; +create view if not exists TARGET.concept as select * from SOURCE.concept; +create view if not exists TARGET.context as select * from SOURCE.context; +create view if not exists TARGET.country as select * from SOURCE.country; +create view if not exists TARGET.countrygdp as select * from SOURCE.countrygdp; +create view if not exists TARGET.creation_date as select * from SOURCE.creation_date; + +create table TARGET.funder stored as parquet as select * from SOURCE.funder where country='IE'; + +create view if not exists TARGET.fundref as select * from SOURCE.fundref; +create view if not exists TARGET.rndexpenditure as select * from SOURCE.rndexpediture; +create view if not exists TARGET.rndgdpexpenditure as select * from SOURCE.rndgdpexpenditure; +create view if not exists TARGET.doctoratestudents as select * from SOURCE.doctoratestudents; +create view if not exists TARGET.totalresearchers as select * from SOURCE.totalresearchers; +create view if not exists TARGET.totalresearchersft as select * from SOURCE.totalresearchersft; +create view if not exists TARGET.hrrst as select * from SOURCE.hrrst; +--create view if not exists TARGET.graduatedoctorates as select * from SOURCE.graduatedoctorates; + +create table TARGET.result_citations stored as parquet as select * from SOURCE.result_citations orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_references_oc stored as parquet as select * from SOURCE.result_references_oc orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_citations_oc stored as parquet as select * from SOURCE.result_citations_oc orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_classifications stored as parquet as select * from SOURCE.result_classifications orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_apc stored as parquet as select * from SOURCE.result_apc orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_concepts stored as parquet as select * from SOURCE.result_concepts orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_datasources stored as parquet as select * from SOURCE.result_datasources orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_fundercount stored as parquet as select * from SOURCE.result_fundercount orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_gold stored as parquet as select * from SOURCE.result_gold orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_greenoa stored as parquet as select * from SOURCE.result_greenoa orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_languages stored as parquet as select * from SOURCE.result_languages orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_licenses stored as parquet as select * from SOURCE.result_licenses orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.licenses_normalized STORED AS PARQUET as select * from SOURCE.licenses_normalized; + +create table TARGET.result_oids stored as parquet as select * from SOURCE.result_oids orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_organization stored as parquet as select * from SOURCE.result_organization orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_peerreviewed stored as parquet as select * from SOURCE.result_peerreviewed orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_pids stored as parquet as select * from SOURCE.result_pids orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_projectcount stored as parquet as select * from SOURCE.result_projectcount orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_projects stored as parquet as select * from SOURCE.result_projects orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_refereed stored as parquet as select * from SOURCE.result_refereed orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_sources stored as parquet as select * from SOURCE.result_sources orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_topics stored as parquet as select * from SOURCE.result_topics orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_fos stored as parquet as select * from SOURCE.result_fos orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_accessroute stored as parquet as select * from SOURCE.result_accessroute orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_instance stored as parquet as select * from SOURCE.result_instance orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_orcid stored as parquet as select * from SOURCE.result_orcid orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create view TARGET.foo1 as select * from SOURCE.result_result rr where rr.source in (select id from TARGET.result); +create view TARGET.foo2 as select * from SOURCE.result_result rr where rr.target in (select id from TARGET.result); +create table TARGET.result_result STORED AS PARQUET as select distinct * from (select * from TARGET.foo1 union all select * from TARGET.foo2) foufou; +drop view TARGET.foo1; +drop view TARGET.foo2; + +-- datasources +create view if not exists TARGET.datasource as select * from SOURCE.datasource; +create view if not exists TARGET.datasource_oids as select * from SOURCE.datasource_oids; +create view if not exists TARGET.datasource_organizations as select * from SOURCE.datasource_organizations; +create view if not exists TARGET.datasource_sources as select * from SOURCE.datasource_sources; + +create table TARGET.datasource_results stored as parquet as select id as result, datasource as id from TARGET.result_datasources; + +-- organizations +create view if not exists TARGET.organization as select * from SOURCE.organization; +create view if not exists TARGET.organization_datasources as select * from SOURCE.organization_datasources; +create view if not exists TARGET.organization_pids as select * from SOURCE.organization_pids; +create view if not exists TARGET.organization_projects as select * from SOURCE.organization_projects; +create view if not exists TARGET.organization_sources as select * from SOURCE.organization_sources; + +-- projects +create view if not exists TARGET.project as select * from SOURCE.project; +create view if not exists TARGET.project_oids as select * from SOURCE.project_oids; +create view if not exists TARGET.project_organizations as select * from SOURCE.project_organizations; +create view if not exists TARGET.project_resultcount as select * from SOURCE.project_resultcount; +create view if not exists TARGET.project_classification as select * from SOURCE.project_classification; +create view if not exists TARGET.project_organization_contribution as select * from SOURCE.project_organization_contribution; + +create table TARGET.project_results stored as parquet as select id as result, project as id from TARGET.result_projects; + + +-- indicators +-- Sprint 1 ---- +create table TARGET.indi_pub_green_oa stored as parquet as select * from SOURCE.indi_pub_green_oa orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_pub_grey_lit stored as parquet as select * from SOURCE.indi_pub_grey_lit orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_pub_doi_from_crossref stored as parquet as select * from SOURCE.indi_pub_doi_from_crossref orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +-- Sprint 2 ---- +create table TARGET.indi_result_has_cc_licence stored as parquet as select * from SOURCE.indi_result_has_cc_licence orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_result_has_cc_licence_url stored as parquet as select * from SOURCE.indi_result_has_cc_licence_url orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_pub_has_abstract stored as parquet as select * from SOURCE.indi_pub_has_abstract orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_result_with_orcid stored as parquet as select * from SOURCE.indi_result_with_orcid orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +---- Sprint 3 ---- +create table TARGET.indi_funded_result_with_fundref stored as parquet as select * from SOURCE.indi_funded_result_with_fundref orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create view TARGET.indi_result_org_collab as select * from SOURCE.indi_result_org_collab; +create view TARGET.indi_result_org_country_collab as select * from SOURCE.indi_result_org_country_collab; +create view TARGET.indi_project_collab_org as select * from SOURCE.indi_project_collab_org; +create view TARGET.indi_project_collab_org_country as select * from SOURCE.indi_project_collab_org_country; +create view TARGET.indi_funder_country_collab as select * from SOURCE.indi_funder_country_collab; +create view TARGET.indi_result_country_collab as select * from SOURCE.indi_result_country_collab; +---- Sprint 4 ---- +create table TARGET.indi_pub_diamond stored as parquet as select * from SOURCE.indi_pub_diamond orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_pub_in_transformative stored as parquet as select * from SOURCE.indi_pub_in_transformative orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_pub_closed_other_open stored as parquet as select * from SOURCE.indi_pub_closed_other_open orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +---- Sprint 5 ---- +create table TARGET.indi_result_no_of_copies stored as parquet as select * from SOURCE.indi_result_no_of_copies orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +---- Sprint 6 ---- +create table TARGET.indi_pub_hybrid_oa_with_cc stored as parquet as select * from SOURCE.indi_pub_hybrid_oa_with_cc orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_pub_bronze_oa stored as parquet as select * from SOURCE.indi_pub_bronze_oa orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_pub_downloads stored as parquet as select * from SOURCE.indi_pub_downloads orig where exists (select 1 from TARGET.result r where r.id=orig.result_id); + +create table TARGET.indi_pub_downloads_datasource stored as parquet as select * from SOURCE.indi_pub_downloads_datasource orig where exists (select 1 from TARGET.result r where r.id=orig.result_id); + +create table TARGET.indi_pub_downloads_year stored as parquet as select * from SOURCE.indi_pub_downloads_year orig where exists (select 1 from TARGET.result r where r.id=orig.result_id); + +create table TARGET.indi_pub_downloads_datasource_year stored as parquet as select * from SOURCE.indi_pub_downloads_datasource_year orig where exists (select 1 from TARGET.result r where r.id=orig.result_id); + +---- Sprint 7 ---- +create table TARGET.indi_pub_gold_oa stored as parquet as select * from SOURCE.indi_pub_gold_oa orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_pub_hybrid stored as parquet as select * from SOURCE.indi_pub_hybrid orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create view TARGET.indi_org_fairness as select * from SOURCE.indi_org_fairness; +create view TARGET.indi_org_fairness_pub_pr as select * from SOURCE.indi_org_fairness_pub_pr; +create view TARGET.indi_org_fairness_pub_year as select * from SOURCE.indi_org_fairness_pub_year; +create view TARGET.indi_org_fairness_pub as select * from SOURCE.indi_org_fairness_pub; +create view TARGET.indi_org_fairness_year as select * from SOURCE.indi_org_fairness_year; +create view TARGET.indi_org_findable_year as select * from SOURCE.indi_org_findable_year; +create view TARGET.indi_org_findable as select * from SOURCE.indi_org_findable; +create view TARGET.indi_org_openess as select * from SOURCE.indi_org_openess; +create view TARGET.indi_org_openess_year as select * from SOURCE.indi_org_openess_year; +create table TARGET.indi_pub_has_preprint stored as parquet as select * from SOURCE.indi_pub_has_preprint orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_pub_in_subscribed stored as parquet as select * from SOURCE.indi_pub_in_subscribed orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_result_with_pid stored as parquet as select * from SOURCE.indi_result_with_pid orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_impact_measures stored as parquet as select * from SOURCE.indi_impact_measures orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_pub_interdisciplinarity stored as parquet as select * from SOURCE.indi_pub_interdisciplinarity orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.result_apc_affiliations stored as parquet as select * from SOURCE.result_apc_affiliations orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_is_project_result_after stored as parquet as select * from SOURCE.indi_is_project_result_after orig where exists (select 1 from TARGET.result r where r.id=orig.result_id); +create view TARGET.indi_is_funder_plan_s as select * from SOURCE.indi_is_funder_plan_s; +create view TARGET.indi_funder_fairness as select * from SOURCE.indi_funder_fairness; +create view TARGET.indi_funder_openess as select * from SOURCE.indi_funder_openess; +create view TARGET.indi_funder_findable as select * from SOURCE.indi_funder_findable; +create view TARGET.indi_ris_fairness as select * from SOURCE.indi_ris_fairness; +create view TARGET.indi_ris_openess as select * from SOURCE.indi_ris_openess; +create view TARGET.indi_ris_findable as select * from SOURCE.indi_ris_findable; +create table TARGET.indi_pub_green_with_license stored as parquet as select * from SOURCE.indi_pub_green_with_license orig where exists (select 1 from TARGET.result r where r.id=orig.id); +create table TARGET.result_country stored as parquet as select * from SOURCE.result_country orig where exists (select 1 from TARGET.result r where r.id=orig.id); +create table TARGET.indi_pub_publicly_funded stored as parquet as select * from SOURCE.indi_pub_publicly_funded orig where exists (select 1 from TARGET.result r where r.id=orig.id); \ No newline at end of file diff --git a/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/workflow.xml b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/workflow.xml new file mode 100644 index 000000000..e49552c60 --- /dev/null +++ b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/workflow.xml @@ -0,0 +1,118 @@ + + + + stats_db_name + the target stats database name + + + graph_db_name + the graph database name + + + monitor_irish_db_name + the target monitor db name + + + monitor_irish_db_prod_name + the name of the production monitor db + + + monitor_irish_db_shadow_name + the name of the shadow monitor db + + + hive_metastore_uris + hive server metastore URIs + + + hive_jdbc_url + hive server jdbc url + + + hive_timeout + the time period, in seconds, after which Hive fails a transaction if a Hive client has not sent a hearbeat. The default value is 300 seconds. + + + hadoop_user_name + user name of the wf owner + + + + + ${jobTracker} + ${nameNode} + + + hive.metastore.uris + ${hive_metastore_uris} + + + hive.txn.timeout + ${hive_timeout} + + + mapred.job.queue.name + analytics + + + + + + + + ${wf:conf('resumeFrom') eq 'Step1-buildIrishMonitorDB'} + ${wf:conf('resumeFrom') eq 'Step2-copyDataToImpalaCluster'} + ${wf:conf('resumeFrom') eq 'Step3-finalizeImpalaCluster'} + + + + + + Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}] + + + + + ${jobTracker} + ${nameNode} + monitor_irish.sh + ${stats_db_name} + ${monitor_irish_db_name} + ${monitor_irish_db_shadow_name} + ${wf:appPath()}/scripts/buildIrishMonitorDB.sql + ${graph_db_name} + monitor_irish.sh + + + + + + + + ${jobTracker} + ${nameNode} + copyDataToImpalaCluster.sh + ${monitor_irish_db_name} + ${hadoop_user_name} + copyDataToImpalaCluster.sh + + + + + + + + ${jobTracker} + ${nameNode} + finalizeImpalaCluster.sh + ${monitor_irish_db_name} + ${monitor_irish_db_prod_name} + ${monitor_irish_db_shadow_name} + finalizeImpalaCluster.sh + + + + + + + -- 2.17.1 From 75bfde043cd131c84055f0e0ca93d03e4a9e2d13 Mon Sep 17 00:00:00 2001 From: dimitrispie Date: Thu, 4 Jan 2024 15:11:04 +0200 Subject: [PATCH 03/22] Historical Snapshots Workflow Create historical snapshots db with parameters: hist_db_name=openaire_beta_historical_snapshots_xxx hist_db_name_prev=openaire_beta_historical_snapshots_xxx (previous run of wf) stats_db_name=openaire_beta_stats_xxx stats_irish_db_name=openaire_beta_stats_monitor_ie_xxx monitor_db_name=openaire_beta_stats_monitor_xxx monitor_db_prod_name=openaire_beta_stats_monitor monitor_irish_db_name=openaire_beta_stats_monitor_ie_xxx monitor_irish_db_prod_name=openaire_beta_stats_monitor_ie hist_db_prod_name=openaire_beta_historical_snapshots hist_db_shadow_name=openaire_beta_historical_snapshots_shadow hist_date=122023 hive_timeout=150000 hadoop_user_name=xxx resumeFrom=CreateDB --- dhp-workflows/dhp-stats-hist-snaps/pom.xml | 32 ++++ .../oozie_app/config-default.xml | 30 ++++ .../oozie_app/copyDataToImpalaCluster.sh | 71 ++++++++ .../oozie_app/finalizeImpalaCluster.sh | 43 +++++ .../stats-hist-snaps/oozie_app/hist_snaps.sh | 27 +++ .../oozie_app/scripts/BuildHistSnapsAll.sql | 82 +++++++++ .../oozie_app/scripts/BuildHistSnapsIrish.sql | 91 ++++++++++ .../oozie_app/scripts/CreateDB.sql | 92 ++++++++++ .../stats-hist-snaps/oozie_app/workflow.xml | 159 ++++++++++++++++++ 9 files changed, 627 insertions(+) create mode 100644 dhp-workflows/dhp-stats-hist-snaps/pom.xml create mode 100644 dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/config-default.xml create mode 100644 dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/copyDataToImpalaCluster.sh create mode 100644 dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/finalizeImpalaCluster.sh create mode 100644 dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/hist_snaps.sh create mode 100644 dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/scripts/BuildHistSnapsAll.sql create mode 100644 dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/scripts/BuildHistSnapsIrish.sql create mode 100644 dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/scripts/CreateDB.sql create mode 100644 dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/workflow.xml diff --git a/dhp-workflows/dhp-stats-hist-snaps/pom.xml b/dhp-workflows/dhp-stats-hist-snaps/pom.xml new file mode 100644 index 000000000..08ddcef49 --- /dev/null +++ b/dhp-workflows/dhp-stats-hist-snaps/pom.xml @@ -0,0 +1,32 @@ + + + + dhp-workflows + eu.dnetlib.dhp + 1.2.4-SNAPSHOT + + 4.0.0 + dhp-stats-hist-snaps + + + org.apache.spark + spark-core_2.11 + + + org.apache.spark + spark-sql_2.11 + + + + + + pl.project13.maven + git-commit-id-plugin + 2.1.11 + + false + + + + + diff --git a/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/config-default.xml b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/config-default.xml new file mode 100644 index 000000000..b2a1322e6 --- /dev/null +++ b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/config-default.xml @@ -0,0 +1,30 @@ + + + jobTracker + ${jobTracker} + + + nameNode + ${nameNode} + + + oozie.use.system.libpath + true + + + oozie.action.sharelib.for.spark + spark2 + + + hive_metastore_uris + thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083 + + + hive_jdbc_url + jdbc:hive2://iis-cdh5-test-m3.ocean.icm.edu.pl:10000/;UseNativeQuery=1;?spark.executor.memory=22166291558;spark.yarn.executor.memoryOverhead=3225;spark.driver.memory=15596411699;spark.yarn.driver.memoryOverhead=1228 + + + oozie.wf.workflow.notification.url + {serviceUrl}/v1/oozieNotification/jobUpdate?jobId=$jobId%26status=$status + + \ No newline at end of file diff --git a/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/copyDataToImpalaCluster.sh b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/copyDataToImpalaCluster.sh new file mode 100644 index 000000000..ef9bb9495 --- /dev/null +++ b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/copyDataToImpalaCluster.sh @@ -0,0 +1,71 @@ +export PYTHON_EGG_CACHE=/home/$(whoami)/.python-eggs +export link_folder=/tmp/impala-shell-python-egg-cache-$(whoami) +if ! [ -L $link_folder ] +then + rm -Rf "$link_folder" + ln -sfn ${PYTHON_EGG_CACHE}${link_folder} ${link_folder} +fi + +export HADOOP_USER_NAME=$2 + +function copydb() { + + + db=$1 + FILE=("hive_wf_tmp_"$RANDOM) + hdfs dfs -mkdir hdfs://impala-cluster-mn1.openaire.eu:8020/tmp/$FILE/ + + # change ownership to impala +# hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db + hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/ + + + # copy the databases from ocean to impala + echo "copying $db" + hadoop distcp -Dmapreduce.map.memory.mb=6144 -pb hdfs://nameservice1/user/hive/warehouse/${db}.db hdfs://impala-cluster-mn1.openaire.eu:8020/tmp/$FILE/ + + hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db + + # drop tables from db + for i in `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} --delimited -q "show tables"`; + do + `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} -q "drop table $i;"`; + done + + # drop views from db + for i in `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} --delimited -q "show tables"`; + do + `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} -q "drop view $i;"`; + done + + # delete the database + impala-shell -i impala-cluster-dn1.openaire.eu -q "drop database if exists ${db} cascade"; + + # create the databases + impala-shell -i impala-cluster-dn1.openaire.eu -q "create database ${db}"; + + impala-shell -q "INVALIDATE METADATA" + echo "creating schema for ${db}" + for (( k = 0; k < 5; k ++ )); do + for i in `impala-shell -d ${db} --delimited -q "show tables"`; + do + impala-shell -d ${db} --delimited -q "show create table $i"; + done | sed 's/"$/;/' | sed 's/^"//' | sed 's/[[:space:]]\date[[:space:]]/`date`/g' | impala-shell --user $HADOOP_USER_NAME -i impala-cluster-dn1.openaire.eu -c -f - + done + + # load the data from /tmp in the respective tables + echo "copying data in tables and computing stats" + for i in `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} --delimited -q "show tables"`; + do + impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} -q "load data inpath '/tmp/$FILE/${db}.db/$i' into table $i"; + impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} -q "compute stats $i"; + done + + # deleting the remaining directory from hdfs +hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -rm -R /tmp/$FILE/${db}.db +} + +MONITOR_DB=$1 +#HADOOP_USER_NAME=$2 +copydb $MONITOR_DB + diff --git a/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/finalizeImpalaCluster.sh b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/finalizeImpalaCluster.sh new file mode 100644 index 000000000..e2277083d --- /dev/null +++ b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/finalizeImpalaCluster.sh @@ -0,0 +1,43 @@ +export PYTHON_EGG_CACHE=/home/$(whoami)/.python-eggs +export link_folder=/tmp/impala-shell-python-egg-cache-$(whoami) +if ! [ -L $link_folder ] +then + rm -Rf "$link_folder" + ln -sfn ${PYTHON_EGG_CACHE}${link_folder} ${link_folder} +fi + +SOURCE=$1 +PRODUCTION=$2 +SHADOW=$3 +MONITOR=$4 +MONITOR_PROD=$5 +MONITOR_IRISH_PROD=$6 +MONITOR_IRISH=$7 + + +echo ${SOURCE} +echo ${PRODUCTION} + +#echo "Updating ${PRODUCTION} monitor database old cluster" +#impala-shell -q "create database if not exists ${PRODUCTION}" +#impala-shell -d ${PRODUCTION} -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}./" | sed "s/$/;/" | impala-shell -c -f - +#impala-shell -d ${SOURCE} -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}.\1 as select * from ${SOURCE}.\1;/" | impala-shell -c -f - + +echo "Updating ${PRODUCTION} historical snapshots database" +impala-shell -i impala-cluster-dn1.openaire.eu -q "create database if not exists ${PRODUCTION}" +impala-shell -i impala-cluster-dn1.openaire.eu -d ${PRODUCTION} -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}./" | sed "s/$/;/" | impala-shell -i impala-cluster-dn1.openaire.eu -c -f - +impala-shell -i impala-cluster-dn1.openaire.eu -d ${SOURCE} -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}.\1 as select * from ${SOURCE}.\1;/" | impala-shell -i impala-cluster-dn1.openaire.eu -c -f - +echo "Production monitor db ready!" + +impala-shell -i impala-cluster-dn1.openaire.eu -q "drop view ${MONITOR_PROD}.historical_snapshots" +impala-shell -i impala-cluster-dn1.openaire.eu -q "drop view ${MONITOR_PROD}.historical_snapshots_fos" + +impala-shell -i impala-cluster-dn1.openaire.eu -q "create view ${MONITOR_PROD}.historical_snapshots as select * from ${SOURCE}.historical_snapshots" +impala-shell -i impala-cluster-dn1.openaire.eu -q "create view ${MONITOR_PROD}.historical_snapshots_fos as select * from ${SOURCE}.historical_snapshots_fos" + +impala-shell -i impala-cluster-dn1.openaire.eu -q "drop view ${MONITOR_IRISH_PROD}.historical_snapshots_irish" +impala-shell -i impala-cluster-dn1.openaire.eu -q "drop view ${MONITOR_IRISH_PROD}.historical_snapshots_irish_fos" + + +impala-shell -i impala-cluster-dn1.openaire.eu -q "create view ${MONITOR_IRISH_PROD}.historical_snapshots_irish as select * from ${SOURCE}.historical_snapshots_irish" +impala-shell -i impala-cluster-dn1.openaire.eu -q "create view ${MONITOR_IRISH_PROD}.historical_snapshots_irish_fos as select * from ${SOURCE}.historical_snapshots_irish" diff --git a/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/hist_snaps.sh b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/hist_snaps.sh new file mode 100644 index 000000000..bcaa7984c --- /dev/null +++ b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/hist_snaps.sh @@ -0,0 +1,27 @@ +export PYTHON_EGG_CACHE=/home/$(whoami)/.python-eggs +export link_folder=/tmp/impala-shell-python-egg-cache-$(whoami) +if ! [ -L $link_folder ] +then + rm -Rf "$link_folder" + ln -sfn ${PYTHON_EGG_CACHE}${link_folder} ${link_folder} +fi + +export SOURCE=$1 +export TARGET=$2 +export SHADOW=$3 +export SCRIPT_PATH=$4 + + +export HIVE_OPTS="-hiveconf mapred.job.queue.name=analytics -hiveconf hive.spark.client.connect.timeout=120000ms -hiveconf hive.spark.client.server.connect.timeout=300000ms -hiveconf spark.executor.memory=19166291558 -hiveconf spark.yarn.executor.memoryOverhead=3225 -hiveconf spark.driver.memory=11596411699 -hiveconf spark.yarn.driver.memoryOverhead=1228" +export HADOOP_USER_NAME="oozie" + +echo "Getting file from " $4 +hdfs dfs -copyToLocal $4 + +#update Monitor DB IRISH +#cat CreateDB.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2/g1" | sed "s/GRAPHDB/$3/g1" > foo +cat buildIrishMonitorDB.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2/g1" > foo +hive $HIVE_OPTS -f foo + +echo "Hive shell finished" + diff --git a/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/scripts/BuildHistSnapsAll.sql b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/scripts/BuildHistSnapsAll.sql new file mode 100644 index 000000000..93d804820 --- /dev/null +++ b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/scripts/BuildHistSnapsAll.sql @@ -0,0 +1,82 @@ +INSERT INTO ${hist_db_name}.historical_snapshots_fos_tmp +SELECT * FROM ${hist_db_name_prev}.historical_snapshots_fos; + +INSERT INTO ${hist_db_name}.historical_snapshots_fos_tmp +select + cast(${hist_date} as STRING), + count(distinct r.id), + r.type, + rf.lvl1, + rf.lvl2, + pf.publicly_funded, + r.access_mode, + r.gold, + r.green, + coalesce(gl.green_with_license,0), + h.is_hybrid, + b.is_bronze_oa, + d.in_diamond_journal, + t.is_transformative, + pr.refereed +from ${stats_db_name}.result r + left outer join ${stats_db_name}.result_fos rf on rf.id=r.id + left outer join ${stats_db_name}.indi_pub_publicly_funded pf on pf.id=r.id + left outer join ${stats_db_name}.indi_pub_green_with_license gl on gl.id=r.id + left outer join ${stats_db_name}.indi_pub_bronze_oa b on b.id=r.id + left outer join ${stats_db_name}.indi_pub_diamond d on d.id=r.id + left outer join ${stats_db_name}.indi_pub_in_transformative t on t.id=r.id + left outer join ${stats_db_name}.indi_pub_hybrid h on h.id=r.id + left outer join ${stats_db_name}.result_refereed pr on pr.id=r.id +group by r.green, r.gold, r.access_mode, r.type, rf.lvl1,rf.lvl2, pf.publicly_funded,r.green, gl.green_with_license,b.is_bronze_oa,d.in_diamond_journal,t.is_transformative,h.is_hybrid,pr.refereed; + +drop table if exists ${hist_db_name}.historical_snapshots_fos purge; + +CREATE TABLE ${hist_db_name}.historical_snapshots_fos STORED AS PARQUET AS +SELECT * FROM ${hist_db_name}.historical_snapshots_fos_tmp; + +drop table if exists ${monitor_db_name}.historical_snapshots_fos purge; + +create table ${monitor_db_name}.historical_snapshots_fos stored as parquet +as select * from ${hist_db_name}.historical_snapshots_fos; + +drop table ${hist_db_name}.historical_snapshots_fos_tmp purge; + +INSERT INTO ${hist_db_name}.historical_snapshots_tmp as +SELECT * FROM ${hist_db_name_prev}.historical_snapshots; + +INSERT INTO ${hist_db_name}.historical_snapshots_tmp +select + cast(${hist_date} as STRING), + count(distinct r.id), + r.type, + pf.publicly_funded, + r.access_mode, + r.gold, + r.green, + coalesce(gl.green_with_license,0), + h.is_hybrid, + b.is_bronze_oa, + d.in_diamond_journal, + t.is_transformative, + pr.refereed +from ${stats_db_name}.result r + left outer join ${stats_db_name}.indi_pub_publicly_funded pf on pf.id=r.id + left outer join ${stats_db_name}.indi_pub_green_with_license gl on gl.id=r.id + left outer join ${stats_db_name}.indi_pub_bronze_oa b on b.id=r.id + left outer join ${stats_db_name}.indi_pub_diamond d on d.id=r.id + left outer join ${stats_db_name}.indi_pub_in_transformative t on t.id=r.id + left outer join ${stats_db_name}.indi_pub_hybrid h on h.id=r.id + left outer join ${stats_db_name}.result_refereed pr on pr.id=r.id +group by r.green, r.gold, r.access_mode, r.type, pf.publicly_funded,r.green, gl.green_with_license,b.is_bronze_oa,d.in_diamond_journal,t.is_transformative,h.is_hybrid,pr.refereed; + +drop table if exists ${hist_db_name}.historical_snapshots purge; + +CREATE TABLE ${hist_db_name}.historical_snapshots STORED AS PARQUET AS +SELECT * FROM ${hist_db_name}.historical_snapshots_tmp; + +drop table if exists ${monitor_db_name}.historical_snapshots purge; + +create table ${monitor_db_name}.historical_snapshots stored as parquet +as select * from ${hist_db_name}.historical_snapshots; + +drop table ${hist_db_name}.historical_snapshots_tmp purge; \ No newline at end of file diff --git a/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/scripts/BuildHistSnapsIrish.sql b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/scripts/BuildHistSnapsIrish.sql new file mode 100644 index 000000000..95e811f64 --- /dev/null +++ b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/scripts/BuildHistSnapsIrish.sql @@ -0,0 +1,91 @@ +INSERT INTO ${hist_db_name}.historical_snapshots_fos_irish_tmp +SELECT * FROM ${hist_db_name_prev}.historical_snapshots_irish_fos; + +INSERT INTO ${hist_db_name}.historical_snapshots_fos_irish_tmp +select + cast(${hist_date} as STRING), + count(distinct r.id), + r.type, + rf.lvl1, + rf.lvl2, + pf.publicly_funded, + r.access_mode, + r.gold, + r.green, + coalesce(gl.green_with_license,0), + h.is_hybrid, + b.is_bronze_oa, + d.in_diamond_journal, + t.is_transformative, + pr.refereed +from ${stats_irish_db_name}.result r + left outer join ${stats_irish_db_name}.result_fos rf on rf.id=r.id + left outer join ${stats_irish_db_name}.indi_pub_publicly_funded pf on pf.id=r.id + left outer join ${stats_irish_db_name}.indi_pub_green_with_license gl on gl.id=r.id + left outer join ${stats_irish_db_name}.indi_pub_bronze_oa b on b.id=r.id + left outer join ${stats_irish_db_name}.indi_pub_diamond d on d.id=r.id + left outer join ${stats_irish_db_name}.indi_pub_in_transformative t on t.id=r.id + left outer join ${stats_irish_db_name}.indi_pub_hybrid h on h.id=r.id + left outer join ${stats_irish_db_name}.result_refereed pr on pr.id=r.id +group by r.green, r.gold, r.access_mode, r.type, rf.lvl1,rf.lvl2, pf.publicly_funded,r.green, gl.green_with_license,b.is_bronze_oa,d.in_diamond_journal,t.is_transformative,h.is_hybrid,pr.refereed; + +drop table if exists ${hist_db_name}.historical_snapshots_irish_fos purge; + +CREATE TABLE ${hist_db_name}.historical_snapshots_irish_fos STORED AS PARQUET AS +SELECT * FROM ${hist_db_name}.historical_snapshots_fos_irish_tmp; + +drop table if exists ${monitor_irish_db_name}.historical_snapshots_irish_fos purge; + +create table ${monitor_irish_db_name}.historical_snapshots_irish_fos stored as parquet +as select * from ${hist_db_name}.historical_snapshots_irish_fos; + +drop table ${hist_db_name}.historical_snapshots_fos_irish_tmp purge; + +INSERT INTO ${hist_db_name}.historical_snapshots_irish_tmp +SELECT * FROM ${hist_db_name_prev}.historical_snapshots_irish; + +INSERT INTO ${hist_db_name}.historical_snapshots_irish_tmp +select + cast(${hist_date} as STRING), + count(distinct r.id), + r.type, + pf.publicly_funded, + r.access_mode, + r.gold, + r.green, + coalesce(gl.green_with_license,0), + h.is_hybrid, + b.is_bronze_oa, + d.in_diamond_journal, + t.is_transformative, + pr.refereed +from ${stats_irish_db_name}.result r + left outer join ${stats_irish_db_name}.indi_pub_publicly_funded pf on pf.id=r.id + left outer join ${stats_irish_db_name}.indi_pub_green_with_license gl on gl.id=r.id + left outer join ${stats_irish_db_name}.indi_pub_bronze_oa b on b.id=r.id + left outer join ${stats_irish_db_name}.indi_pub_diamond d on d.id=r.id + left outer join ${stats_irish_db_name}.indi_pub_in_transformative t on t.id=r.id + left outer join ${stats_irish_db_name}.indi_pub_hybrid h on h.id=r.id + left outer join ${stats_irish_db_name}.result_refereed pr on pr.id=r.id +group by r.green, r.gold, r.access_mode, r.type, pf.publicly_funded,r.green, gl.green_with_license,b.is_bronze_oa,d.in_diamond_journal,t.is_transformative,h.is_hybrid,pr.refereed; + + +drop table if exists ${hist_db_name}.historical_snapshots_irish purge; + +CREATE TABLE ${hist_db_name}.historical_snapshots_irish STORED AS PARQUET AS +SELECT * FROM ${hist_db_name}.historical_snapshots_irish_tmp; + +drop table if exists ${monitor_irish_db_name}.historical_snapshots_irish purge; + +create table ${monitor_irish_db_name}.historical_snapshots_irish stored as parquet +as select * from ${hist_db_name}.historical_snapshots_irish; + +drop table ${hist_db_name}.historical_snapshots_irish_tmp purge; + + +drop table if exists ${monitor_irish_db_name}.historical_snapshots_irish_fos purge; + +create table ${monitor_irish_db_name}.historical_snapshots_irish_fos stored as parquet +as select * from ${hist_db_name}.historical_snapshots_irish_fos; + +drop table ${hist_db_name}.historical_snapshots_fos_irish_tmp purge; \ No newline at end of file diff --git a/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/scripts/CreateDB.sql b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/scripts/CreateDB.sql new file mode 100644 index 000000000..18af135bf --- /dev/null +++ b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/scripts/CreateDB.sql @@ -0,0 +1,92 @@ +-------------------------------------------------------------- +-------------------------------------------------------------- +-- Historical Snapshots database creation +-------------------------------------------------------------- +-------------------------------------------------------------- + +DROP database IF EXISTS ${hist_db_name} CASCADE; +CREATE database ${hist_db_name}; + +drop table if exists ${hist_db_name}.historical_snapshots_fos_tmp purge; + +CREATE TABLE ${hist_db_name}.historical_snapshots_fos_tmp +( + hist_date STRING, + total INT, + type STRING, + lvl1 STRING, + lvl2 STRING, + publicly_funded INT, + accessrights STRING, + gold INT, + green INT, + green_with_license INT, + hybrid INT, + bronze INT, + diamond INT, + transformative INT, + peer_reviewed STRING +) +CLUSTERED BY (hist_date) INTO 100 buckets stored as orc tblproperties ('transactional' = 'true'); + +drop table if exists ${hist_db_name}.historical_snapshots_fos_irish_tmp purge; + +CREATE TABLE ${hist_db_name}.historical_snapshots_fos_irish_tmp +( + hist_date STRING, + total INT, + type STRING, + lvl1 STRING, + lvl2 STRING, + publicly_funded INT, + accessrights STRING, + gold INT, + green INT, + green_with_license INT, + hybrid INT, + bronze INT, + diamond INT, + transformative INT, + peer_reviewed STRING +) +CLUSTERED BY (hist_date) INTO 100 buckets stored as orc tblproperties ('transactional' = 'true'); + +drop table if exists ${hist_db_name}.historical_snapshots_tmp purge; + +CREATE TABLE ${hist_db_name}.historical_snapshots_tmp +( + hist_date STRING, + total INT, + type STRING, + publicly_funded INT, + accessrights STRING, + gold INT, + green INT, + green_with_license INT, + hybrid INT, + bronze INT, + diamond INT, + transformative INT, + peer_reviewed STRING +) +CLUSTERED BY (hist_date) INTO 100 buckets stored as orc tblproperties ('transactional' = 'true'); + +drop table if exists ${hist_db_name}.historical_snapshots_irish_tmp purge; + +CREATE TABLE ${hist_db_name}.historical_snapshots_irish_tmp +( + hist_date STRING, + total INT, + type STRING, + publicly_funded INT, + accessrights STRING, + gold INT, + green INT, + green_with_license INT, + hybrid INT, + bronze INT, + diamond INT, + transformative INT, + peer_reviewed STRING +) +CLUSTERED BY (hist_date) INTO 100 buckets stored as orc tblproperties ('transactional' = 'true'); \ No newline at end of file diff --git a/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/workflow.xml b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/workflow.xml new file mode 100644 index 000000000..3245e0859 --- /dev/null +++ b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/workflow.xml @@ -0,0 +1,159 @@ + + + + hist_db_name + the target hist database name + + + hist_db_name_prev + the hist database name of previous_month + + + + stats_db_name + the stats db name + + + stats_irish_db_name + the stats irish db name + + + monitor_db_name + the monitor db name + + + monitor_irish_db_name + the irish monitor db name + + + hist_db_prod_name + the production db + + + hist_db_shadow_name + the production shadow db + + + hist_date + the snaps date + + + hive_metastore_uris + hive server metastore URIs + + + hive_jdbc_url + hive server jdbc url + + + hive_timeout + the time period, in seconds, after which Hive fails a transaction if a Hive client has not sent a hearbeat. The default value is 300 seconds. + + + hadoop_user_name + user name of the wf owner + + + + + ${jobTracker} + ${nameNode} + + + hive.metastore.uris + ${hive_metastore_uris} + + + hive.txn.timeout + ${hive_timeout} + + + mapred.job.queue.name + analytics + + + + + + + + ${wf:conf('resumeFrom') eq 'CreateDB'} + ${wf:conf('resumeFrom') eq 'BuildHistSnaps'} + ${wf:conf('resumeFrom') eq 'BuildHistSnapsIrish'} + ${wf:conf('resumeFrom') eq 'Step2-copyDataToImpalaCluster'} + ${wf:conf('resumeFrom') eq 'Step3-finalizeImpalaCluster'} + + + + + + Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}] + + + + + ${hive_jdbc_url} + + hist_db_name=${hist_db_name} + + + + + + + + ${hive_jdbc_url} + + hist_db_name=${hist_db_name} + hist_db_name_prev=${hist_db_name_prev} + stats_db_name=${stats_db_name} + monitor_db_name=${monitor_db_name} + hist_date=${hist_date} + + + + + + + + ${hive_jdbc_url} + + hist_db_name=${hist_db_name} + hist_db_name_prev=${hist_db_name_prev} + stats_irish_db_name=${stats_irish_db_name} + monitor_irish_db_name=${monitor_irish_db_name} + hist_date=${hist_date} + + + + + + + ${jobTracker} + ${nameNode} + copyDataToImpalaCluster.sh + ${hist_db_name} + ${hadoop_user_name} + copyDataToImpalaCluster.sh + + + + + + + ${jobTracker} + ${nameNode} + finalizeImpalaCluster.sh + ${hist_db_name} + ${hist_db_prod_name} + ${monitor_db_name} + ${monitor_db_name} + ${monitor_irish_db_name} + finalizeImpalaCluster.sh + + + + + + + -- 2.17.1 From 6b823100aede7c41f4dfa645394082f7d95d514f Mon Sep 17 00:00:00 2001 From: dimitrispie Date: Sun, 7 Jan 2024 22:54:39 +0200 Subject: [PATCH 04/22] Update buildIrishMonitorDB.sql New indicators added --- .../oozie_app/scripts/buildIrishMonitorDB.sql | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/scripts/buildIrishMonitorDB.sql b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/scripts/buildIrishMonitorDB.sql index db76fdab3..eb632ad19 100644 --- a/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/scripts/buildIrishMonitorDB.sql +++ b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/scripts/buildIrishMonitorDB.sql @@ -219,6 +219,12 @@ create view TARGET.indi_funder_findable as select * from SOURCE.indi_funder_find create view TARGET.indi_ris_fairness as select * from SOURCE.indi_ris_fairness; create view TARGET.indi_ris_openess as select * from SOURCE.indi_ris_openess; create view TARGET.indi_ris_findable as select * from SOURCE.indi_ris_findable; + create table TARGET.indi_pub_green_with_license stored as parquet as select * from SOURCE.indi_pub_green_with_license orig where exists (select 1 from TARGET.result r where r.id=orig.id); create table TARGET.result_country stored as parquet as select * from SOURCE.result_country orig where exists (select 1 from TARGET.result r where r.id=orig.id); -create table TARGET.indi_pub_publicly_funded stored as parquet as select * from SOURCE.indi_pub_publicly_funded orig where exists (select 1 from TARGET.result r where r.id=orig.id); \ No newline at end of file +create table TARGET.indi_pub_publicly_funded stored as parquet as select * from SOURCE.indi_pub_publicly_funded orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_result_oa_with_license stored as parquet as select * from SOURCE.indi_result_oa_with_license orig where exists (select 1 from TARGET.result r where r.id=orig.id); +create table TARGET.indi_result_oa_without_license stored as parquet as select * from SOURCE.indi_result_oa_without_license orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_result_under_transformative stored as parquet as select * from SOURCE.indi_result_under_transformative orig where exists (select 1 from TARGET.result r where r.id=orig.id); \ No newline at end of file -- 2.17.1 From 2e4cab026c7707549414eb9be872e3f91c83a592 Mon Sep 17 00:00:00 2001 From: Antonis Lempesis Date: Mon, 8 Jan 2024 16:01:26 +0200 Subject: [PATCH 05/22] fixed the result_country definition --- .../scripts/step16-createIndicatorsTables.sql | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql index f8213030e..daef003df 100755 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql @@ -1210,6 +1210,10 @@ where ri.license is not null and datasource.type like '%Repository%' and datasou on p.id= tmp.id; create table ${stats_db_name}.result_country stored as parquet as -select distinct ro.id, o.country from ${stats_db_name}.result_organization ro -join ${stats_db_name}.organization o on o.id=ro.organization -join ${stats_db_name}.funder f on f.country=o.country; \ No newline at end of file +select distinct ro.id, coalesce(o.country, f.