[Stats wf] #372, #405 to production #406

Merged
claudio.atzori merged 28 commits from antonis.lempesis/dnet-hadoop:beta into master 2024-03-26 12:18:27 +01:00
37 changed files with 2949 additions and 778 deletions

View File

@ -0,0 +1,32 @@
<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<parent>
<artifactId>dhp-workflows</artifactId>
<groupId>eu.dnetlib.dhp</groupId>
<version>1.2.4-SNAPSHOT</version>
</parent>
<modelVersion>4.0.0</modelVersion>
<artifactId>dhp-stats-hist-snaps</artifactId>
<dependencies>
<dependency>
<groupId>org.apache.spark</groupId>
<artifactId>spark-core_2.11</artifactId>
</dependency>
<dependency>
<groupId>org.apache.spark</groupId>
<artifactId>spark-sql_2.11</artifactId>
</dependency>
</dependencies>
<build>
<plugins>
<plugin>
<groupId>pl.project13.maven</groupId>
<artifactId>git-commit-id-plugin</artifactId>
<version>2.1.11</version>
<configuration>
<failOnNoGitDirectory>false</failOnNoGitDirectory>
</configuration>
</plugin>
</plugins>
</build>
</project>

View File

@ -0,0 +1,30 @@
<configuration>
<property>
<name>jobTracker</name>
<value>${jobTracker}</value>
</property>
<property>
<name>nameNode</name>
<value>${nameNode}</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>hive_metastore_uris</name>
<value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
</property>
<property>
<name>hive_jdbc_url</name>
<value>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</value>
</property>
<property>
<name>oozie.wf.workflow.notification.url</name>
<value>{serviceUrl}/v1/oozieNotification/jobUpdate?jobId=$jobId%26status=$status</value>
</property>
</configuration>

View File

@ -0,0 +1,83 @@
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
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 ${IMPALA_HDFS_NODE}/tmp/$FILE/
# change ownership to impala
# hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db
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 ${IMPALA_HDFS_NODE}/tmp/$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

View File

@ -0,0 +1,41 @@
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_PROD=$4
MONITOR_IRISH_PROD=$5
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"

View File

@ -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"

View File

@ -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;

View File

@ -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;

View File

@ -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');

View File

@ -0,0 +1,159 @@
<workflow-app name="Stats Hist Snapshots" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>hist_db_name</name>
<description>the target hist database name</description>
</property>
<property>
<name>hist_db_name_prev</name>
<description>the hist database name of previous_month</description>
</property>
<property>
<name>stats_db_name</name>
<description>the stats db name</description>
</property>
<property>
<name>stats_irish_db_name</name>
<description>the stats irish db name</description>
</property>
<property>
<name>monitor_db_name</name>
<description>the monitor db name</description>
</property>
<property>
<name>monitor_irish_db_name</name>
<description>the irish monitor db name</description>
</property>
<property>
<name>hist_db_prod_name</name>
<description>the production db</description>
</property>
<property>
<name>hist_db_shadow_name</name>
<description>the production shadow db</description>
</property>
<property>
<name>hist_date</name>
<description>the snaps date</description>
</property>
<property>
<name>hive_metastore_uris</name>
<description>hive server metastore URIs</description>
</property>
<property>
<name>hive_jdbc_url</name>
<description>hive server jdbc url</description>
</property>
<property>
<name>hive_timeout</name>
<description>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.</description>
</property>
<property>
<name>hadoop_user_name</name>
<description>user name of the wf owner</description>
</property>
</parameters>
<global>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
<property>
<name>hive.txn.timeout</name>
<value>${hive_timeout}</value>
</property>
<property>
<name>mapred.job.queue.name</name>
<value>analytics</value>
</property>
</configuration>
</global>
<start to="resume_from"/>
<decision name="resume_from">
<switch>
<case to="CreateDB">${wf:conf('resumeFrom') eq 'CreateDB'}</case>
<case to="BuildHistSnaps">${wf:conf('resumeFrom') eq 'BuildHistSnaps'}</case>
<case to="BuildHistSnapsIrish">${wf:conf('resumeFrom') eq 'BuildHistSnapsIrish'}</case>
<case to="Step2-copyDataToImpalaCluster">${wf:conf('resumeFrom') eq 'Step2-copyDataToImpalaCluster'}</case>
<case to="Step3-finalizeImpalaCluster">${wf:conf('resumeFrom') eq 'Step3-finalizeImpalaCluster'}</case>
<default to="BuildHistSnaps"/>
</switch>
</decision>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="CreateDB">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/CreateDB.sql</script>
<param>hist_db_name=${hist_db_name}</param>
</hive2>
<ok to="BuildHistSnaps"/>
<error to="Kill"/>
</action>
<action name="BuildHistSnaps">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/BuildHistSnapsAll.sql</script>
<param>hist_db_name=${hist_db_name}</param>
<param>hist_db_name_prev=${hist_db_name_prev}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>monitor_db_name=${monitor_db_name}</param>
<param>hist_date=${hist_date}</param>
</hive2>
<ok to="BuildHistSnapsIrish"/>
<error to="Kill"/>
</action>
<action name="BuildHistSnapsIrish">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/BuildHistSnapsIrish.sql</script>
<param>hist_db_name=${hist_db_name}</param>
<param>hist_db_name_prev=${hist_db_name_prev}</param>
<param>stats_irish_db_name=${stats_irish_db_name}</param>
<param>monitor_irish_db_name=${monitor_irish_db_name}</param>
<param>hist_date=${hist_date}</param>
</hive2>
<ok to="Step2-copyDataToImpalaCluster"/>
<error to="Kill"/>
</action>
<action name="Step2-copyDataToImpalaCluster">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>copyDataToImpalaCluster.sh</exec>
<argument>${hist_db_name}</argument>
<argument>${hadoop_user_name}</argument>
<file>copyDataToImpalaCluster.sh</file>
</shell>
<ok to="Step3-finalizeImpalaCluster"/>
<error to="Kill"/>
</action>
<action name="Step3-finalizeImpalaCluster">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>finalizeImpalaCluster.sh</exec>
<argument>${hist_db_name}</argument>
<argument>${hist_db_prod_name}</argument>
<argument>${hist_db_shadow_name}</argument>
<argument>${monitor_db_prod_name}</argument>
<argument>${monitor_irish_db_prod_name}</argument>
<file>finalizeImpalaCluster.sh</file>
</shell>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -0,0 +1,32 @@
<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<parent>
<artifactId>dhp-workflows</artifactId>
<groupId>eu.dnetlib.dhp</groupId>
<version>1.2.4-SNAPSHOT</version>
</parent>
<modelVersion>4.0.0</modelVersion>
<artifactId>dhp-stats-monitor-irish</artifactId>
<dependencies>
<dependency>
<groupId>org.apache.spark</groupId>
<artifactId>spark-core_2.11</artifactId>
</dependency>
<dependency>
<groupId>org.apache.spark</groupId>
<artifactId>spark-sql_2.11</artifactId>
</dependency>
</dependencies>
<build>
<plugins>
<plugin>
<groupId>pl.project13.maven</groupId>
<artifactId>git-commit-id-plugin</artifactId>
<version>2.1.11</version>
<configuration>
<failOnNoGitDirectory>false</failOnNoGitDirectory>
</configuration>
</plugin>
</plugins>
</build>
</project>

View File

@ -0,0 +1,30 @@
<configuration>
<property>
<name>jobTracker</name>
<value>${jobTracker}</value>
</property>
<property>
<name>nameNode</name>
<value>${nameNode}</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>hive_metastore_uris</name>
<value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
</property>
<property>
<name>hive_jdbc_url</name>
<value>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</value>
</property>
<property>
<name>oozie.wf.workflow.notification.url</name>
<value>{serviceUrl}/v1/oozieNotification/jobUpdate?jobId=$jobId%26status=$status</value>
</property>
</configuration>

View File

@ -0,0 +1,81 @@
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
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 ${IMPALA_HDFS_NODE}/tmp/$FILE/
# change ownership to impala
# hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db
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 ${IMPALA_HDFS_NODE}/tmp/$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

View File

@ -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!"