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; -- 2.17.1 From 8b2cbb611e07a54e2e1ba8d2b6c68b8f98fee0a2 Mon Sep 17 00:00:00 2001 From: dimitrispie Date: Tue, 9 Jan 2024 00:40:56 +0200 Subject: [PATCH 06/22] Changes to beta db names --- .../stats-hist-snaps/oozie_app/finalizeImpalaCluster.sh | 6 ++---- .../dhp/oa/graph/stats-hist-snaps/oozie_app/workflow.xml | 6 +++--- 2 files changed, 5 insertions(+), 7 deletions(-) diff --git a/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/finalizeImpalaCluster.sh b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/finalizeImpalaCluster.sh index e2277083d..d780f103b 100644 --- a/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/finalizeImpalaCluster.sh +++ b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/finalizeImpalaCluster.sh @@ -9,10 +9,8 @@ fi SOURCE=$1 PRODUCTION=$2 SHADOW=$3 -MONITOR=$4 -MONITOR_PROD=$5 -MONITOR_IRISH_PROD=$6 -MONITOR_IRISH=$7 +MONITOR_PROD=$4 +MONITOR_IRISH_PROD=$5 echo ${SOURCE} diff --git a/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/workflow.xml b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/workflow.xml index 3245e0859..8846fcdd7 100644 --- a/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/workflow.xml +++ b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/workflow.xml @@ -146,9 +146,9 @@ finalizeImpalaCluster.sh ${hist_db_name} ${hist_db_prod_name} - ${monitor_db_name} - ${monitor_db_name} - ${monitor_irish_db_name} + ${hist_db_shadow_name} + ${monitor_db_prod_name} + ${monitor_irish_db_prod_name} finalizeImpalaCluster.sh -- 2.17.1 From b920307bdd733d8a9a97067888a89cdec6bdbc46 Mon Sep 17 00:00:00 2001 From: dimitrispie Date: Tue, 9 Jan 2024 00:47:09 +0200 Subject: [PATCH 07/22] Changes to indicators --- .../scripts/step16-createIndicatorsTables.sql | 40 ++++++++++++++++--- .../scripts/step20-createMonitorDB.sql | 6 +++ .../scripts/step20-createMonitorDBAll.sql | 6 +++ .../scripts/step21-createObservatoryDB.sql | 3 +- 4 files changed, 48 insertions(+), 7 deletions(-) diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql index f8213030e..eab91c99c 100755 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql @@ -58,7 +58,7 @@ select distinct r.id, (case when lic='' or lic is null then 0 else 1 end) as has from ${stats_db_name}.result r left outer join (select r.id, license.type as lic from ${stats_db_name}.result r join ${stats_db_name}.result_licenses as license on license.id = r.id - where lower(license.type) LIKE '%creativecommons.org%' OR lower(license.type) LIKE '%cc-%') tmp + where lower(license.type) LIKE '%creativecommons.org%' OR license.type LIKE '%CC%') tmp on r.id= tmp.id; drop table if exists ${stats_db_name}.indi_result_has_cc_licence_url purge; @@ -1189,7 +1189,7 @@ union all select pf.name from stats_ext.insitutions_for_publicly_funded pf join ${stats_db_name}.project p on p.funder=pf.name union all -select pf.name from stats_ext.insitutions_for_publicly_funded pf +select op.name from stats_ext.insitutions_for_publicly_funded pf join org_names_pids op on (op.name=pf.name or op.pid=pf.ror) and pf.publicly_funded='yes') foo) select distinct p.id, coalesce(publicly_funded, 0) as publicly_funded @@ -1210,6 +1210,36 @@ where ri.license is not null and datasource.type like '%Repository%' and datasou on p.id= tmp.id; create table ${stats_db_name}.result_country stored as parquet as -select distinct ro.id, o.country from ${stats_db_name}.result_organization ro -join ${stats_db_name}.organization o on o.id=ro.organization -join ${stats_db_name}.funder f on f.country=o.country; \ No newline at end of file +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; + +create table ${stats_db_name}.indi_result_oa_with_license stored as parquet as +select distinct r.id, coalesce(oa_with_license,0) as oa_with_license +from ${stats_db_name}.result r +left outer join (select distinct r.id, 1 as oa_with_license from ${stats_db_name}.result r +join ${stats_db_name}.result_licenses rl on rl.id=r.id where r.bestlicence='Open Access') tmp on r.id=tmp.id; + +create table ${stats_db_name}.indi_result_oa_without_license stored as parquet as +with without_license as +(select distinct id from ${stats_db_name}.indi_result_oa_with_license +where oa_with_license=0) +select distinct r.id, coalesce(oa_without_license,0) as oa_without_license +from ${stats_db_name}.result r +left outer join (select distinct r.id, 1 as oa_without_license +from ${stats_db_name}.result r +join without_license wl on wl.id=r.id +where r.bestlicence='Open Access') tmp on r.id=tmp.id; + +create table ${stats_db_name}.indi_result_under_transformative stored as parquet as +with transformative_dois as +(select distinct doi from stats_ext.transformative_facts) +select distinct r.id, coalesce(under_transformative,0) as under_transformative +from ${stats_db_name}.result r +left outer join (select distinct rp.id, 1 as under_transformative +from ${stats_db_name}.result_pids rp join ${stats_db_name}.result r on r.id=rp.id +join transformative_dois td on td.doi=rp.pid) tmp on r.id=tmp.id; \ No newline at end of file diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB.sql index a99a75798..4abb6bdbc 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB.sql @@ -181,5 +181,11 @@ create view TARGET.indi_funder_findable as select * from SOURCE.indi_funder_find create view TARGET.indi_ris_fairness as select * from SOURCE.indi_ris_fairness; create view TARGET.indi_ris_openess as select * from SOURCE.indi_ris_openess; create view TARGET.indi_ris_findable as select * from SOURCE.indi_ris_findable; + create table TARGET.indi_pub_green_with_license stored as parquet as select * from SOURCE.indi_pub_green_with_license orig where exists (select 1 from TARGET.result r where r.id=orig.id); create table TARGET.result_country stored as parquet as select * from SOURCE.result_country orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_result_oa_with_license stored as parquet as select * from SOURCE.indi_result_oa_with_license orig where exists (select 1 from TARGET.result r where r.id=orig.id); +create table TARGET.indi_result_oa_without_license stored as parquet as select * from SOURCE.indi_result_oa_without_license orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_result_under_transformative stored as parquet as select * from SOURCE.indi_result_under_transformative orig where exists (select 1 from TARGET.result r where r.id=orig.id); diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDBAll.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDBAll.sql index 671a9ea9e..b4f87a184 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDBAll.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDBAll.sql @@ -263,6 +263,12 @@ create view TARGET.indi_funder_findable as select * from SOURCE.indi_funder_find create view TARGET.indi_ris_fairness as select * from SOURCE.indi_ris_fairness; create view TARGET.indi_ris_openess as select * from SOURCE.indi_ris_openess; create view TARGET.indi_ris_findable as select * from SOURCE.indi_ris_findable; + create table TARGET.indi_pub_green_with_license stored as parquet as select * from SOURCE.indi_pub_green_with_license orig where exists (select 1 from TARGET.result r where r.id=orig.id); create table TARGET.result_country stored as parquet as select * from SOURCE.result_country orig where exists (select 1 from TARGET.result r where r.id=orig.id); +create table TARGET.indi_result_oa_with_license stored as parquet as select * from SOURCE.indi_result_oa_with_license orig where exists (select 1 from TARGET.result r where r.id=orig.id); +create table TARGET.indi_result_oa_without_license stored as parquet as select * from SOURCE.indi_result_oa_without_license orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create table TARGET.indi_result_under_transformative stored as parquet as select * from SOURCE.indi_result_under_transformative orig where exists (select 1 from TARGET.result r where r.id=orig.id); + diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step21-createObservatoryDB.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step21-createObservatoryDB.sql index 2e6f0711c..66620ac38 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step21-createObservatoryDB.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step21-createObservatoryDB.sql @@ -2,9 +2,8 @@ create table ${observatory_db_name}.result_cc_licence stored as parquet as select r.id, coalesce(rln.count, 0) > 0 as cc_licence from ${stats_db_name}.result r left outer join ( - select rl.id, sum(case when lower(rln.normalized) like 'cc-%' then 1 else 0 end) as count + select rl.id, sum(case when rl.type like 'CC%' then 1 else 0 end) as count from ${stats_db_name}.result_licenses rl - left outer join ${stats_db_name}.licenses_normalized rln on rl.type=rln.license group by rl.id ) rln on rln.id=r.id; -- 2.17.1 From e024718f73bcb42d67ed2fd78f5a1ccdff246935 Mon Sep 17 00:00:00 2001 From: Antonis Lempesis Date: Wed, 10 Jan 2024 22:25:50 +0100 Subject: [PATCH 08/22] creating result_instances even when no pids exist for the instance --- .../dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15_5.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15_5.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15_5.sql index 2c606fb92..d895c12f6 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15_5.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15_5.sql @@ -50,7 +50,7 @@ select distinct r.* from ( select substr(r.id, 4) as id, inst.accessright.classname as accessright, inst.accessright.openaccessroute as accessright_uw, substr(inst.collectedfrom.key, 4) as collectedfrom, substr(inst.hostedby.key, 4) as hostedby, inst.dateofacceptance.value as dateofacceptance, inst.license.value as license, p.qualifier.classname as pidtype, p.value as pid - from ${openaire_db_name}.result r lateral view explode(r.instance) instances as inst lateral view explode(inst.pid) pids as p) r + from ${openaire_db_name}.result r lateral view explode(r.instance) instances as inst lateral view outer explode(inst.pid) pids as p) r join ${stats_db_name}.result res on res.id=r.id; DROP TABLE IF EXISTS ${stats_db_name}.result_apc purge; -- 2.17.1 From fd43b0e84a6355d885346b01092b6e1789564370 Mon Sep 17 00:00:00 2001 From: Antonis Lempesis Date: Thu, 25 Jan 2024 15:06:34 +0100 Subject: [PATCH 09/22] max mem of joins (hive.mapjoin.followby.gby.localtask.max.memory.usage) now 80%, up from 55%. --- .../eu/dnetlib/dhp/oa/graph/stats/oozie_app/workflow.xml | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/workflow.xml b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/workflow.xml index cbf97944d..f15f22320 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/workflow.xml +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/workflow.xml @@ -78,6 +78,10 @@ hive.txn.timeout ${hive_timeout} + + hive.mapjoin.followby.gby.localtask.max.memory.usage + 0.80 + mapred.job.queue.name analytics -- 2.17.1 From c548796463814f1637d0d1c47120ba01845e1ef2 Mon Sep 17 00:00:00 2001 From: antleb Date: Fri, 26 Jan 2024 02:04:48 +0200 Subject: [PATCH 10/22] Changed step16-createIndicatorsTables to use a spark oozie action instead of hive --- .../scripts/step16-createIndicatorsTables.sql | 424 +++++++++--------- .../dhp/oa/graph/stats/oozie_app/workflow.xml | 66 ++- 2 files changed, 268 insertions(+), 222 deletions(-) diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql index eab91c99c..4461e58b4 100755 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql @@ -1,5 +1,5 @@ -- Sprint 1 ---- -drop table if exists ${stats_db_name}.indi_pub_green_oa purge; +drop table if exists ${stats_db_name}.indi_pub_green_oa purge; /*EOS*/ --create table if not exists ${stats_db_name}.indi_pub_green_oa stored as parquet as --select distinct p.id, coalesce(green_oa, 0) as green_oa @@ -24,9 +24,9 @@ from ${stats_db_name}.publication p where datasource.type like '%Repository%' and (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and datasource.name!='Other') tmp - on p.id= tmp.id; + on p.id= tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_grey_lit purge; +drop table if exists ${stats_db_name}.indi_pub_grey_lit purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_grey_lit stored as parquet as select distinct p.id, coalesce(grey_lit, 0) as grey_lit @@ -37,9 +37,9 @@ from ${stats_db_name}.publication p join ${stats_db_name}.result_classifications rt on rt.id = p.id where rt.type not in ('Article','Part of book or chapter of book','Book','Doctoral thesis','Master thesis','Data Paper', 'Thesis', 'Bachelor thesis', 'Conference object') and not exists (select 1 from ${stats_db_name}.result_classifications rc where type ='Other literature type' - and rc.id=p.id)) tmp on p.id=tmp.id; + and rc.id=p.id)) tmp on p.id=tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_doi_from_crossref purge; +drop table if exists ${stats_db_name}.indi_pub_doi_from_crossref purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_doi_from_crossref stored as parquet as select distinct p.id, coalesce(doi_from_crossref, 0) as doi_from_crossref @@ -48,10 +48,10 @@ from ${stats_db_name}.publication p (select ri.id, 1 as doi_from_crossref from ${stats_db_name}.result_instance ri join ${stats_db_name}.datasource d on d.id = ri.collectedfrom where pidtype='Digital Object Identifier' and d.name ='Crossref') tmp - on tmp.id=p.id; + on tmp.id=p.id; /*EOS*/ -- Sprint 2 ---- -drop table if exists ${stats_db_name}.indi_result_has_cc_licence purge; +drop table if exists ${stats_db_name}.indi_result_has_cc_licence purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_result_has_cc_licence stored as parquet as select distinct r.id, (case when lic='' or lic is null then 0 else 1 end) as has_cc_license @@ -59,9 +59,9 @@ from ${stats_db_name}.result r left outer join (select r.id, license.type as lic from ${stats_db_name}.result r join ${stats_db_name}.result_licenses as license on license.id = r.id where lower(license.type) LIKE '%creativecommons.org%' OR license.type LIKE '%CC%') tmp - on r.id= tmp.id; + on r.id= tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_result_has_cc_licence_url purge; +drop table if exists ${stats_db_name}.indi_result_has_cc_licence_url purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_result_has_cc_licence_url stored as parquet as select distinct r.id, case when lic_host='' or lic_host is null then 0 else 1 end as has_cc_license_url @@ -70,32 +70,32 @@ from ${stats_db_name}.result r from ${stats_db_name}.result r join ${stats_db_name}.result_licenses as license on license.id = r.id WHERE lower(parse_url(license.type, "HOST")) = "creativecommons.org") tmp - on r.id= tmp.id; + on r.id= tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_has_abstract purge; +drop table if exists ${stats_db_name}.indi_pub_has_abstract purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_has_abstract stored as parquet as select distinct publication.id, cast(coalesce(abstract, true) as int) has_abstract -from ${stats_db_name}.publication; +from ${stats_db_name}.publication; /*EOS*/ -drop table if exists ${stats_db_name}.indi_result_with_orcid purge; +drop table if exists ${stats_db_name}.indi_result_with_orcid purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_result_with_orcid stored as parquet as select distinct r.id, coalesce(has_orcid, 0) as has_orcid from ${stats_db_name}.result r left outer join (select id, 1 as has_orcid from ${stats_db_name}.result_orcid) tmp - on r.id= tmp.id; + on r.id= tmp.id; /*EOS*/ ---- Sprint 3 ---- -drop table if exists ${stats_db_name}.indi_funded_result_with_fundref purge; +drop table if exists ${stats_db_name}.indi_funded_result_with_fundref purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_funded_result_with_fundref stored as parquet as select distinct r.result as id, coalesce(fundref, 0) as fundref from ${stats_db_name}.project_results r left outer join (select distinct result, 1 as fundref from ${stats_db_name}.project_results where provenance='Harvested') tmp - on r.result= tmp.result; + on r.result= tmp.result; /*EOS*/ -- create table indi_result_org_collab stored as parquet as -- select o1.organization org1, o2.organization org2, count(distinct o1.id) as collaborations @@ -106,64 +106,64 @@ from ${stats_db_name}.project_results r -- compute stats indi_result_org_collab; -- create TEMPORARY TABLE ${stats_db_name}.tmp AS SELECT ro.organization organization, ro.id, o.name from ${stats_db_name}.result_organization ro -join ${stats_db_name}.organization o on o.id=ro.organization where o.name is not null; +join ${stats_db_name}.organization o on o.id=ro.organization where o.name is not null; /*EOS*/ -drop table if exists ${stats_db_name}.indi_result_org_collab purge; +drop table if exists ${stats_db_name}.indi_result_org_collab purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_result_org_collab stored as parquet as select o1.organization org1, o1.name org1name1, o2.organization org2, o2.name org2name2, count(o1.id) as collaborations from ${stats_db_name}.tmp as o1 join ${stats_db_name}.tmp as o2 where o1.id=o2.id and o1.organization!=o2.organization and o1.name!=o2.name -group by o1.organization, o2.organization, o1.name, o2.name; +group by o1.organization, o2.organization, o1.name, o2.name; /*EOS*/ -drop table if exists ${stats_db_name}.tmp purge; +drop table if exists ${stats_db_name}.tmp purge; /*EOS*/ create TEMPORARY TABLE ${stats_db_name}.tmp AS select distinct ro.organization organization, ro.id, o.name, o.country from ${stats_db_name}.result_organization ro -join ${stats_db_name}.organization o on o.id=ro.organization where country <> 'UNKNOWN' and o.name is not null; +join ${stats_db_name}.organization o on o.id=ro.organization where country <> 'UNKNOWN' and o.name is not null; /*EOS*/ -drop table if exists ${stats_db_name}.indi_result_org_country_collab purge; +drop table if exists ${stats_db_name}.indi_result_org_country_collab purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_result_org_country_collab stored as parquet as select o1.organization org1,o1.name org1name1, o2.country country2, count(o1.id) as collaborations from ${stats_db_name}.tmp as o1 join ${stats_db_name}.tmp as o2 on o1.id=o2.id where o1.id=o2.id and o1.country!=o2.country -group by o1.organization, o1.id, o1.name, o2.country; +group by o1.organization, o1.id, o1.name, o2.country; /*EOS*/ -drop table if exists ${stats_db_name}.tmp purge; +drop table if exists ${stats_db_name}.tmp purge; /*EOS*/ create TEMPORARY TABLE ${stats_db_name}.tmp AS select o.id organization, o.name, ro.project as project from ${stats_db_name}.organization o - join ${stats_db_name}.organization_projects ro on o.id=ro.id where o.name is not null; + join ${stats_db_name}.organization_projects ro on o.id=ro.id where o.name is not null; /*EOS*/ -drop table if exists ${stats_db_name}.indi_project_collab_org purge; +drop table if exists ${stats_db_name}.indi_project_collab_org purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_project_collab_org stored as parquet as select o1.organization org1,o1.name orgname1, o2.organization org2, o2.name orgname2, count(distinct o1.project) as collaborations from ${stats_db_name}.tmp as o1 join ${stats_db_name}.tmp as o2 on o1.project=o2.project where o1.organization<>o2.organization and o1.name<>o2.name -group by o1.name,o2.name, o1.organization, o2.organization; +group by o1.name,o2.name, o1.organization, o2.organization; /*EOS*/ -drop table if exists ${stats_db_name}.tmp purge; +drop table if exists ${stats_db_name}.tmp purge; /*EOS*/ create TEMPORARY TABLE ${stats_db_name}.tmp AS select o.id organization, o.name, o.country , ro.project as project from ${stats_db_name}.organization o join ${stats_db_name}.organization_projects ro on o.id=ro.id - and o.country <> 'UNKNOWN' and o.name is not null; + and o.country <> 'UNKNOWN' and o.name is not null; /*EOS*/ -drop table if exists ${stats_db_name}.indi_project_collab_org_country purge; +drop table if exists ${stats_db_name}.indi_project_collab_org_country purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_project_collab_org_country stored as parquet as select o1.organization org1,o1.name org1name, o2.country country2, count(distinct o1.project) as collaborations from ${stats_db_name}.tmp as o1 join ${stats_db_name}.tmp as o2 on o1.project=o2.project where o1.organization<>o2.organization and o1.country<>o2.country -group by o1.organization, o2.country, o1.name; +group by o1.organization, o2.country, o1.name; /*EOS*/ -drop table if exists ${stats_db_name}.tmp purge; +drop table if exists ${stats_db_name}.tmp purge; /*EOS*/ -drop table if exists ${stats_db_name}.indi_funder_country_collab purge; +drop table if exists ${stats_db_name}.indi_funder_country_collab purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_funder_country_collab stored as parquet as with tmp as (select funder, project, country from ${stats_db_name}.organization_projects op @@ -174,26 +174,26 @@ select f1.funder, f1.country as country1, f2.country as country2, count(distinct from tmp as f1 join tmp as f2 on f1.project=f2.project where f1.country<>f2.country -group by f1.funder, f2.country, f1.country; +group by f1.funder, f2.country, f1.country; /*EOS*/ create TEMPORARY TABLE ${stats_db_name}.tmp AS select distinct country, ro.id as result from ${stats_db_name}.organization o join ${stats_db_name}.result_organization ro on o.id=ro.organization - where country <> 'UNKNOWN' and o.name is not null; + where country <> 'UNKNOWN' and o.name is not null; /*EOS*/ -drop table if exists ${stats_db_name}.indi_result_country_collab purge; +drop table if exists ${stats_db_name}.indi_result_country_collab purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_result_country_collab stored as parquet as select o1.country country1, o2.country country2, count(o1.result) as collaborations from ${stats_db_name}.tmp as o1 join ${stats_db_name}.tmp as o2 on o1.result=o2.result where o1.country<>o2.country -group by o1.country, o2.country; +group by o1.country, o2.country; /*EOS*/ -drop table if exists ${stats_db_name}.tmp purge; +drop table if exists ${stats_db_name}.tmp purge; /*EOS*/ ---- Sprint 4 ---- -drop table if exists ${stats_db_name}.indi_pub_diamond purge; +drop table if exists ${stats_db_name}.indi_pub_diamond purge; /*EOS*/ --create table if not exists ${stats_db_name}.indi_pub_diamond stored as parquet as --select distinct pd.id, coalesce(in_diamond_journal, 0) as in_diamond_journal @@ -212,9 +212,9 @@ left outer join (select pd.id, 1 as in_diamond_journal from ${stats_db_name}.pub join ${stats_db_name}.datasource d on d.id=pd.datasource join STATS_EXT.plan_s_jn ps where (ps.issn_print=d.issn_printed and ps.issn_online=d.issn_online) and (ps.journal_is_in_doaj=true or ps.journal_is_oa=true) and ps.has_apc=false) tmp -on pd.id=tmp.id; +on pd.id=tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_in_transformative purge; +drop table if exists ${stats_db_name}.indi_pub_in_transformative purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_in_transformative stored as parquet as select distinct pd.id, coalesce(is_transformative, 0) as is_transformative @@ -224,9 +224,9 @@ from ${stats_db_name}.publication pd join ${stats_db_name}.datasource d on d.id=pd.datasource join STATS_EXT.plan_s_jn ps where (ps.issn_print=d.issn_printed and ps.issn_online=d.issn_online) and ps.is_transformative_journal=true) tmp - on pd.id=tmp.id; + on pd.id=tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_closed_other_open purge; +drop table if exists ${stats_db_name}.indi_pub_closed_other_open purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_closed_other_open stored as parquet as select distinct ri.id, coalesce(pub_closed_other_open, 0) as pub_closed_other_open from ${stats_db_name}.result_instance ri @@ -236,53 +236,53 @@ select distinct ri.id, coalesce(pub_closed_other_open, 0) as pub_closed_other_op join ${stats_db_name}.datasource d on ri.hostedby=d.id where d.type like '%Journal%' and ri.accessright='Closed Access' and (p.bestlicence='Open Access' or p.bestlicence='Open Source')) tmp - on tmp.id=ri.id; + on tmp.id=ri.id; /*EOS*/ ---- Sprint 5 ---- -drop table if exists ${stats_db_name}.indi_result_no_of_copies purge; +drop table if exists ${stats_db_name}.indi_result_no_of_copies purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_result_no_of_copies stored as parquet as -select id, count(id) as number_of_copies from ${stats_db_name}.result_instance group by id; +select id, count(id) as number_of_copies from ${stats_db_name}.result_instance group by id; /*EOS*/ ---- Sprint 6 ---- -drop table if exists ${stats_db_name}.indi_pub_downloads purge; +drop table if exists ${stats_db_name}.indi_pub_downloads purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_downloads stored as parquet as SELECT result_id, sum(downloads) no_downloads from openaire_prod_usage_stats.usage_stats join ${stats_db_name}.publication on result_id=id where downloads>0 GROUP BY result_id -order by no_downloads desc; +order by no_downloads desc; /*EOS*/ --ANALYZE TABLE ${stats_db_name}.indi_pub_downloads COMPUTE STATISTICS; -drop table if exists ${stats_db_name}.indi_pub_downloads_datasource purge; +drop table if exists ${stats_db_name}.indi_pub_downloads_datasource purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_downloads_datasource stored as parquet as SELECT result_id, repository_id, sum(downloads) no_downloads from openaire_prod_usage_stats.usage_stats join ${stats_db_name}.publication on result_id=id where downloads>0 GROUP BY result_id, repository_id -order by result_id; +order by result_id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_downloads_year purge; +drop table if exists ${stats_db_name}.indi_pub_downloads_year purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_downloads_year stored as parquet as SELECT result_id, cast(substring(us.`date`, 1,4) as int) as `year`, sum(downloads) no_downloads from openaire_prod_usage_stats.usage_stats us join ${stats_db_name}.publication on result_id=id where downloads>0 -GROUP BY result_id, substring(us.`date`, 1,4); +GROUP BY result_id, substring(us.`date`, 1,4); /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_downloads_datasource_year purge; +drop table if exists ${stats_db_name}.indi_pub_downloads_datasource_year purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_downloads_datasource_year stored as parquet as SELECT result_id, cast(substring(us.`date`, 1,4) as int) as `year`, repository_id, sum(downloads) no_downloads from openaire_prod_usage_stats.usage_stats us join ${stats_db_name}.publication on result_id=id where downloads>0 -GROUP BY result_id, repository_id, substring(us.`date`, 1,4); +GROUP BY result_id, repository_id, substring(us.`date`, 1,4); /*EOS*/ ---- Sprint 7 ---- -drop table if exists ${stats_db_name}.indi_pub_gold_oa purge; +drop table if exists ${stats_db_name}.indi_pub_gold_oa purge; /*EOS*/ --create table if not exists ${stats_db_name}.indi_pub_gold_oa stored as parquet as -- WITH gold_oa AS ( SELECT @@ -381,9 +381,9 @@ left outer join ( select pd.id, 1 as is_gold FROM ${stats_db_name}.publication_datasources pd 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; + 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; +drop table if exists ${stats_db_name}.indi_pub_hybrid_oa_with_cc purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_hybrid_oa_with_cc stored as parquet as WITH hybrid_oa AS ( @@ -414,9 +414,9 @@ FROM ${stats_db_name}.publication_datasources pd JOIN hybrid_oa ON issn.issn = hybrid_oa.issn JOIN ${stats_db_name}.indi_result_has_cc_licence cc on pd.id=cc.id JOIN ${stats_db_name}.indi_pub_gold_oa ga on pd.id=ga.id - where cc.has_cc_license=1 and ga.is_gold=0) tmp on pd.id=tmp.id; + where cc.has_cc_license=1 and ga.is_gold=0) tmp on pd.id=tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_hybrid purge; +drop table if exists ${stats_db_name}.indi_pub_hybrid purge; /*EOS*/ --create table if not exists ${stats_db_name}.indi_pub_hybrid stored as parquet as -- WITH gold_oa AS ( SELECT @@ -489,9 +489,9 @@ join ${stats_db_name}.result_accessroute ra on ra.id=pd.id join ${stats_db_name}.datasource d on d.id=ri.hostedby where indi_gold.is_gold=0 and ((d.type like '%Journal%' and ri.accessright!='Closed Access' and ri.accessright!='Restricted' and ri.license is not null) or ra.accessroute='hybrid'))tmp -on pd.id=tmp.id; +on pd.id=tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_fairness purge; +drop table if exists ${stats_db_name}.indi_org_fairness purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_fairness stored as parquet as --return results with PIDs, and rich metadata group by organization @@ -509,7 +509,7 @@ create table if not exists ${stats_db_name}.indi_org_fairness stored as parquet --return results_fair/all_results select allresults.organization, result_fair.no_result_fair/allresults.no_allresults org_fairness from allresults - join result_fair on result_fair.organization=allresults.organization; + join result_fair on result_fair.organization=allresults.organization; /*EOS*/ CREATE TEMPORARY table ${stats_db_name}.result_fair as select ro.organization organization, count(distinct ro.id) no_result_fair @@ -519,44 +519,44 @@ select ro.organization organization, count(distinct ro.id) no_result_fair join ${stats_db_name}.indi_pub_grey_lit gl on gl.id=p.id where (title is not null) and (publisher is not null) and (abstract=true) and (year is not null) and (authors>0) and cast(year as int)>2003 and dc.doi_from_crossref=1 and gl.grey_lit=0 - group by ro.organization; + group by ro.organization; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allresults as select ro.organization, count(distinct ro.id) no_allresults from ${stats_db_name}.result_organization ro join ${stats_db_name}.publication p on p.id=ro.id where cast(year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_fairness_pub_pr purge; +drop table if exists ${stats_db_name}.indi_org_fairness_pub_pr purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_fairness_pub_pr stored as parquet as select ar.organization, rf.no_result_fair/ar.no_allresults org_fairness from ${stats_db_name}.allresults ar - join ${stats_db_name}.result_fair rf on rf.organization=ar.organization; + join ${stats_db_name}.result_fair rf on rf.organization=ar.organization; /*EOS*/ -DROP table ${stats_db_name}.result_fair purge; -DROP table ${stats_db_name}.allresults purge; +DROP table ${stats_db_name}.result_fair purge; /*EOS*/ +DROP table ${stats_db_name}.allresults purge; /*EOS*/ CREATE TEMPORARY table ${stats_db_name}.result_fair as select year, ro.organization organization, count(distinct ro.id) no_result_fair from ${stats_db_name}.result_organization ro join ${stats_db_name}.result p on p.id=ro.id where (title is not null) and (publisher is not null) and (abstract=true) and (year is not null) and (authors>0) and cast(year as int)>2003 - group by ro.organization, year; + group by ro.organization, year; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allresults as select year, ro.organization, count(distinct ro.id) no_allresults from ${stats_db_name}.result_organization ro join ${stats_db_name}.result p on p.id=ro.id where cast(year as int)>2003 - group by ro.organization, year; + group by ro.organization, year; /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_fairness_pub_year purge; +drop table if exists ${stats_db_name}.indi_org_fairness_pub_year purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_fairness_pub_year stored as parquet as select cast(allresults.year as int) year, allresults.organization, result_fair.no_result_fair/allresults.no_allresults org_fairness from ${stats_db_name}.allresults - join ${stats_db_name}.result_fair on result_fair.organization=allresults.organization and result_fair.year=allresults.year; + join ${stats_db_name}.result_fair on result_fair.organization=allresults.organization and result_fair.year=allresults.year; /*EOS*/ -DROP table ${stats_db_name}.result_fair purge; -DROP table ${stats_db_name}.allresults purge; +DROP table ${stats_db_name}.result_fair purge; /*EOS*/ +DROP table ${stats_db_name}.allresults purge; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.result_fair as select ro.organization organization, count(distinct ro.id) no_result_fair @@ -564,92 +564,92 @@ CREATE TEMPORARY TABLE ${stats_db_name}.result_fair as join ${stats_db_name}.result p on p.id=ro.id where (title is not null) and (publisher is not null) and (abstract=true) and (year is not null) and (authors>0) and cast(year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allresults as select ro.organization, count(distinct ro.id) no_allresults from ${stats_db_name}.result_organization ro join ${stats_db_name}.result p on p.id=ro.id where cast(year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_fairness_pub purge; +drop table if exists ${stats_db_name}.indi_org_fairness_pub purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_fairness_pub as select ar.organization, rf.no_result_fair/ar.no_allresults org_fairness from ${stats_db_name}.allresults ar join ${stats_db_name}.result_fair rf -on rf.organization=ar.organization; +on rf.organization=ar.organization; /*EOS*/ -DROP table ${stats_db_name}.result_fair purge; -DROP table ${stats_db_name}.allresults purge; +DROP table ${stats_db_name}.result_fair purge; /*EOS*/ +DROP table ${stats_db_name}.allresults purge; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.result_fair as select year, ro.organization organization, count(distinct ro.id) no_result_fair from ${stats_db_name}.result_organization ro join ${stats_db_name}.result r on r.id=ro.id join ${stats_db_name}.result_pids rp on r.id=rp.id where (title is not null) and (publisher is not null) and (abstract=true) and (year is not null) and (authors>0) and cast(year as int)>2003 - group by ro.organization, year; + group by ro.organization, year; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allresults as select year, ro.organization, count(distinct ro.id) no_allresults from ${stats_db_name}.result_organization ro join ${stats_db_name}.result r on r.id=ro.id where cast(year as int)>2003 - group by ro.organization, year; + group by ro.organization, year; /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_fairness_year purge; +drop table if exists ${stats_db_name}.indi_org_fairness_year purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_fairness_year stored as parquet as select cast(allresults.year as int) year, allresults.organization, result_fair.no_result_fair/allresults.no_allresults