View File

@ -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"

View File

@ -0,0 +1,241 @@
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 (
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;
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);
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);

View File

@ -0,0 +1,118 @@
<workflow-app name="Irish Monitor Update" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>stats_db_name</name>
<description>the target stats database name</description>
</property>
<property>
<name>graph_db_name</name>
<description>the graph database name</description>
</property>
<property>
<name>monitor_irish_db_name</name>
<description>the target monitor db name</description>
</property>
<property>
<name>monitor_irish_db_prod_name</name>
<description>the name of the production monitor db</description>
</property>
<property>
<name>monitor_irish_db_shadow_name</name>
<description>the name of the shadow monitor db</description>
</property>
<property>
<name>hive_metastore_uris</name>
<description>hive server metastore URIs</description>
</property>
<property>
<name>hive_jdbc_url</name>
<description>hive server jdbc url</description>
</property>
<property>
<name>hive_timeout</name>
<description>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.</description>
</property>
<property>
<name>hadoop_user_name</name>
<description>user name of the wf owner</description>
</property>
</parameters>
<global>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
<property>
<name>hive.txn.timeout</name>
<value>${hive_timeout}</value>
</property>
<property>
<name>mapred.job.queue.name</name>
<value>analytics</value>
</property>
</configuration>
</global>
<start to="resume_from"/>
<decision name="resume_from">
<switch>
<case to="Step1-buildIrishMonitorDB">${wf:conf('resumeFrom') eq 'Step1-buildIrishMonitorDB'}</case>
<case to="Step2-copyDataToImpalaCluster">${wf:conf('resumeFrom') eq 'Step2-copyDataToImpalaCluster'}</case>
<case to="Step3-finalizeImpalaCluster">${wf:conf('resumeFrom') eq 'Step3-finalizeImpalaCluster'}</case>
<default to="Step1-buildIrishMonitorDB"/>
</switch>
</decision>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="Step1-buildIrishMonitorDB">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>monitor_irish.sh</exec>
<argument>${stats_db_name}</argument>
<argument>${monitor_irish_db_name}</argument>
<argument>${monitor_irish_db_shadow_name}</argument>
<argument>${wf:appPath()}/scripts/buildIrishMonitorDB.sql</argument>
<argument>${graph_db_name}</argument>
<file>monitor_irish.sh</file>
</shell>
<ok to="Step2-copyDataToImpalaCluster"/>
<error to="Kill"/>
</action>
<action name="Step2-copyDataToImpalaCluster">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>copyDataToImpalaCluster.sh</exec>
<argument>${monitor_irish_db_name}</argument>
<argument>${hadoop_user_name}</argument>
<file>copyDataToImpalaCluster.sh</file>
</shell>
<ok to="Step3-finalizeImpalaCluster"/>
<error to="Kill"/>
</action>
<action name="Step3-finalizeImpalaCluster">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>finalizeImpalaCluster.sh</exec>
<argument>${monitor_irish_db_name}</argument>
<argument>${monitor_irish_db_prod_name}</argument>
<argument>${monitor_irish_db_shadow_name}</argument>
<file>finalizeImpalaCluster.sh</file>
</shell>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -0,0 +1,32 @@
<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<parent>
<artifactId>dhp-workflows</artifactId>
<groupId>eu.dnetlib.dhp</groupId>
<version>1.2.4-SNAPSHOT</version>
</parent>
<modelVersion>4.0.0</modelVersion>
<artifactId>dhp-stats-monitor-update</artifactId>
<dependencies>
<dependency>
<groupId>org.apache.spark</groupId>
<artifactId>spark-core_2.11</artifactId>
</dependency>
<dependency>
<groupId>org.apache.spark</groupId>
<artifactId>spark-sql_2.11</artifactId>
</dependency>
</dependencies>
<build>
<plugins>
<plugin>
<groupId>pl.project13.maven</groupId>
<artifactId>git-commit-id-plugin</artifactId>
<version>2.1.11</version>
<configuration>
<failOnNoGitDirectory>false</failOnNoGitDirectory>
</configuration>
</plugin>
</plugins>
</build>
</project>

View File

@ -0,0 +1,30 @@
<configuration>
<property>
<name>jobTracker</name>
<value>${jobTracker}</value>
</property>
<property>
<name>nameNode</name>
<value>${nameNode}</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
<property>
<name>hive_metastore_uris</name>
<value>thrift://iis-cdh5-test-m3.ocean.icm.edu.pl:9083</value>
</property>
<property>
<name>hive_jdbc_url</name>
<value>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</value>
</property>
<property>
<name>oozie.wf.workflow.notification.url</name>
<value>{serviceUrl}/v1/oozieNotification/jobUpdate?jobId=$jobId%26status=$status</value>
</property>
</configuration>

View File

@ -0,0 +1,86 @@
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
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"
export HADOOP_USER_NAME='dimitris.pierrakos'
db=$1
FILE=("hive_wf_tmp_"$RANDOM)
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
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 ${IMPALA_HDFS_NODE}/tmp/$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

View File

@ -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!"

View File

@ -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!"

View File

@ -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);

View File

@ -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;

View File

@ -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;

View File

@ -0,0 +1,111 @@
<workflow-app name="Stats Monitor Update" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>stats_db_name</name>
<description>the target stats database name</description>
</property>
<property>
<name>monitor_db_name</name>
<description>the target monitor db name</description>
</property>
<property>
<name>monitor_db_shadow_name</name>
<description>the name of the shadow monitor db</description>
</property>
<property>
<name>hive_metastore_uris</name>
<description>hive server metastore URIs</description>
</property>
<property>
<name>hive_jdbc_url</name>
<description>hive server jdbc url</description>
</property>
<property>
<name>hive_timeout</name>
<description>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.</description>
</property>
<property>
<name>hadoop_user_name</name>
<description>user name of the wf owner</description>
</property>
</parameters>
<global>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<configuration>
<property>
<name>hive.metastore.uris</name>
<value>${hive_metastore_uris}</value>
</property>
<property>
<name>hive.txn.timeout</name>
<value>${hive_timeout}</value>
</property>
<property>
<name>mapred.job.queue.name</name>
<value>analytics</value>
</property>
</configuration>
</global>
<start to="resume_from"/>
<decision name="resume_from">
<switch>
<case to="Step1-updateMonitorDB">${wf:conf('resumeFrom') eq 'Step1-updateMonitorDB'}</case>
<case to="Step2-copyDataToImpalaCluster">${wf:conf('resumeFrom') eq 'Step2-copyDataToImpalaCluster'}</case>
<case to="Step3-finalizeImpalaCluster">${wf:conf('resumeFrom') eq 'Step3-finalizeImpalaCluster'}</case>
<default to="Step1-updateMonitorDB"/>
</switch>
</decision>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="Step1-updateMonitorDB">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>monitor.sh</exec>
<argument>${stats_db_name}</argument>
<argument>${monitor_db_name}</argument>
<argument>${monitor_db_shadow_name}</argument>
<argument>${wf:appPath()}/scripts/updateMonitorDB_institutions.sql</argument>
<argument>${wf:appPath()}/scripts/updateMonitorDB.sql</argument>
<argument>${wf:appPath()}/scripts/updateMonitorDBAll.sql</argument>
<file>monitor.sh</file>
</shell>
<ok to="End"/>
<error to="Kill"/>
</action>
<action name="Step2-copyDataToImpalaCluster">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>copyDataToImpalaCluster.sh</exec>
<argument>${monitor_db_name}</argument>
<argument>${hadoop_user_name}</argument>
<file>copyDataToImpalaCluster.sh</file>
</shell>
<ok to="End"/>
<error to="Kill"/>
</action>
<action name="Step3-finalizeImpalaCluster">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>finalizeImpalaCluster.sh</exec>
<argument>${monitor_db_name}</argument>
<argument>${monitor_db_prod_name}</argument>
<argument>${monitor_db_shadow_name}</argument>
<file>finalizeImpalaCluster.sh</file>
</shell>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>