org_fairness from ${stats_db_name}.allresults - join ${stats_db_name}.result_fair on result_fair.organization=allresults.organization and cast(result_fair.year as int)=cast(allresults.year as int); + join ${stats_db_name}.result_fair on result_fair.organization=allresults.organization and cast(result_fair.year as int)=cast(allresults.year as int); /*EOS*/ -DROP table ${stats_db_name}.result_fair purge; -DROP table ${stats_db_name}.allresults purge; +DROP table ${stats_db_name}.result_fair purge; /*EOS*/ +DROP table ${stats_db_name}.allresults purge; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.result_with_pid as select year, ro.organization, count(distinct rp.id) no_result_with_pid from ${stats_db_name}.result_organization ro join ${stats_db_name}.result_pids rp on rp.id=ro.id join ${stats_db_name}.result r on r.id=rp.id where cast(year as int) >2003 - group by ro.organization, year; + group by ro.organization, year; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allresults as select year, ro.organization, count(distinct ro.id) no_allresults from ${stats_db_name}.result_organization ro join ${stats_db_name}.result r on r.id=ro.id where cast(year as int) >2003 - group by ro.organization, year; + group by ro.organization, year; /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_findable_year purge; +drop table if exists ${stats_db_name}.indi_org_findable_year purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_findable_year stored as parquet as select cast(allresults.year as int) year, allresults.organization, result_with_pid.no_result_with_pid/allresults.no_allresults org_findable from ${stats_db_name}.allresults - join ${stats_db_name}.result_with_pid on result_with_pid.organization=allresults.organization and cast(result_with_pid.year as int)=cast(allresults.year as int); + join ${stats_db_name}.result_with_pid on result_with_pid.organization=allresults.organization and cast(result_with_pid.year as int)=cast(allresults.year as int); /*EOS*/ -DROP table ${stats_db_name}.result_with_pid purge; -DROP table ${stats_db_name}.allresults purge; +DROP table ${stats_db_name}.result_with_pid purge; /*EOS*/ +DROP table ${stats_db_name}.allresults purge; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.result_with_pid as select ro.organization, count(distinct rp.id) no_result_with_pid from ${stats_db_name}.result_organization ro join ${stats_db_name}.result_pids rp on rp.id=ro.id join ${stats_db_name}.result r on r.id=rp.id where cast(year as int) >2003 - group by ro.organization; + group by ro.organization; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allresults as select ro.organization, count(distinct ro.id) no_allresults from ${stats_db_name}.result_organization ro join ${stats_db_name}.result r on r.id=ro.id where cast(year as int) >2003 - group by ro.organization; + group by ro.organization; /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_findable purge; +drop table if exists ${stats_db_name}.indi_org_findable purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_findable stored as parquet as select allresults.organization, result_with_pid.no_result_with_pid/allresults.no_allresults org_findable from ${stats_db_name}.allresults - join ${stats_db_name}.result_with_pid on result_with_pid.organization=allresults.organization; + join ${stats_db_name}.result_with_pid on result_with_pid.organization=allresults.organization; /*EOS*/ -DROP table ${stats_db_name}.result_with_pid purge; -DROP table ${stats_db_name}.allresults purge; +DROP table ${stats_db_name}.result_with_pid purge; /*EOS*/ +DROP table ${stats_db_name}.allresults purge; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.pubs_oa as SELECT ro.organization, count(distinct r.id) no_oapubs FROM ${stats_db_name}.publication r @@ -657,7 +657,7 @@ SELECT ro.organization, count(distinct r.id) no_oapubs FROM ${stats_db_name}.pub join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.datasets_oa as SELECT ro.organization, count(distinct r.id) no_oadatasets FROM ${stats_db_name}.dataset r @@ -665,7 +665,7 @@ SELECT ro.organization, count(distinct r.id) no_oadatasets FROM ${stats_db_name} join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.software_oa as SELECT ro.organization, count(distinct r.id) no_oasoftware FROM ${stats_db_name}.software r @@ -673,41 +673,41 @@ SELECT ro.organization, count(distinct r.id) no_oasoftware FROM ${stats_db_name} join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allpubs as SELECT ro.organization, count(ro.id) no_allpubs FROM ${stats_db_name}.result_organization ro join ${stats_db_name}.publication ps on ps.id=ro.id where cast(ps.year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.alldatasets as SELECT ro.organization, count(ro.id) no_alldatasets FROM ${stats_db_name}.result_organization ro join ${stats_db_name}.dataset ps on ps.id=ro.id where cast(ps.year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allsoftware as SELECT ro.organization, count(ro.id) no_allsoftware FROM ${stats_db_name}.result_organization ro join ${stats_db_name}.software ps on ps.id=ro.id where cast(ps.year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allpubsshare as select pubs_oa.organization, pubs_oa.no_oapubs/allpubs.no_allpubs p from ${stats_db_name}.allpubs - join ${stats_db_name}.pubs_oa on allpubs.organization=pubs_oa.organization; + join ${stats_db_name}.pubs_oa on allpubs.organization=pubs_oa.organization; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.alldatasetssshare as select datasets_oa.organization, datasets_oa.no_oadatasets/alldatasets.no_alldatasets d from ${stats_db_name}.alldatasets - join ${stats_db_name}.datasets_oa on alldatasets.organization=datasets_oa.organization; + join ${stats_db_name}.datasets_oa on alldatasets.organization=datasets_oa.organization; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allsoftwaresshare as select software_oa.organization, software_oa.no_oasoftware/allsoftware.no_allsoftware s from ${stats_db_name}.allsoftware - join ${stats_db_name}.software_oa on allsoftware.organization=software_oa.organization; + join ${stats_db_name}.software_oa on allsoftware.organization=software_oa.organization; /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_openess purge; +drop table if exists ${stats_db_name}.indi_org_openess purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_openess stored as parquet as select allpubsshare.organization, @@ -719,17 +719,17 @@ select allpubsshare.organization, on tmp1.organization=allpubsshare.organization left outer join (select organization,s from ${stats_db_name}.allsoftwaresshare) tmp2 - on tmp2.organization=allpubsshare.organization; + on tmp2.organization=allpubsshare.organization; /*EOS*/ -DROP TABLE ${stats_db_name}.pubs_oa purge; -DROP TABLE ${stats_db_name}.datasets_oa purge; -DROP TABLE ${stats_db_name}.software_oa purge; -DROP TABLE ${stats_db_name}.allpubs purge; -DROP TABLE ${stats_db_name}.alldatasets purge; -DROP TABLE ${stats_db_name}.allsoftware purge; -DROP TABLE ${stats_db_name}.allpubsshare purge; -DROP TABLE ${stats_db_name}.alldatasetssshare purge; -DROP TABLE ${stats_db_name}.allsoftwaresshare purge; +DROP TABLE ${stats_db_name}.pubs_oa purge; /*EOS*/ +DROP TABLE ${stats_db_name}.datasets_oa purge; /*EOS*/ +DROP TABLE ${stats_db_name}.software_oa purge; /*EOS*/ +DROP TABLE ${stats_db_name}.allpubs purge; /*EOS*/ +DROP TABLE ${stats_db_name}.alldatasets purge; /*EOS*/ +DROP TABLE ${stats_db_name}.allsoftware purge; /*EOS*/ +DROP TABLE ${stats_db_name}.allpubsshare purge; /*EOS*/ +DROP TABLE ${stats_db_name}.alldatasetssshare purge; /*EOS*/ +DROP TABLE ${stats_db_name}.allsoftwaresshare purge; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.pubs_oa AS SELECT r.year, ro.organization, count(distinct r.id) no_oapubs FROM ${stats_db_name}.publication r @@ -737,7 +737,7 @@ SELECT r.year, ro.organization, count(distinct r.id) no_oapubs FROM ${stats_db_n join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 - group by ro.organization,r.year; + group by ro.organization,r.year; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.datasets_oa AS SELECT r.year,ro.organization, count(distinct r.id) no_oadatasets FROM ${stats_db_name}.dataset r @@ -745,7 +745,7 @@ SELECT r.year,ro.organization, count(distinct r.id) no_oadatasets FROM ${stats_d join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 - group by ro.organization, r.year; + group by ro.organization, r.year; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.software_oa AS SELECT r.year,ro.organization, count(distinct r.id) no_oasoftware FROM ${stats_db_name}.software r @@ -753,38 +753,38 @@ SELECT r.year,ro.organization, count(distinct r.id) no_oasoftware FROM ${stats_d join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 - group by ro.organization, r.year; + group by ro.organization, r.year; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allpubs as SELECT p.year,ro.organization organization, count(ro.id) no_allpubs FROM ${stats_db_name}.result_organization ro join ${stats_db_name}.publication p on p.id=ro.id where cast(p.year as int)>2003 - group by ro.organization, p.year; + group by ro.organization, p.year; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.alldatasets as SELECT d.year, ro.organization organization, count(ro.id) no_alldatasets FROM ${stats_db_name}.result_organization ro join ${stats_db_name}.dataset d on d.id=ro.id where cast(d.year as int)>2003 - group by ro.organization, d.year; + group by ro.organization, d.year; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allsoftware as SELECT s.year,ro.organization organization, count(ro.id) no_allsoftware FROM ${stats_db_name}.result_organization ro join ${stats_db_name}.software s on s.id=ro.id where cast(s.year as int)>2003 - group by ro.organization, s.year; + group by ro.organization, s.year; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allpubsshare as select allpubs.year, pubs_oa.organization, pubs_oa.no_oapubs/allpubs.no_allpubs p from ${stats_db_name}.allpubs - join ${stats_db_name}.pubs_oa on allpubs.organization=pubs_oa.organization where cast(allpubs.year as INT)=cast(pubs_oa.year as int); + join ${stats_db_name}.pubs_oa on allpubs.organization=pubs_oa.organization where cast(allpubs.year as INT)=cast(pubs_oa.year as int); /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.alldatasetssshare as select alldatasets.year, datasets_oa.organization, datasets_oa.no_oadatasets/alldatasets.no_alldatasets d from ${stats_db_name}.alldatasets - join ${stats_db_name}.datasets_oa on alldatasets.organization=datasets_oa.organization where cast(alldatasets.year as INT)=cast(datasets_oa.year as int); + join ${stats_db_name}.datasets_oa on alldatasets.organization=datasets_oa.organization where cast(alldatasets.year as INT)=cast(datasets_oa.year as int); /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allsoftwaresshare as select allsoftware.year, software_oa.organization, software_oa.no_oasoftware/allsoftware.no_allsoftware s from ${stats_db_name}.allsoftware - join ${stats_db_name}.software_oa on allsoftware.organization=software_oa.organization where cast(allsoftware.year as INT)=cast(software_oa.year as int); + join ${stats_db_name}.software_oa on allsoftware.organization=software_oa.organization where cast(allsoftware.year as INT)=cast(software_oa.year as int); /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_openess_year purge; +drop table if exists ${stats_db_name}.indi_org_openess_year purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_openess_year stored as parquet as select cast(allpubsshare.year as int) year, allpubsshare.organization, @@ -796,19 +796,19 @@ select cast(allpubsshare.year as int) year, allpubsshare.organization, on tmp1.organization=allpubsshare.organization and tmp1.year=allpubsshare.year left outer join (select cast(year as int), organization,s from ${stats_db_name}.allsoftwaresshare) tmp2 - on tmp2.organization=allpubsshare.organization and cast(tmp2.year as int)=cast(allpubsshare.year as int); + on tmp2.organization=allpubsshare.organization and cast(tmp2.year as int)=cast(allpubsshare.year as int); /*EOS*/ -DROP TABLE ${stats_db_name}.pubs_oa purge; -DROP TABLE ${stats_db_name}.datasets_oa purge; -DROP TABLE ${stats_db_name}.software_oa purge; -DROP TABLE ${stats_db_name}.allpubs purge; -DROP TABLE ${stats_db_name}.alldatasets purge; -DROP TABLE ${stats_db_name}.allsoftware purge; -DROP TABLE ${stats_db_name}.allpubsshare purge; -DROP TABLE ${stats_db_name}.alldatasetssshare purge; -DROP TABLE ${stats_db_name}.allsoftwaresshare purge; +DROP TABLE ${stats_db_name}.pubs_oa purge; /*EOS*/ +DROP TABLE ${stats_db_name}.datasets_oa purge; /*EOS*/ +DROP TABLE ${stats_db_name}.software_oa purge; /*EOS*/ +DROP TABLE ${stats_db_name}.allpubs purge; /*EOS*/ +DROP TABLE ${stats_db_name}.alldatasets purge; /*EOS*/ +DROP TABLE ${stats_db_name}.allsoftware purge; /*EOS*/ +DROP TABLE ${stats_db_name}.allpubsshare purge; /*EOS*/ +DROP TABLE ${stats_db_name}.alldatasetssshare purge; /*EOS*/ +DROP TABLE ${stats_db_name}.allsoftwaresshare purge; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_has_preprint purge; +drop table if exists ${stats_db_name}.indi_pub_has_preprint purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_has_preprint stored as parquet as select distinct p.id, coalesce(has_preprint, 0) as has_preprint @@ -817,8 +817,8 @@ from ${stats_db_name}.publication_classifications p select p.id, 1 as has_preprint from ${stats_db_name}.publication_classifications p where p.type='Preprint') tmp - on p.id= tmp.id; -drop table if exists ${stats_db_name}.indi_pub_in_subscribed purge; + on p.id= tmp.id; /*EOS*/ +drop table if exists ${stats_db_name}.indi_pub_in_subscribed purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_in_subscribed stored as parquet as select distinct p.id, coalesce(is_subscription, 0) as is_subscription @@ -829,9 +829,9 @@ from ${stats_db_name}.publication p join ${stats_db_name}.indi_pub_hybrid h on p.id=h.id join ${stats_db_name}.indi_pub_in_transformative t on p.id=t.id where g.is_gold=0 and h.is_hybrid=0 and t.is_transformative=0) tmp - on p.id=tmp.id; + on p.id=tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_result_with_pid purge; +drop table if exists ${stats_db_name}.indi_result_with_pid purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_result_with_pid as select distinct p.id, coalesce(result_with_pid, 0) as result_with_pid @@ -839,13 +839,13 @@ from ${stats_db_name}.result p left outer join ( select p.id, 1 as result_with_pid from ${stats_db_name}.result_pids p) tmp - on p.id= tmp.id; + on p.id= tmp.id; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.pub_fos_totals as select rf.id, count(distinct lvl3) totals from ${stats_db_name}.result_fos rf -group by rf.id; +group by rf.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_interdisciplinarity purge; +drop table if exists ${stats_db_name}.indi_pub_interdisciplinarity purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_interdisciplinarity as select distinct p.id as id, coalesce(is_interdisciplinary, 0) @@ -853,11 +853,11 @@ as is_interdisciplinary from ${stats_db_name}.pub_fos_totals p left outer join ( select pub_fos_totals.id, 1 as is_interdisciplinary from ${stats_db_name}.pub_fos_totals -where totals>1) tmp on p.id=tmp.id; +where totals>1) tmp on p.id=tmp.id; /*EOS*/ -drop table ${stats_db_name}.pub_fos_totals purge; +drop table ${stats_db_name}.pub_fos_totals purge; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_bronze_oa purge; +drop table if exists ${stats_db_name}.indi_pub_bronze_oa purge; /*EOS*/ --create table if not exists ${stats_db_name}.indi_pub_bronze_oa stored as parquet as --select distinct p.id, coalesce(is_bronze_oa,0) as is_bronze_oa @@ -883,16 +883,16 @@ join ${stats_db_name}.datasource d on d.id=ri.hostedby where indi_gold.is_gold=0 and indi_hybrid.is_hybrid=0 and ((d.type like '%Journal%' and ri.accessright!='Closed Access' and ri.accessright!='Restricted' and ri.license is null) or ra.accessroute='bronze')) tmp -on pd.id=tmp.id; +on pd.id=tmp.id; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.project_year_result_year as select p.id project_id, acronym, r.id result_id, r.year, p.end_year from ${stats_db_name}.project p join ${stats_db_name}.result_projects rp on p.id=rp.project join ${stats_db_name}.result r on r.id=rp.id -where p.end_year is NOT NULL and r.year is not null; +where p.end_year is NOT NULL and r.year is not null; /*EOS*/ -drop table if exists ${stats_db_name}.indi_is_project_result_after purge; +drop table if exists ${stats_db_name}.indi_is_project_result_after purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_is_project_result_after stored as parquet as select pry.project_id, pry.acronym, pry.result_id, @@ -900,21 +900,21 @@ coalesce(is_project_result_after, 0) as is_project_result_after from ${stats_db_name}.project_year_result_year pry left outer join (select pry.project_id, pry.acronym, pry.result_id, 1 as is_project_result_after from ${stats_db_name}.project_year_result_year pry -where pry.year>pry.end_year) tmp on pry.result_id=tmp.result_id; +where pry.year>pry.end_year) tmp on pry.result_id=tmp.result_id; /*EOS*/ -drop table ${stats_db_name}.project_year_result_year purge; +drop table ${stats_db_name}.project_year_result_year purge; /*EOS*/ -drop table ${stats_db_name}.indi_is_funder_plan_s purge; +drop table ${stats_db_name}.indi_is_funder_plan_s purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_is_funder_plan_s stored as parquet as select distinct f.id, f.name, coalesce(is_funder_plan_s, 0) as is_funder_plan_s from ${stats_db_name}.funder f left outer join (select id, name, 1 as is_funder_plan_s from ${stats_db_name}.funder join stats_ext.plan_s_short on c_o_alition_s_organisation_funder=name) tmp - on f.name= tmp.name; + on f.name= tmp.name; /*EOS*/ --Funder Fairness -drop table ${stats_db_name}.indi_funder_fairness purge; +drop table ${stats_db_name}.indi_funder_fairness purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_funder_fairness stored as parquet as with result_fair as @@ -930,10 +930,10 @@ create table if not exists ${stats_db_name}.indi_funder_fairness stored as parqu group by p.funder) select allresults.funder, result_fair.no_result_fair/allresults.no_allresults funder_fairness from allresults - join result_fair on result_fair.funder=allresults.funder; + join result_fair on result_fair.funder=allresults.funder; /*EOS*/ --RIs Fairness -drop table ${stats_db_name}.indi_ris_fairness purge; +drop table ${stats_db_name}.indi_ris_fairness purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_ris_fairness stored as parquet as with result_contexts as @@ -953,7 +953,7 @@ allresults as group by rc.ri_initiative) select allresults.ri_initiative, result_fair.no_result_fair/allresults.no_allresults ris_fairness from allresults - join result_fair on result_fair.ri_initiative=allresults.ri_initiative; + join result_fair on result_fair.ri_initiative=allresults.ri_initiative; /*EOS*/ --Funder Openess @@ -964,7 +964,7 @@ join ${stats_db_name}.publication r on r.id=rp.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 -group by p.funder; +group by p.funder; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.datasets_oa as @@ -974,7 +974,7 @@ join ${stats_db_name}.dataset r on r.id=rp.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 -group by p.funder; +group by p.funder; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.software_oa as select p.funder funder, count(distinct rp.id) no_oasoftware from ${stats_db_name}.result_projects rp @@ -983,44 +983,44 @@ join ${stats_db_name}.software r on r.id=rp.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 -group by p.funder; +group by p.funder; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allpubs as select p.funder funder, count(distinct rp.id) no_allpubs from ${stats_db_name}.result_projects rp join ${stats_db_name}.project p on p.id=rp.project join ${stats_db_name}.publication r on r.id=rp.id where cast(r.year as int)>2003 -group by p.funder; +group by p.funder; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.alldatasets as select p.funder funder, count(distinct rp.id) no_alldatasets from ${stats_db_name}.result_projects rp join ${stats_db_name}.project p on p.id=rp.project join ${stats_db_name}.dataset r on r.id=rp.id where cast(r.year as int)>2003 -group by p.funder; +group by p.funder; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allsoftware as select p.funder funder, count(distinct rp.id) no_allsoftware from ${stats_db_name}.result_projects rp join ${stats_db_name}.project p on p.id=rp.project join ${stats_db_name}.software r on r.id=rp.id where cast(r.year as int)>2003 -group by p.funder; +group by p.funder; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allpubsshare as select pubs_oa.funder, pubs_oa.no_oapubs/allpubs.no_allpubs p from ${stats_db_name}.allpubs - join ${stats_db_name}.pubs_oa on allpubs.funder=pubs_oa.funder; + join ${stats_db_name}.pubs_oa on allpubs.funder=pubs_oa.funder; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.alldatasetssshare as select datasets_oa.funder, datasets_oa.no_oadatasets/alldatasets.no_alldatasets d from ${stats_db_name}.alldatasets - join ${stats_db_name}.datasets_oa on alldatasets.funder=datasets_oa.funder; + join ${stats_db_name}.datasets_oa on alldatasets.funder=datasets_oa.funder; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allsoftwaresshare as select software_oa.funder, software_oa.no_oasoftware/allsoftware.no_allsoftware s from ${stats_db_name}.allsoftware - join ${stats_db_name}.software_oa on allsoftware.funder=software_oa.funder; + join ${stats_db_name}.software_oa on allsoftware.funder=software_oa.funder; /*EOS*/ -drop table ${stats_db_name}.indi_funder_openess purge; +drop table ${stats_db_name}.indi_funder_openess purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_funder_openess stored as parquet as select allpubsshare.funder, @@ -1032,17 +1032,17 @@ select allpubsshare.funder, on tmp1.funder=allpubsshare.funder left outer join (select funder,s from ${stats_db_name}.allsoftwaresshare) tmp2 - on tmp2.funder=allpubsshare.funder; + on tmp2.funder=allpubsshare.funder; /*EOS*/ -DROP TABLE ${stats_db_name}.pubs_oa purge; -DROP TABLE ${stats_db_name}.datasets_oa purge; -DROP TABLE ${stats_db_name}.software_oa purge; -DROP TABLE ${stats_db_name}.allpubs purge; -DROP TABLE ${stats_db_name}.alldatasets purge; -DROP TABLE ${stats_db_name}.allsoftware purge; -DROP TABLE ${stats_db_name}.allpubsshare purge; -DROP TABLE ${stats_db_name}.alldatasetssshare purge; -DROP TABLE ${stats_db_name}.allsoftwaresshare purge; +DROP TABLE ${stats_db_name}.pubs_oa purge; /*EOS*/ +DROP TABLE ${stats_db_name}.datasets_oa purge; /*EOS*/ +DROP TABLE ${stats_db_name}.software_oa purge; /*EOS*/ +DROP TABLE ${stats_db_name}.allpubs purge; /*EOS*/ +DROP TABLE ${stats_db_name}.alldatasets purge; /*EOS*/ +DROP TABLE ${stats_db_name}.allsoftware purge; /*EOS*/ +DROP TABLE ${stats_db_name}.allpubsshare purge; /*EOS*/ +DROP TABLE ${stats_db_name}.alldatasetssshare purge; /*EOS*/ +DROP TABLE ${stats_db_name}.allsoftwaresshare purge; /*EOS*/ --RIs Openess @@ -1050,7 +1050,7 @@ CREATE TEMPORARY TABLE ${stats_db_name}.result_contexts as select distinct rc.id, context.name ri_initiative from ${stats_db_name}.result_concepts rc join ${stats_db_name}.concept on concept.id=rc.concept join ${stats_db_name}.category on category.id=concept.category -join ${stats_db_name}.context on context.id=category.context; +join ${stats_db_name}.context on context.id=category.context; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.pubs_oa as select rp.ri_initiative ri_initiative, count(distinct rp.id) no_oapubs from ${stats_db_name}.result_contexts rp @@ -1058,7 +1058,7 @@ join ${stats_db_name}.publication r on r.id=rp.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 -group by rp.ri_initiative; +group by rp.ri_initiative; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.datasets_oa as select rp.ri_initiative ri_initiative, count(distinct rp.id) no_oadatasets from ${stats_db_name}.result_contexts rp @@ -1066,7 +1066,7 @@ join ${stats_db_name}.dataset r on r.id=rp.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 -group by rp.ri_initiative; +group by rp.ri_initiative; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.software_oa as select rp.ri_initiative ri_initiative, count(distinct rp.id) no_oasoftware from ${stats_db_name}.result_contexts rp @@ -1074,41 +1074,41 @@ join ${stats_db_name}.software r on r.id=rp.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 -group by rp.ri_initiative; +group by rp.ri_initiative; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allpubs as select rp.ri_initiative ri_initiative, count(distinct rp.id) no_allpubs from ${stats_db_name}.result_contexts rp join ${stats_db_name}.publication r on r.id=rp.id where cast(r.year as int)>2003 -group by rp.ri_initiative; +group by rp.ri_initiative; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.alldatasets as select rp.ri_initiative ri_initiative, count(distinct rp.id) no_alldatasets from ${stats_db_name}.result_contexts rp join ${stats_db_name}.dataset r on r.id=rp.id where cast(r.year as int)>2003 -group by rp.ri_initiative; +group by rp.ri_initiative; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allsoftware as select rp.ri_initiative ri_initiative, count(distinct rp.id) no_allsoftware from ${stats_db_name}.result_contexts rp join ${stats_db_name}.software r on r.id=rp.id where cast(r.year as int)>2003 -group by rp.ri_initiative; +group by rp.ri_initiative; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allpubsshare as select pubs_oa.ri_initiative, pubs_oa.no_oapubs/allpubs.no_allpubs p from ${stats_db_name}.allpubs - join ${stats_db_name}.pubs_oa on allpubs.ri_initiative=pubs_oa.ri_initiative; + join ${stats_db_name}.pubs_oa on allpubs.ri_initiative=pubs_oa.ri_initiative; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.alldatasetssshare as select datasets_oa.ri_initiative, datasets_oa.no_oadatasets/alldatasets.no_alldatasets d from ${stats_db_name}.alldatasets - join ${stats_db_name}.datasets_oa on alldatasets.ri_initiative=datasets_oa.ri_initiative; + join ${stats_db_name}.datasets_oa on alldatasets.ri_initiative=datasets_oa.ri_initiative; /*EOS*/ CREATE TEMPORARY TABLE ${stats_db_name}.allsoftwaresshare as select software_oa.ri_initiative, software_oa.no_oasoftware/allsoftware.no_allsoftware s from ${stats_db_name}.allsoftware - join ${stats_db_name}.software_oa on allsoftware.ri_initiative=software_oa.ri_initiative; + join ${stats_db_name}.software_oa on allsoftware.ri_initiative=software_oa.ri_initiative; /*EOS*/ -drop table ${stats_db_name}.indi_ris_openess purge; +drop table ${stats_db_name}.indi_ris_openess purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_ris_openess stored as parquet as select allpubsshare.ri_initiative, @@ -1120,21 +1120,21 @@ select allpubsshare.ri_initiative, on tmp1.ri_initiative=allpubsshare.ri_initiative left outer join (select ri_initiative,s from ${stats_db_name}.allsoftwaresshare) tmp2 - on tmp2.ri_initiative=allpubsshare.ri_initiative; + on tmp2.ri_initiative=allpubsshare.ri_initiative; /*EOS*/ -DROP TABLE ${stats_db_name}.result_contexts purge; -DROP TABLE ${stats_db_name}.pubs_oa purge; -DROP TABLE ${stats_db_name}.datasets_oa purge; -DROP TABLE ${stats_db_name}.software_oa purge; -DROP TABLE ${stats_db_name}.allpubs purge; -DROP TABLE ${stats_db_name}.alldatasets purge; -DROP TABLE ${stats_db_name}.allsoftware purge; -DROP TABLE ${stats_db_name}.allpubsshare purge; -DROP TABLE ${stats_db_name}.alldatasetssshare purge; -DROP TABLE ${stats_db_name}.allsoftwaresshare purge; +DROP TABLE ${stats_db_name}.result_contexts purge; /*EOS*/ +DROP TABLE ${stats_db_name}.pubs_oa purge; /*EOS*/ +DROP TABLE ${stats_db_name}.datasets_oa purge; /*EOS*/ +DROP TABLE ${stats_db_name}.software_oa purge; /*EOS*/ +DROP TABLE ${stats_db_name}.allpubs purge; /*EOS*/ +DROP TABLE ${stats_db_name}.alldatasets purge; /*EOS*/ +DROP TABLE ${stats_db_name}.allsoftware purge; /*EOS*/ +DROP TABLE ${stats_db_name}.allpubsshare purge; /*EOS*/ +DROP TABLE ${stats_db_name}.alldatasetssshare purge; /*EOS*/ +DROP TABLE ${stats_db_name}.allsoftwaresshare purge; /*EOS*/ --Funder Findability -drop table ${stats_db_name}.indi_funder_findable purge; +drop table ${stats_db_name}.indi_funder_findable purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_funder_findable stored as parquet as with result_findable as @@ -1151,10 +1151,10 @@ with result_findable as group by p.funder) select allresults.funder, result_findable.no_result_findable/allresults.no_allresults funder_findable from allresults - join result_findable on result_findable.funder=allresults.funder; + join result_findable on result_findable.funder=allresults.funder; /*EOS*/ --RIs Findability -drop table ${stats_db_name}.indi_ris_findable purge; +drop table ${stats_db_name}.indi_ris_findable purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_ris_findable stored as parquet as with result_contexts as @@ -1175,7 +1175,7 @@ allresults as group by rc.ri_initiative) select allresults.ri_initiative, result_findable.no_result_findable/allresults.no_allresults ris_findable from allresults - join result_findable on result_findable.ri_initiative=allresults.ri_initiative; + join result_findable on result_findable.ri_initiative=allresults.ri_initiative; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_publicly_funded stored as parquet as with org_names_pids as @@ -1197,7 +1197,7 @@ from ${stats_db_name}.publication p left outer join ( select distinct ro.id, 1 as publicly_funded from ${stats_db_name}.result_organization ro join ${stats_db_name}.organization o on o.id=ro.organization -join publicly_funded_orgs pfo on o.name=pfo.name) tmp on p.id=tmp.id; +join publicly_funded_orgs pfo on o.name=pfo.name) tmp on p.id=tmp.id; /*EOS*/ create table ${stats_db_name}.indi_pub_green_with_license stored as parquet as select distinct p.id, coalesce(green_with_license, 0) as green_with_license @@ -1207,7 +1207,7 @@ select distinct p.id, 1 as green_with_license from ${stats_db_name}.publication join ${stats_db_name}.result_instance ri on ri.id = p.id join ${stats_db_name}.datasource on datasource.id = ri.hostedby where ri.license is not null and datasource.type like '%Repository%' and datasource.name!='Other') tmp -on p.id= tmp.id; +on p.id= tmp.id; /*EOS*/ create table ${stats_db_name}.result_country stored as parquet as select distinct ro.id, coalesce(o.country, f.country) as country @@ -1216,13 +1216,13 @@ 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; +where coalesce(o.country, f.country) IS NOT NULL; /*EOS*/ create table ${stats_db_name}.indi_result_oa_with_license stored as parquet as select distinct r.id, coalesce(oa_with_license,0) as oa_with_license from ${stats_db_name}.result r left outer join (select distinct r.id, 1 as oa_with_license from ${stats_db_name}.result r -join ${stats_db_name}.result_licenses rl on rl.id=r.id where r.bestlicence='Open Access') tmp on r.id=tmp.id; +join ${stats_db_name}.result_licenses rl on rl.id=r.id where r.bestlicence='Open Access') tmp on r.id=tmp.id; /*EOS*/ create table ${stats_db_name}.indi_result_oa_without_license stored as parquet as with without_license as @@ -1233,7 +1233,7 @@ from ${stats_db_name}.result r left outer join (select distinct r.id, 1 as oa_without_license from ${stats_db_name}.result r join without_license wl on wl.id=r.id -where r.bestlicence='Open Access') tmp on r.id=tmp.id; +where r.bestlicence='Open Access') tmp on r.id=tmp.id; /*EOS*/ create table ${stats_db_name}.indi_result_under_transformative stored as parquet as with transformative_dois as @@ -1242,4 +1242,4 @@ select distinct r.id, coalesce(under_transformative,0) as under_transformative from ${stats_db_name}.result r left outer join (select distinct rp.id, 1 as under_transformative from ${stats_db_name}.result_pids rp join ${stats_db_name}.result r on r.id=rp.id -join transformative_dois td on td.doi=rp.pid) tmp on r.id=tmp.id; \ No newline at end of file +join transformative_dois td on td.doi=rp.pid) tmp on r.id=tmp.id; /*EOS*/ \ No newline at end of file diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/workflow.xml b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/workflow.xml index f15f22320..6618381c5 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/workflow.xml +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/workflow.xml @@ -64,6 +64,26 @@ hadoop_user_name user name of the wf owner + + + sparkSqlWarehouseDir + + + + sparkClusterOpts + --conf spark.network.timeout=600 --conf spark.extraListeners= --conf spark.sql.queryExecutionListeners= --conf spark.yarn.historyServer.address=http://iis-cdh5-test-m3.ocean.icm.edu.pl:18088 --conf spark.eventLog.dir=hdfs://nameservice1/user/spark/applicationHistory + spark cluster-wide options + + + sparkResourceOpts + --executor-memory=3G --conf spark.executor.memoryOverhead=3G --executor-cores=6 --driver-memory=8G --driver-cores=4 + spark resource options + + + sparkApplicationOpts + --conf spark.sql.shuffle.partitions=3840 + spark resource options + @@ -82,10 +102,14 @@ hive.mapjoin.followby.gby.localtask.max.memory.usage 0.80 - - mapred.job.queue.name - analytics - + + oozie.action.sharelib.for.spark + ${oozieActionShareLibForSpark2} + + + mapred.job.queue.name + analytics + @@ -321,13 +345,35 @@ + + + + + + + + + + + - - ${hive_jdbc_url} - - stats_db_name=${stats_db_name} - external_stats_db_name=${external_stats_db_name} - + + yarn + cluster + Step16-createIndicatorsTables + eu.dnetlib.dhp.oozie.RunSQLSparkJob + dhp-dedup-openaire-${projectVersion}.jar + + --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir} + ${sparkClusterOpts} + ${sparkResourceOpts} + ${sparkApplicationOpts} + + --hiveMetastoreUris${hive_metastore_uris} + --sqleu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql + --stats_db_name${stats_db_name} + --external_stats_db_name${external_stats_db_name} + -- 2.17.1 From 078df0b4d1f9afcf254235bc944ed3e26d6235a6 Mon Sep 17 00:00:00 2001 From: Giambattista Bloisi Date: Fri, 26 Jan 2024 20:19:52 +0100 Subject: [PATCH 11/22] Use SparkSQL in place of Hive for executing step16-createIndicatorsTables.sql of stats update wf --- dhp-workflows/dhp-stats-update/pom.xml | 5 + .../scripts/step16-createIndicatorsTables.sql | 627 +++++++++--------- .../dhp/oa/graph/stats/oozie_app/workflow.xml | 199 +++--- 3 files changed, 436 insertions(+), 395 deletions(-) diff --git a/dhp-workflows/dhp-stats-update/pom.xml b/dhp-workflows/dhp-stats-update/pom.xml index f491b5868..cc15b8a15 100644 --- a/dhp-workflows/dhp-stats-update/pom.xml +++ b/dhp-workflows/dhp-stats-update/pom.xml @@ -8,6 +8,11 @@ 4.0.0 dhp-stats-update + + eu.dnetlib.dhp + dhp-common + ${project.version} + org.apache.spark spark-core_${scala.binary.version} diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql index 5aa14e2c2..f13b2500c 100755 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql @@ -1,5 +1,5 @@ -- Sprint 1 ---- -drop table if exists ${stats_db_name}.indi_pub_green_oa purge; +drop table if exists ${stats_db_name}.indi_pub_green_oa purge; /*EOS*/ --create table if not exists ${stats_db_name}.indi_pub_green_oa stored as parquet as --select distinct p.id, coalesce(green_oa, 0) as green_oa @@ -24,9 +24,9 @@ from ${stats_db_name}.publication p where datasource.type like '%Repository%' and (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and datasource.name!='Other') tmp - on p.id= tmp.id; + on p.id= tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_grey_lit purge; +drop table if exists ${stats_db_name}.indi_pub_grey_lit purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_grey_lit stored as parquet as select distinct p.id, coalesce(grey_lit, 0) as grey_lit @@ -37,9 +37,9 @@ from ${stats_db_name}.publication p join ${stats_db_name}.result_classifications rt on rt.id = p.id where rt.type not in ('Article','Part of book or chapter of book','Book','Doctoral thesis','Master thesis','Data Paper', 'Thesis', 'Bachelor thesis', 'Conference object') and not exists (select 1 from ${stats_db_name}.result_classifications rc where type ='Other literature type' - and rc.id=p.id)) tmp on p.id=tmp.id; + and rc.id=p.id)) tmp on p.id=tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_doi_from_crossref purge; +drop table if exists ${stats_db_name}.indi_pub_doi_from_crossref purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_doi_from_crossref stored as parquet as select distinct p.id, coalesce(doi_from_crossref, 0) as doi_from_crossref @@ -48,10 +48,10 @@ from ${stats_db_name}.publication p (select ri.id, 1 as doi_from_crossref from ${stats_db_name}.result_instance ri join ${stats_db_name}.datasource d on d.id = ri.collectedfrom where pidtype='Digital Object Identifier' and d.name ='Crossref') tmp - on tmp.id=p.id; + on tmp.id=p.id; /*EOS*/ -- Sprint 2 ---- -drop table if exists ${stats_db_name}.indi_result_has_cc_licence purge; +drop table if exists ${stats_db_name}.indi_result_has_cc_licence purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_result_has_cc_licence stored as parquet as select distinct r.id, (case when lic='' or lic is null then 0 else 1 end) as has_cc_license @@ -59,9 +59,9 @@ from ${stats_db_name}.result r left outer join (select r.id, license.type as lic from ${stats_db_name}.result r join ${stats_db_name}.result_licenses as license on license.id = r.id where lower(license.type) LIKE '%creativecommons.org%' OR lower(license.type) LIKE '%cc-%') tmp - on r.id= tmp.id; + on r.id= tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_result_has_cc_licence_url purge; +drop table if exists ${stats_db_name}.indi_result_has_cc_licence_url purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_result_has_cc_licence_url stored as parquet as select distinct r.id, case when lic_host='' or lic_host is null then 0 else 1 end as has_cc_license_url @@ -70,32 +70,32 @@ from ${stats_db_name}.result r from ${stats_db_name}.result r join ${stats_db_name}.result_licenses as license on license.id = r.id WHERE lower(parse_url(license.type, "HOST")) = "creativecommons.org") tmp - on r.id= tmp.id; + on r.id= tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_has_abstract purge; +drop table if exists ${stats_db_name}.indi_pub_has_abstract purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_has_abstract stored as parquet as select distinct publication.id, cast(coalesce(abstract, true) as int) has_abstract -from ${stats_db_name}.publication; +from ${stats_db_name}.publication; /*EOS*/ -drop table if exists ${stats_db_name}.indi_result_with_orcid purge; +drop table if exists ${stats_db_name}.indi_result_with_orcid purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_result_with_orcid stored as parquet as select distinct r.id, coalesce(has_orcid, 0) as has_orcid from ${stats_db_name}.result r left outer join (select id, 1 as has_orcid from ${stats_db_name}.result_orcid) tmp - on r.id= tmp.id; + on r.id= tmp.id; /*EOS*/ ---- Sprint 3 ---- -drop table if exists ${stats_db_name}.indi_funded_result_with_fundref purge; +drop table if exists ${stats_db_name}.indi_funded_result_with_fundref purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_funded_result_with_fundref stored as parquet as select distinct r.result as id, coalesce(fundref, 0) as fundref from ${stats_db_name}.project_results r left outer join (select distinct result, 1 as fundref from ${stats_db_name}.project_results where provenance='Harvested') tmp - on r.result= tmp.result; + on r.result= tmp.result; /*EOS*/ -- create table indi_result_org_collab stored as parquet as -- select o1.organization org1, o2.organization org2, count(distinct o1.id) as collaborations @@ -105,65 +105,65 @@ from ${stats_db_name}.project_results r -- -- compute stats indi_result_org_collab; -- -create TEMPORARY TABLE ${stats_db_name}.tmp AS SELECT ro.organization organization, ro.id, o.name from ${stats_db_name}.result_organization ro -join ${stats_db_name}.organization o on o.id=ro.organization where o.name is not null; +create TEMPORARY VIEW tmp AS SELECT ro.organization organization, ro.id, o.name from ${stats_db_name}.result_organization ro +join ${stats_db_name}.organization o on o.id=ro.organization where o.name is not null; /*EOS*/ -drop table if exists ${stats_db_name}.indi_result_org_collab purge; +drop table if exists ${stats_db_name}.indi_result_org_collab purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_result_org_collab stored as parquet as select o1.organization org1, o1.name org1name1, o2.organization org2, o2.name org2name2, count(o1.id) as collaborations -from ${stats_db_name}.tmp as o1 -join ${stats_db_name}.tmp as o2 where o1.id=o2.id and o1.organization!