View File

@ -8,6 +8,11 @@
<modelVersion>4.0.0</modelVersion>
<artifactId>dhp-stats-update</artifactId>
<dependencies>
<dependency>
<groupId>eu.dnetlib.dhp</groupId>
<artifactId>dhp-common</artifactId>
<version>${project.version}</version>
</dependency>
<dependency>
<groupId>org.apache.spark</groupId>
<artifactId>spark-core_${scala.binary.version}</artifactId>

View File

@ -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

View File

@ -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

View File

@ -7,32 +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;
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;
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;
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;
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
@ -60,4 +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;

View File

@ -50,13 +50,13 @@ 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;
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

View File

@ -180,4 +180,12 @@ 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;
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);

View File

@ -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;
@ -264,4 +264,11 @@ 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);

View File

@ -60,5 +60,8 @@ 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
'openorgs____::64badd35233ba2cd4946368ef2f4cf57', -- University of Vienna
'openorgs____::7501d66d2297a963ebfb075c43fff88e' -- Royal Institute of Technology
))) foo;

View File

@ -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;

View File

@ -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;

View File

@ -64,6 +64,26 @@
<name>hadoop_user_name</name>
<description>user name of the wf owner</description>
</property>
<property>
<name>sparkSqlWarehouseDir</name>
</property>
<!-- General oozie workflow properties -->
<property>
<name>sparkClusterOpts</name>
<value>--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</value>
<description>spark cluster-wide options</description>
</property>
<property>
<name>sparkResourceOpts</name>
<value>--executor-memory=6G --conf spark.executor.memoryOverhead=4G --executor-cores=6 --driver-memory=8G --driver-cores=4</value>
<description>spark resource options</description>
</property>
<property>
<name>sparkApplicationOpts</name>
<value>--conf spark.sql.shuffle.partitions=3840</value>
<description>spark resource options</description>
</property>
</parameters>
<global>
@ -75,17 +95,21 @@
<value>${hive_metastore_uris}</value>
</property>
<property>
<name>hive.txn.timeout</name>
<value>${hive_timeout}</value>
<name>hive.txn.timeout</name>
<value>${hive_timeout}</value>
</property>
<property>
<name>hive.mapjoin.followby.gby.localtask.max.memory.usage</name>
<value>0.80</value>
</property>
<property>
<name>mapred.job.queue.name</name>
<value>analytics</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>${oozieActionShareLibForSpark2}</value>
</property>
<property>
<name>mapred.job.queue.name</name>
<value>analytics</value>
</property>
</configuration>
</global>
@ -133,164 +157,164 @@
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step1.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step2"/>
<error to="Kill"/>
</action>
<action name="Step2">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step2.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step3"/>
<error to="Kill"/>
</action>
<action name="Step3">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step3.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step4"/>
<error to="Kill"/>
</action>
<action name="Step4">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step4.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step5"/>
<error to="Kill"/>
</action>
<action name="Step5">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step5.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step6"/>
<error to="Kill"/>
</action>
<action name="Step6">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step6.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step7"/>
<error to="Kill"/>
</action>
<action name="Step7">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step7.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step8"/>
<error to="Kill"/>
</action>
<action name="Step8">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step8.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step9"/>
<error to="Kill"/>
</action>
<action name="Step9">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step9.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step10"/>