=o2.organization and o1.name!=o2.name -group by o1.organization, o2.organization, o1.name, o2.name; +from tmp as o1 +join tmp as o2 where o1.id=o2.id and o1.organization!=o2.organization and o1.name!=o2.name +group by o1.organization, o2.organization, o1.name, o2.name; /*EOS*/ -drop table if exists ${stats_db_name}.tmp purge; +DROP VIEW if exists tmp; /*EOS*/ -create TEMPORARY TABLE ${stats_db_name}.tmp AS +create TEMPORARY VIEW tmp AS select distinct ro.organization organization, ro.id, o.name, o.country from ${stats_db_name}.result_organization ro -join ${stats_db_name}.organization o on o.id=ro.organization where country <> 'UNKNOWN' and o.name is not null; +join ${stats_db_name}.organization o on o.id=ro.organization where country <> 'UNKNOWN' and o.name is not null; /*EOS*/ -drop table if exists ${stats_db_name}.indi_result_org_country_collab purge; +drop table if exists ${stats_db_name}.indi_result_org_country_collab purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_result_org_country_collab stored as parquet as select o1.organization org1,o1.name org1name1, o2.country country2, count(o1.id) as collaborations -from ${stats_db_name}.tmp as o1 join ${stats_db_name}.tmp as o2 on o1.id=o2.id +from tmp as o1 join tmp as o2 on o1.id=o2.id where o1.id=o2.id and o1.country!=o2.country -group by o1.organization, o1.id, o1.name, o2.country; +group by o1.organization, o1.id, o1.name, o2.country; /*EOS*/ -drop table if exists ${stats_db_name}.tmp purge; +drop table if exists tmp purge; /*EOS*/ -create TEMPORARY TABLE ${stats_db_name}.tmp AS +create TEMPORARY VIEW tmp AS select o.id organization, o.name, ro.project as project from ${stats_db_name}.organization o - join ${stats_db_name}.organization_projects ro on o.id=ro.id where o.name is not null; + join ${stats_db_name}.organization_projects ro on o.id=ro.id where o.name is not null; /*EOS*/ -drop table if exists ${stats_db_name}.indi_project_collab_org purge; +drop table if exists ${stats_db_name}.indi_project_collab_org purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_project_collab_org stored as parquet as select o1.organization org1,o1.name orgname1, o2.organization org2, o2.name orgname2, count(distinct o1.project) as collaborations -from ${stats_db_name}.tmp as o1 - join ${stats_db_name}.tmp as o2 on o1.project=o2.project +from tmp as o1 + join tmp as o2 on o1.project=o2.project where o1.organization<>o2.organization and o1.name<>o2.name -group by o1.name,o2.name, o1.organization, o2.organization; +group by o1.name,o2.name, o1.organization, o2.organization; /*EOS*/ -drop table if exists ${stats_db_name}.tmp purge; +DROP VIEW if exists tmp; /*EOS*/ -create TEMPORARY TABLE ${stats_db_name}.tmp AS +create TEMPORARY VIEW tmp AS select o.id organization, o.name, o.country , ro.project as project from ${stats_db_name}.organization o join ${stats_db_name}.organization_projects ro on o.id=ro.id - and o.country <> 'UNKNOWN' and o.name is not null; + and o.country <> 'UNKNOWN' and o.name is not null; /*EOS*/ -drop table if exists ${stats_db_name}.indi_project_collab_org_country purge; +drop table if exists ${stats_db_name}.indi_project_collab_org_country purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_project_collab_org_country stored as parquet as select o1.organization org1,o1.name org1name, o2.country country2, count(distinct o1.project) as collaborations -from ${stats_db_name}.tmp as o1 - join ${stats_db_name}.tmp as o2 on o1.project=o2.project +from tmp as o1 + join tmp as o2 on o1.project=o2.project where o1.organization<>o2.organization and o1.country<>o2.country -group by o1.organization, o2.country, o1.name; +group by o1.organization, o2.country, o1.name; /*EOS*/ -drop table if exists ${stats_db_name}.tmp purge; +DROP VIEW if exists tmp; /*EOS*/ -drop table if exists ${stats_db_name}.indi_funder_country_collab purge; +drop table if exists ${stats_db_name}.indi_funder_country_collab purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_funder_country_collab stored as parquet as with tmp as (select funder, project, country from ${stats_db_name}.organization_projects op @@ -174,26 +174,26 @@ select f1.funder, f1.country as country1, f2.country as country2, count(distinct from tmp as f1 join tmp as f2 on f1.project=f2.project where f1.country<>f2.country -group by f1.funder, f2.country, f1.country; +group by f1.funder, f2.country, f1.country; /*EOS*/ -create TEMPORARY TABLE ${stats_db_name}.tmp AS +create TEMPORARY VIEW tmp AS select distinct country, ro.id as result from ${stats_db_name}.organization o join ${stats_db_name}.result_organization ro on o.id=ro.organization - where country <> 'UNKNOWN' and o.name is not null; + where country <> 'UNKNOWN' and o.name is not null; /*EOS*/ -drop table if exists ${stats_db_name}.indi_result_country_collab purge; +drop table if exists ${stats_db_name}.indi_result_country_collab purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_result_country_collab stored as parquet as select o1.country country1, o2.country country2, count(o1.result) as collaborations -from ${stats_db_name}.tmp as o1 - join ${stats_db_name}.tmp as o2 on o1.result=o2.result +from tmp as o1 + join tmp as o2 on o1.result=o2.result where o1.country<>o2.country -group by o1.country, o2.country; +group by o1.country, o2.country; /*EOS*/ -drop table if exists ${stats_db_name}.tmp purge; +DROP VIEW if exists tmp; /*EOS*/ ---- Sprint 4 ---- -drop table if exists ${stats_db_name}.indi_pub_diamond purge; +drop table if exists ${stats_db_name}.indi_pub_diamond purge; /*EOS*/ --create table if not exists ${stats_db_name}.indi_pub_diamond stored as parquet as --select distinct pd.id, coalesce(in_diamond_journal, 0) as in_diamond_journal @@ -212,9 +212,9 @@ left outer join (select pd.id, 1 as in_diamond_journal from ${stats_db_name}.pub join ${stats_db_name}.datasource d on d.id=pd.datasource join STATS_EXT.plan_s_jn ps where (ps.issn_print=d.issn_printed and ps.issn_online=d.issn_online) and (ps.journal_is_in_doaj=true or ps.journal_is_oa=true) and ps.has_apc=false) tmp -on pd.id=tmp.id; +on pd.id=tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_in_transformative purge; +drop table if exists ${stats_db_name}.indi_pub_in_transformative purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_in_transformative stored as parquet as select distinct pd.id, coalesce(is_transformative, 0) as is_transformative @@ -224,9 +224,9 @@ from ${stats_db_name}.publication pd join ${stats_db_name}.datasource d on d.id=pd.datasource join STATS_EXT.plan_s_jn ps where (ps.issn_print=d.issn_printed and ps.issn_online=d.issn_online) and ps.is_transformative_journal=true) tmp - on pd.id=tmp.id; + on pd.id=tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_closed_other_open purge; +drop table if exists ${stats_db_name}.indi_pub_closed_other_open purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_closed_other_open stored as parquet as select distinct ri.id, coalesce(pub_closed_other_open, 0) as pub_closed_other_open from ${stats_db_name}.result_instance ri @@ -236,53 +236,53 @@ select distinct ri.id, coalesce(pub_closed_other_open, 0) as pub_closed_other_op join ${stats_db_name}.datasource d on ri.hostedby=d.id where d.type like '%Journal%' and ri.accessright='Closed Access' and (p.bestlicence='Open Access' or p.bestlicence='Open Source')) tmp - on tmp.id=ri.id; + on tmp.id=ri.id; /*EOS*/ ---- Sprint 5 ---- -drop table if exists ${stats_db_name}.indi_result_no_of_copies purge; +drop table if exists ${stats_db_name}.indi_result_no_of_copies purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_result_no_of_copies stored as parquet as -select id, count(id) as number_of_copies from ${stats_db_name}.result_instance group by id; +select id, count(id) as number_of_copies from ${stats_db_name}.result_instance group by id; /*EOS*/ ---- Sprint 6 ---- -drop table if exists ${stats_db_name}.indi_pub_downloads purge; +drop table if exists ${stats_db_name}.indi_pub_downloads purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_downloads stored as parquet as SELECT result_id, sum(downloads) no_downloads from openaire_prod_usage_stats.usage_stats join ${stats_db_name}.publication on result_id=id where downloads>0 GROUP BY result_id -order by no_downloads desc; +order by no_downloads desc; /*EOS*/ --ANALYZE TABLE ${stats_db_name}.indi_pub_downloads COMPUTE STATISTICS; -drop table if exists ${stats_db_name}.indi_pub_downloads_datasource purge; +drop table if exists ${stats_db_name}.indi_pub_downloads_datasource purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_downloads_datasource stored as parquet as SELECT result_id, repository_id, sum(downloads) no_downloads from openaire_prod_usage_stats.usage_stats join ${stats_db_name}.publication on result_id=id where downloads>0 GROUP BY result_id, repository_id -order by result_id; +order by result_id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_downloads_year purge; +drop table if exists ${stats_db_name}.indi_pub_downloads_year purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_downloads_year stored as parquet as SELECT result_id, cast(substring(us.`date`, 1,4) as int) as `year`, sum(downloads) no_downloads from openaire_prod_usage_stats.usage_stats us join ${stats_db_name}.publication on result_id=id where downloads>0 -GROUP BY result_id, substring(us.`date`, 1,4); +GROUP BY result_id, substring(us.`date`, 1,4); /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_downloads_datasource_year purge; +drop table if exists ${stats_db_name}.indi_pub_downloads_datasource_year purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_downloads_datasource_year stored as parquet as SELECT result_id, cast(substring(us.`date`, 1,4) as int) as `year`, repository_id, sum(downloads) no_downloads from openaire_prod_usage_stats.usage_stats us join ${stats_db_name}.publication on result_id=id where downloads>0 -GROUP BY result_id, repository_id, substring(us.`date`, 1,4); +GROUP BY result_id, repository_id, substring(us.`date`, 1,4); /*EOS*/ ---- Sprint 7 ---- -drop table if exists ${stats_db_name}.indi_pub_gold_oa purge; +drop table if exists ${stats_db_name}.indi_pub_gold_oa purge; /*EOS*/ --create table if not exists ${stats_db_name}.indi_pub_gold_oa stored as parquet as -- WITH gold_oa AS ( SELECT @@ -381,9 +381,9 @@ left outer join ( select pd.id, 1 as is_gold FROM ${stats_db_name}.publication_datasources pd 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; + 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; +drop table if exists ${stats_db_name}.indi_pub_hybrid_oa_with_cc purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_hybrid_oa_with_cc stored as parquet as WITH hybrid_oa AS ( @@ -414,9 +414,9 @@ FROM ${stats_db_name}.publication_datasources pd JOIN hybrid_oa ON issn.issn = hybrid_oa.issn JOIN ${stats_db_name}.indi_result_has_cc_licence cc on pd.id=cc.id JOIN ${stats_db_name}.indi_pub_gold_oa ga on pd.id=ga.id - where cc.has_cc_license=1 and ga.is_gold=0) tmp on pd.id=tmp.id; + where cc.has_cc_license=1 and ga.is_gold=0) tmp on pd.id=tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_hybrid purge; +drop table if exists ${stats_db_name}.indi_pub_hybrid purge; /*EOS*/ --create table if not exists ${stats_db_name}.indi_pub_hybrid stored as parquet as -- WITH gold_oa AS ( SELECT @@ -489,9 +489,9 @@ join ${stats_db_name}.result_accessroute ra on ra.id=pd.id join ${stats_db_name}.datasource d on d.id=ri.hostedby where indi_gold.is_gold=0 and ((d.type like '%Journal%' and ri.accessright!='Closed Access' and ri.accessright!='Restricted' and ri.license is not null) or ra.accessroute='hybrid'))tmp -on pd.id=tmp.id; +on pd.id=tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_fairness purge; +drop table if exists ${stats_db_name}.indi_org_fairness purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_fairness stored as parquet as --return results with PIDs, and rich metadata group by organization @@ -509,9 +509,9 @@ create table if not exists ${stats_db_name}.indi_org_fairness stored as parquet --return results_fair/all_results select allresults.organization, result_fair.no_result_fair/allresults.no_allresults org_fairness from allresults - join result_fair on result_fair.organization=allresults.organization; + join result_fair on result_fair.organization=allresults.organization; /*EOS*/ -CREATE TEMPORARY table ${stats_db_name}.result_fair as +CREATE TEMPORARY VIEW result_fair as select ro.organization organization, count(distinct ro.id) no_result_fair from ${stats_db_name}.result_organization ro join ${stats_db_name}.publication p on p.id=ro.id @@ -519,296 +519,296 @@ select ro.organization organization, count(distinct ro.id) no_result_fair join ${stats_db_name}.indi_pub_grey_lit gl on gl.id=p.id where (title is not null) and (publisher is not null) and (abstract=true) and (year is not null) and (authors>0) and cast(year as int)>2003 and dc.doi_from_crossref=1 and gl.grey_lit=0 - group by ro.organization; + group by ro.organization; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allresults as +CREATE TEMPORARY VIEW allresults as select ro.organization, count(distinct ro.id) no_allresults from ${stats_db_name}.result_organization ro join ${stats_db_name}.publication p on p.id=ro.id where cast(year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_fairness_pub_pr purge; +drop table if exists ${stats_db_name}.indi_org_fairness_pub_pr purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_fairness_pub_pr stored as parquet as select ar.organization, rf.no_result_fair/ar.no_allresults org_fairness -from ${stats_db_name}.allresults ar - join ${stats_db_name}.result_fair rf on rf.organization=ar.organization; +from allresults ar + join result_fair rf on rf.organization=ar.organization; /*EOS*/ -DROP table ${stats_db_name}.result_fair purge; -DROP table ${stats_db_name}.allresults purge; +DROP VIEW result_fair; /*EOS*/ +DROP VIEW allresults; /*EOS*/ -CREATE TEMPORARY table ${stats_db_name}.result_fair as +CREATE TEMPORARY VIEW result_fair as select year, ro.organization organization, count(distinct ro.id) no_result_fair from ${stats_db_name}.result_organization ro join ${stats_db_name}.result p on p.id=ro.id where (title is not null) and (publisher is not null) and (abstract=true) and (year is not null) and (authors>0) and cast(year as int)>2003 - group by ro.organization, year; + group by ro.organization, year; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allresults as select year, ro.organization, count(distinct ro.id) no_allresults from ${stats_db_name}.result_organization ro +CREATE TEMPORARY VIEW allresults as select year, ro.organization, count(distinct ro.id) no_allresults from ${stats_db_name}.result_organization ro join ${stats_db_name}.result p on p.id=ro.id where cast(year as int)>2003 - group by ro.organization, year; + group by ro.organization, year; /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_fairness_pub_year purge; +drop table if exists ${stats_db_name}.indi_org_fairness_pub_year purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_fairness_pub_year stored as parquet as select cast(allresults.year as int) year, allresults.organization, result_fair.no_result_fair/allresults.no_allresults org_fairness -from ${stats_db_name}.allresults - join ${stats_db_name}.result_fair on result_fair.organization=allresults.organization and result_fair.year=allresults.year; +from allresults + join result_fair on result_fair.organization=allresults.organization and result_fair.year=allresults.year; /*EOS*/ -DROP table ${stats_db_name}.result_fair purge; -DROP table ${stats_db_name}.allresults purge; +DROP VIEW result_fair; /*EOS*/ +DROP VIEW allresults; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.result_fair as +CREATE TEMPORARY VIEW result_fair as select ro.organization organization, count(distinct ro.id) no_result_fair from ${stats_db_name}.result_organization ro join ${stats_db_name}.result p on p.id=ro.id where (title is not null) and (publisher is not null) and (abstract=true) and (year is not null) and (authors>0) and cast(year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allresults as +CREATE TEMPORARY VIEW allresults as select ro.organization, count(distinct ro.id) no_allresults from ${stats_db_name}.result_organization ro join ${stats_db_name}.result p on p.id=ro.id where cast(year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_fairness_pub purge; +drop table if exists ${stats_db_name}.indi_org_fairness_pub purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_fairness_pub as select ar.organization, rf.no_result_fair/ar.no_allresults org_fairness -from ${stats_db_name}.allresults ar join ${stats_db_name}.result_fair rf -on rf.organization=ar.organization; +from allresults ar join result_fair rf +on rf.organization=ar.organization; /*EOS*/ -DROP table ${stats_db_name}.result_fair purge; -DROP table ${stats_db_name}.allresults purge; +DROP VIEW result_fair; /*EOS*/ +DROP VIEW allresults; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.result_fair as +CREATE TEMPORARY VIEW result_fair as select year, ro.organization organization, count(distinct ro.id) no_result_fair from ${stats_db_name}.result_organization ro join ${stats_db_name}.result r on r.id=ro.id join ${stats_db_name}.result_pids rp on r.id=rp.id where (title is not null) and (publisher is not null) and (abstract=true) and (year is not null) and (authors>0) and cast(year as int)>2003 - group by ro.organization, year; + group by ro.organization, year; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allresults as +CREATE TEMPORARY VIEW allresults as select year, ro.organization, count(distinct ro.id) no_allresults from ${stats_db_name}.result_organization ro join ${stats_db_name}.result r on r.id=ro.id where cast(year as int)>2003 - group by ro.organization, year; + group by ro.organization, year; /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_fairness_year purge; +drop table if exists ${stats_db_name}.indi_org_fairness_year purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_fairness_year stored as parquet as select cast(allresults.year as int) year, allresults.organization, result_fair.no_result_fair/allresults.no_allresults org_fairness - from ${stats_db_name}.allresults - join ${stats_db_name}.result_fair on result_fair.organization=allresults.organization and cast(result_fair.year as int)=cast(allresults.year as int); + from allresults + join result_fair on result_fair.organization=allresults.organization and cast(result_fair.year as int)=cast(allresults.year as int); /*EOS*/ -DROP table ${stats_db_name}.result_fair purge; -DROP table ${stats_db_name}.allresults purge; +DROP VIEW result_fair; /*EOS*/ +DROP VIEW allresults; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.result_with_pid as +CREATE TEMPORARY VIEW result_with_pid as select year, ro.organization, count(distinct rp.id) no_result_with_pid from ${stats_db_name}.result_organization ro join ${stats_db_name}.result_pids rp on rp.id=ro.id join ${stats_db_name}.result r on r.id=rp.id where cast(year as int) >2003 - group by ro.organization, year; + group by ro.organization, year; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allresults as +CREATE TEMPORARY VIEW allresults as select year, ro.organization, count(distinct ro.id) no_allresults from ${stats_db_name}.result_organization ro join ${stats_db_name}.result r on r.id=ro.id where cast(year as int) >2003 - group by ro.organization, year; + group by ro.organization, year; /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_findable_year purge; +drop table if exists ${stats_db_name}.indi_org_findable_year purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_findable_year stored as parquet as select cast(allresults.year as int) year, allresults.organization, result_with_pid.no_result_with_pid/allresults.no_allresults org_findable -from ${stats_db_name}.allresults - join ${stats_db_name}.result_with_pid on result_with_pid.organization=allresults.organization and cast(result_with_pid.year as int)=cast(allresults.year as int); +from allresults + join result_with_pid on result_with_pid.organization=allresults.organization and cast(result_with_pid.year as int)=cast(allresults.year as int); /*EOS*/ -DROP table ${stats_db_name}.result_with_pid purge; -DROP table ${stats_db_name}.allresults purge; +DROP VIEW result_with_pid; /*EOS*/ +DROP VIEW allresults; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.result_with_pid as +CREATE TEMPORARY VIEW result_with_pid as select ro.organization, count(distinct rp.id) no_result_with_pid from ${stats_db_name}.result_organization ro join ${stats_db_name}.result_pids rp on rp.id=ro.id join ${stats_db_name}.result r on r.id=rp.id where cast(year as int) >2003 - group by ro.organization; + group by ro.organization; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allresults as +CREATE TEMPORARY VIEW allresults as select ro.organization, count(distinct ro.id) no_allresults from ${stats_db_name}.result_organization ro join ${stats_db_name}.result r on r.id=ro.id where cast(year as int) >2003 - group by ro.organization; + group by ro.organization; /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_findable purge; +drop table if exists ${stats_db_name}.indi_org_findable purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_findable stored as parquet as select allresults.organization, result_with_pid.no_result_with_pid/allresults.no_allresults org_findable -from ${stats_db_name}.allresults - join ${stats_db_name}.result_with_pid on result_with_pid.organization=allresults.organization; +from allresults + join result_with_pid on result_with_pid.organization=allresults.organization; /*EOS*/ -DROP table ${stats_db_name}.result_with_pid purge; -DROP table ${stats_db_name}.allresults purge; +DROP VIEW result_with_pid; /*EOS*/ +DROP VIEW allresults; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.pubs_oa as +CREATE TEMPORARY VIEW pubs_oa as SELECT ro.organization, count(distinct r.id) no_oapubs FROM ${stats_db_name}.publication r join ${stats_db_name}.result_organization ro on ro.id=r.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.datasets_oa as +CREATE TEMPORARY VIEW datasets_oa as SELECT ro.organization, count(distinct r.id) no_oadatasets FROM ${stats_db_name}.dataset r join ${stats_db_name}.result_organization ro on ro.id=r.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.software_oa as +CREATE TEMPORARY VIEW software_oa as SELECT ro.organization, count(distinct r.id) no_oasoftware FROM ${stats_db_name}.software r join ${stats_db_name}.result_organization ro on ro.id=r.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allpubs as +CREATE TEMPORARY VIEW allpubs as SELECT ro.organization, count(ro.id) no_allpubs FROM ${stats_db_name}.result_organization ro join ${stats_db_name}.publication ps on ps.id=ro.id where cast(ps.year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.alldatasets as +CREATE TEMPORARY VIEW alldatasets as SELECT ro.organization, count(ro.id) no_alldatasets FROM ${stats_db_name}.result_organization ro join ${stats_db_name}.dataset ps on ps.id=ro.id where cast(ps.year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allsoftware as +CREATE TEMPORARY VIEW allsoftware as SELECT ro.organization, count(ro.id) no_allsoftware FROM ${stats_db_name}.result_organization ro join ${stats_db_name}.software ps on ps.id=ro.id where cast(ps.year as int)>2003 - group by ro.organization; + group by ro.organization; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allpubsshare as -select pubs_oa.organization, pubs_oa.no_oapubs/allpubs.no_allpubs p from ${stats_db_name}.allpubs - join ${stats_db_name}.pubs_oa on allpubs.organization=pubs_oa.organization; +CREATE TEMPORARY VIEW allpubsshare as +select pubs_oa.organization, pubs_oa.no_oapubs/allpubs.no_allpubs p from allpubs + join pubs_oa on allpubs.organization=pubs_oa.organization; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.alldatasetssshare as +CREATE TEMPORARY VIEW alldatasetssshare as select datasets_oa.organization, datasets_oa.no_oadatasets/alldatasets.no_alldatasets d - from ${stats_db_name}.alldatasets - join ${stats_db_name}.datasets_oa on alldatasets.organization=datasets_oa.organization; + from alldatasets + join datasets_oa on alldatasets.organization=datasets_oa.organization; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allsoftwaresshare as +CREATE TEMPORARY VIEW allsoftwaresshare as select software_oa.organization, software_oa.no_oasoftware/allsoftware.no_allsoftware s - from ${stats_db_name}.allsoftware - join ${stats_db_name}.software_oa on allsoftware.organization=software_oa.organization; + from allsoftware + join software_oa on allsoftware.organization=software_oa.organization; /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_openess purge; +drop table if exists ${stats_db_name}.indi_org_openess purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_openess stored as parquet as select allpubsshare.organization, (p+if(isnull(s),0,s)+if(isnull(d),0,d))/(1+(case when s is null then 0 else 1 end) +(case when d is null then 0 else 1 end)) - org_openess FROM ${stats_db_name}.allpubsshare + org_openess FROM allpubsshare left outer join (select organization,d from - ${stats_db_name}.alldatasetssshare) tmp1 + alldatasetssshare) tmp1 on tmp1.organization=allpubsshare.organization left outer join (select organization,s from - ${stats_db_name}.allsoftwaresshare) tmp2 - on tmp2.organization=allpubsshare.organization; + allsoftwaresshare) tmp2 + on tmp2.organization=allpubsshare.organization; /*EOS*/ -DROP TABLE ${stats_db_name}.pubs_oa purge; -DROP TABLE ${stats_db_name}.datasets_oa purge; -DROP TABLE ${stats_db_name}.software_oa purge; -DROP TABLE ${stats_db_name}.allpubs purge; -DROP TABLE ${stats_db_name}.alldatasets purge; -DROP TABLE ${stats_db_name}.allsoftware purge; -DROP TABLE ${stats_db_name}.allpubsshare purge; -DROP TABLE ${stats_db_name}.alldatasetssshare purge; -DROP TABLE ${stats_db_name}.allsoftwaresshare purge; +DROP VIEW pubs_oa; /*EOS*/ +DROP VIEW datasets_oa; /*EOS*/ +DROP VIEW software_oa; /*EOS*/ +DROP VIEW allpubs; /*EOS*/ +DROP VIEW alldatasets; /*EOS*/ +DROP VIEW allsoftware; /*EOS*/ +DROP VIEW allpubsshare; /*EOS*/ +DROP VIEW alldatasetssshare; /*EOS*/ +DROP VIEW allsoftwaresshare; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.pubs_oa AS +CREATE TEMPORARY VIEW pubs_oa AS SELECT r.year, ro.organization, count(distinct r.id) no_oapubs FROM ${stats_db_name}.publication r join ${stats_db_name}.result_organization ro on ro.id=r.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 - group by ro.organization,r.year; + group by ro.organization,r.year; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.datasets_oa AS +CREATE TEMPORARY VIEW datasets_oa AS SELECT r.year,ro.organization, count(distinct r.id) no_oadatasets FROM ${stats_db_name}.dataset r join ${stats_db_name}.result_organization ro on ro.id=r.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 - group by ro.organization, r.year; + group by ro.organization, r.year; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.software_oa AS +CREATE TEMPORARY VIEW software_oa AS SELECT r.year,ro.organization, count(distinct r.id) no_oasoftware FROM ${stats_db_name}.software r join ${stats_db_name}.result_organization ro on ro.id=r.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 - group by ro.organization, r.year; + group by ro.organization, r.year; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allpubs as +CREATE TEMPORARY VIEW allpubs as SELECT p.year,ro.organization organization, count(ro.id) no_allpubs FROM ${stats_db_name}.result_organization ro join ${stats_db_name}.publication p on p.id=ro.id where cast(p.year as int)>2003 - group by ro.organization, p.year; + group by ro.organization, p.year; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.alldatasets as +CREATE TEMPORARY VIEW alldatasets as SELECT d.year, ro.organization organization, count(ro.id) no_alldatasets FROM ${stats_db_name}.result_organization ro join ${stats_db_name}.dataset d on d.id=ro.id where cast(d.year as int)>2003 - group by ro.organization, d.year; + group by ro.organization, d.year; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allsoftware as +CREATE TEMPORARY VIEW allsoftware as SELECT s.year,ro.organization organization, count(ro.id) no_allsoftware FROM ${stats_db_name}.result_organization ro join ${stats_db_name}.software s on s.id=ro.id where cast(s.year as int)>2003 - group by ro.organization, s.year; + group by ro.organization, s.year; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allpubsshare as -select allpubs.year, pubs_oa.organization, pubs_oa.no_oapubs/allpubs.no_allpubs p from ${stats_db_name}.allpubs - join ${stats_db_name}.pubs_oa on allpubs.organization=pubs_oa.organization where cast(allpubs.year as INT)=cast(pubs_oa.year as int); +CREATE TEMPORARY VIEW allpubsshare as +select allpubs.year, pubs_oa.organization, pubs_oa.no_oapubs/allpubs.no_allpubs p from allpubs + join pubs_oa on allpubs.organization=pubs_oa.organization where cast(allpubs.year as INT)=cast(pubs_oa.year as int); /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.alldatasetssshare as +CREATE TEMPORARY VIEW alldatasetssshare as select alldatasets.year, datasets_oa.organization, datasets_oa.no_oadatasets/alldatasets.no_alldatasets d - from ${stats_db_name}.alldatasets - join ${stats_db_name}.datasets_oa on alldatasets.organization=datasets_oa.organization where cast(alldatasets.year as INT)=cast(datasets_oa.year as int); + from alldatasets + join datasets_oa on alldatasets.organization=datasets_oa.organization where cast(alldatasets.year as INT)=cast(datasets_oa.year as int); /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allsoftwaresshare as +CREATE TEMPORARY VIEW allsoftwaresshare as select allsoftware.year, software_oa.organization, software_oa.no_oasoftware/allsoftware.no_allsoftware s - from ${stats_db_name}.allsoftware - join ${stats_db_name}.software_oa on allsoftware.organization=software_oa.organization where cast(allsoftware.year as INT)=cast(software_oa.year as int); + from allsoftware + join software_oa on allsoftware.organization=software_oa.organization where cast(allsoftware.year as INT)=cast(software_oa.year as int); /*EOS*/ -drop table if exists ${stats_db_name}.indi_org_openess_year purge; +drop table if exists ${stats_db_name}.indi_org_openess_year purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_org_openess_year stored as parquet as select cast(allpubsshare.year as int) year, allpubsshare.organization, (p+if(isnull(s),0,s)+if(isnull(d),0,d))/(1+(case when s is null then 0 else 1 end) +(case when d is null then 0 else 1 end)) - org_openess FROM ${stats_db_name}.allpubsshare + org_openess FROM allpubsshare left outer join (select cast(year as int), organization,d from - ${stats_db_name}.alldatasetssshare) tmp1 + alldatasetssshare) tmp1 on tmp1.organization=allpubsshare.organization and tmp1.year=allpubsshare.year left outer join (select cast(year as int), organization,s from - ${stats_db_name}.allsoftwaresshare) tmp2 - on tmp2.organization=allpubsshare.organization and cast(tmp2.year as int)=cast(allpubsshare.year as int); + allsoftwaresshare) tmp2 + on tmp2.organization=allpubsshare.organization and cast(tmp2.year as int)=cast(allpubsshare.year as int); /*EOS*/ -DROP TABLE ${stats_db_name}.pubs_oa purge; -DROP TABLE ${stats_db_name}.datasets_oa purge; -DROP TABLE ${stats_db_name}.software_oa purge; -DROP TABLE ${stats_db_name}.allpubs purge; -DROP TABLE ${stats_db_name}.alldatasets purge; -DROP TABLE ${stats_db_name}.allsoftware purge; -DROP TABLE ${stats_db_name}.allpubsshare purge; -DROP TABLE ${stats_db_name}.alldatasetssshare purge; -DROP TABLE ${stats_db_name}.allsoftwaresshare purge; +DROP VIEW pubs_oa; /*EOS*/ +DROP VIEW datasets_oa; /*EOS*/ +DROP VIEW software_oa; /*EOS*/ +DROP VIEW allpubs; /*EOS*/ +DROP VIEW alldatasets; /*EOS*/ +DROP VIEW allsoftware; /*EOS*/ +DROP VIEW allpubsshare; /*EOS*/ +DROP VIEW alldatasetssshare; /*EOS*/ +DROP VIEW allsoftwaresshare; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_has_preprint purge; +drop table if exists ${stats_db_name}.indi_pub_has_preprint purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_has_preprint stored as parquet as select distinct p.id, coalesce(has_preprint, 0) as has_preprint @@ -817,8 +817,8 @@ from ${stats_db_name}.publication_classifications p select p.id, 1 as has_preprint from ${stats_db_name}.publication_classifications p where p.type='Preprint') tmp - on p.id= tmp.id; -drop table if exists ${stats_db_name}.indi_pub_in_subscribed purge; + on p.id= tmp.id; /*EOS*/ +drop table if exists ${stats_db_name}.indi_pub_in_subscribed purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_in_subscribed stored as parquet as select distinct p.id, coalesce(is_subscription, 0) as is_subscription @@ -829,9 +829,9 @@ from ${stats_db_name}.publication p join ${stats_db_name}.indi_pub_hybrid h on p.id=h.id join ${stats_db_name}.indi_pub_in_transformative t on p.id=t.id where g.is_gold=0 and h.is_hybrid=0 and t.is_transformative=0) tmp - on p.id=tmp.id; + on p.id=tmp.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_result_with_pid purge; +drop table if exists ${stats_db_name}.indi_result_with_pid purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_result_with_pid as select distinct p.id, coalesce(result_with_pid, 0) as result_with_pid @@ -839,25 +839,25 @@ from ${stats_db_name}.result p left outer join ( select p.id, 1 as result_with_pid from ${stats_db_name}.result_pids p) tmp - on p.id= tmp.id; + on p.id= tmp.id; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.pub_fos_totals as +CREATE TEMPORARY VIEW pub_fos_totals as select rf.id, count(distinct lvl3) totals from ${stats_db_name}.result_fos rf -group by rf.id; +group by rf.id; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_interdisciplinarity purge; +drop table if exists ${stats_db_name}.indi_pub_interdisciplinarity purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_interdisciplinarity as select distinct p.id as id, coalesce(is_interdisciplinary, 0) as is_interdisciplinary -from ${stats_db_name}.pub_fos_totals p +from pub_fos_totals p left outer join ( -select pub_fos_totals.id, 1 as is_interdisciplinary from ${stats_db_name}.pub_fos_totals -where totals>1) tmp on p.id=tmp.id; +select pub_fos_totals.id, 1 as is_interdisciplinary from pub_fos_totals +where totals>1) tmp on p.id=tmp.id; /*EOS*/ -drop table ${stats_db_name}.pub_fos_totals purge; +drop view pub_fos_totals; /*EOS*/ -drop table if exists ${stats_db_name}.indi_pub_bronze_oa purge; +drop table if exists ${stats_db_name}.indi_pub_bronze_oa purge; /*EOS*/ --create table if not exists ${stats_db_name}.indi_pub_bronze_oa stored as parquet as --select distinct p.id, coalesce(is_bronze_oa,0) as is_bronze_oa @@ -883,38 +883,38 @@ join ${stats_db_name}.datasource d on d.id=ri.hostedby where indi_gold.is_gold=0 and indi_hybrid.is_hybrid=0 and ((d.type like '%Journal%' and ri.accessright!='Closed Access' and ri.accessright!='Restricted' and ri.license is null) or ra.accessroute='bronze')) tmp -on pd.id=tmp.id; +on pd.id=tmp.id; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.project_year_result_year as +CREATE TEMPORARY VIEW project_year_result_year as select p.id project_id, acronym, r.id result_id, r.year, p.end_year from ${stats_db_name}.project p join ${stats_db_name}.result_projects rp on p.id=rp.project join ${stats_db_name}.result r on r.id=rp.id -where p.end_year is NOT NULL and r.year is not null; +where p.end_year is NOT NULL and r.year is not null; /*EOS*/ -drop table if exists ${stats_db_name}.indi_is_project_result_after purge; +drop table if exists ${stats_db_name}.indi_is_project_result_after purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_is_project_result_after stored as parquet as select pry.project_id, pry.acronym, pry.result_id, coalesce(is_project_result_after, 0) as is_project_result_after -from ${stats_db_name}.project_year_result_year pry +from project_year_result_year pry left outer join (select pry.project_id, pry.acronym, pry.result_id, 1 as is_project_result_after -from ${stats_db_name}.project_year_result_year pry -where pry.year>pry.end_year) tmp on pry.result_id=tmp.result_id; +from project_year_result_year pry +where pry.year>pry.end_year) tmp on pry.result_id=tmp.result_id; /*EOS*/ -drop table ${stats_db_name}.project_year_result_year purge; +drop view project_year_result_year; /*EOS*/ -drop table ${stats_db_name}.indi_is_funder_plan_s purge; +drop table if exists ${stats_db_name}.indi_is_funder_plan_s purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_is_funder_plan_s stored as parquet as select distinct f.id, f.name, coalesce(is_funder_plan_s, 0) as is_funder_plan_s from ${stats_db_name}.funder f left outer join (select id, name, 1 as is_funder_plan_s from ${stats_db_name}.funder join stats_ext.plan_s_short on c_o_alition_s_organisation_funder=name) tmp - on f.name= tmp.name; + on f.name= tmp.name; /*EOS*/ --Funder Fairness -drop table ${stats_db_name}.indi_funder_fairness purge; +drop table if exists ${stats_db_name}.indi_funder_fairness purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_funder_fairness stored as parquet as with result_fair as @@ -930,10 +930,10 @@ create table if not exists ${stats_db_name}.indi_funder_fairness stored as parqu group by p.funder) select allresults.funder, result_fair.no_result_fair/allresults.no_allresults funder_fairness from allresults - join result_fair on result_fair.funder=allresults.funder; + join result_fair on result_fair.funder=allresults.funder; /*EOS*/ --RIs Fairness -drop table ${stats_db_name}.indi_ris_fairness purge; +drop table if exists ${stats_db_name}.indi_ris_fairness purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_ris_fairness stored as parquet as with result_contexts as @@ -953,188 +953,188 @@ allresults as group by rc.ri_initiative) select allresults.ri_initiative, result_fair.no_result_fair/allresults.no_allresults ris_fairness from allresults - join result_fair on result_fair.ri_initiative=allresults.ri_initiative; + join result_fair on result_fair.ri_initiative=allresults.ri_initiative; /*EOS*/ --Funder Openess -CREATE TEMPORARY TABLE ${stats_db_name}.pubs_oa as +CREATE TEMPORARY VIEW pubs_oa as select p.funder funder, count(distinct rp.id) no_oapubs from ${stats_db_name}.result_projects rp join ${stats_db_name}.project p on p.id=rp.project join ${stats_db_name}.publication r on r.id=rp.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 -group by p.funder; +group by p.funder; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.datasets_oa as +CREATE TEMPORARY VIEW datasets_oa as select p.funder funder, count(distinct rp.id) no_oadatasets from ${stats_db_name}.result_projects rp join ${stats_db_name}.project p on p.id=rp.project join ${stats_db_name}.dataset r on r.id=rp.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 -group by p.funder; +group by p.funder; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.software_oa as +CREATE TEMPORARY VIEW software_oa as select p.funder funder, count(distinct rp.id) no_oasoftware from ${stats_db_name}.result_projects rp join ${stats_db_name}.project p on p.id=rp.project join ${stats_db_name}.software r on r.id=rp.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 -group by p.funder; +group by p.funder; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allpubs as +CREATE TEMPORARY VIEW allpubs as select p.funder funder, count(distinct rp.id) no_allpubs from ${stats_db_name}.result_projects rp join ${stats_db_name}.project p on p.id=rp.project join ${stats_db_name}.publication r on r.id=rp.id where cast(r.year as int)>2003 -group by p.funder; +group by p.funder; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.alldatasets as +CREATE TEMPORARY VIEW alldatasets as select p.funder funder, count(distinct rp.id) no_alldatasets from ${stats_db_name}.result_projects rp join ${stats_db_name}.project p on p.id=rp.project join ${stats_db_name}.dataset r on r.id=rp.id where cast(r.year as int)>2003 -group by p.funder; +group by p.funder; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allsoftware as +CREATE TEMPORARY VIEW allsoftware as select p.funder funder, count(distinct rp.id) no_allsoftware from ${stats_db_name}.result_projects rp join ${stats_db_name}.project p on p.id=rp.project join ${stats_db_name}.software r on r.id=rp.id where cast(r.year as int)>2003 -group by p.funder; +group by p.funder; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allpubsshare as -select pubs_oa.funder, pubs_oa.no_oapubs/allpubs.no_allpubs p from ${stats_db_name}.allpubs - join ${stats_db_name}.pubs_oa on allpubs.funder=pubs_oa.funder; +CREATE TEMPORARY VIEW allpubsshare as +select pubs_oa.funder, pubs_oa.no_oapubs/allpubs.no_allpubs p from allpubs + join pubs_oa on allpubs.funder=pubs_oa.funder; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.alldatasetssshare as +CREATE TEMPORARY VIEW alldatasetssshare as select datasets_oa.funder, datasets_oa.no_oadatasets/alldatasets.no_alldatasets d - from ${stats_db_name}.alldatasets - join ${stats_db_name}.datasets_oa on alldatasets.funder=datasets_oa.funder; + from alldatasets + join datasets_oa on alldatasets.funder=datasets_oa.funder; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allsoftwaresshare as +CREATE TEMPORARY VIEW allsoftwaresshare as select software_oa.funder, software_oa.no_oasoftware/allsoftware.no_allsoftware s - from ${stats_db_name}.allsoftware - join ${stats_db_name}.software_oa on allsoftware.funder=software_oa.funder; + from allsoftware + join software_oa on allsoftware.funder=software_oa.funder; /*EOS*/ -drop table ${stats_db_name}.indi_funder_openess purge; +drop table if exists ${stats_db_name}.indi_funder_openess purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_funder_openess stored as parquet as select allpubsshare.funder, (p+if(isnull(s),0,s)+if(isnull(d),0,d))/(1+(case when s is null then 0 else 1 end) +(case when d is null then 0 else 1 end)) - funder_openess FROM ${stats_db_name}.allpubsshare + funder_openess FROM allpubsshare left outer join (select funder,d from - ${stats_db_name}.alldatasetssshare) tmp1 + alldatasetssshare) tmp1 on tmp1.funder=allpubsshare.funder left outer join (select funder,s from - ${stats_db_name}.allsoftwaresshare) tmp2 - on tmp2.funder=allpubsshare.funder; + allsoftwaresshare) tmp2 + on tmp2.funder=allpubsshare.funder; /*EOS*/ -DROP TABLE ${stats_db_name}.pubs_oa purge; -DROP TABLE ${stats_db_name}.datasets_oa purge; -DROP TABLE ${stats_db_name}.software_oa purge; -DROP TABLE ${stats_db_name}.allpubs purge; -DROP TABLE ${stats_db_name}.alldatasets purge; -DROP TABLE ${stats_db_name}.allsoftware purge; -DROP TABLE ${stats_db_name}.allpubsshare purge; -DROP TABLE ${stats_db_name}.alldatasetssshare purge; -DROP TABLE ${stats_db_name}.allsoftwaresshare purge; +DROP VIEW pubs_oa; /*EOS*/ +DROP VIEW datasets_oa; /*EOS*/ +DROP VIEW software_oa; /*EOS*/ +DROP VIEW allpubs; /*EOS*/ +DROP VIEW alldatasets; /*EOS*/ +DROP VIEW allsoftware; /*EOS*/ +DROP VIEW allpubsshare; /*EOS*/ +DROP VIEW alldatasetssshare; /*EOS*/ +DROP VIEW allsoftwaresshare; /*EOS*/ --RIs Openess -CREATE TEMPORARY TABLE ${stats_db_name}.result_contexts as +CREATE TEMPORARY VIEW result_contexts as select distinct rc.id, context.name ri_initiative from ${stats_db_name}.result_concepts rc join ${stats_db_name}.concept on concept.id=rc.concept join ${stats_db_name}.category on category.id=concept.category -join ${stats_db_name}.context on context.id=category.context; +join ${stats_db_name}.context on context.id=category.context; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.pubs_oa as -select rp.ri_initiative ri_initiative, count(distinct rp.id) no_oapubs from ${stats_db_name}.result_contexts rp +CREATE TEMPORARY VIEW pubs_oa as +select rp.ri_initiative ri_initiative, count(distinct rp.id) no_oapubs from result_contexts rp join ${stats_db_name}.publication r on r.id=rp.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 -group by rp.ri_initiative; +group by rp.ri_initiative; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.datasets_oa as -select rp.ri_initiative ri_initiative, count(distinct rp.id) no_oadatasets from ${stats_db_name}.result_contexts rp +CREATE TEMPORARY VIEW datasets_oa as +select rp.ri_initiative ri_initiative, count(distinct rp.id) no_oadatasets from result_contexts rp join ${stats_db_name}.dataset r on r.id=rp.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 -group by rp.ri_initiative; +group by rp.ri_initiative; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.software_oa as -select rp.ri_initiative ri_initiative, count(distinct rp.id) no_oasoftware from ${stats_db_name}.result_contexts rp +CREATE TEMPORARY VIEW software_oa as +select rp.ri_initiative ri_initiative, count(distinct rp.id) no_oasoftware from result_contexts rp join ${stats_db_name}.software r on r.id=rp.id join ${stats_db_name}.result_instance ri on ri.id=r.id where (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and cast(r.year as int)>2003 -group by rp.ri_initiative; +group by rp.ri_initiative; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allpubs as -select rp.ri_initiative ri_initiative, count(distinct rp.id) no_allpubs from ${stats_db_name}.result_contexts rp +CREATE TEMPORARY VIEW allpubs as +select rp.ri_initiative ri_initiative, count(distinct rp.id) no_allpubs from result_contexts rp join ${stats_db_name}.publication r on r.id=rp.id where cast(r.year as int)>2003 -group by rp.ri_initiative; +group by rp.ri_initiative; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.alldatasets as -select rp.ri_initiative ri_initiative, count(distinct rp.id) no_alldatasets from ${stats_db_name}.result_contexts rp +CREATE TEMPORARY VIEW alldatasets as +select rp.ri_initiative ri_initiative, count(distinct rp.id) no_alldatasets from result_contexts rp join ${stats_db_name}.dataset r on r.id=rp.id where cast(r.year as int)>2003 -group by rp.ri_initiative; +group by rp.ri_initiative; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allsoftware as -select rp.ri_initiative ri_initiative, count(distinct rp.id) no_allsoftware from ${stats_db_name}.result_contexts rp +CREATE TEMPORARY VIEW allsoftware as +select rp.ri_initiative ri_initiative, count(distinct rp.id) no_allsoftware from result_contexts rp join ${stats_db_name}.software r on r.id=rp.id where cast(r.year as int)>2003 -group by rp.ri_initiative; +group by rp.ri_initiative; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allpubsshare as -select pubs_oa.ri_initiative, pubs_oa.no_oapubs/allpubs.no_allpubs p from ${stats_db_name}.allpubs - join ${stats_db_name}.pubs_oa on allpubs.ri_initiative=pubs_oa.ri_initiative; +CREATE TEMPORARY VIEW allpubsshare as +select pubs_oa.ri_initiative, pubs_oa.no_oapubs/allpubs.no_allpubs p from allpubs + join pubs_oa on allpubs.ri_initiative=pubs_oa.ri_initiative; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.alldatasetssshare as +CREATE TEMPORARY VIEW alldatasetssshare as select datasets_oa.ri_initiative, datasets_oa.no_oadatasets/alldatasets.no_alldatasets d - from ${stats_db_name}.alldatasets - join ${stats_db_name}.datasets_oa on alldatasets.ri_initiative=datasets_oa.ri_initiative; + from alldatasets + join datasets_oa on alldatasets.ri_initiative=datasets_oa.ri_initiative; /*EOS*/ -CREATE TEMPORARY TABLE ${stats_db_name}.allsoftwaresshare as +CREATE TEMPORARY VIEW allsoftwaresshare as select software_oa.ri_initiative, software_oa.no_oasoftware/allsoftware.no_allsoftware s - from ${stats_db_name}.allsoftware - join ${stats_db_name}.software_oa on allsoftware.ri_initiative=software_oa.ri_initiative; + from allsoftware + join software_oa on allsoftware.ri_initiative=software_oa.ri_initiative; /*EOS*/ -drop table ${stats_db_name}.indi_ris_openess purge; +drop table if exists ${stats_db_name}.indi_ris_openess purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_ris_openess stored as parquet as select allpubsshare.ri_initiative, (p+if(isnull(s),0,s)+if(isnull(d),0,d))/(1+(case when s is null then 0 else 1 end) +(case when d is null then 0 else 1 end)) - ris_openess FROM ${stats_db_name}.allpubsshare + ris_openess FROM allpubsshare left outer join (select ri_initiative,d from - ${stats_db_name}.alldatasetssshare) tmp1 + alldatasetssshare) tmp1 on tmp1.ri_initiative=allpubsshare.ri_initiative left outer join (select ri_initiative,s from - ${stats_db_name}.allsoftwaresshare) tmp2 - on tmp2.ri_initiative=allpubsshare.ri_initiative; + allsoftwaresshare) tmp2 + on tmp2.ri_initiative=allpubsshare.ri_initiative; /*EOS*/ -DROP TABLE ${stats_db_name}.result_contexts purge; -DROP TABLE ${stats_db_name}.pubs_oa purge; -DROP TABLE ${stats_db_name}.datasets_oa purge; -DROP TABLE ${stats_db_name}.software_oa purge; -DROP TABLE ${stats_db_name}.allpubs purge; -DROP TABLE ${stats_db_name}.alldatasets purge; -DROP TABLE ${stats_db_name}.allsoftware purge; -DROP TABLE ${stats_db_name}.allpubsshare purge; -DROP TABLE ${stats_db_name}.alldatasetssshare purge; -DROP TABLE ${stats_db_name}.allsoftwaresshare purge; +DROP VIEW result_contexts; /*EOS*/ +DROP VIEW pubs_oa; /*EOS*/ +DROP VIEW datasets_oa; /*EOS*/ +DROP VIEW software_oa; /*EOS*/ +DROP VIEW allpubs; /*EOS*/ +DROP VIEW alldatasets; /*EOS*/ +DROP VIEW allsoftware; /*EOS*/ +DROP VIEW allpubsshare; /*EOS*/ +DROP VIEW alldatasetssshare; /*EOS*/ +DROP VIEW allsoftwaresshare; /*EOS*/ --Funder Findability -drop table ${stats_db_name}.indi_funder_findable purge; +drop table if exists ${stats_db_name}.indi_funder_findable purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_funder_findable stored as parquet as with result_findable as @@ -1151,10 +1151,10 @@ with result_findable as group by p.funder) select allresults.funder, result_findable.no_result_findable/allresults.no_allresults funder_findable from allresults - join result_findable on result_findable.funder=allresults.funder; + join result_findable on result_findable.funder=allresults.funder; /*EOS*/ --RIs Findability -drop table ${stats_db_name}.indi_ris_findable purge; +drop table if exists ${stats_db_name}.indi_ris_findable purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_ris_findable stored as parquet as with result_contexts as @@ -1175,7 +1175,7 @@ allresults as group by rc.ri_initiative) select allresults.ri_initiative, result_findable.no_result_findable/allresults.no_allresults ris_findable from allresults - join result_findable on result_findable.ri_initiative=allresults.ri_initiative; + join result_findable on result_findable.ri_initiative=allresults.ri_initiative; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_publicly_funded stored as parquet as with org_names_pids as @@ -1195,6 +1195,7 @@ and pf.publicly_funded='yes') foo) select distinct p.id, coalesce(publicly_funded, 0) as publicly_funded from ${stats_db_name}.publication p left outer join ( -select distinct ro.id, 1 as publicly_funded from result_organization ro +select distinct ro.id, 1 as publicly_funded from ${stats_db_name}.result_organization ro join ${stats_db_name}.organization o on o.id=ro.organization -join publicly_funded_orgs pfo on o.name=pfo.name) tmp on p.id=tmp.id; \ No newline at end of file +join publicly_funded_orgs pfo on o.name=pfo.name) tmp on p.id=tmp.id; /*EOS*/ + diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/workflow.xml b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/workflow.xml index f15f22320..709de6595 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/workflow.xml +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/workflow.xml @@ -64,6 +64,26 @@ hadoop_user_name user name of the wf owner + + + sparkSqlWarehouseDir + + + + sparkClusterOpts + --conf spark.network.timeout=600 --conf spark.extraListeners= --conf spark.sql.queryExecutionListeners= --conf spark.yarn.historyServer.address=http://iis-cdh5-test-m3.ocean.icm.edu.pl:18088 --conf spark.eventLog.dir=hdfs://nameservice1/user/spark/applicationHistory + spark cluster-wide options + + + sparkResourceOpts + --executor-memory=6G --conf spark.executor.memoryOverhead=4G --executor-cores=6 --driver-memory=8G --driver-cores=4 + spark resource options + + + sparkApplicationOpts + --conf spark.sql.shuffle.partitions=3840 + spark resource options + @@ -75,17 +95,21 @@ ${hive_metastore_uris} - hive.txn.timeout - ${hive_timeout} + hive.txn.timeout + ${hive_timeout} hive.mapjoin.followby.gby.localtask.max.memory.usage 0.80 - - mapred.job.queue.name - analytics - + + oozie.action.sharelib.for.spark + ${oozieActionShareLibForSpark2} + + + mapred.job.queue.name + analytics + @@ -133,164 +157,164 @@ ${hive_jdbc_url} - stats_db_name=${stats_db_name} - openaire_db_name=${openaire_db_name} + stats_db_name=${stats_db_name} + openaire_db_name=${openaire_db_name} - + ${hive_jdbc_url} - stats_db_name=${stats_db_name} - openaire_db_name=${openaire_db_name} + stats_db_name=${stats_db_name} + openaire_db_name=${openaire_db_name} - + ${hive_jdbc_url} - stats_db_name=${stats_db_name} - openaire_db_name=${openaire_db_name} + stats_db_name=${stats_db_name} + openaire_db_name=${openaire_db_name} - + - + ${hive_jdbc_url} - stats_db_name=${stats_db_name} - openaire_db_name=${openaire_db_name} + stats_db_name=${stats_db_name} + openaire_db_name=${openaire_db_name} - + - + ${hive_jdbc_url} - stats_db_name=${stats_db_name} - openaire_db_name=${openaire_db_name} + stats_db_name=${stats_db_name} + openaire_db_name=${openaire_db_name} - + - + ${hive_jdbc_url} - stats_db_name=${stats_db_name} - openaire_db_name=${openaire_db_name} + stats_db_name=${stats_db_name} + openaire_db_name=${openaire_db_name} - + - + ${hive_jdbc_url} - stats_db_name=${stats_db_name} - openaire_db_name=${openaire_db_name} + stats_db_name=${stats_db_name} + openaire_db_name=${openaire_db_name} - + - + ${hive_jdbc_url} - stats_db_name=${stats_db_name} - openaire_db_name=${openaire_db_name} + stats_db_name=${stats_db_name} + openaire_db_name=${openaire_db_name} - + - + ${hive_jdbc_url} - stats_db_name=${stats_db_name} - openaire_db_name=${openaire_db_name} + stats_db_name=${stats_db_name} + openaire_db_name=${openaire_db_name} - + - + ${hive_jdbc_url} - stats_db_name=${stats_db_name} - openaire_db_name=${openaire_db_name} - external_stats_db_name=${external_stats_db_name} + stats_db_name=${stats_db_name} + openaire_db_name=${openaire_db_name} + external_stats_db_name=${external_stats_db_name} - + - + ${hive_jdbc_url} - stats_db_name=${stats_db_name} - openaire_db_name=${openaire_db_name} - external_stats_db_name=${external_stats_db_name} + stats_db_name=${stats_db_name} + openaire_db_name=${openaire_db_name} + external_stats_db_name=${external_stats_db_name} - - + + ${hive_jdbc_url} - stats_db_name=${stats_db_name} - openaire_db_name=${openaire_db_name} + stats_db_name=${stats_db_name} + openaire_db_name=${openaire_db_name} - + ${hive_jdbc_url} - stats_db_name=${stats_db_name} - openaire_db_name=${openaire_db_name} + stats_db_name=${stats_db_name} + openaire_db_name=${openaire_db_name} - + ${hive_jdbc_url} - stats_db_name=${stats_db_name} - openaire_db_name=${openaire_db_name} + stats_db_name=${stats_db_name} + openaire_db_name=${openaire_db_name} - + ${hive_jdbc_url} - stats_db_name=${stats_db_name} - openaire_db_name=${openaire_db_name} + stats_db_name=${stats_db_name} + openaire_db_name=${openaire_db_name} @@ -322,12 +346,23 @@ - - ${hive_jdbc_url} - - stats_db_name=${stats_db_name} - external_stats_db_name=${external_stats_db_name} - + + yarn + cluster + Step16-createIndicatorsTables + eu.dnetlib.dhp.oozie.RunSQLSparkJob + dhp-stats-update-${projectVersion}.jar + + --conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir} + ${sparkClusterOpts} + ${sparkResourceOpts} + ${sparkApplicationOpts} + + --hiveMetastoreUris${hive_metastore_uris} + --sqleu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql + --stats_db_name${stats_db_name} + --external_stats_db_name${external_stats_db_name} + @@ -387,18 +422,18 @@ - - - - - - - - - - - - + + + + + + + + + + + + @@ -443,8 +478,8 @@ ${jobTracker} ${nameNode} copyDataToImpalaCluster.sh - - + + ${stats_db_name} ${monitor_db_name} ${observatory_db_name} @@ -505,4 +540,4 @@ - + \ No newline at end of file -- 2.17.1 From a512ead447a02fd708a1dbb772b63ab3d92bca9c Mon Sep 17 00:00:00 2001 From: antleb Date: Tue, 30 Jan 2024 16:54:47 +0200 Subject: [PATCH 12/22] changed orcid ids to all capital --- .../eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step13.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step13.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step13.sql index 6493fa7d0..8c1dbdc4d 100755 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step13.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step13.sql @@ -69,7 +69,7 @@ SELECT * FROM ${stats_db_name}.otherresearchproduct_sources; DROP TABLE IF EXISTS ${stats_db_name}.result_orcid purge; 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 +select distinct res.id, upper(regexp_replace(res.orcid, 'http://orcid.org/' ,'')) as orcid from ( SELECT substr(res.id, 4) as id, auth_pid.value as orcid FROM ${openaire_db_name}.result res -- 2.17.1 From dd4c27f4f3bf191ad7f1067d75841bc8395d6068 Mon Sep 17 00:00:00 2001 From: antleb Date: Thu, 8 Feb 2024 12:57:57 +0200 Subject: [PATCH 13/22] added 2 new institutions in monitor --- .../oozie_app/scripts/step20-createMonitorDB_institutions.sql | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB_institutions.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB_institutions.sql index 5c347eb6b..c99407ad8 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB_institutions.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB_institutions.sql @@ -61,5 +61,7 @@ create table TARGET.result stored as parquet as 'openorgs____::8839b55dae0c84d56fd533f52d5d483a', -- Leibniz Institute of Ecological Urban and Regional Development 'openorgs____::526468206bca24c1c90da6a312295cf4', -- Cyprus University of Technology 'openorgs____::b5ca9d4340e26454e367e2908ef3872f', -- Alma Mater Studiorum University of Bologna - 'openorgs____::a6340e6ecf60f6bba163659df985b0f2' -- TU Dresden + 'openorgs____::a6340e6ecf60f6bba163659df985b0f2', -- TU Dresden + 'openorgs____::64badd35233ba2cd4946368ef2f4cf57', -- University of Vienna + 'openorgs____::7501d66d2297a963ebfb075c43fff88e' -- Royal Institute of Technology ))) foo; \ No newline at end of file -- 2.17.1 From 316d585c8a4f6cd7dbd2b7b69a6e5e7b8bd3ccef Mon Sep 17 00:00:00 2001 From: Antonis Lempesis Date: Thu, 7 Mar 2024 02:07:59 +0200 Subject: [PATCH 14/22] using distinct apcs per publication to avoid huge sums --- .../dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15_5.sql | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15_5.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15_5.sql index 2c606fb92..c0f39aefb 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15_5.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15_5.sql @@ -56,7 +56,7 @@ join ${stats_db_name}.result res on res.id=r.id; DROP TABLE IF EXISTS ${stats_db_name}.result_apc purge; create table if not exists ${stats_db_name}.result_apc STORED AS PARQUET as -select r.id, r.amount, r.currency +select distinct r.id, r.amount, r.currency from ( select substr(r.id, 4) as id, cast(inst.processingchargeamount.value as float) as amount, inst.processingchargecurrency.value as currency from ${openaire_db_name}.result r lateral view explode(r.instance) instances as inst) r -- 2.17.1 From 3c7972034207566b7664d05a078fa1ac3dd20671 Mon Sep 17 00:00:00 2001 From: antleb Date: Thu, 7 Mar 2024 14:08:57 +0200 Subject: [PATCH 15/22] fixed the irish result subset --- .../oozie_app/scripts/buildIrishMonitorDB.sql | 23 ++++++++++++++----- 1 file changed, 17 insertions(+), 6 deletions(-) diff --git a/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/scripts/buildIrishMonitorDB.sql b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/scripts/buildIrishMonitorDB.sql index eb632ad19..3f0922020 100644 --- a/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/scripts/buildIrishMonitorDB.sql +++ b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/scripts/buildIrishMonitorDB.sql @@ -27,12 +27,23 @@ select distinct xpath_string(fundingtree[0].value, '//funder/name') as funder fr drop table if exists TARGET.result; create table TARGET.result stored as parquet as - select distinct * from SOURCE.result r where r.id in (select ro.id from SOURCE.result_organization ro - left outer join SOURCE.organization o on o.id=ro.organization - left outer join SOURCE.result_projects rp on rp.id=ro.id - left outer join SOURCE.project p on p.id=rp.project - join TARGET.irish_funders irf - where o.country='IE' or p.funder=irf.funder); +select distinct * from ( + select r.* + from SOURCE.result r + join SOURCE.result_projects rp on rp.id=r.id + join SOURCE.project p on p.id=rp.project + join openaire_prod_stats_monitor_ie_20231226b.irish_funders irf on irf.funder=p.funder + union all + select r.* + from SOURCE.result r + join SOURCE.result_organization ro on ro.id=r.id + join SOURCE.organization o on o.id=ro.organization and o.country='IE' + union all + select r.* + from SOURCE.result r + join SOURCE.result_pids pid on pid.id=r.id + join stats_ext.transformative_facts tf on tf.doi=pid.pid + ) foo; create view if not exists TARGET.category as select * from SOURCE.category; create view if not exists TARGET.concept as select * from SOURCE.concept; -- 2.17.1 From f74c7e8689f2a05b5ec9e85324a705438a3f683f Mon Sep 17 00:00:00 2001 From: antleb Date: Tue, 12 Mar 2024 02:13:04 +0200 Subject: [PATCH 16/22] selecting distinct peer_reviewed --- .../graph/stats/oozie_app/scripts/step15.sql | 150 +++++++----------- 1 file changed, 61 insertions(+), 89 deletions(-) diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15.sql index 343438cf4..ce6b6cc2f 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15.sql @@ -7,104 +7,76 @@ ------------------------------------------------------ DROP TABLE IF EXISTS ${stats_db_name}.publication_refereed purge; - ---CREATE TABLE IF NOT EXISTS ${stats_db_name}.publication_refereed STORED AS PARQUET as ---select substr(r.id, 4) as id, inst.refereed.classname as refereed ---from ${openaire_db_name}.publication r lateral view explode(r.instance) instances as inst ---where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE; CREATE TABLE IF NOT EXISTS ${stats_db_name}.publication_refereed STORED AS PARQUET as -with peer_reviewed as -(select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed -from ${openaire_db_name}.publication r lateral -view explode(r.instance) instances as inst -where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE -and inst.refereed.classname='peerReviewed'), -non_peer_reviewed as -(select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed -from ${openaire_db_name}.publication r lateral -view explode(r.instance) instances as inst -where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE -and inst.refereed.classname='nonPeerReviewed') -select peer_reviewed.* from peer_reviewed -union all -select non_peer_reviewed.* from non_peer_reviewed -left join peer_reviewed on peer_reviewed.id=non_peer_reviewed.id -where peer_reviewed.id is null; +with peer_reviewed as ( + select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed + from ${openaire_db_name}.publication r lateral view explode(r.instance) instances as inst + where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE and inst.refereed.classname='peerReviewed'), +non_peer_reviewed as ( + select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed + from ${openaire_db_name}.publication r lateral view explode(r.instance) instances as inst + where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE and inst.refereed.classname='nonPeerReviewed') +select distinct * +from ( + select peer_reviewed.* from peer_reviewed + union all + select non_peer_reviewed.* from non_peer_reviewed + left join peer_reviewed on peer_reviewed.id=non_peer_reviewed.id + where peer_reviewed.id is null) pr; DROP TABLE IF EXISTS ${stats_db_name}.dataset_refereed purge; - ---CREATE TABLE IF NOT EXISTS ${stats_db_name}.dataset_refereed STORED AS PARQUET as ---select substr(r.id, 4) as id, inst.refereed.classname as refereed ---from ${openaire_db_name}.dataset r lateral view explode(r.instance) instances as inst ---where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE; CREATE TABLE IF NOT EXISTS ${stats_db_name}.dataset_refereed STORED AS PARQUET as -with peer_reviewed as -(select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed -from ${openaire_db_name}.dataset r lateral -view explode(r.instance) instances as inst -where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE -and inst.refereed.classname='peerReviewed'), -non_peer_reviewed as -(select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed -from ${openaire_db_name}.dataset r lateral -view explode(r.instance) instances as inst -where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE -and inst.refereed.classname='nonPeerReviewed') -select peer_reviewed.* from peer_reviewed -union all -select non_peer_reviewed.* from non_peer_reviewed -left join peer_reviewed on peer_reviewed.id=non_peer_reviewed.id -where peer_reviewed.id is null; +with peer_reviewed as ( + select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed + from ${openaire_db_name}.dataset r lateral view explode(r.instance) instances as inst + where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE and inst.refereed.classname='peerReviewed'), +non_peer_reviewed as ( + select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed + from ${openaire_db_name}.dataset r lateral view explode(r.instance) instances as inst + where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE and inst.refereed.classname='nonPeerReviewed') +select distinct * +from ( + select peer_reviewed.* from peer_reviewed + union all + select non_peer_reviewed.* from non_peer_reviewed + left join peer_reviewed on peer_reviewed.id=non_peer_reviewed.id + where peer_reviewed.id is null) pr; DROP TABLE IF EXISTS ${stats_db_name}.software_refereed purge; - ---CREATE TABLE IF NOT EXISTS ${stats_db_name}.software_refereed STORED AS PARQUET as ---select substr(r.id, 4) as id, inst.refereed.classname as refereed ---from ${openaire_db_name}.software r lateral view explode(r.instance) instances as inst ---where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE; CREATE TABLE IF NOT EXISTS ${stats_db_name}.software_refereed STORED AS PARQUET as -with peer_reviewed as -(select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed -from ${openaire_db_name}.software r lateral -view explode(r.instance) instances as inst -where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE -and inst.refereed.classname='peerReviewed'), -non_peer_reviewed as -(select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed -from ${openaire_db_name}.software r lateral -view explode(r.instance) instances as inst -where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE -and inst.refereed.classname='nonPeerReviewed') -select peer_reviewed.* from peer_reviewed -union all -select non_peer_reviewed.* from non_peer_reviewed -left join peer_reviewed on peer_reviewed.id=non_peer_reviewed.id -where peer_reviewed.id is null; +with peer_reviewed as ( + select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed + from ${openaire_db_name}.software r lateral view explode(r.instance) instances as inst + where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE and inst.refereed.classname='peerReviewed'), +non_peer_reviewed as ( + select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed + from ${openaire_db_name}.software r lateral view explode(r.instance) instances as inst + where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE and inst.refereed.classname='nonPeerReviewed') +select distinct * +from ( + select peer_reviewed.* from peer_reviewed + union all + select non_peer_reviewed.* from non_peer_reviewed + left join peer_reviewed on peer_reviewed.id=non_peer_reviewed.id + where peer_reviewed.id is null) pr; DROP TABLE IF EXISTS ${stats_db_name}.otherresearchproduct_refereed purge; - ---CREATE TABLE IF NOT EXISTS ${stats_db_name}.otherresearchproduct_refereed STORED AS PARQUET as ---select substr(r.id, 4) as id, inst.refereed.classname as refereed ---from ${openaire_db_name}.otherresearchproduct r lateral view explode(r.instance) instances as inst ---where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE; CREATE TABLE IF NOT EXISTS ${stats_db_name}.otherresearchproduct_refereed STORED AS PARQUET as -with peer_reviewed as -(select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed -from ${openaire_db_name}.otherresearchproduct r lateral -view explode(r.instance) instances as inst -where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE -and inst.refereed.classname='peerReviewed'), -non_peer_reviewed as -(select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed -from ${openaire_db_name}.otherresearchproduct r lateral -view explode(r.instance) instances as inst -where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE -and inst.refereed.classname='nonPeerReviewed') -select peer_reviewed.* from peer_reviewed -union all -select non_peer_reviewed.* from non_peer_reviewed -left join peer_reviewed on peer_reviewed.id=non_peer_reviewed.id -where peer_reviewed.id is null; +with peer_reviewed as ( + select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed + from ${openaire_db_name}.otherresearchproduct r lateral view explode(r.instance) instances as inst + where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE and inst.refereed.classname='peerReviewed'), +non_peer_reviewed as ( + select distinct substr(r.id, 4) as id, inst.refereed.classname as refereed + from ${openaire_db_name}.otherresearchproduct r lateral view explode(r.instance) instances as inst + where r.datainfo.deletedbyinference=false and r.datainfo.invisible = FALSE and inst.refereed.classname='nonPeerReviewed') +select distinct * +from ( + select peer_reviewed.* from peer_reviewed + union all + select non_peer_reviewed.* from non_peer_reviewed + left join peer_reviewed on peer_reviewed.id=non_peer_reviewed.id + where peer_reviewed.id is null) pr; CREATE VIEW IF NOT EXISTS ${stats_db_name}.result_refereed as select * from ${stats_db_name}.publication_refereed @@ -132,4 +104,4 @@ rel.properties[1].value apc_currency from ${openaire_db_name}.relation rel join ${openaire_db_name}.organization o on o.id=rel.source join ${openaire_db_name}.result r on r.id=rel.target -where rel.subreltype = 'affiliation' and rel.datainfo.deletedbyinference = false and size(rel.properties)>0; +where rel.subreltype = 'affiliation' and rel.datainfo.deletedbyinference = false and size(rel.properties)>0; \ No newline at end of file -- 2.17.1 From a3a570e9a030f2a6e60fc8a600d4ff5b600ae806 Mon Sep 17 00:00:00 2001 From: dimitrispie Date: Tue, 19 Mar 2024 09:42:21 +0200 Subject: [PATCH 17/22] Commit monitor-updates-wf --- .../dhp-stats-monitor-update/pom.xml | 32 ++ .../oozie_app/config-default.xml | 30 ++ .../oozie_app/copyDataToImpalaCluster.sh | 75 +++++ .../oozie_app/finalizeImpalaCluster.sh | 57 ++++ .../graph/stats-monitor/oozie_app/monitor.sh | 60 ++++ .../oozie_app/scripts/updateMonitorDB.sql | 278 ++++++++++++++++ .../oozie_app/scripts/updateMonitorDBAll.sql | 297 ++++++++++++++++++ .../scripts/updateMonitorDB_institutions.sql | 67 ++++ .../stats-monitor/oozie_app/workflow.xml | 111 +++++++ 9 files changed, 1007 insertions(+) create mode 100644 dhp-workflows/dhp-stats-monitor-update/pom.xml create mode 100644 dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/config-default.xml create mode 100644 dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/copyDataToImpalaCluster.sh create mode 100644 dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/finalizeImpalaCluster.sh create mode 100644 dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/monitor.sh create mode 100644 dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/scripts/updateMonitorDB.sql create mode 100644 dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/scripts/updateMonitorDBAll.sql create mode 100644 dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/scripts/updateMonitorDB_institutions.sql create mode 100644 dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/workflow.xml diff --git a/dhp-workflows/dhp-stats-monitor-update/pom.xml b/dhp-workflows/dhp-stats-monitor-update/pom.xml new file mode 100644 index 000000000..f63c7568a --- /dev/null +++ b/dhp-workflows/dhp-stats-monitor-update/pom.xml @@ -0,0 +1,32 @@ + + + + dhp-workflows + eu.dnetlib.dhp + 1.2.4-SNAPSHOT + + 4.0.0 + dhp-stats-monitor-update + + + org.apache.spark + spark-core_2.11 + + + org.apache.spark + spark-sql_2.11 + + + + + + pl.project13.maven + git-commit-id-plugin + 2.1.11 + + false + + + + + diff --git a/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/config-default.xml b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/config-default.xml new file mode 100644 index 000000000..b2a1322e6 --- /dev/null +++ b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/config-default.xml @@ -0,0 +1,30 @@ + + + jobTracker + ${jobTracker} + + + nameNode + ${nameNode} + + + oozie.use.system.libpath + true + + + oozie.action.sharelib.for.spark + spark2 + + + hive_metastore_uris + thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083 + + + hive_jdbc_url + jdbc:hive2://iis-cdh5-test-m3.ocean.icm.edu.pl:10000/;UseNativeQuery=1;?spark.executor.memory=22166291558;spark.yarn.executor.memoryOverhead=3225;spark.driver.memory=15596411699;spark.yarn.driver.memoryOverhead=1228 + + + oozie.wf.workflow.notification.url + {serviceUrl}/v1/oozieNotification/jobUpdate?jobId=$jobId%26status=$status + + \ No newline at end of file diff --git a/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/copyDataToImpalaCluster.sh b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/copyDataToImpalaCluster.sh new file mode 100644 index 000000000..1587f7152 --- /dev/null +++ b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/copyDataToImpalaCluster.sh @@ -0,0 +1,75 @@ +export PYTHON_EGG_CACHE=/home/$(whoami)/.python-eggs +export link_folder=/tmp/impala-shell-python-egg-cache-$(whoami) +if ! [ -L $link_folder ] +then + rm -Rf "$link_folder" + ln -sfn ${PYTHON_EGG_CACHE}${link_folder} ${link_folder} +fi + +#export HADOOP_USER_NAME=$2 + +function copydb() { + + export HADOOP_USER="dimitris.pierrakos" + export HADOOP_USER_NAME='dimitris.pierrakos' + + db=$1 + FILE=("hive_wf_tmp_"$RANDOM) + hdfs dfs -mkdir hdfs://impala-cluster-mn1.openaire.eu:8020/tmp/$FILE/ + + # change ownership to impala +# hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db + hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/ + + + # copy the databases from ocean to impala + echo "copying $db" + hadoop distcp -Dmapreduce.map.memory.mb=6144 -pb hdfs://nameservice1/user/hive/warehouse/${db}.db hdfs://impala-cluster-mn1.openaire.eu:8020/tmp/$FILE/ + + hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db + + # drop tables from db + for i in `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} --delimited -q "show tables"`; + do + `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} -q "drop table $i;"`; + done + + # drop views from db + for i in `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} --delimited -q "show tables"`; + do + `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} -q "drop view $i;"`; + done + + # delete the database + impala-shell -i impala-cluster-dn1.openaire.eu -q "drop database if exists ${db} cascade"; + + # create the databases + impala-shell -i impala-cluster-dn1.openaire.eu -q "create database ${db}"; + + impala-shell -q "INVALIDATE METADATA" + echo "creating schema for ${db}" + for (( k = 0; k < 5; k ++ )); do + for i in `impala-shell -d ${db} --delimited -q "show tables"`; + do + impala-shell -d ${db} --delimited -q "show create table $i"; + done | sed 's/"$/;/' | sed 's/^"//' | sed 's/[[:space:]]\date[[:space:]]/`date`/g' | impala-shell --user $HADOOP_USER_NAME -i impala-cluster-dn1.openaire.eu -c -f - + done + + # load the data from /tmp in the respective tables + echo "copying data in tables and computing stats" + for i in `impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} --delimited -q "show tables"`; + do + impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} -q "load data inpath '/tmp/$FILE/${db}.db/$i' into table $i"; + impala-shell -i impala-cluster-dn1.openaire.eu -d ${db} -q "compute stats $i"; + done + + # deleting the remaining directory from hdfs +hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -rm -R /tmp/$FILE/${db}.db +} + +MONITOR_DB=$1 +#HADOOP_USER_NAME=$2 + +copydb $MONITOR_DB'_institutions' +copydb $MONITOR_DB + diff --git a/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/finalizeImpalaCluster.sh b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/finalizeImpalaCluster.sh new file mode 100644 index 000000000..cb5452154 --- /dev/null +++ b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/finalizeImpalaCluster.sh @@ -0,0 +1,57 @@ +export PYTHON_EGG_CACHE=/home/$(whoami)/.python-eggs +export link_folder=/tmp/impala-shell-python-egg-cache-$(whoami) +if ! [ -L $link_folder ] +then + rm -Rf "$link_folder" + ln -sfn ${PYTHON_EGG_CACHE}${link_folder} ${link_folder} +fi +# +#function createShadowDB() { +# SOURCE=$1 +# SHADOW=$2 +# +# # drop views from db +# for i in `impala-shell -i impala-cluster-dn1.openaire.eu -d ${SHADOW} --delimited -q "show tables"`; +# do +# `impala-shell -i impala-cluster-dn1.openaire.eu -d ${SHADOW} -q "drop view $i;"`; +# done +# +# impala-shell -i impala-cluster-dn1.openaire.eu -q "drop database ${SHADOW} CASCADE"; +# impala-shell -i impala-cluster-dn1.openaire.eu -q "create database if not exists ${SHADOW}"; +## impala-shell -i impala-cluster-dn1.openaire.eu -d ${SHADOW} -q "show tables" | sed "s/^/drop view if exists ${SHADOW}./" | sed "s/$/;/" | impala-shell -i impala-cluster-dn1.openaire.eu -f - +# impala-shell -i impala-cluster-dn1.openaire.eu -d ${SOURCE} -q "show tables" --delimited | sed "s/\(.