<error to="Kill"/>
</action>
<action name="Step10">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step10.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>external_stats_db_name=${external_stats_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>external_stats_db_name=${external_stats_db_name}</param>
</hive2>
<ok to="Step11"/>
<error to="Kill"/>
</action>
</action>
<action name="Step11">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step11.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>external_stats_db_name=${external_stats_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>external_stats_db_name=${external_stats_db_name}</param>
</hive2>
<ok to="Step12"/>
<error to="Kill"/>
</action>
</action>
<action name="Step12">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step12.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step13"/>
<error to="Kill"/>
</action>
<action name="Step13">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step13.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step14"/>
<error to="Kill"/>
</action>
<action name="Step14">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step14.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step15"/>
<error to="Kill"/>
</action>
<action name="Step15">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step15.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step15_5"/>
<error to="Kill"/>
@ -321,13 +345,35 @@
<error to="Kill"/>
</action>
<!-- <action name="Step16-createIndicatorsTables">-->
<!-- <hive2 xmlns="uri:oozie:hive2-action:0.1">-->
<!-- <jdbc-url>${hive_jdbc_url}</jdbc-url>-->
<!-- <script>scripts/step16-createIndicatorsTables.sql</script>-->
<!-- <param>stats_db_name=${stats_db_name}</param>-->
<!-- <param>external_stats_db_name=${external_stats_db_name}</param>-->
<!-- </hive2>-->
<!-- <ok to="Step16_1-definitions"/>-->
<!-- <error to="Kill"/>-->
<!-- </action>-->
<action name="Step16-createIndicatorsTables">
<hive2 xmlns="uri:oozie:hive2-action:0.1">
<jdbc-url>${hive_jdbc_url}</jdbc-url>
<script>scripts/step16-createIndicatorsTables.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>external_stats_db_name=${external_stats_db_name}</param>
</hive2>
<spark xmlns="uri:oozie:spark-action:0.2">
<master>yarn</master>
<mode>cluster</mode>
<name>Step16-createIndicatorsTables</name>
<class>eu.dnetlib.dhp.oozie.RunSQLSparkJob</class>
<jar>dhp-stats-update-${projectVersion}.jar</jar>
<spark-opts>
--conf spark.sql.warehouse.dir=${sparkSqlWarehouseDir}
${sparkClusterOpts}
${sparkResourceOpts}
${sparkApplicationOpts}
</spark-opts>
<arg>--hiveMetastoreUris</arg><arg>${hive_metastore_uris}</arg>
<arg>--sql</arg><arg>eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql</arg>
<arg>--stats_db_name</arg><arg>${stats_db_name}</arg>
<arg>--external_stats_db_name</arg><arg>${external_stats_db_name}</arg>
</spark>
<ok to="Step16_1-definitions"/>
<error to="Kill"/>
</action>
@ -387,18 +433,18 @@
<error to="Kill"/>
</action>
<!-- <action name="step20-createMonitorDB-post">-->
<!-- <shell xmlns="uri:oozie:shell-action:0.1">-->
<!-- <job-tracker>${jobTracker}</job-tracker>-->
<!-- <name-node>${nameNode}</name-node>-->
<!-- <exec>monitor-post.sh</exec>-->
<!-- <argument>${monitor_db_name}</argument>-->
<!-- <argument>${monitor_db_shadow_name}</argument>-->
<!-- <file>monitor-post.sh</file>-->
<!-- </shell>-->
<!-- <ok to="step21-createObservatoryDB-pre"/>-->
<!-- <error to="Kill"/>-->
<!-- </action>-->
<!-- <action name="step20-createMonitorDB-post">-->
<!-- <shell xmlns="uri:oozie:shell-action:0.1">-->
<!-- <job-tracker>${jobTracker}</job-tracker>-->
<!-- <name-node>${nameNode}</name-node>-->
<!-- <exec>monitor-post.sh</exec>-->
<!-- <argument>${monitor_db_name}</argument>-->
<!-- <argument>${monitor_db_shadow_name}</argument>-->
<!-- <file>monitor-post.sh</file>-->
<!-- </shell>-->
<!-- <ok to="step21-createObservatoryDB-pre"/>-->
<!-- <error to="Kill"/>-->
<!-- </action>-->
<action name="step21-createObservatoryDB-pre">
<shell xmlns="uri:oozie:shell-action:0.1">
@ -443,8 +489,8 @@
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>copyDataToImpalaCluster.sh</exec>
<!-- <env-var>HADOOP_USER_NAME=${wf:user()}</env-var>-->
<!-- <argument>${external_stats_db_name}</argument>-->
<!-- <env-var>HADOOP_USER_NAME=${wf:user()}</env-var>-->
<!-- <argument>${external_stats_db_name}</argument>-->
<argument>${stats_db_name}</argument>
<argument>${monitor_db_name}</argument>
<argument>${observatory_db_name}</argument>
@ -505,4 +551,4 @@
</action>
<end name="End"/>
</workflow-app>
</workflow-app>