*\)/create view ${SHADOW}.\1 as select * from ${SOURCE}.\1;/" | impala-shell -i impala-cluster-dn1.openaire.eu -f - +#} +# +#MONITOR_DB=$1 +#MONITOR_DB_SHADOW=$2 +# +#createShadowDB $MONITOR_DB'_institutions' $MONITOR_DB'_institutions_shadow' +#createShadowDB $MONITOR_DB $MONITOR_DB'_shadow' + +SOURCE=$1 +PRODUCTION=$2 +echo ${SOURCE} +echo ${PRODUCTION} + +#echo "Updating ${PRODUCTION} monitor database old cluster" +#impala-shell -q "create database if not exists ${PRODUCTION}" +#impala-shell -d ${PRODUCTION} -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}./" | sed "s/$/;/" | impala-shell -c -f - +#impala-shell -d ${SOURCE} -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}.\1 as select * from ${SOURCE}.\1;/" | impala-shell -c -f - +# +#echo "Updating ${PRODUCTION}_institutions database old cluster" +#impala-shell -q "create database if not exists ${PRODUCTION}_institutions" +#impala-shell -d ${PRODUCTION}_institutions -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}_institutions./" | sed "s/$/;/" | impala-shell -c -f - +#impala-shell -d ${SOURCE}_institutions -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}_institutions.\1 as select * from ${SOURCE}_institutions.\1;/" | impala-shell -c -f - +#echo "Production insitutions db ready!" + +echo "Updating ${PRODUCTION} monitor database" +impala-shell -i impala-cluster-dn1.openaire.eu -q "create database if not exists ${PRODUCTION}" +impala-shell -i impala-cluster-dn1.openaire.eu -d ${PRODUCTION} -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}./" | sed "s/$/;/" | impala-shell -i impala-cluster-dn1.openaire.eu -c -f - +impala-shell -i impala-cluster-dn1.openaire.eu -d ${SOURCE} -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}.\1 as select * from ${SOURCE}.\1;/" | impala-shell -i impala-cluster-dn1.openaire.eu -c -f - +echo "Production monitor db ready!" + +echo "Updating ${PRODUCTION}_institutions database" +impala-shell -i impala-cluster-dn1.openaire.eu -q "create database if not exists ${PRODUCTION}_institutions" +impala-shell -i impala-cluster-dn1.openaire.eu -d ${PRODUCTION}_institutions -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}_institutions./" | sed "s/$/;/" | impala-shell -i impala-cluster-dn1.openaire.eu -c -f - +impala-shell -i impala-cluster-dn1.openaire.eu -d ${SOURCE}_institutions -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}_institutions.\1 as select * from ${SOURCE}_institutions.\1;/" | impala-shell -i impala-cluster-dn1.openaire.eu -c -f - +echo "Production insitutions db ready!" \ No newline at end of file diff --git a/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/monitor.sh b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/monitor.sh new file mode 100644 index 000000000..50f5983e4 --- /dev/null +++ b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/monitor.sh @@ -0,0 +1,60 @@ +export PYTHON_EGG_CACHE=/home/$(whoami)/.python-eggs +export link_folder=/tmp/impala-shell-python-egg-cache-$(whoami) +if ! [ -L $link_folder ] +then + rm -Rf "$link_folder" + ln -sfn ${PYTHON_EGG_CACHE}${link_folder} ${link_folder} +fi + +export SOURCE=$1 +export TARGET=$2 +export SHADOW=$3 +export SCRIPT_PATH=$4 +export SCRIPT_PATH2=$5 +export SCRIPT_PATH2=$6 + +export HIVE_OPTS="-hiveconf mapred.job.queue.name=analytics -hiveconf hive.spark.client.connect.timeout=120000ms -hiveconf hive.spark.client.server.connect.timeout=300000ms -hiveconf spark.executor.memory=19166291558 -hiveconf spark.yarn.executor.memoryOverhead=3225 -hiveconf spark.driver.memory=11596411699 -hiveconf spark.yarn.driver.memoryOverhead=1228" +export HADOOP_USER_NAME="oozie" + +echo "Getting file from " $4 +hdfs dfs -copyToLocal $4 + +echo "Getting file from " $5 +hdfs dfs -copyToLocal $5 + +echo "Getting file from " $6 +hdfs dfs -copyToLocal $6 + +#update Monitor DB +cat updateMonitorDBAll.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2/g1" > foo +hive $HIVE_OPTS -f foo + +#update Institutions DB +cat updateMonitorDB_institutions.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2_institutions/g1" > foo +hive $HIVE_OPTS -f foo +cat updateMonitorDB.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2_institutions/g1" > foo +hive $HIVE_OPTS -f foo + + + +echo "Hive shell finished" + +#echo "Updating shadow monitor insitutions database" +#hive -e "drop database if exists ${SHADOW}_institutions cascade" +#hive -e "create database if not exists ${SHADOW}_institutions" +#hive $HIVE_OPTS --database ${2}_institutions -e "show tables" | grep -v WARN | sed "s/\(.*\)/create view ${SHADOW}_institutions.\1 as select * from ${2}_institutions.\1;/" > foo +#hive -f foo +#echo "Shadow db monitor insitutions ready!" +# +##update Monitor DB +#cat updateMonitorDBAll.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2/g1" > foo +#hive $HIVE_OPTS -f foo +# +#echo "Hive shell finished" +# +#echo "Updating shadow monitor database" +#hive -e "drop database if exists ${SHADOW} cascade" +#hive -e "create database if not exists ${SHADOW}" +#hive $HIVE_OPTS --database ${2} -e "show tables" | grep -v WARN | sed "s/\(.*\)/create view ${SHADOW}.\1 as select * from ${2}.\1;/" > foo +#hive -f foo +#echo "Shadow db monitor insitutions ready!" diff --git a/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/scripts/updateMonitorDB.sql b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/scripts/updateMonitorDB.sql new file mode 100644 index 000000000..321fba87a --- /dev/null +++ b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/scripts/updateMonitorDB.sql @@ -0,0 +1,278 @@ +--drop database if exists TARGET cascade; +--create database if not exists TARGET; +-- +--create view if not exists TARGET.category as select * from SOURCE.category; +--create view if not exists TARGET.concept as select * from SOURCE.concept; +--create view if not exists TARGET.context as select * from SOURCE.context; +--create view if not exists TARGET.country as select * from SOURCE.country; +--create view if not exists TARGET.countrygdp as select * from SOURCE.countrygdp; +--create view if not exists TARGET.creation_date as select * from SOURCE.creation_date; +--create view if not exists TARGET.funder as select * from SOURCE.funder; +--create view if not exists TARGET.fundref as select * from SOURCE.fundref; +--create view if not exists TARGET.rndexpenditure as select * from SOURCE.rndexpediture; +--create view if not exists TARGET.rndgdpexpenditure as select * from SOURCE.rndgdpexpenditure; +--create view if not exists TARGET.doctoratestudents as select * from SOURCE.doctoratestudents; +--create view if not exists TARGET.totalresearchers as select * from SOURCE.totalresearchers; +--create view if not exists TARGET.totalresearchersft as select * from SOURCE.totalresearchersft; +--create view if not exists TARGET.hrrst as select * from SOURCE.hrrst; +-- +--create table TARGET.result stored as parquet as +-- select distinct * from ( +-- select * from SOURCE.result r where exists (select 1 from SOURCE.result_projects rp join SOURCE.project p on rp.project=p.id where rp.id=r.id) +-- union all +-- select * from SOURCE.result r where exists (select 1 from SOURCE.result_concepts rc where rc.id=r.id) +-- union all +-- select * from SOURCE.result r where exists (select 1 from SOURCE.result_organization ro where ro.id=r.id and ro.organization in ( +-- 'openorgs____::b84450f9864182c67b8611b5593f4250', --"Athena Research and Innovation Center In Information Communication & Knowledge Technologies', --ARC" +-- 'openorgs____::d41cf6bd4ab1b1362a44397e0b95c975', --National Research Council +-- 'openorgs____::d2a09b9d5eabb10c95f9470e172d05d2', --??? Not exists ?? +-- 'openorgs____::d169c7407dd417152596908d48c11460', --Masaryk University +-- 'openorgs____::1ec924b1759bb16d0a02f2dad8689b21', --University of Belgrade +-- 'openorgs____::0ae431b820e4c33db8967fbb2b919150', --University of Helsinki +-- 'openorgs____::759d59f05d77188faee99b7493b46805', --University of Minho +-- 'openorgs____::cad284878801b9465fa51a95b1d779db', --Universidad Politécnica de Madrid +-- 'openorgs____::eadc8da90a546e98c03f896661a2e4d4', --University of Göttingen +-- 'openorgs____::c0286313e36479eff8676dba9b724b40', --National and Kapodistrian University of Athens +-- -- 'openorgs____::c80a8243a5e5c620d7931c88d93bf17a', --Université Paris Diderot +-- 'openorgs____::c08634f0a6b0081c3dc6e6c93a4314f3', --Bielefeld University +-- 'openorgs____::6fc85e4a8f7ecaf4b0c738d010e967ea', --University of Southern Denmark +-- 'openorgs____::3d6122f87f9a97a99d8f6e3d73313720', --Humboldt-Universität zu Berlin +-- 'openorgs____::16720ada63d0fa8ca41601feae7d1aa5', --TU Darmstadt +-- 'openorgs____::ccc0a066b56d2cfaf90c2ae369df16f5', --KU Leuven +-- 'openorgs____::4c6f119632adf789746f0a057ed73e90', --University of the Western Cape +-- 'openorgs____::ec3665affa01aeafa28b7852c4176dbd', --Rudjer Boskovic Institute +-- 'openorgs____::5f31346d444a7f06a28c880fb170b0f6', --Ghent University +-- 'openorgs____::2dbe47117fd5409f9c61620813456632', --University of Luxembourg +-- 'openorgs____::6445d7758d3a40c4d997953b6632a368', --National Institute of Informatics (NII) +-- 'openorgs____::b77c01aa15de3675da34277d48de2ec1', -- Valencia Catholic University Saint Vincent Martyr +-- 'openorgs____::7fe2f66cdc43983c6b24816bfe9cf6a0', -- Unviersity of Warsaw +-- 'openorgs____::15e7921fc50d9aa1229a82a84429419e', -- University Of Thessaly +-- 'openorgs____::11f7919dadc8f8a7251af54bba60c956', -- Technical University of Crete +-- 'openorgs____::84f0c5f5dbb6daf42748485924efde4b', -- University of Piraeus +-- 'openorgs____::4ac562f0376fce3539504567649cb373', -- University of Patras +-- 'openorgs____::3e8d1f8c3f6cd7f418b09f1f58b4873b', -- Aristotle University of Thessaloniki +-- 'openorgs____::3fcef6e1c469c10f2a84b281372c9814', -- World Bank +-- 'openorgs____::1698a2eb1885ef8adb5a4a969e745ad3', -- École des Ponts ParisTech +-- 'openorgs____::e15adb13c4dadd49de4d35c39b5da93a', -- Nanyang Technological University +-- 'openorgs____::4b34103bde246228fcd837f5f1bf4212', -- Autonomous University of Barcelona +-- 'openorgs____::72ec75fcfc4e0df1a76dc4c49007fceb', -- McMaster University +-- 'openorgs____::51c7fc556e46381734a25a6fbc3fd398', -- University of Modena and Reggio Emilia +-- 'openorgs____::235d7f9ad18ecd7e6dc62ea4990cb9db', -- Bilkent University +-- 'openorgs____::31f2fa9e05b49d4cf40a19c3fed8eb06', -- Saints Cyril and Methodius University of Skopje +-- 'openorgs____::db7686f30f22cbe73a4fde872ce812a6', -- University of Milan +-- 'openorgs____::b8b8ca674452579f3f593d9f5e557483', -- University College Cork +-- 'openorgs____::38d7097854736583dde879d12dacafca' -- Brown University +-- 'openorgs____::57784c9e047e826fefdb1ef816120d92', --Arts et Métiers ParisTech +-- 'openorgs____::2530baca8a15936ba2e3297f2bce2e7e', -- University of Cape Town +-- 'openorgs____::d11f981828c485cd23d93f7f24f24db1', -- Technological University Dublin +-- 'openorgs____::5e6bf8962665cdd040341171e5c631d8', -- Delft University of Technology +-- 'openorgs____::846cb428d3f52a445f7275561a7beb5d', -- University of Manitoba +-- 'openorgs____::eb391317ed0dc684aa81ac16265de041', -- Universitat Rovira i Virgili +-- 'openorgs____::66aa9fc2fceb271423dfabcc38752dc0', -- Lund University +-- 'openorgs____::3cff625a4370d51e08624cc586138b2f' -- IMT Atlantique +-- ) )) foo; +-- +--ANALYZE TABLE TARGET.result COMPUTE STATISTICS; + +create view if not exists TARGET.category as select * from SOURCE.category; +create view if not exists TARGET.concept as select * from SOURCE.concept; +create view if not exists TARGET.context as select * from SOURCE.context; +create view if not exists TARGET.country as select * from SOURCE.country; +create view if not exists TARGET.countrygdp as select * from SOURCE.countrygdp; +create view if not exists TARGET.creation_date as select * from SOURCE.creation_date; +create view if not exists TARGET.funder as select * from SOURCE.funder; +create view if not exists TARGET.fundref as select * from SOURCE.fundref; +create view if not exists TARGET.rndexpenditure as select * from SOURCE.rndexpediture; +create view if not exists TARGET.rndgdpexpenditure as select * from SOURCE.rndgdpexpenditure; +create view if not exists TARGET.doctoratestudents as select * from SOURCE.doctoratestudents; +create view if not exists TARGET.totalresearchers as select * from SOURCE.totalresearchers; +create view if not exists TARGET.totalresearchersft as select * from SOURCE.totalresearchersft; +create view if not exists TARGET.hrrst as select * from SOURCE.hrrst; +--create view if not exists TARGET.graduatedoctorates as select * from SOURCE.graduatedoctorates; + +create table TARGET.result_citations stored as parquet as select * from SOURCE.result_citations orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_citations COMPUTE STATISTICS; + +create table TARGET.result_references_oc stored as parquet as select * from SOURCE.result_references_oc orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_references_oc COMPUTE STATISTICS; + +create table TARGET.result_citations_oc stored as parquet as select * from SOURCE.result_citations_oc orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_citations_oc COMPUTE STATISTICS; + +create table TARGET.result_classifications stored as parquet as select * from SOURCE.result_classifications orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_classifications COMPUTE STATISTICS; + +create table TARGET.result_apc stored as parquet as select * from SOURCE.result_apc orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_apc COMPUTE STATISTICS; + +create table TARGET.result_concepts stored as parquet as select * from SOURCE.result_concepts orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_concepts COMPUTE STATISTICS; + +create table TARGET.result_datasources stored as parquet as select * from SOURCE.result_datasources orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_datasources COMPUTE STATISTICS; + +create table TARGET.result_fundercount stored as parquet as select * from SOURCE.result_fundercount orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_fundercount COMPUTE STATISTICS; + +create table TARGET.result_gold stored as parquet as select * from SOURCE.result_gold orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_gold COMPUTE STATISTICS; + +create table TARGET.result_greenoa stored as parquet as select * from SOURCE.result_greenoa orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_greenoa COMPUTE STATISTICS; + +create table TARGET.result_languages stored as parquet as select * from SOURCE.result_languages orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_languages COMPUTE STATISTICS; + +create table TARGET.result_licenses stored as parquet as select * from SOURCE.result_licenses orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_licenses COMPUTE STATISTICS; + +create table TARGET.licenses_normalized STORED AS PARQUET as select * from SOURCE.licenses_normalized; +--ANALYZE TABLE TARGET.licenses_normalized COMPUTE STATISTICS; + +create table TARGET.result_oids stored as parquet as select * from SOURCE.result_oids orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_oids COMPUTE STATISTICS; + +create table TARGET.result_organization stored as parquet as select * from SOURCE.result_organization orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_organization COMPUTE STATISTICS; + +create table TARGET.result_peerreviewed stored as parquet as select * from SOURCE.result_peerreviewed orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_peerreviewed COMPUTE STATISTICS; + +create table TARGET.result_pids stored as parquet as select * from SOURCE.result_pids orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_pids COMPUTE STATISTICS; + +create table TARGET.result_projectcount stored as parquet as select * from SOURCE.result_projectcount orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_projectcount COMPUTE STATISTICS; + +create table TARGET.result_projects stored as parquet as select * from SOURCE.result_projects orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_projects COMPUTE STATISTICS; + +create table TARGET.result_refereed stored as parquet as select * from SOURCE.result_refereed orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_refereed COMPUTE STATISTICS; + +create table TARGET.result_sources stored as parquet as select * from SOURCE.result_sources orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_sources COMPUTE STATISTICS; + +create table TARGET.result_topics stored as parquet as select * from SOURCE.result_topics orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_topics COMPUTE STATISTICS; + +create table TARGET.result_fos stored as parquet as select * from SOURCE.result_fos orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_fos COMPUTE STATISTICS; + +create table TARGET.result_accessroute stored as parquet as select * from SOURCE.result_accessroute orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_accessroute COMPUTE STATISTICS; + +create table TARGET.result_instance stored as parquet as select * from SOURCE.result_instance orig where exists (select 1 from TARGET.result r where r.id=orig.id); +create table TARGET.result_orcid stored as parquet as select * from SOURCE.result_orcid orig where exists (select 1 from TARGET.result r where r.id=orig.id); + +create view TARGET.foo1 as select * from SOURCE.result_result rr where rr.source in (select id from TARGET.result); +create view TARGET.foo2 as select * from SOURCE.result_result rr where rr.target in (select id from TARGET.result); +create table TARGET.result_result STORED AS PARQUET as select distinct * from (select * from TARGET.foo1 union all select * from TARGET.foo2) foufou; +drop view TARGET.foo1; +drop view TARGET.foo2; +--ANALYZE TABLE TARGET.result_result COMPUTE STATISTICS; + +-- datasources +create view if not exists TARGET.datasource as select * from SOURCE.datasource; +create view if not exists TARGET.datasource_oids as select * from SOURCE.datasource_oids; +create view if not exists TARGET.datasource_organizations as select * from SOURCE.datasource_organizations; +create view if not exists TARGET.datasource_sources as select * from SOURCE.datasource_sources; + +create table TARGET.datasource_results stored as parquet as select id as result, datasource as id from TARGET.result_datasources; +--ANALYZE TABLE TARGET.datasource_results COMPUTE STATISTICS; + +-- organizations +create view if not exists TARGET.organization as select * from SOURCE.organization; +create view if not exists TARGET.organization_datasources as select * from SOURCE.organization_datasources; +create view if not exists TARGET.organization_pids as select * from SOURCE.organization_pids; +create view if not exists TARGET.organization_projects as select * from SOURCE.organization_projects; +create view if not exists TARGET.organization_sources as select * from SOURCE.organization_sources; + +-- projects +create view if not exists TARGET.project as select * from SOURCE.project; +create view if not exists TARGET.project_oids as select * from SOURCE.project_oids; +create view if not exists TARGET.project_organizations as select * from SOURCE.project_organizations; +create view if not exists TARGET.project_resultcount as select * from SOURCE.project_resultcount; +create view if not exists TARGET.project_classification as select * from SOURCE.project_classification; +create view if not exists TARGET.project_organization_contribution as select * from SOURCE.project_organization_contribution; + +create table TARGET.project_results stored as parquet as select id as result, project as id from TARGET.result_projects; +--ANALYZE TABLE TARGET.project_results COMPUTE STATISTICS; + +-- indicators +-- Sprint 1 ---- +create table TARGET.indi_pub_green_oa stored as parquet as select * from SOURCE.indi_pub_green_oa orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_green_oa COMPUTE STATISTICS; +create table TARGET.indi_pub_grey_lit stored as parquet as select * from SOURCE.indi_pub_grey_lit orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_grey_lit COMPUTE STATISTICS; +create table TARGET.indi_pub_doi_from_crossref stored as parquet as select * from SOURCE.indi_pub_doi_from_crossref orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_doi_from_crossref COMPUTE STATISTICS; +-- Sprint 2 ---- +create table TARGET.indi_result_has_cc_licence stored as parquet as select * from SOURCE.indi_result_has_cc_licence orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_result_has_cc_licence COMPUTE STATISTICS; +create table TARGET.indi_result_has_cc_licence_url stored as parquet as select * from SOURCE.indi_result_has_cc_licence_url orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_result_has_cc_licence_url COMPUTE STATISTICS; +create table TARGET.indi_pub_has_abstract stored as parquet as select * from SOURCE.indi_pub_has_abstract orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_has_abstract COMPUTE STATISTICS; +create table TARGET.indi_result_with_orcid stored as parquet as select * from SOURCE.indi_result_with_orcid orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_result_with_orcid COMPUTE STATISTICS; +---- Sprint 3 ---- +create table TARGET.indi_funded_result_with_fundref stored as parquet as select * from SOURCE.indi_funded_result_with_fundref orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_funded_result_with_fundref COMPUTE STATISTICS; +create view TARGET.indi_result_org_collab as select * from SOURCE.indi_result_org_collab; +create view TARGET.indi_result_org_country_collab as select * from SOURCE.indi_result_org_country_collab; +create view TARGET.indi_project_collab_org as select * from SOURCE.indi_project_collab_org; +create view TARGET.indi_project_collab_org_country as select * from SOURCE.indi_project_collab_org_country; +create view TARGET.indi_funder_country_collab as select * from SOURCE.indi_funder_country_collab; +create view TARGET.indi_result_country_collab as select * from SOURCE.indi_result_country_collab; +---- Sprint 4 ---- +create table TARGET.indi_pub_diamond stored as parquet as select * from SOURCE.indi_pub_diamond orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_diamond COMPUTE STATISTICS; +create table TARGET.indi_pub_in_transformative stored as parquet as select * from SOURCE.indi_pub_in_transformative orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_in_transformative COMPUTE STATISTICS; +create table TARGET.indi_pub_closed_other_open stored as parquet as select * from SOURCE.indi_pub_closed_other_open orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_closed_other_open COMPUTE STATISTICS; +---- Sprint 5 ---- +create table TARGET.indi_result_no_of_copies stored as parquet as select * from SOURCE.indi_result_no_of_copies orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_result_no_of_copies COMPUTE STATISTICS; +---- Sprint 6 ---- +create table TARGET.indi_pub_hybrid_oa_with_cc stored as parquet as select * from SOURCE.indi_pub_hybrid_oa_with_cc orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_hybrid_oa_with_cc COMPUTE STATISTICS; +create table TARGET.indi_pub_bronze_oa stored as parquet as select * from SOURCE.indi_pub_bronze_oa orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_bronze_oa COMPUTE STATISTICS; +create table TARGET.indi_pub_downloads stored as parquet as select * from SOURCE.indi_pub_downloads orig where exists (select 1 from TARGET.result r where r.id=orig.result_id); +--ANALYZE TABLE TARGET.indi_pub_downloads COMPUTE STATISTICS; +create table TARGET.indi_pub_downloads_datasource stored as parquet as select * from SOURCE.indi_pub_downloads_datasource orig where exists (select 1 from TARGET.result r where r.id=orig.result_id); +--ANALYZE TABLE TARGET.indi_pub_downloads_datasource COMPUTE STATISTICS; +create table TARGET.indi_pub_downloads_year stored as parquet as select * from SOURCE.indi_pub_downloads_year orig where exists (select 1 from TARGET.result r where r.id=orig.result_id); +--ANALYZE TABLE TARGET.indi_pub_downloads_year COMPUTE STATISTICS; +create table TARGET.indi_pub_downloads_datasource_year stored as parquet as select * from SOURCE.indi_pub_downloads_datasource_year orig where exists (select 1 from TARGET.result r where r.id=orig.result_id); +--ANALYZE TABLE TARGET.indi_pub_downloads_datasource_year COMPUTE STATISTICS; +---- Sprint 7 ---- +create table TARGET.indi_pub_gold_oa stored as parquet as select * from SOURCE.indi_pub_gold_oa orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_gold_oa COMPUTE STATISTICS; +create table TARGET.indi_pub_hybrid stored as parquet as select * from SOURCE.indi_pub_hybrid orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_hybrid COMPUTE STATISTICS; +create view TARGET.indi_org_fairness as select * from SOURCE.indi_org_fairness; +create view TARGET.indi_org_fairness_pub_pr as select * from SOURCE.indi_org_fairness_pub_pr; +create view TARGET.indi_org_fairness_pub_year as select * from SOURCE.indi_org_fairness_pub_year; +create view TARGET.indi_org_fairness_pub as select * from SOURCE.indi_org_fairness_pub; +create view TARGET.indi_org_fairness_year as select * from SOURCE.indi_org_fairness_year; +create view TARGET.indi_org_findable_year as select * from SOURCE.indi_org_findable_year; +create view TARGET.indi_org_findable as select * from SOURCE.indi_org_findable; +create view TARGET.indi_org_openess as select * from SOURCE.indi_org_openess; +create view TARGET.indi_org_openess_year as select * from SOURCE.indi_org_openess_year; +create table TARGET.indi_pub_has_preprint stored as parquet as select * from SOURCE.indi_pub_has_preprint orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_has_preprint COMPUTE STATISTICS; +create table TARGET.indi_pub_in_subscribed stored as parquet as select * from SOURCE.indi_pub_in_subscribed orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_in_subscribed COMPUTE STATISTICS; +create table TARGET.indi_result_with_pid stored as parquet as select * from SOURCE.indi_result_with_pid orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_result_with_pid COMPUTE STATISTICS; +create table TARGET.indi_impact_measures stored as parquet as select * from SOURCE.indi_impact_measures orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_impact_measures COMPUTE STATISTICS; +create table TARGET.indi_pub_interdisciplinarity stored as parquet as select * from SOURCE.indi_pub_interdisciplinarity orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_interdisciplinarity COMPUTE STATISTICS; +create table TARGET.result_apc_affiliations stored as parquet as select * from SOURCE.result_apc_affiliations orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_apc_affiliations COMPUTE STATISTICS; +create table TARGET.indi_is_project_result_after stored as parquet as select * from SOURCE.indi_is_project_result_after orig where exists (select 1 from TARGET.result r where r.id=orig.result_id); +create table TARGET.indi_is_funder_plan_s stored as parquet as select * from SOURCE.indi_is_funder_plan_s orig where exists (select 1 from TARGET.result r where r.id=orig.id); diff --git a/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/scripts/updateMonitorDBAll.sql b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/scripts/updateMonitorDBAll.sql new file mode 100644 index 000000000..35ab42029 --- /dev/null +++ b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/scripts/updateMonitorDBAll.sql @@ -0,0 +1,297 @@ +drop database if exists TARGET cascade; +create database if not exists TARGET; + +create view if not exists TARGET.category as select * from SOURCE.category; +create view if not exists TARGET.concept as select * from SOURCE.concept; +create view if not exists TARGET.context as select * from SOURCE.context; +create view if not exists TARGET.country as select * from SOURCE.country; +create view if not exists TARGET.countrygdp as select * from SOURCE.countrygdp; +create view if not exists TARGET.creation_date as select * from SOURCE.creation_date; +create view if not exists TARGET.funder as select * from SOURCE.funder; +create view if not exists TARGET.fundref as select * from SOURCE.fundref; +create view if not exists TARGET.rndexpenditure as select * from SOURCE.rndexpediture; +create view if not exists TARGET.rndgdpexpenditure as select * from SOURCE.rndgdpexpenditure; +create view if not exists TARGET.doctoratestudents as select * from SOURCE.doctoratestudents; +create view if not exists TARGET.totalresearchers as select * from SOURCE.totalresearchers; +create view if not exists TARGET.totalresearchersft as select * from SOURCE.totalresearchersft; +create view if not exists TARGET.hrrst as select * from SOURCE.hrrst; +--create view if not exists TARGET.graduatedoctorates as select * from SOURCE.graduatedoctorates; + +create table TARGET.result stored as parquet as + select distinct * from ( + select * from SOURCE.result r where exists (select 1 from SOURCE.result_projects rp join SOURCE.project p on rp.project=p.id where rp.id=r.id) + union all + select * from SOURCE.result r where exists (select 1 from SOURCE.result_concepts rc where rc.id=r.id) + union all + select * from SOURCE.result r where exists (select 1 from SOURCE.result_organization ro where ro.id=r.id and ro.organization in ( + 'openorgs____::b84450f9864182c67b8611b5593f4250', --"Athena Research and Innovation Center In Information Communication & Knowledge Technologies', --ARC" + 'openorgs____::d41cf6bd4ab1b1362a44397e0b95c975', --National Research Council + 'openorgs____::d2a09b9d5eabb10c95f9470e172d05d2', --??? Not exists ?? + 'openorgs____::d169c7407dd417152596908d48c11460', --Masaryk University + 'openorgs____::1ec924b1759bb16d0a02f2dad8689b21', --University of Belgrade + 'openorgs____::0ae431b820e4c33db8967fbb2b919150', --University of Helsinki + 'openorgs____::759d59f05d77188faee99b7493b46805', --University of Minho + 'openorgs____::cad284878801b9465fa51a95b1d779db', --Universidad Politécnica de Madrid + 'openorgs____::eadc8da90a546e98c03f896661a2e4d4', --University of Göttingen + 'openorgs____::c0286313e36479eff8676dba9b724b40', --National and Kapodistrian University of Athens + -- 'openorgs____::c80a8243a5e5c620d7931c88d93bf17a', --Université Paris Diderot + 'openorgs____::c08634f0a6b0081c3dc6e6c93a4314f3', --Bielefeld University + 'openorgs____::6fc85e4a8f7ecaf4b0c738d010e967ea', --University of Southern Denmark + 'openorgs____::3d6122f87f9a97a99d8f6e3d73313720', --Humboldt-Universität zu Berlin + 'openorgs____::16720ada63d0fa8ca41601feae7d1aa5', --TU Darmstadt + 'openorgs____::ccc0a066b56d2cfaf90c2ae369df16f5', --KU Leuven + 'openorgs____::4c6f119632adf789746f0a057ed73e90', --University of the Western Cape + 'openorgs____::ec3665affa01aeafa28b7852c4176dbd', --Rudjer Boskovic Institute + 'openorgs____::5f31346d444a7f06a28c880fb170b0f6', --Ghent University + 'openorgs____::2dbe47117fd5409f9c61620813456632', --University of Luxembourg + 'openorgs____::6445d7758d3a40c4d997953b6632a368', --National Institute of Informatics (NII) + 'openorgs____::b77c01aa15de3675da34277d48de2ec1', -- Valencia Catholic University Saint Vincent Martyr + 'openorgs____::7fe2f66cdc43983c6b24816bfe9cf6a0', -- Unviersity of Warsaw + 'openorgs____::15e7921fc50d9aa1229a82a84429419e', -- University Of Thessaly + 'openorgs____::11f7919dadc8f8a7251af54bba60c956', -- Technical University of Crete + 'openorgs____::84f0c5f5dbb6daf42748485924efde4b', -- University of Piraeus + 'openorgs____::4ac562f0376fce3539504567649cb373', -- University of Patras + 'openorgs____::3e8d1f8c3f6cd7f418b09f1f58b4873b', -- Aristotle University of Thessaloniki + 'openorgs____::3fcef6e1c469c10f2a84b281372c9814', -- World Bank + 'openorgs____::1698a2eb1885ef8adb5a4a969e745ad3', -- École des Ponts ParisTech + 'openorgs____::e15adb13c4dadd49de4d35c39b5da93a', -- Nanyang Technological University + 'openorgs____::4b34103bde246228fcd837f5f1bf4212', -- Autonomous University of Barcelona + 'openorgs____::72ec75fcfc4e0df1a76dc4c49007fceb', -- McMaster University + 'openorgs____::51c7fc556e46381734a25a6fbc3fd398', -- University of Modena and Reggio Emilia + 'openorgs____::235d7f9ad18ecd7e6dc62ea4990cb9db', -- Bilkent University + 'openorgs____::31f2fa9e05b49d4cf40a19c3fed8eb06', -- Saints Cyril and Methodius University of Skopje + 'openorgs____::db7686f30f22cbe73a4fde872ce812a6', -- University of Milan + 'openorgs____::b8b8ca674452579f3f593d9f5e557483', -- University College Cork + 'openorgs____::38d7097854736583dde879d12dacafca', -- Brown University + 'openorgs____::57784c9e047e826fefdb1ef816120d92', --Arts et Métiers ParisTech + 'openorgs____::2530baca8a15936ba2e3297f2bce2e7e', -- University of Cape Town + 'openorgs____::d11f981828c485cd23d93f7f24f24db1', -- Technological University Dublin + 'openorgs____::5e6bf8962665cdd040341171e5c631d8', -- Delft University of Technology + 'openorgs____::846cb428d3f52a445f7275561a7beb5d', -- University of Manitoba + 'openorgs____::eb391317ed0dc684aa81ac16265de041', -- Universitat Rovira i Virgili + 'openorgs____::66aa9fc2fceb271423dfabcc38752dc0', -- Lund University + 'openorgs____::3cff625a4370d51e08624cc586138b2f', -- IMT Atlantique + 'openorgs____::c0b262bd6eab819e4c994914f9c010e2', -- National Institute of Geophysics and Volcanology + 'openorgs____::1624ff7c01bb641b91f4518539a0c28a', -- Vrije Universiteit Amsterdam + 'openorgs____::4d4051b56708688235252f1d8fddb8c1', -- Iscte - Instituto Universitário de Lisboa + 'openorgs____::ab4ac74c35fa5dada770cf08e5110fab', -- Universidade Católica Portuguesa + 'openorgs____::4d4051b56708688235252f1d8fddb8c1', -- Iscte - Instituto Universitário de Lisboa + 'openorgs____::5d55fb216b14691cf68218daf5d78cd9', -- Munster Technological University + 'openorgs____::0fccc7640f0cb44d5cd1b06b312a06b9', -- Cardiff University + 'openorgs____::8839b55dae0c84d56fd533f52d5d483a', -- Leibniz Institute of Ecological Urban and Regional Development + 'openorgs____::526468206bca24c1c90da6a312295cf4', -- Cyprus University of Technology + 'openorgs____::b5ca9d4340e26454e367e2908ef3872f', -- Alma Mater Studiorum University of Bologna + 'openorgs____::a6340e6ecf60f6bba163659df985b0f2' -- TU Dresden + ))) foo; + +--ANALYZE TABLE TARGET.result COMPUTE STATISTICS; + +create view if not exists TARGET.category as select * from SOURCE.category; +create view if not exists TARGET.concept as select * from SOURCE.concept; +create view if not exists TARGET.context as select * from SOURCE.context; +create view if not exists TARGET.country as select * from SOURCE.country; +create view if not exists TARGET.countrygdp as select * from SOURCE.countrygdp; +create view if not exists TARGET.creation_date as select * from SOURCE.creation_date; +create view if not exists TARGET.funder as select * from SOURCE.funder; +create view if not exists TARGET.fundref as select * from SOURCE.fundref; +create view if not exists TARGET.rndexpenditure as select * from SOURCE.rndexpediture; +create view if not exists TARGET.rndgdpexpenditure as select * from SOURCE.rndgdpexpenditure; +create view if not exists TARGET.doctoratestudents as select * from SOURCE.doctoratestudents; +create view if not exists TARGET.totalresearchers as select * from SOURCE.totalresearchers; +create view if not exists TARGET.totalresearchersft as select * from SOURCE.totalresearchersft; +create view if not exists TARGET.hrrst as select * from SOURCE.hrrst; +--create view if not exists TARGET.graduatedoctorates as select * from SOURCE.graduatedoctorates; + +create table TARGET.result_citations stored as parquet as select * from SOURCE.result_citations orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_citations COMPUTE STATISTICS; + +create table TARGET.result_references_oc stored as parquet as select * from SOURCE.result_references_oc orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_references_oc COMPUTE STATISTICS; + +create table TARGET.result_citations_oc stored as parquet as select * from SOURCE.result_citations_oc orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_citations_oc COMPUTE STATISTICS; + +create table TARGET.result_classifications stored as parquet as select * from SOURCE.result_classifications orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_classifications COMPUTE STATISTICS; + +create table TARGET.result_apc stored as parquet as select * from SOURCE.result_apc orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_apc COMPUTE STATISTICS; + +create table TARGET.result_concepts stored as parquet as select * from SOURCE.result_concepts orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_concepts COMPUTE STATISTICS; + +create table TARGET.result_datasources stored as parquet as select * from SOURCE.result_datasources orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_datasources COMPUTE STATISTICS; + +create table TARGET.result_fundercount stored as parquet as select * from SOURCE.result_fundercount orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_fundercount COMPUTE STATISTICS; + +create table TARGET.result_gold stored as parquet as select * from SOURCE.result_gold orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_gold COMPUTE STATISTICS; + +create table TARGET.result_greenoa stored as parquet as select * from SOURCE.result_greenoa orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_greenoa COMPUTE STATISTICS; + +create table TARGET.result_languages stored as parquet as select * from SOURCE.result_languages orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_languages COMPUTE STATISTICS; + +create table TARGET.result_licenses stored as parquet as select * from SOURCE.result_licenses orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_licenses COMPUTE STATISTICS; + +create table TARGET.licenses_normalized STORED AS PARQUET as select * from SOURCE.licenses_normalized; +--ANALYZE TABLE TARGET.licenses_normalized COMPUTE STATISTICS; + +create table TARGET.result_oids stored as parquet as select * from SOURCE.result_oids orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_oids COMPUTE STATISTICS; + +create table TARGET.result_organization stored as parquet as select * from SOURCE.result_organization orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_organization COMPUTE STATISTICS; + +create table TARGET.result_peerreviewed stored as parquet as select * from SOURCE.result_peerreviewed orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_peerreviewed COMPUTE STATISTICS; + +create table TARGET.result_pids stored as parquet as select * from SOURCE.result_pids orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_pids COMPUTE STATISTICS; + +create table TARGET.result_projectcount stored as parquet as select * from SOURCE.result_projectcount orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_projectcount COMPUTE STATISTICS; + +create table TARGET.result_projects stored as parquet as select * from SOURCE.result_projects orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_projects COMPUTE STATISTICS; + +create table TARGET.result_refereed stored as parquet as select * from SOURCE.result_refereed orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_refereed COMPUTE STATISTICS; + +create table TARGET.result_sources stored as parquet as select * from SOURCE.result_sources orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_sources COMPUTE STATISTICS; + +create table TARGET.result_topics stored as parquet as select * from SOURCE.result_topics orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_topics COMPUTE STATISTICS; + +create table TARGET.result_fos stored as parquet as select * from SOURCE.result_fos orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_fos COMPUTE STATISTICS; + +create table TARGET.result_accessroute stored as parquet as select * from SOURCE.result_accessroute orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_accessroute COMPUTE STATISTICS; + +create table TARGET.result_instance stored as parquet as select * from SOURCE.result_instance orig where exists (select 1 from TARGET.result r where r.id=orig.id); +create table TARGET.result_orcid stored as parquet as select * from SOURCE.result_orcid orig where exists (select 1 from TARGET.result r where r.id=orig.id); + + +create view TARGET.foo1 as select * from SOURCE.result_result rr where rr.source in (select id from TARGET.result); +create view TARGET.foo2 as select * from SOURCE.result_result rr where rr.target in (select id from TARGET.result); +create table TARGET.result_result STORED AS PARQUET as select distinct * from (select * from TARGET.foo1 union all select * from TARGET.foo2) foufou; +drop view TARGET.foo1; +drop view TARGET.foo2; +--ANALYZE TABLE TARGET.result_result COMPUTE STATISTICS; + +-- datasources +create view if not exists TARGET.datasource as select * from SOURCE.datasource; +create view if not exists TARGET.datasource_oids as select * from SOURCE.datasource_oids; +create view if not exists TARGET.datasource_organizations as select * from SOURCE.datasource_organizations; +create view if not exists TARGET.datasource_sources as select * from SOURCE.datasource_sources; + +create table TARGET.datasource_results stored as parquet as select id as result, datasource as id from TARGET.result_datasources; +--ANALYZE TABLE TARGET.datasource_results COMPUTE STATISTICS; + +-- organizations +create view if not exists TARGET.organization as select * from SOURCE.organization; +create view if not exists TARGET.organization_datasources as select * from SOURCE.organization_datasources; +create view if not exists TARGET.organization_pids as select * from SOURCE.organization_pids; +create view if not exists TARGET.organization_projects as select * from SOURCE.organization_projects; +create view if not exists TARGET.organization_sources as select * from SOURCE.organization_sources; + +-- projects +create view if not exists TARGET.project as select * from SOURCE.project; +create view if not exists TARGET.project_oids as select * from SOURCE.project_oids; +create view if not exists TARGET.project_organizations as select * from SOURCE.project_organizations; +create view if not exists TARGET.project_resultcount as select * from SOURCE.project_resultcount; +create view if not exists TARGET.project_classification as select * from SOURCE.project_classification; +create view if not exists TARGET.project_organization_contribution as select * from SOURCE.project_organization_contribution; + +create table TARGET.project_results stored as parquet as select id as result, project as id from TARGET.result_projects; +--ANALYZE TABLE TARGET.project_results COMPUTE STATISTICS; + +-- indicators +-- Sprint 1 ---- +create table TARGET.indi_pub_green_oa stored as parquet as select * from SOURCE.indi_pub_green_oa orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_green_oa COMPUTE STATISTICS; +create table TARGET.indi_pub_grey_lit stored as parquet as select * from SOURCE.indi_pub_grey_lit orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_grey_lit COMPUTE STATISTICS; +create table TARGET.indi_pub_doi_from_crossref stored as parquet as select * from SOURCE.indi_pub_doi_from_crossref orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_doi_from_crossref COMPUTE STATISTICS; +-- Sprint 2 ---- +create table TARGET.indi_result_has_cc_licence stored as parquet as select * from SOURCE.indi_result_has_cc_licence orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_result_has_cc_licence COMPUTE STATISTICS; +create table TARGET.indi_result_has_cc_licence_url stored as parquet as select * from SOURCE.indi_result_has_cc_licence_url orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_result_has_cc_licence_url COMPUTE STATISTICS; +create table TARGET.indi_pub_has_abstract stored as parquet as select * from SOURCE.indi_pub_has_abstract orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_has_abstract COMPUTE STATISTICS; +create table TARGET.indi_result_with_orcid stored as parquet as select * from SOURCE.indi_result_with_orcid orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_result_with_orcid COMPUTE STATISTICS; +---- Sprint 3 ---- +create table TARGET.indi_funded_result_with_fundref stored as parquet as select * from SOURCE.indi_funded_result_with_fundref orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_funded_result_with_fundref COMPUTE STATISTICS; +create view TARGET.indi_result_org_collab as select * from SOURCE.indi_result_org_collab; +create view TARGET.indi_result_org_country_collab as select * from SOURCE.indi_result_org_country_collab; +create view TARGET.indi_project_collab_org as select * from SOURCE.indi_project_collab_org; +create view TARGET.indi_project_collab_org_country as select * from SOURCE.indi_project_collab_org_country; +create view TARGET.indi_funder_country_collab as select * from SOURCE.indi_funder_country_collab; +create view TARGET.indi_result_country_collab as select * from SOURCE.indi_result_country_collab; +---- Sprint 4 ---- +create table TARGET.indi_pub_diamond stored as parquet as select * from SOURCE.indi_pub_diamond orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_diamond COMPUTE STATISTICS; +create table TARGET.indi_pub_in_transformative stored as parquet as select * from SOURCE.indi_pub_in_transformative orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_in_transformative COMPUTE STATISTICS; +create table TARGET.indi_pub_closed_other_open stored as parquet as select * from SOURCE.indi_pub_closed_other_open orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_closed_other_open COMPUTE STATISTICS; +---- Sprint 5 ---- +create table TARGET.indi_result_no_of_copies stored as parquet as select * from SOURCE.indi_result_no_of_copies orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_result_no_of_copies COMPUTE STATISTICS; +---- Sprint 6 ---- +create table TARGET.indi_pub_hybrid_oa_with_cc stored as parquet as select * from SOURCE.indi_pub_hybrid_oa_with_cc orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_hybrid_oa_with_cc COMPUTE STATISTICS; +create table TARGET.indi_pub_bronze_oa stored as parquet as select * from SOURCE.indi_pub_bronze_oa orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_bronze_oa COMPUTE STATISTICS; +create table TARGET.indi_pub_downloads stored as parquet as select * from SOURCE.indi_pub_downloads orig where exists (select 1 from TARGET.result r where r.id=orig.result_id); +--ANALYZE TABLE TARGET.indi_pub_downloads COMPUTE STATISTICS; +create table TARGET.indi_pub_downloads_datasource stored as parquet as select * from SOURCE.indi_pub_downloads_datasource orig where exists (select 1 from TARGET.result r where r.id=orig.result_id); +--ANALYZE TABLE TARGET.indi_pub_downloads_datasource COMPUTE STATISTICS; +create table TARGET.indi_pub_downloads_year stored as parquet as select * from SOURCE.indi_pub_downloads_year orig where exists (select 1 from TARGET.result r where r.id=orig.result_id); +--ANALYZE TABLE TARGET.indi_pub_downloads_year COMPUTE STATISTICS; +create table TARGET.indi_pub_downloads_datasource_year stored as parquet as select * from SOURCE.indi_pub_downloads_datasource_year orig where exists (select 1 from TARGET.result r where r.id=orig.result_id); +--ANALYZE TABLE TARGET.indi_pub_downloads_datasource_year COMPUTE STATISTICS; +---- Sprint 7 ---- +create table TARGET.indi_pub_gold_oa stored as parquet as select * from SOURCE.indi_pub_gold_oa orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_gold_oa COMPUTE STATISTICS; +create table TARGET.indi_pub_hybrid stored as parquet as select * from SOURCE.indi_pub_hybrid orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_hybrid COMPUTE STATISTICS; +create view TARGET.indi_org_fairness as select * from SOURCE.indi_org_fairness; +create view TARGET.indi_org_fairness_pub_pr as select * from SOURCE.indi_org_fairness_pub_pr; +create view TARGET.indi_org_fairness_pub_year as select * from SOURCE.indi_org_fairness_pub_year; +create view TARGET.indi_org_fairness_pub as select * from SOURCE.indi_org_fairness_pub; +create view TARGET.indi_org_fairness_year as select * from SOURCE.indi_org_fairness_year; +create view TARGET.indi_org_findable_year as select * from SOURCE.indi_org_findable_year; +create view TARGET.indi_org_findable as select * from SOURCE.indi_org_findable; +create view TARGET.indi_org_openess as select * from SOURCE.indi_org_openess; +create view TARGET.indi_org_openess_year as select * from SOURCE.indi_org_openess_year; +create table TARGET.indi_pub_has_preprint stored as parquet as select * from SOURCE.indi_pub_has_preprint orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_has_preprint COMPUTE STATISTICS; +create table TARGET.indi_pub_in_subscribed stored as parquet as select * from SOURCE.indi_pub_in_subscribed orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_in_subscribed COMPUTE STATISTICS; +create table TARGET.indi_result_with_pid stored as parquet as select * from SOURCE.indi_result_with_pid orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_result_with_pid COMPUTE STATISTICS; +create table TARGET.indi_impact_measures stored as parquet as select * from SOURCE.indi_impact_measures orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_impact_measures COMPUTE STATISTICS; +create table TARGET.indi_pub_interdisciplinarity stored as parquet as select * from SOURCE.indi_pub_interdisciplinarity orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.indi_pub_interdisciplinarity COMPUTE STATISTICS; +create table TARGET.result_apc_affiliations stored as parquet as select * from SOURCE.result_apc_affiliations orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--ANALYZE TABLE TARGET.result_apc_affiliations COMPUTE STATISTICS; +--create table TARGET.indi_is_project_result_after stored as parquet as select * from SOURCE.indi_is_project_result_after orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--create table TARGET.indi_is_funder_plan_s stored as parquet as select * from SOURCE.indi_is_funder_plan_s orig where exists (select 1 from TARGET.result r where r.id=orig.id); +--create view TARGET.indi_funder_fairness as select * from SOURCE.indi_funder_fairness; +--create view TARGET.indi_funder_openess as select * from SOURCE.indi_funder_openess; +--create view TARGET.indi_funder_findable as select * from SOURCE.indi_funder_findable; +--create view TARGET.indi_ris_fairness as select * from SOURCE.indi_ris_fairness; +--create view TARGET.indi_ris_openess as select * from SOURCE.indi_ris_openess; +--create view TARGET.indi_ris_findable as select * from SOURCE.indi_ris_findable; \ No newline at end of file diff --git a/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/scripts/updateMonitorDB_institutions.sql b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/scripts/updateMonitorDB_institutions.sql new file mode 100644 index 000000000..5ab8c88b5 --- /dev/null +++ b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/scripts/updateMonitorDB_institutions.sql @@ -0,0 +1,67 @@ +drop database if exists TARGET cascade; +create database if not exists TARGET; + +create table TARGET.result stored as parquet as + select distinct * from ( + select * from SOURCE.result r where exists (select 1 from SOURCE.result_organization ro where ro.id=r.id and ro.organization in ( + 'openorgs____::b84450f9864182c67b8611b5593f4250', --"Athena Research and Innovation Center In Information Communication & Knowledge Technologies', --ARC" + 'openorgs____::d41cf6bd4ab1b1362a44397e0b95c975', --National Research Council + 'openorgs____::d2a09b9d5eabb10c95f9470e172d05d2', --??? Not exists ?? + 'openorgs____::d169c7407dd417152596908d48c11460', --Masaryk University + 'openorgs____::1ec924b1759bb16d0a02f2dad8689b21', --University of Belgrade + 'openorgs____::0ae431b820e4c33db8967fbb2b919150', --University of Helsinki + 'openorgs____::759d59f05d77188faee99b7493b46805', --University of Minho + 'openorgs____::cad284878801b9465fa51a95b1d779db', --Universidad Politécnica de Madrid + 'openorgs____::eadc8da90a546e98c03f896661a2e4d4', --University of Göttingen + 'openorgs____::c0286313e36479eff8676dba9b724b40', --National and Kapodistrian University of Athens + -- 'openorgs____::c80a8243a5e5c620d7931c88d93bf17a', --Université Paris Diderot + 'openorgs____::c08634f0a6b0081c3dc6e6c93a4314f3', --Bielefeld University + 'openorgs____::6fc85e4a8f7ecaf4b0c738d010e967ea', --University of Southern Denmark + 'openorgs____::3d6122f87f9a97a99d8f6e3d73313720', --Humboldt-Universität zu Berlin + 'openorgs____::16720ada63d0fa8ca41601feae7d1aa5', --TU Darmstadt + 'openorgs____::ccc0a066b56d2cfaf90c2ae369df16f5', --KU Leuven + 'openorgs____::4c6f119632adf789746f0a057ed73e90', --University of the Western Cape + 'openorgs____::ec3665affa01aeafa28b7852c4176dbd', --Rudjer Boskovic Institute + 'openorgs____::5f31346d444a7f06a28c880fb170b0f6', --Ghent University + 'openorgs____::2dbe47117fd5409f9c61620813456632', --University of Luxembourg + 'openorgs____::6445d7758d3a40c4d997953b6632a368', --National Institute of Informatics (NII) + 'openorgs____::b77c01aa15de3675da34277d48de2ec1', -- Valencia Catholic University Saint Vincent Martyr + 'openorgs____::7fe2f66cdc43983c6b24816bfe9cf6a0', -- Unviersity of Warsaw + 'openorgs____::15e7921fc50d9aa1229a82a84429419e', -- University Of Thessaly + 'openorgs____::11f7919dadc8f8a7251af54bba60c956', -- Technical University of Crete + 'openorgs____::84f0c5f5dbb6daf42748485924efde4b', -- University of Piraeus + 'openorgs____::4ac562f0376fce3539504567649cb373', -- University of Patras + 'openorgs____::3e8d1f8c3f6cd7f418b09f1f58b4873b', -- Aristotle University of Thessaloniki + 'openorgs____::3fcef6e1c469c10f2a84b281372c9814', -- World Bank + 'openorgs____::1698a2eb1885ef8adb5a4a969e745ad3', -- École des Ponts ParisTech + 'openorgs____::e15adb13c4dadd49de4d35c39b5da93a', -- Nanyang Technological University + 'openorgs____::4b34103bde246228fcd837f5f1bf4212', -- Autonomous University of Barcelona + 'openorgs____::72ec75fcfc4e0df1a76dc4c49007fceb', -- McMaster University + 'openorgs____::51c7fc556e46381734a25a6fbc3fd398', -- University of Modena and Reggio Emilia + 'openorgs____::235d7f9ad18ecd7e6dc62ea4990cb9db', -- Bilkent University + 'openorgs____::31f2fa9e05b49d4cf40a19c3fed8eb06', -- Saints Cyril and Methodius University of Skopje + 'openorgs____::db7686f30f22cbe73a4fde872ce812a6', -- University of Milan + 'openorgs____::b8b8ca674452579f3f593d9f5e557483', -- University College Cork + 'openorgs____::38d7097854736583dde879d12dacafca', -- Brown University + 'openorgs____::57784c9e047e826fefdb1ef816120d92', --Arts et Métiers ParisTech + 'openorgs____::2530baca8a15936ba2e3297f2bce2e7e', -- University of Cape Town + 'openorgs____::d11f981828c485cd23d93f7f24f24db1', -- Technological University Dublin + 'openorgs____::5e6bf8962665cdd040341171e5c631d8', -- Delft University of Technology + 'openorgs____::846cb428d3f52a445f7275561a7beb5d', -- University of Manitoba + 'openorgs____::eb391317ed0dc684aa81ac16265de041', -- Universitat Rovira i Virgili + 'openorgs____::66aa9fc2fceb271423dfabcc38752dc0', -- Lund University + 'openorgs____::3cff625a4370d51e08624cc586138b2f', -- IMT Atlantique + 'openorgs____::c0b262bd6eab819e4c994914f9c010e2', -- National Institute of Geophysics and Volcanology + 'openorgs____::1624ff7c01bb641b91f4518539a0c28a', -- Vrije Universiteit Amsterdam + 'openorgs____::4d4051b56708688235252f1d8fddb8c1', --Iscte - Instituto Universitário de Lisboa + 'openorgs____::ab4ac74c35fa5dada770cf08e5110fab', -- Universidade Católica Portuguesa + 'openorgs____::4d4051b56708688235252f1d8fddb8c1', -- Iscte - Instituto Universitário de Lisboa + 'openorgs____::5d55fb216b14691cf68218daf5d78cd9', -- Munster Technological University + 'openorgs____::0fccc7640f0cb44d5cd1b06b312a06b9', -- Cardiff University + 'openorgs____::8839b55dae0c84d56fd533f52d5d483a', -- Leibniz Institute of Ecological Urban and Regional Development + 'openorgs____::526468206bca24c1c90da6a312295cf4', -- Cyprus University of Technology + 'openorgs____::b5ca9d4340e26454e367e2908ef3872f', -- Alma Mater Studiorum University of Bologna + 'openorgs____::a6340e6ecf60f6bba163659df985b0f2' -- TU Dresden + ))) foo; + +--ANALYZE TABLE TARGET.result COMPUTE STATISTICS; \ No newline at end of file diff --git a/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/workflow.xml b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/workflow.xml new file mode 100644 index 000000000..0542c0c87 --- /dev/null +++ b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/workflow.xml @@ -0,0 +1,111 @@ + + + + stats_db_name + the target stats database name + + + monitor_db_name + the target monitor db name + + + monitor_db_shadow_name + the name of the shadow monitor db + + + hive_metastore_uris + hive server metastore URIs + + + hive_jdbc_url + hive server jdbc url + + + hive_timeout + the time period, in seconds, after which Hive fails a transaction if a Hive client has not sent a hearbeat. The default value is 300 seconds. + + + hadoop_user_name + user name of the wf owner + + + + + ${jobTracker} + ${nameNode} + + + hive.metastore.uris + ${hive_metastore_uris} + + + hive.txn.timeout + ${hive_timeout} + + + mapred.job.queue.name + analytics + + + + + + + + ${wf:conf('resumeFrom') eq 'Step1-updateMonitorDB'} + ${wf:conf('resumeFrom') eq 'Step2-copyDataToImpalaCluster'} + ${wf:conf('resumeFrom') eq 'Step3-finalizeImpalaCluster'} + + + + + + Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}] + + + + + ${jobTracker} + ${nameNode} + monitor.sh + ${stats_db_name} + ${monitor_db_name} + ${monitor_db_shadow_name} + ${wf:appPath()}/scripts/updateMonitorDB_institutions.sql + ${wf:appPath()}/scripts/updateMonitorDB.sql + ${wf:appPath()}/scripts/updateMonitorDBAll.sql + monitor.sh + + + + + + + + ${jobTracker} + ${nameNode} + copyDataToImpalaCluster.sh + ${monitor_db_name} + ${hadoop_user_name} + copyDataToImpalaCluster.sh + + + + + + + + ${jobTracker} + ${nameNode} + finalizeImpalaCluster.sh + ${monitor_db_name} + ${monitor_db_prod_name} + ${monitor_db_shadow_name} + finalizeImpalaCluster.sh + + + + + + + -- 2.17.1 From 9521625a07dfbe584c0b5a57a3d3dd76affca636 Mon Sep 17 00:00:00 2001 From: Antonis Lempesis Date: Thu, 21 Mar 2024 11:45:08 +0200 Subject: [PATCH 18/22] code cleanup --- .../scripts/step16-createIndicatorsTables.sql | 63 +++---------------- 1 file changed, 7 insertions(+), 56 deletions(-) diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql index 8402cb364..d2cdf43b6 100755 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql @@ -58,7 +58,7 @@ select distinct r.id, (case when lic='' or lic is null then 0 else 1 end) as has from ${stats_db_name}.result r left outer join (select r.id, license.type as lic from ${stats_db_name}.result r join ${stats_db_name}.result_licenses as license on license.id = r.id - where lower(license.type) LIKE '%creativecommons.org%' OR lower(license.type) LIKE '%cc-%') tmp + where lower(license.type) LIKE '%creativecommons.org%' OR lower(license.type) LIKE '%cc %') tmp on r.id= tmp.id; /*EOS*/ drop table if exists ${stats_db_name}.indi_result_has_cc_licence_url purge; /*EOS*/ @@ -856,19 +856,6 @@ drop view pub_fos_totals; /*EOS*/ drop table if exists ${stats_db_name}.indi_pub_bronze_oa purge; /*EOS*/ ---create table if not exists ${stats_db_name}.indi_pub_bronze_oa stored as parquet as ---select distinct p.id, coalesce(is_bronze_oa,0) as is_bronze_oa ---from ${stats_db_name}.publication p ---left outer join ---(select p.id, 1 as is_bronze_oa from ${stats_db_name}.publication p ---join ${stats_db_name}.indi_result_has_cc_licence cc on cc.id=p.id ---join ${stats_db_name}.indi_pub_gold_oa ga on ga.id=p.id ---join ${stats_db_name}.result_instance ri on ri.id=p.id ---join ${stats_db_name}.datasource d on d.id=ri.hostedby ---where cc.has_cc_license=0 and ga.is_gold=0 ---and (d.type='Journal' or d.type='Journal Aggregator/Publisher') ---and ri.accessright='Open Access') tmp on tmp.id=p.id; - create table ${stats_db_name}.indi_pub_bronze_oa stored as parquet as select distinct pd.id,coalesce(is_bronze_oa,0) is_bronze_oa from ${stats_db_name}.publication pd left outer join (select pd.id, 1 as is_bronze_oa from ${stats_db_name}.publication pd @@ -1239,47 +1226,11 @@ where r.bestlicence='Open Access') tmp on r.id=tmp.id; /*EOS*/ drop table if exists ${stats_db_name}.indi_result_under_transformative purge; /*EOS*/ create table ${stats_db_name}.indi_result_under_transformative stored as parquet as -with transformative_dois as -(select distinct doi from stats_ext.transformative_facts) +with transformative_dois as ( + select distinct doi from stats_ext.transformative_facts) select distinct r.id, coalesce(under_transformative,0) as under_transformative from ${stats_db_name}.result r -left outer join (select distinct rp.id, 1 as under_transformative -from ${stats_db_name}.result_pids rp join ${stats_db_name}.result r on r.id=rp.id -join transformative_dois td on td.doi=rp.pid) tmp on r.id=tmp.id; /*EOS*/ - -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*/ - -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 -select distinct r.id, coalesce(oa_with_license,0) as oa_with_license -from ${stats_db_name}.result r -left outer join (select distinct r.id, 1 as oa_with_license from ${stats_db_name}.result r -join ${stats_db_name}.result_licenses rl on rl.id=r.id where r.bestlicence='Open Access') tmp on r.id=tmp.id; /*EOS*/ - -drop table if exists ${stats_db_name}.indi_result_oa_without_license purge; /*EOS*/ -create table ${stats_db_name}.indi_result_oa_without_license stored as parquet as -with without_license as -(select distinct id from ${stats_db_name}.indi_result_oa_with_license -where oa_with_license=0) -select distinct r.id, coalesce(oa_without_license,0) as oa_without_license -from ${stats_db_name}.result r -left outer join (select distinct r.id, 1 as oa_without_license -from ${stats_db_name}.result r -join without_license wl on wl.id=r.id -where r.bestlicence='Open Access') tmp on r.id=tmp.id; /*EOS*/ - -drop table if exists ${stats_db_name}.indi_result_under_transformative purge; /*EOS*/ -create table ${stats_db_name}.indi_result_under_transformative stored as parquet as -with transformative_dois as -(select distinct doi from stats_ext.transformative_facts) -select distinct r.id, coalesce(under_transformative,0) as under_transformative -from ${stats_db_name}.result r -left outer join (select distinct rp.id, 1 as under_transformative -from ${stats_db_name}.result_pids rp join ${stats_db_name}.result r on r.id=rp.id -join transformative_dois td on td.doi=rp.pid) tmp on r.id=tmp.id; /*EOS*/ \ No newline at end of file +left outer join ( + select distinct rp.id, 1 as under_transformative + from ${stats_db_name}.result_pids rp join ${stats_db_name}.result r on r.id=rp.id + join transformative_dois td on td.doi=rp.pid) tmp on r.id=tmp.id; /*EOS*/ \ No newline at end of file -- 2.17.1 From 07f634a46d8ba71082799014c3dc9258b464fb65 Mon Sep 17 00:00:00 2001 From: antleb Date: Thu, 21 Mar 2024 12:44:30 +0200 Subject: [PATCH 19/22] code cleanup --- .../scripts/step16-createIndicatorsTables.sql | 227 ++++-------------- 1 file changed, 43 insertions(+), 184 deletions(-) diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql index 8402cb364..3758954d4 100755 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql @@ -58,7 +58,7 @@ select distinct r.id, (case when lic='' or lic is null then 0 else 1 end) as has from ${stats_db_name}.result r left outer join (select r.id, license.type as lic from ${stats_db_name}.result r join ${stats_db_name}.result_licenses as license on license.id = r.id - where lower(license.type) LIKE '%creativecommons.org%' OR lower(license.type) LIKE '%cc-%') tmp + where lower(license.type) LIKE '%creativecommons.org%' OR lower(license.type) LIKE '%cc %') tmp on r.id= tmp.id; /*EOS*/ drop table if exists ${stats_db_name}.indi_result_has_cc_licence_url purge; /*EOS*/ @@ -130,7 +130,7 @@ from tmp as o1 join tmp as o2 on o1.id=o2.id where o1.id=o2.id and o1.country!=o2.country group by o1.organization, o1.id, o1.name, o2.country; /*EOS*/ -drop table if exists tmp purge; /*EOS*/ +DROP VIEW if exists tmp; /*EOS*/ create TEMPORARY VIEW tmp AS select o.id organization, o.name, ro.project as project from ${stats_db_name}.organization o @@ -195,48 +195,40 @@ DROP VIEW if exists tmp; /*EOS*/ ---- Sprint 4 ---- drop table if exists ${stats_db_name}.indi_pub_diamond purge; /*EOS*/ ---create table if not exists ${stats_db_name}.indi_pub_diamond stored as parquet as ---select distinct pd.id, coalesce(in_diamond_journal, 0) as in_diamond_journal ---from ${stats_db_name}.publication_datasources pd --- left outer join ( --- select pd.id, 1 as in_diamond_journal from ${stats_db_name}.publication_datasources pd --- join ${stats_db_name}.datasource d on d.id=pd.datasource --- join STATS_EXT.plan_s_jn ps where (ps.issn_print=d.issn_printed and ps.issn_online=d.issn_online) --- and (ps.journal_is_in_doaj=true or ps.journal_is_oa=true) and ps.has_apc=false) tmp --- on pd.id=tmp.id; - create table if not exists ${stats_db_name}.indi_pub_diamond stored as parquet as select distinct pd.id, coalesce(in_diamond_journal, 0) as in_diamond_journal from ${stats_db_name}.publication_datasources pd -left outer join (select pd.id, 1 as in_diamond_journal from ${stats_db_name}.publication_datasources pd -join ${stats_db_name}.datasource d on d.id=pd.datasource -join STATS_EXT.plan_s_jn ps where (ps.issn_print=d.issn_printed and ps.issn_online=d.issn_online) -and (ps.journal_is_in_doaj=true or ps.journal_is_oa=true) and ps.has_apc=false) tmp -on pd.id=tmp.id; /*EOS*/ +left outer join ( + select pd.id, 1 as in_diamond_journal + from ${stats_db_name}.publication_datasources pd + join ${stats_db_name}.datasource d on d.id=pd.datasource + join STATS_EXT.plan_s_jn ps where (ps.issn_print=d.issn_printed and ps.issn_online=d.issn_online) + and (ps.journal_is_in_doaj=true or ps.journal_is_oa=true) and ps.has_apc=false) tmp on pd.id=tmp.id; /*EOS*/ drop table if exists ${stats_db_name}.indi_pub_in_transformative purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_in_transformative stored as parquet as select distinct pd.id, coalesce(is_transformative, 0) as is_transformative from ${stats_db_name}.publication pd - left outer join ( - select pd.id, 1 as is_transformative from ${stats_db_name}.publication_datasources pd - join ${stats_db_name}.datasource d on d.id=pd.datasource - join STATS_EXT.plan_s_jn ps where (ps.issn_print=d.issn_printed and ps.issn_online=d.issn_online) - and ps.is_transformative_journal=true) tmp - on pd.id=tmp.id; /*EOS*/ +left outer join ( + select pd.id, 1 as is_transformative + from ${stats_db_name}.publication_datasources pd + join ${stats_db_name}.datasource d on d.id=pd.datasource + join STATS_EXT.plan_s_jn ps where (ps.issn_print=d.issn_printed and ps.issn_online=d.issn_online) + and ps.is_transformative_journal=true) tmp on pd.id=tmp.id; /*EOS*/ drop table if exists ${stats_db_name}.indi_pub_closed_other_open purge; /*EOS*/ create table if not exists ${stats_db_name}.indi_pub_closed_other_open stored as parquet as -select distinct ri.id, coalesce(pub_closed_other_open, 0) as pub_closed_other_open from ${stats_db_name}.result_instance ri - left outer join - (select ri.id, 1 as pub_closed_other_open from ${stats_db_name}.result_instance ri - join ${stats_db_name}.publication p on p.id=ri.id - join ${stats_db_name}.datasource d on ri.hostedby=d.id - where d.type like '%Journal%' and ri.accessright='Closed Access' and - (p.bestlicence='Open Access' or p.bestlicence='Open Source')) tmp - on tmp.id=ri.id; /*EOS*/ +select distinct ri.id, coalesce(pub_closed_other_open, 0) as pub_closed_other_open +from ${stats_db_name}.result_instance ri +left outer join ( + select ri.id, 1 as pub_closed_other_open + from ${stats_db_name}.result_instance ri + join ${stats_db_name}.publication p on p.id=ri.id + join ${stats_db_name}.datasource d on ri.hostedby=d.id + where d.type like '%Journal%' and ri.accessright='Closed Access' and + (p.bestlicence='Open Access' or p.bestlicence='Open Source')) tmp on tmp.id=ri.id; /*EOS*/ ---- Sprint 5 ---- drop table if exists ${stats_db_name}.indi_result_no_of_copies purge; /*EOS*/ @@ -284,77 +276,6 @@ GROUP BY result_id, repository_id, substring(us.`date`, 1,4); /*EOS*/ ---- Sprint 7 ---- drop table if exists ${stats_db_name}.indi_pub_gold_oa purge; /*EOS*/ ---create table if not exists ${stats_db_name}.indi_pub_gold_oa stored as parquet as --- WITH gold_oa AS ( SELECT --- issn_l, --- journal_is_in_doaj, --- journal_is_oa, --- issn_1 as issn --- FROM --- STATS_EXT.oa_journals --- WHERE --- issn_1 != "" --- UNION --- ALL SELECT --- issn_l, --- journal_is_in_doaj, --- journal_is_oa, --- issn_2 as issn --- FROM --- STATS_EXT.oa_journals --- WHERE --- issn_2 != "" ), issn AS ( SELECT --- * --- FROM ---( SELECT --- id, --- issn_printed as issn --- FROM --- ${stats_db_name}.datasource --- WHERE --- issn_printed IS NOT NULL --- UNION ALL --- SELECT --- id, --- issn_online as issn --- FROM --- ${stats_db_name}.datasource --- WHERE --- issn_online IS NOT NULL or id like '%doajarticles%') as issn --- WHERE --- LENGTH(issn) > 7) ---SELECT --- DISTINCT pd.id, coalesce(is_gold, 0) as is_gold ---FROM --- ${stats_db_name}.publication_datasources pd --- left outer join( --- select pd.id, 1 as is_gold FROM ${stats_db_name}.publication_datasources pd --- JOIN issn on issn.id=pd.datasource --- JOIN gold_oa on issn.issn = gold_oa.issn) tmp --- on pd.id=tmp.id; - ---create table if not exists ${stats_db_name}.indi_pub_gold_oa stored as parquet as ---with gold_oa as ( ---SELECT issn,issn_l from stats_ext.issn_gold_oa_dataset_v5), ---issn AS (SELECT * FROM ---(SELECT id,issn_printed as issn FROM ${stats_db_name}.datasource ---WHERE issn_printed IS NOT NULL ---UNION ALL ---SELECT id, issn_online as issn FROM ${stats_db_name}.datasource ---WHERE issn_online IS NOT NULL or id like '%doajarticles%') as issn ---WHERE LENGTH(issn) > 7), ---alljournals AS(select issn, issn_l from stats_ext.alljournals ---where journal_is_in_doaj=true or journal_is_oa=true) ---SELECT DISTINCT pd.id, coalesce(is_gold, 0) as is_gold ---FROM ${stats_db_name}.publication_datasources pd ---left outer join ( ---select pd.id, 1 as is_gold FROM ${stats_db_name}.publication_datasources pd ---JOIN issn on issn.id=pd.datasource ---JOIN gold_oa on issn.issn = gold_oa.issn ---join alljournals on issn.issn=alljournals.issn ---left outer join ${stats_db_name}.result_instance ri on ri.id=pd.id ---and ri.accessright!='Closed Access' and ri.accessright_uw='gold') tmp ---on pd.id=tmp.id; create table if not exists ${stats_db_name}.indi_pub_gold_oa stored as parquet as with gold_oa as ( select distinct issn from ( @@ -407,89 +328,25 @@ create table if not exists ${stats_db_name}.indi_pub_hybrid_oa_with_cc stored as WHERE LENGTH(issn) > 7) SELECT DISTINCT pd.id, coalesce(is_hybrid_oa, 0) as is_hybrid_oa FROM ${stats_db_name}.publication_datasources pd - LEFT OUTER JOIN ( +LEFT OUTER JOIN ( SELECT pd.id, 1 as is_hybrid_oa from ${stats_db_name}.publication_datasources pd - JOIN ${stats_db_name}.datasource d on d.id=pd.datasource - JOIN issn on issn.id=pd.datasource - JOIN hybrid_oa ON issn.issn = hybrid_oa.issn - JOIN ${stats_db_name}.indi_result_has_cc_licence cc on pd.id=cc.id - JOIN ${stats_db_name}.indi_pub_gold_oa ga on pd.id=ga.id - where cc.has_cc_license=1 and ga.is_gold=0) tmp on pd.id=tmp.id; /*EOS*/ + JOIN ${stats_db_name}.datasource d on d.id=pd.datasource + JOIN issn on issn.id=pd.datasource + JOIN hybrid_oa ON issn.issn = hybrid_oa.issn + JOIN ${stats_db_name}.indi_result_has_cc_licence cc on pd.id=cc.id + JOIN ${stats_db_name}.indi_pub_gold_oa ga on pd.id=ga.id where cc.has_cc_license=1 and ga.is_gold=0) tmp on pd.id=tmp.id; /*EOS*/ drop table if exists ${stats_db_name}.indi_pub_hybrid purge; /*EOS*/ ---create table if not exists ${stats_db_name}.indi_pub_hybrid stored as parquet as --- WITH gold_oa AS ( SELECT --- issn_l, --- journal_is_in_doaj, --- journal_is_oa, --- issn_1 as issn, --- has_apc --- FROM --- STATS_EXT.oa_journals --- WHERE --- issn_1 != "" --- UNION --- ALL SELECT --- issn_l, --- journal_is_in_doaj, --- journal_is_oa, --- issn_2 as issn, --- has_apc --- FROM --- STATS_EXT.oa_journals --- WHERE --- issn_2 != "" ), issn AS ( SELECT --- * --- FROM ---( SELECT --- id, --- issn_printed as issn --- FROM --- ${stats_db_name}.datasource --- WHERE --- issn_printed IS NOT NULL --- UNION ALL --- SELECT --- id, --- issn_online as issn --- FROM --- ${stats_db_name}.datasource --- WHERE --- issn_online IS NOT NULL or id like '%doajarticles%') as issn --- WHERE --- LENGTH(issn) > 7) ---select distinct pd.id, coalesce(is_hybrid, 0) as is_hybrid ---from ${stats_db_name}.publication_datasources pd --- left outer join ( --- select pd.id, 1 as is_hybrid from ${stats_db_name}.publication_datasources pd --- join ${stats_db_name}.datasource d on d.id=pd.datasource --- join issn on issn.id=pd.datasource --- join gold_oa on issn.issn=gold_oa.issn --- where (gold_oa.journal_is_in_doaj=false or gold_oa.journal_is_oa=false))tmp --- on pd.id=tmp.id; - ---create table if not exists ${stats_db_name}.indi_pub_hybrid stored as parquet as ---select distinct pd.id,coalesce(is_hybrid,0) is_hybrid from ${stats_db_name}.publication_datasources pd ---left outer join (select pd.id, 1 as is_hybrid from ${stats_db_name}.publication_datasources pd ---join ${stats_db_name}.datasource d on pd.datasource=d.id ---join ${stats_db_name}.result_instance ri on ri.id=pd.id ---join ${stats_db_name}.indi_pub_gold_oa indi_gold on indi_gold.id=pd.id ---join ${stats_db_name}.result_accessroute ra on ra.id=pd.id ---where d.type like '%Journal%' and ri.accessright!='Closed Access' and (ri.accessright_uw!='gold' ---or indi_gold.is_gold=0) and (ra.accessroute='hybrid' or ri.license is not null)) tmp ---on pd.id=tmp.id; - create table if not exists ${stats_db_name}.indi_pub_hybrid stored as parquet as select distinct pd.id,coalesce(is_hybrid,0) is_hybrid from ${stats_db_name}.publication pd -left outer join (select pd.id, 1 as is_hybrid from ${stats_db_name}.publication pd -join ${stats_db_name}.result_instance ri on ri.id=pd.id -join ${stats_db_name}.indi_pub_gold_oa indi_gold on indi_gold.id=pd.id -join ${stats_db_name}.result_accessroute ra on ra.id=pd.id -join ${stats_db_name}.datasource d on d.id=ri.hostedby -where indi_gold.is_gold=0 and ((d.type like '%Journal%' and ri.accessright!='Closed Access' and ri.accessright!='Restricted' and ri.license is not null) or -ra.accessroute='hybrid'))tmp -on pd.id=tmp.id; /*EOS*/ +left outer join ( + select pd.id, 1 as is_hybrid from ${stats_db_name}.publication pd + join ${stats_db_name}.result_instance ri on ri.id=pd.id + join ${stats_db_name}.indi_pub_gold_oa indi_gold on indi_gold.id=pd.id + join ${stats_db_name}.result_accessroute ra on ra.id=pd.id + join ${stats_db_name}.datasource d on d.id=ri.hostedby + where indi_gold.is_gold=0 and ((d.type like '%Journal%' and ri.accessright!='Closed Access' and ri.accessright!='Restricted' and ri.license is not null) or ra.accessroute='hybrid')) tmp on pd.id=tmp.id; /*EOS*/ drop table if exists ${stats_db_name}.indi_org_fairness purge; /*EOS*/ @@ -534,6 +391,9 @@ select ar.organization, rf.no_result_fair/ar.no_allresults org_fairness from allresults ar join result_fair rf on rf.organization=ar.organization; /*EOS*/ +DROP VIEW result_fair; +DROP VIEW allresults; + CREATE TEMPORARY VIEW result_fair as select year, ro.organization organization, count(distinct ro.id) no_result_fair from ${stats_db_name}.result_organization ro join ${stats_db_name}.result p on p.id=ro.id @@ -1201,11 +1061,10 @@ create table ${stats_db_name}.indi_pub_green_with_license stored as parquet as select distinct p.id, coalesce(green_with_license, 0) as green_with_license from ${stats_db_name}.publication p left outer join ( -select distinct p.id, 1 as green_with_license from ${stats_db_name}.publication p -join ${stats_db_name}.result_instance ri on ri.id = p.id -join ${stats_db_name}.datasource on datasource.id = ri.hostedby -where ri.license is not null and datasource.type like '%Repository%' and datasource.name!='Other') tmp -on p.id= tmp.id; /*EOS*/ + select distinct p.id, 1 as green_with_license from ${stats_db_name}.publication p + join ${stats_db_name}.result_instance ri on ri.id = p.id + join ${stats_db_name}.datasource on datasource.id = ri.hostedby + where ri.license is not null and datasource.type like '%Repository%' and datasource.name!='Other') tmp on p.id= tmp.id; /*EOS*/ drop table if exists ${stats_db_name}.result_country purge; /*EOS*/ -- 2.17.1 From 4c40c96e30fc69917ab27e7a635c810b3d3a9b39 Mon Sep 17 00:00:00 2001 From: antleb Date: Fri, 22 Mar 2024 10:16:49 +0200 Subject: [PATCH 20/22] code cleanup --- .../scripts/step16-createIndicatorsTables.sql | 418 ++++++++---------- 1 file changed, 181 insertions(+), 237 deletions(-) diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql index becb85cf5..70cde6481 100755 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql @@ -1,311 +1,255 @@ -- Sprint 1 ---- drop table if exists ${stats_db_name}.indi_pub_green_oa purge; /*EOS*/ - ---create table if not exists ${stats_db_name}.indi_pub_green_oa stored as parquet as ---select distinct p.id, coalesce(green_oa, 0) as green_oa ---from ${stats_db_name}.publication p --- left outer join ( --- select p.id, 1 as green_oa --- from ${stats_db_name}.publication p --- join ${stats_db_name}.result_instance ri on ri.id = p.id --- join ${stats_db_name}.datasource on datasource.id = ri.hostedby --- where datasource.type like '%Repository%' --- and (ri.accessright = 'Open Access' --- or ri.accessright = 'Embargo' or ri.accessright = 'Open Source')) tmp --- on p.id= tmp.id; create table if not exists ${stats_db_name}.indi_pub_green_oa stored as parquet as select distinct p.id, coalesce(green_oa, 0) as green_oa from ${stats_db_name}.publication p - left outer join ( +left outer join ( select p.id, 1 as green_oa from ${stats_db_name}.publication p - join ${stats_db_name}.result_instance ri on ri.id = p.id - join ${stats_db_name}.datasource on datasource.id = ri.hostedby - where datasource.type like '%Repository%' - and (ri.accessright = 'Open Access' - or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and datasource.name!='Other') tmp - on p.id= tmp.id; /*EOS*/ + join ${stats_db_name}.result_instance ri on ri.id = p.id + join ${stats_db_name}.datasource on datasource.id = ri.hostedby + where datasource.type like '%Repository%' and (ri.accessright = 'Open Access' or ri.accessright = 'Embargo' or ri.accessright = 'Open Source') and datasource.name!='Other') tmp on p.id= tmp.id; /*EOS*/ drop table if exists ${stats_db_name}.indi_pub_grey_lit purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_pub_grey_lit stored as parquet as select distinct p.id, coalesce(grey_lit, 0) as grey_lit from ${stats_db_name}.publication p - left outer join ( +left outer join ( select p.id, 1 as grey_lit from ${stats_db_name}.publication p - join ${stats_db_name}.result_classifications rt on rt.id = p.id - where rt.type not in ('Article','Part of book or chapter of book','Book','Doctoral thesis','Master thesis','Data Paper', 'Thesis', 'Bachelor thesis', 'Conference object') and - not exists (select 1 from ${stats_db_name}.result_classifications rc where type ='Other literature type' - and rc.id=p.id)) tmp on p.id=tmp.id; /*EOS*/ + join ${stats_db_name}.result_classifications rt on rt.id = p.id + where rt.type not in ('Article','Part of book or chapter of book','Book','Doctoral thesis','Master thesis','Data Paper', 'Thesis', 'Bachelor thesis', 'Conference object') + and not exists (select 1 from ${stats_db_name}.result_classifications rc where type ='Other literature type' and rc.id=p.id)) tmp on p.id=tmp.id; /*EOS*/ drop table if exists ${stats_db_name}.indi_pub_doi_from_crossref purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_pub_doi_from_crossref stored as parquet as select distinct p.id, coalesce(doi_from_crossref, 0) as doi_from_crossref from ${stats_db_name}.publication p - left outer join - (select ri.id, 1 as doi_from_crossref from ${stats_db_name}.result_instance ri - join ${stats_db_name}.datasource d on d.id = ri.collectedfrom - where pidtype='Digital Object Identifier' and d.name ='Crossref') tmp - on tmp.id=p.id; /*EOS*/ +left outer join ( + select ri.id, 1 as doi_from_crossref from ${stats_db_name}.result_instance ri + join ${stats_db_name}.datasource d on d.id = ri.collectedfrom + where pidtype='Digital Object Identifier' and d.name ='Crossref') tmp on tmp.id=p.id; /*EOS*/ -- Sprint 2 ---- drop table if exists ${stats_db_name}.indi_result_has_cc_licence purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_result_has_cc_licence stored as parquet as select distinct r.id, (case when lic='' or lic is null then 0 else 1 end) as has_cc_license from ${stats_db_name}.result r -left outer join (select r.id, license.type as lic from ${stats_db_name}.result r - join ${stats_db_name}.result_licenses as license on license.id = r.id - where lower(license.type) LIKE '%creativecommons.org%' OR lower(license.type) LIKE '%cc %') tmp - on r.id= tmp.id; /*EOS*/ +left outer join ( + select r.id, license.type as lic from ${stats_db_name}.result r + join ${stats_db_name}.result_licenses as license on license.id = r.id + where lower(license.type) LIKE '%creativecommons.org%' OR lower(license.type) LIKE '%cc %') tmp on r.id= tmp.id; /*EOS*/ drop table if exists ${stats_db_name}.indi_result_has_cc_licence_url purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_result_has_cc_licence_url stored as parquet as select distinct r.id, case when lic_host='' or lic_host is null then 0 else 1 end as has_cc_license_url from ${stats_db_name}.result r - left outer join (select r.id, lower(parse_url(license.type, "HOST")) as lic_host - from ${stats_db_name}.result r - join ${stats_db_name}.result_licenses as license on license.id = r.id - WHERE lower(parse_url(license.type, "HOST")) = "creativecommons.org") tmp - on r.id= tmp.id; /*EOS*/ +left outer join ( + select r.id, lower(parse_url(license.type, "HOST")) as lic_host + from ${stats_db_name}.result r + join ${stats_db_name}.result_licenses as license on license.id = r.id + WHERE lower(parse_url(license.type, "HOST")) = "creativecommons.org") tmp on r.id= tmp.id; /*EOS*/ drop table if exists ${stats_db_name}.indi_pub_has_abstract purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_pub_has_abstract stored as parquet as select distinct publication.id, cast(coalesce(abstract, true) as int) has_abstract from ${stats_db_name}.publication; /*EOS*/ drop table if exists ${stats_db_name}.indi_result_with_orcid purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_result_with_orcid stored as parquet as select distinct r.id, coalesce(has_orcid, 0) as has_orcid from ${stats_db_name}.result r - left outer join (select id, 1 as has_orcid from ${stats_db_name}.result_orcid) tmp - on r.id= tmp.id; /*EOS*/ +left outer join ( + select id, 1 as has_orcid from ${stats_db_name}.result_orcid) tmp on r.id= tmp.id; /*EOS*/ + ---- Sprint 3 ---- - drop table if exists ${stats_db_name}.indi_funded_result_with_fundref purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_funded_result_with_fundref stored as parquet as select distinct r.result as id, coalesce(fundref, 0) as fundref from ${stats_db_name}.project_results r - left outer join (select distinct result, 1 as fundref from ${stats_db_name}.project_results - where provenance='Harvested') tmp - on r.result= tmp.result; /*EOS*/ - --- create table indi_result_org_collab stored as parquet as --- select o1.organization org1, o2.organization org2, count(distinct o1.id) as collaborations --- from result_organization as o1 --- join result_organization as o2 on o1.id=o2.id and o1.organization!=o2.organization --- group by o1.organization, o2.organization; --- --- compute stats indi_result_org_collab; --- -create TEMPORARY VIEW tmp AS SELECT ro.organization organization, ro.id, o.name from ${stats_db_name}.result_organization ro -join ${stats_db_name}.organization o on o.id=ro.organization where o.name is not null; /*EOS*/ +left outer join ( + select distinct result, 1 as fundref from ${stats_db_name}.project_results where provenance='Harvested') tmp on r.result= tmp.result; /*EOS*/ drop table if exists ${stats_db_name}.indi_result_org_collab purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_result_org_collab stored as parquet as -select o1.organization org1, o1.name org1name1, o2.organization org2, o2.name org2name2, count(o1.id) as collaborations -from tmp as o1 -join tmp as o2 where o1.id=o2.id and o1.organization!=o2.organization and o1.name!=o2.name -group by o1.organization, o2.organization, o1.name, o2.name; /*EOS*/ - -DROP VIEW if exists tmp; /*EOS*/ - -create TEMPORARY VIEW tmp AS -select distinct ro.organization organization, ro.id, o.name, o.country from ${stats_db_name}.result_organization ro -join ${stats_db_name}.organization o on o.id=ro.organization where country <> 'UNKNOWN' and o.name is not null; /*EOS*/ + WITH tmp AS ( + SELECT ro.organization organization, ro.id, o.name + from ${stats_db_name}.result_organization ro + join ${stats_db_name}.organization o on o.id=ro.organization where o.name is not null) + select o1.organization org1, o1.name org1name1, o2.organization org2, o2.name org2name2, count(o1.id) as collaborations + from tmp as o1 + join tmp as o2 where o1.id=o2.id and o1.organization!=o2.organization and o1.name!=o2.name + group by o1.organization, o2.organization, o1.name, o2.name; /*EOS*/ drop table if exists ${stats_db_name}.indi_result_org_country_collab purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_result_org_country_collab stored as parquet as -select o1.organization org1,o1.name org1name1, o2.country country2, count(o1.id) as collaborations -from tmp as o1 join tmp as o2 on o1.id=o2.id -where o1.id=o2.id and o1.country!=o2.country -group by o1.organization, o1.id, o1.name, o2.country; /*EOS*/ - -DROP VIEW if exists tmp; /*EOS*/ - -create TEMPORARY VIEW tmp AS -select o.id organization, o.name, ro.project as project from ${stats_db_name}.organization o - join ${stats_db_name}.organization_projects ro on o.id=ro.id where o.name is not null; /*EOS*/ + WITH tmp AS ( + select distinct ro.organization organization, ro.id, o.name, o.country + from ${stats_db_name}.result_organization ro + join ${stats_db_name}.organization o on o.id=ro.organization + where country <> 'UNKNOWN' and o.name is not null) + select o1.organization org1,o1.name org1name1, o2.country country2, count(o1.id) as collaborations + from tmp as o1 join tmp as o2 on o1.id=o2.id + where o1.id=o2.id and o1.country!=o2.country + group by o1.organization, o1.id, o1.name, o2.country; /*EOS*/ drop table if exists ${stats_db_name}.indi_project_collab_org purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_project_collab_org stored as parquet as -select o1.organization org1,o1.name orgname1, o2.organization org2, o2.name orgname2, count(distinct o1.project) as collaborations -from tmp as o1 - join tmp as o2 on o1.project=o2.project -where o1.organization<>o2.organization and o1.name<>o2.name -group by o1.name,o2.name, o1.organization, o2.organization; /*EOS*/ - -DROP VIEW if exists tmp; /*EOS*/ - -create TEMPORARY VIEW tmp AS -select o.id organization, o.name, o.country , ro.project as project from ${stats_db_name}.organization o - join ${stats_db_name}.organization_projects ro on o.id=ro.id - and o.country <> 'UNKNOWN' and o.name is not null; /*EOS*/ + WITH tmp AS ( + select o.id organization, o.name, ro.project as project + from ${stats_db_name}.organization o + join ${stats_db_name}.organization_projects ro on o.id=ro.id where o.name is not null) + select o1.organization org1,o1.name orgname1, o2.organization org2, o2.name orgname2, count(distinct o1.project) as collaborations + from tmp as o1 + join tmp as o2 on o1.project=o2.project + where o1.organization<>o2.organization and o1.name<>o2.name + group by o1.name,o2.name, o1.organization, o2.organization; /*EOS*/ drop table if exists ${stats_db_name}.indi_project_collab_org_country purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_project_collab_org_country stored as parquet as -select o1.organization org1,o1.name org1name, o2.country country2, count(distinct o1.project) as collaborations -from tmp as o1 - join tmp as o2 on o1.project=o2.project -where o1.organization<>o2.organization and o1.country<>o2.country -group by o1.organization, o2.country, o1.name; /*EOS*/ - -DROP VIEW if exists tmp; /*EOS*/ + WITH tmp AS ( + select o.id organization, o.name, o.country , ro.project as project + from ${stats_db_name}.organization o + join ${stats_db_name}.organization_projects ro on o.id=ro.id and o.country <> 'UNKNOWN' and o.name is not null) + select o1.organization org1,o1.name org1name, o2.country country2, count(distinct o1.project) as collaborations + from tmp as o1 + join tmp as o2 on o1.project=o2.project + where o1.organization<>o2.organization and o1.country<>o2.country + group by o1.organization, o2.country, o1.name; /*EOS*/ drop table if exists ${stats_db_name}.indi_funder_country_collab purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_funder_country_collab stored as parquet as with tmp as (select funder, project, country from ${stats_db_name}.organization_projects op join ${stats_db_name}.organization o on o.id=op.id join ${stats_db_name}.project p on p.id=op.project where country <> 'UNKNOWN') -select f1.funder, f1.country as country1, f2.country as country2, count(distinct f1.project) as collaborations -from tmp as f1 - join tmp as f2 on f1.project=f2.project -where f1.country<>f2.country -group by f1.funder, f2.country, f1.country; /*EOS*/ - -create TEMPORARY VIEW tmp AS -select distinct country, ro.id as result from ${stats_db_name}.organization o - join ${stats_db_name}.result_organization ro on o.id=ro.organization - where country <> 'UNKNOWN' and o.name is not null; /*EOS*/ + select f1.funder, f1.country as country1, f2.country as country2, count(distinct f1.project) as collaborations + from tmp as f1 + join tmp as f2 on f1.project=f2.project + where f1.country<>f2.country + group by f1.funder, f2.country, f1.country; /*EOS*/ drop table if exists ${stats_db_name}.indi_result_country_collab purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_result_country_collab stored as parquet as -select o1.country country1, o2.country country2, count(o1.result) as collaborations -from tmp as o1 - join tmp as o2 on o1.result=o2.result -where o1.country<>o2.country -group by o1.country, o2.country; /*EOS*/ + WITH tmp AS ( + select distinct country, ro.id as result from ${stats_db_name}.organization o + join ${stats_db_name}.result_organization ro on o.id=ro.organization + where country <> 'UNKNOWN' and o.name is not null) + select o1.country country1, o2.country country2, count(o1.result) as collaborations + from tmp as o1 + join tmp as o2 on o1.result=o2.result + where o1.country<>o2.country + group by o1.country, o2.country; /*EOS*/ -DROP VIEW if exists tmp; /*EOS*/ ---- Sprint 4 ---- drop table if exists ${stats_db_name}.indi_pub_diamond purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_pub_diamond stored as parquet as -select distinct pd.id, coalesce(in_diamond_journal, 0) as in_diamond_journal -from ${stats_db_name}.publication_datasources pd -left outer join ( - select pd.id, 1 as in_diamond_journal + select distinct pd.id, coalesce(in_diamond_journal, 0) as in_diamond_journal from ${stats_db_name}.publication_datasources pd - join ${stats_db_name}.datasource d on d.id=pd.datasource - join STATS_EXT.plan_s_jn ps where (ps.issn_print=d.issn_printed and ps.issn_online=d.issn_online) - and (ps.journal_is_in_doaj=true or ps.journal_is_oa=true) and ps.has_apc=false) tmp on pd.id=tmp.id; /*EOS*/ + left outer join ( + select pd.id, 1 as in_diamond_journal + from ${stats_db_name}.publication_datasources pd + join ${stats_db_name}.datasource d on d.id=pd.datasource + join STATS_EXT.plan_s_jn ps where (ps.issn_print=d.issn_printed and ps.issn_online=d.issn_online) + and (ps.journal_is_in_doaj=true or ps.journal_is_oa=true) and ps.has_apc=false) tmp on pd.id=tmp.id; /*EOS*/ drop table if exists ${stats_db_name}.indi_pub_in_transformative purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_pub_in_transformative stored as parquet as -select distinct pd.id, coalesce(is_transformative, 0) as is_transformative -from ${stats_db_name}.publication pd -left outer join ( - select pd.id, 1 as is_transformative - from ${stats_db_name}.publication_datasources pd - join ${stats_db_name}.datasource d on d.id=pd.datasource - join STATS_EXT.plan_s_jn ps where (ps.issn_print=d.issn_printed and ps.issn_online=d.issn_online) - and ps.is_transformative_journal=true) tmp on pd.id=tmp.id; /*EOS*/ + select distinct pd.id, coalesce(is_transformative, 0) as is_transformative + from ${stats_db_name}.publication pd + left outer join ( + select pd.id, 1 as is_transformative + from ${stats_db_name}.publication_datasources pd + join ${stats_db_name}.datasource d on d.id=pd.datasource + join STATS_EXT.plan_s_jn ps where (ps.issn_print=d.issn_printed and ps.issn_online=d.issn_online) + and ps.is_transformative_journal=true) tmp on pd.id=tmp.id; /*EOS*/ drop table if exists ${stats_db_name}.indi_pub_closed_other_open purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_pub_closed_other_open stored as parquet as -select distinct ri.id, coalesce(pub_closed_other_open, 0) as pub_closed_other_open -from ${stats_db_name}.result_instance ri -left outer join ( - select ri.id, 1 as pub_closed_other_open + select distinct ri.id, coalesce(pub_closed_other_open, 0) as pub_closed_other_open from ${stats_db_name}.result_instance ri - join ${stats_db_name}.publication p on p.id=ri.id - join ${stats_db_name}.datasource d on ri.hostedby=d.id - where d.type like '%Journal%' and ri.accessright='Closed Access' and - (p.bestlicence='Open Access' or p.bestlicence='Open Source')) tmp on tmp.id=ri.id; /*EOS*/ + left outer join ( + select ri.id, 1 as pub_closed_other_open + from ${stats_db_name}.result_instance ri + join ${stats_db_name}.publication p on p.id=ri.id + join ${stats_db_name}.datasource d on ri.hostedby=d.id + where d.type like '%Journal%' and ri.accessright='Closed Access' and + (p.bestlicence='Open Access' or p.bestlicence='Open Source')) tmp on tmp.id=ri.id; /*EOS*/ + ---- Sprint 5 ---- drop table if exists ${stats_db_name}.indi_result_no_of_copies purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_result_no_of_copies stored as parquet as -select id, count(id) as number_of_copies from ${stats_db_name}.result_instance group by id; /*EOS*/ + select id, count(id) as number_of_copies + from ${stats_db_name}.result_instance + group by id; /*EOS*/ ---- Sprint 6 ---- drop table if exists ${stats_db_name}.indi_pub_downloads purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_pub_downloads stored as parquet as -SELECT result_id, sum(downloads) no_downloads from openaire_prod_usage_stats.usage_stats - join ${stats_db_name}.publication on result_id=id -where downloads>0 -GROUP BY result_id -order by no_downloads desc; /*EOS*/ - ---ANALYZE TABLE ${stats_db_name}.indi_pub_downloads COMPUTE STATISTICS; + SELECT result_id, sum(downloads) no_downloads + from openaire_prod_usage_stats.usage_stats + join ${stats_db_name}.publication on result_id=id + where downloads>0 + GROUP BY result_id; /*EOS*/ drop table if exists ${stats_db_name}.indi_pub_downloads_datasource purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_pub_downloads_datasource stored as parquet as -SELECT result_id, repository_id, sum(downloads) no_downloads from openaire_prod_usage_stats.usage_stats - join ${stats_db_name}.publication on result_id=id -where downloads>0 -GROUP BY result_id, repository_id -order by result_id; /*EOS*/ + SELECT result_id, repository_id, sum(downloads) no_downloads + from openaire_prod_usage_stats.usage_stats + join ${stats_db_name}.publication on result_id=id + where downloads>0 + GROUP BY result_id, repository_id; /*EOS*/ drop table if exists ${stats_db_name}.indi_pub_downloads_year purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_pub_downloads_year stored as parquet as -SELECT result_id, cast(substring(us.`date`, 1,4) as int) as `year`, sum(downloads) no_downloads -from openaire_prod_usage_stats.usage_stats us -join ${stats_db_name}.publication on result_id=id where downloads>0 -GROUP BY result_id, substring(us.`date`, 1,4); /*EOS*/ + SELECT result_id, cast(substring(us.`date`, 1,4) as int) as `year`, sum(downloads) no_downloads + from openaire_prod_usage_stats.usage_stats us + join ${stats_db_name}.publication on result_id=id where downloads>0 + GROUP BY result_id, substring(us.`date`, 1,4); /*EOS*/ drop table if exists ${stats_db_name}.indi_pub_downloads_datasource_year purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_pub_downloads_datasource_year stored as parquet as -SELECT result_id, cast(substring(us.`date`, 1,4) as int) as `year`, repository_id, sum(downloads) no_downloads from openaire_prod_usage_stats.usage_stats us -join ${stats_db_name}.publication on result_id=id -where downloads>0 -GROUP BY result_id, repository_id, substring(us.`date`, 1,4); /*EOS*/ + SELECT result_id, cast(substring(us.`date`, 1,4) as int) as `year`, repository_id, sum(downloads) no_downloads + from openaire_prod_usage_stats.usage_stats us + join ${stats_db_name}.publication on result_id=id + where downloads>0 + GROUP BY result_id, repository_id, substring(us.`date`, 1,4); /*EOS*/ + ---- Sprint 7 ---- drop table if exists ${stats_db_name}.indi_pub_gold_oa purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_pub_gold_oa stored as parquet as -with gold_oa as ( -select distinct issn from ( - SELECT issn_l as issn from stats_ext.issn_gold_oa_dataset_v5 - UNION ALL - SELECT issn as issn from stats_ext.issn_gold_oa_dataset_v5 - UNION ALL - select issn from stats_ext.alljournals where journal_is_in_doaj=true or journal_is_oa=true - UNION ALL - select issn_l as issn from stats_ext.alljournals where journal_is_in_doaj=true or journal_is_oa=true) foo), -dd as ( -select distinct * from ( - select id, issn_printed as issn from ${stats_db_name}.datasource d where d.id like '%doajarticles%' - UNION ALL - select id, issn_online as issn from ${stats_db_name}.datasource d where d.id like '%doajarticles%' - UNION ALL - select id, issn_printed as issn from ${stats_db_name}.datasource d join gold_oa on gold_oa.issn=d.issn_printed - 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 DISTINCT pd.id, coalesce(is_gold, 0) as is_gold -FROM ${stats_db_name}.publication_datasources pd -left outer join ( - select pd.id, 1 as is_gold - FROM ${stats_db_name}.publication_datasources pd - 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*/ + with gold_oa as ( + select distinct issn from ( + SELECT issn_l as issn from stats_ext.issn_gold_oa_dataset_v5 + UNION ALL + SELECT issn as issn from stats_ext.issn_gold_oa_dataset_v5 + UNION ALL + select issn from stats_ext.alljournals where journal_is_in_doaj=true or journal_is_oa=true + UNION ALL + select issn_l as issn from stats_ext.alljournals where journal_is_in_doaj=true or journal_is_oa=true) foo), + dd as ( + select distinct * from ( + select id, issn_printed as issn from ${stats_db_name}.datasource d where d.id like '%doajarticles%' + UNION ALL + select id, issn_online as issn from ${stats_db_name}.datasource d where d.id like '%doajarticles%' + UNION ALL + select id, issn_printed as issn from ${stats_db_name}.datasource d join gold_oa on gold_oa.issn=d.issn_printed + 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 DISTINCT pd.id, coalesce(is_gold, 0) as is_gold + FROM ${stats_db_name}.publication_datasources pd + left outer join ( + select pd.id, 1 as is_gold + FROM ${stats_db_name}.publication_datasources pd + 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*/ drop table if exists ${stats_db_name}.indi_pub_hybrid_oa_with_cc purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_pub_hybrid_oa_with_cc stored as parquet as WITH hybrid_oa AS ( SELECT issn_l, journal_is_in_doaj, journal_is_oa, issn_print as issn @@ -316,28 +260,27 @@ create table if not exists ${stats_db_name}.indi_pub_hybrid_oa_with_cc stored as FROM STATS_EXT.plan_s_jn WHERE issn_online != "" and (journal_is_in_doaj = FALSE OR journal_is_oa = FALSE)), issn AS ( - SELECT * - FROM ( - SELECT id, issn_printed as issn - FROM ${stats_db_name}.datasource - WHERE issn_printed IS NOT NULL - UNION ALL - SELECT id,issn_online as issn - FROM ${stats_db_name}.datasource - WHERE issn_online IS NOT NULL ) as issn - WHERE LENGTH(issn) > 7) -SELECT DISTINCT pd.id, coalesce(is_hybrid_oa, 0) as is_hybrid_oa -FROM ${stats_db_name}.publication_datasources pd -LEFT OUTER JOIN ( - SELECT pd.id, 1 as is_hybrid_oa from ${stats_db_name}.publication_datasources pd - JOIN ${stats_db_name}.datasource d on d.id=pd.datasource - JOIN issn on issn.id=pd.datasource - JOIN hybrid_oa ON issn.issn = hybrid_oa.issn - JOIN ${stats_db_name}.indi_result_has_cc_licence cc on pd.id=cc.id - JOIN ${stats_db_name}.indi_pub_gold_oa ga on pd.id=ga.id where cc.has_cc_license=1 and ga.is_gold=0) tmp on pd.id=tmp.id; /*EOS*/ + SELECT * + FROM ( + SELECT id, issn_printed as issn + FROM ${stats_db_name}.datasource + WHERE issn_printed IS NOT NULL + UNION ALL + SELECT id,issn_online as issn + FROM ${stats_db_name}.datasource + WHERE issn_online IS NOT NULL ) as issn + WHERE LENGTH(issn) > 7) + SELECT DISTINCT pd.id, coalesce(is_hybrid_oa, 0) as is_hybrid_oa + FROM ${stats_db_name}.publication_datasources pd + LEFT OUTER JOIN ( + SELECT pd.id, 1 as is_hybrid_oa from ${stats_db_name}.publication_datasources pd + JOIN ${stats_db_name}.datasource d on d.id=pd.datasource + JOIN issn on issn.id=pd.datasource + JOIN hybrid_oa ON issn.issn = hybrid_oa.issn + JOIN ${stats_db_name}.indi_result_has_cc_licence cc on pd.id=cc.id + JOIN ${stats_db_name}.indi_pub_gold_oa ga on pd.id=ga.id where cc.has_cc_license=1 and ga.is_gold=0) tmp on pd.id=tmp.id; /*EOS*/ drop table if exists ${stats_db_name}.indi_pub_hybrid purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_pub_hybrid stored as parquet as select distinct pd.id,coalesce(is_hybrid,0) is_hybrid from ${stats_db_name}.publication pd left outer join ( @@ -349,24 +292,25 @@ left outer join ( where indi_gold.is_gold=0 and ((d.type like '%Journal%' and ri.accessright!='Closed Access' and ri.accessright!='Restricted' and ri.license is not null) or ra.accessroute='hybrid')) tmp on pd.id=tmp.id; /*EOS*/ drop table if exists ${stats_db_name}.indi_org_fairness purge; /*EOS*/ - create table if not exists ${stats_db_name}.indi_org_fairness stored as parquet as --return results with PIDs, and rich metadata group by organization - with result_fair as - (select ro.organization organization, count(distinct ro.id) no_result_fair from ${stats_db_name}.result_organization ro - join ${stats_db_name}.result r on r.id=ro.id + with result_fair as ( + select ro.organization organization, count(distinct ro.id) no_result_fair + from ${stats_db_name}.result_organization ro + join ${stats_db_name}.result r on r.id=ro.id --join result_pids rp on r.id=rp.id - where (title is not null) and (publisher is not null) and (abstract=true) and (year is not null) and (authors>0) and cast(year as int)>2003 - group by ro.organization), + where (title is not null) and (publisher is not null) and (abstract=true) and (year is not null) and (authors>0) and cast(year as int)>2003 + group by ro.organization), --return all results group by organization - allresults as (select ro.organization, count(distinct ro.id) no_allresults from ${stats_db_name}.result_organization ro - join ${stats_db_name}.result r on r.id=ro.id - where cast(year as int)>2003 - group by ro.organization) + allresults as ( + select ro.organization, count(distinct ro.id) no_allresults from ${stats_db_name}.result_organization ro + join ${stats_db_name}.result r on r.id=ro.id + where cast(year as int)>2003 + group by ro.organization) --return results_fair/all_results -select allresults.organization, result_fair.no_result_fair/allresults.no_allresults org_fairness -from allresults - join result_fair on result_fair.organization=allresults.organization; /*EOS*/ + select allresults.organization, result_fair.no_result_fair/allresults.no_allresults org_fairness + from allresults + join result_fair on result_fair.organization=allresults.organization; /*EOS*/ CREATE TEMPORARY VIEW result_fair as select ro.organization organization, count(distinct ro.id) no_result_fair -- 2.17.1 From 92cc27e7eb9bdb47fa65e4b3ed20228b66dcd829 Mon Sep 17 00:00:00 2001 From: LSmyrnaios Date: Tue, 26 Mar 2024 12:34:11 +0200 Subject: [PATCH 21/22] Use the ACTIVE HDFS NODE for Impala cluster, in "copyDataToImpalaCluster.sh" script. --- .../stats/oozie_app/copyDataToImpalaCluster.sh | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/copyDataToImpalaCluster.sh b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/copyDataToImpalaCluster.sh index 18ff6dca8..1996c0b03 100644 --- a/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/copyDataToImpalaCluster.sh +++ b/dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/copyDataToImpalaCluster.sh @@ -6,16 +6,27 @@ then ln -sfn ${PYTHON_EGG_CACHE}${link_folder} ${link_folder} fi +IMPALA_HDFS_NODE='' +if hdfs dfs -test -e hdfs://impala-cluster-mn1.openaire.eu >/dev/null 2>&1; then + IMPALA_HDFS_NODE='hdfs://impala-cluster-mn1.openaire.eu:8020' +elif hdfs dfs -test -e hdfs://impala-cluster-mn2.openaire.eu >/dev/null 2>&1; then + IMPALA_HDFS_NODE='hdfs://impala-cluster-mn2.openaire.eu:8020' +else + echo -e "\n\nPROBLEM WHEN SETTING THE HDFS-NODE FOR IMPALA CLUSTER!\n\n" + exit 1 +fi +echo "Active IMPALA HDFS Node: ${IMPALA_HDFS_NODE}" + export HADOOP_USER_NAME=$6 export PROD_USAGE_STATS_DB="openaire_prod_usage_stats" function copydb() { db=$1 FILE=("hive_wf_tmp_"$RANDOM) - hdfs dfs -mkdir hdfs://impala-cluster-mn1.openaire.eu:8020/tmp/$FILE/ + hdfs dfs -mkdir ${IMPALA_HDFS_NODE}/tmp/$FILE/ # copy the databases from ocean to impala echo "copying $db" - hadoop distcp -Dmapreduce.map.memory.mb=6144 -pb hdfs://nameservice1/user/hive/warehouse/${db}.db hdfs://impala-cluster-mn1.openaire.eu:8020/tmp/$FILE/ + hadoop distcp -Dmapreduce.map.memory.mb=6144 -pb hdfs://nameservice1/user/hive/warehouse/${db}.db ${IMPALA_HDFS_NODE}/tmp/$FILE/ # change ownership to impala hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db -- 2.17.1 From bc8c97182d5f495c6d7ceb0947bcdaa83bb14251 Mon Sep 17 00:00:00 2001 From: LSmyrnaios Date: Tue, 26 Mar 2024 13:01:12 +0200 Subject: [PATCH 22/22] Automatically select the ACTIVE HDFS NODE for Impala cluster, in all "copyDataToImpalaCluster.sh" scripts. --- .../oozie_app/copyDataToImpalaCluster.sh | 16 ++++++++++++++-- .../oozie_app/copyDataToImpalaCluster.sh | 15 +++++++++++++-- .../oozie_app/copyDataToImpalaCluster.sh | 15 +++++++++++++-- 3 files changed, 40 insertions(+), 6 deletions(-) diff --git a/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/copyDataToImpalaCluster.sh b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/copyDataToImpalaCluster.sh index ef9bb9495..ef80d0094 100644 --- a/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/copyDataToImpalaCluster.sh +++ b/dhp-workflows/dhp-stats-hist-snaps/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-hist-snaps/oozie_app/copyDataToImpalaCluster.sh @@ -8,12 +8,24 @@ fi export HADOOP_USER_NAME=$2 +IMPALA_HDFS_NODE='' +if hdfs dfs -test -e hdfs://impala-cluster-mn1.openaire.eu >/dev/null 2>&1; then + IMPALA_HDFS_NODE='hdfs://impala-cluster-mn1.openaire.eu:8020' +elif hdfs dfs -test -e hdfs://impala-cluster-mn2.openaire.eu >/dev/null 2>&1; then + IMPALA_HDFS_NODE='hdfs://impala-cluster-mn2.openaire.eu:8020' +else + echo -e "\n\nPROBLEM WHEN SETTING THE HDFS-NODE FOR IMPALA CLUSTER!\n\n" + exit 1 +fi +echo "Active IMPALA HDFS Node: ${IMPALA_HDFS_NODE}" + + function copydb() { db=$1 FILE=("hive_wf_tmp_"$RANDOM) - hdfs dfs -mkdir hdfs://impala-cluster-mn1.openaire.eu:8020/tmp/$FILE/ + hdfs dfs -mkdir ${IMPALA_HDFS_NODE}/tmp/$FILE/ # change ownership to impala # hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db @@ -22,7 +34,7 @@ function copydb() { # copy the databases from ocean to impala echo "copying $db" - hadoop distcp -Dmapreduce.map.memory.mb=6144 -pb hdfs://nameservice1/user/hive/warehouse/${db}.db hdfs://impala-cluster-mn1.openaire.eu:8020/tmp/$FILE/ + hadoop distcp -Dmapreduce.map.memory.mb=6144 -pb hdfs://nameservice1/user/hive/warehouse/${db}.db ${IMPALA_HDFS_NODE}/tmp/$FILE/ hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db diff --git a/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/copyDataToImpalaCluster.sh b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/copyDataToImpalaCluster.sh index 78c44ed20..e7d183ddb 100644 --- a/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/copyDataToImpalaCluster.sh +++ b/dhp-workflows/dhp-stats-monitor-irish/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor-irish/oozie_app/copyDataToImpalaCluster.sh @@ -8,11 +8,22 @@ fi export HADOOP_USER_NAME=$2 +IMPALA_HDFS_NODE='' +if hdfs dfs -test -e hdfs://impala-cluster-mn1.openaire.eu >/dev/null 2>&1; then + IMPALA_HDFS_NODE='hdfs://impala-cluster-mn1.openaire.eu:8020' +elif hdfs dfs -test -e hdfs://impala-cluster-mn2.openaire.eu >/dev/null 2>&1; then + IMPALA_HDFS_NODE='hdfs://impala-cluster-mn2.openaire.eu:8020' +else + echo -e "\n\nPROBLEM WHEN SETTING THE HDFS-NODE FOR IMPALA CLUSTER!\n\n" + exit 1 +fi +echo "Active IMPALA HDFS Node: ${IMPALA_HDFS_NODE}" + function copydb() { db=$1 FILE=("hive_wf_tmp_"$RANDOM) - hdfs dfs -mkdir hdfs://impala-cluster-mn1.openaire.eu:8020/tmp/$FILE/ + hdfs dfs -mkdir ${IMPALA_HDFS_NODE}/tmp/$FILE/ # change ownership to impala # hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db @@ -21,7 +32,7 @@ function copydb() { # copy the databases from ocean to impala echo "copying $db" - hadoop distcp -Dmapreduce.map.memory.mb=6144 -pb hdfs://nameservice1/user/hive/warehouse/${db}.db hdfs://impala-cluster-mn1.openaire.eu:8020/tmp/$FILE/ + hadoop distcp -Dmapreduce.map.memory.mb=6144 -pb hdfs://nameservice1/user/hive/warehouse/${db}.db ${IMPALA_HDFS_NODE}/tmp/$FILE/ hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db diff --git a/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/copyDataToImpalaCluster.sh b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/copyDataToImpalaCluster.sh index 1587f7152..148d9b0b6 100644 --- a/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/copyDataToImpalaCluster.sh +++ b/dhp-workflows/dhp-stats-monitor-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats-monitor/oozie_app/copyDataToImpalaCluster.sh @@ -8,6 +8,17 @@ fi #export HADOOP_USER_NAME=$2 +IMPALA_HDFS_NODE='' +if hdfs dfs -test -e hdfs://impala-cluster-mn1.openaire.eu >/dev/null 2>&1; then + IMPALA_HDFS_NODE='hdfs://impala-cluster-mn1.openaire.eu:8020' +elif hdfs dfs -test -e hdfs://impala-cluster-mn2.openaire.eu >/dev/null 2>&1; then + IMPALA_HDFS_NODE='hdfs://impala-cluster-mn2.openaire.eu:8020' +else + echo -e "\n\nPROBLEM WHEN SETTING THE HDFS-NODE FOR IMPALA CLUSTER!\n\n" + exit 1 +fi +echo "Active IMPALA HDFS Node: ${IMPALA_HDFS_NODE}" + function copydb() { export HADOOP_USER="dimitris.pierrakos" @@ -15,7 +26,7 @@ function copydb() { db=$1 FILE=("hive_wf_tmp_"$RANDOM) - hdfs dfs -mkdir hdfs://impala-cluster-mn1.openaire.eu:8020/tmp/$FILE/ + hdfs dfs -mkdir ${IMPALA_HDFS_NODE}/tmp/$FILE/φ # change ownership to impala # hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db @@ -24,7 +35,7 @@ function copydb() { # copy the databases from ocean to impala echo "copying $db" - hadoop distcp -Dmapreduce.map.memory.mb=6144 -pb hdfs://nameservice1/user/hive/warehouse/${db}.db hdfs://impala-cluster-mn1.openaire.eu:8020/tmp/$FILE/ + hadoop distcp -Dmapreduce.map.memory.mb=6144 -pb hdfs://nameservice1/user/hive/warehouse/${db}.db ${IMPALA_HDFS_NODE}/tmp/$FILE/ hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db -- 2.17.1