1
0
Fork 0

[stats wf] merged workflows from branch beta

This commit is contained in:
Claudio Atzori 2023-06-27 14:32:48 +02:00
parent b93e1541aa
commit e10ce92fe5
26 changed files with 1146 additions and 647 deletions

View File

@ -10,7 +10,7 @@ export SOURCE=$1
export PRODUCTION=$2
echo "Updating ${PRODUCTION} database"
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 -
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 -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 -c -f -
echo "Production db ready!"

View File

@ -0,0 +1,38 @@
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 PRODUCTION=$2
echo "Updating ${PRODUCTION}'_funded' database"
impala-shell -i impala-cluster-dn1.openaire.eu -q "create database if not exists ${PRODUCTION}'_funded'"
impala-shell -i impala-cluster-dn1.openaire.eu -d ${PRODUCTION}'_funded' -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}'_funded'./" | sed "s/$/;/" | impala-shell -c -f -
impala-shell -i impala-cluster-dn1.openaire.eu -d ${SOURCE}'_funded' -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}'_funded'.\1 as select * from ${SOURCE}'_funded'.\1;/" | impala-shell -c -f -
echo "Production funded 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 -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 -c -f -
echo "Production insitutions db ready!"
echo "Updating ${PRODUCTION}'_ris_tail' database"
impala-shell -i impala-cluster-dn1.openaire.eu -q "create database if not exists ${PRODUCTION}'_ris_tail'"
impala-shell -i impala-cluster-dn1.openaire.eu -d ${PRODUCTION}'_ris_tail' -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}'_ris_tail'./" | sed "s/$/;/" | impala-shell -c -f -
impala-shell -i impala-cluster-dn1.openaire.eu -d ${SOURCE}'_RIs_tail' -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}'_ris_tail'.\1 as select * from ${SOURCE}'_ris_tail'.\1;/" | impala-shell -c -f -
echo "Production RIS tail db ready!"
contexts="knowmad::other dh-ch::other enermaps::other gotriple::other neanias-atmospheric::other rural-digital-europe::other covid-19::other aurora::other neanias-space::other north-america-studies::other north-american-studies::other eutopia::other"
for i in ${contexts}
do
tmp=`echo "$i" | sed 's/'-'/'_'/g' | sed 's/'::'/'_'/g'`
impala-shell -i impala-cluster-dn1.openaire.eu -q "create database if not exists ${PRODUCTION}'_'${tmp}"
impala-shell -i impala-cluster-dn1.openaire.eu -d ${PRODUCTION}'_'${tmp} -q "show tables" --delimited | sed "s/^/drop view if exists ${PRODUCTION}'_'${tmp}./" | sed "s/$/;/" | impala-shell -c -f -
impala-shell -i impala-cluster-dn1.openaire.eu -d ${SOURCE}'_'${tmp} -q "show tables" --delimited | sed "s/\(.*\)/create view ${PRODUCTION}'_'${tmp}.\1 as select * from ${SOURCE}'_'${tmp}.\1;/" | impala-shell -c -f -
echo "Production ${tmp} db ready!"
done

View File

@ -80,10 +80,10 @@
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>updateProductionViews.sh</exec>
<exec>updateProductionViewsMonitor.sh</exec>
<argument>${monitor_db_name}</argument>
<argument>${monitor_db_production_name}</argument>
<file>updateProductionViews.sh</file>
<file>updateProductionViewsMonitor.sh</file>
</shell>
<ok to="updateObservatoryViews"/>
<error to="Kill"/>

View File

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

View File

@ -9,6 +9,8 @@ fi
CONTEXT_API=$1
TARGET_DB=$2
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=4831838208 -hiveconf spark.yarn.executor.memoryOverhead=450"
TMP=/tmp/stats-update-`tr -dc A-Za-z0-9 </dev/urandom | head -c 6`
echo "Downloading context ids"
@ -29,21 +31,20 @@ hdfs dfs -copyFromLocal categories.csv ${TMP}
hdfs dfs -copyFromLocal concepts.csv ${TMP}
hdfs dfs -chmod -R 777 ${TMP}
export HADOOP_USER="oozie"
export HADOOP_USER_NAME="oozie"
echo "Creating and populating impala tables"
impala-shell -q "invalidate metadata"
impala-shell -d ${TARGET_DB} -q "invalidate metadata"
impala-shell -q "create table ${TARGET_DB}.context (id string, name string) row format delimited fields terminated by ','"
impala-shell -q "create table ${TARGET_DB}.category (context string, id string, name string) row format delimited fields terminated by ','"
impala-shell -q "create table ${TARGET_DB}.concept (category string, id string, name string) row format delimited fields terminated by ','"
impala-shell -d ${TARGET_DB} -q "invalidate metadata"
impala-shell -q "load data inpath '${TMP}/contexts.csv' into table ${TARGET_DB}.context"
impala-shell -q "load data inpath '${TMP}/categories.csv' into table ${TARGET_DB}.category"
impala-shell -q "load data inpath '${TMP}/concepts.csv' into table ${TARGET_DB}.concept"
hive $HIVE_OPTS -e "create table ${TARGET_DB}.context (id string, name string) row format delimited fields terminated by ','"
hive $HIVE_OPTS -e "create table ${TARGET_DB}.category (context string, id string, name string) row format delimited fields terminated by ','"
hive $HIVE_OPTS -e "create table ${TARGET_DB}.concept (category string, id string, name string) row format delimited fields terminated by ','"
hive $HIVE_OPTS -e "load data inpath '${TMP}/contexts.csv' into table ${TARGET_DB}.context"
hive $HIVE_OPTS -e "load data inpath '${TMP}/categories.csv' into table ${TARGET_DB}.category"
hive $HIVE_OPTS -e "load data inpath '${TMP}/concepts.csv' into table ${TARGET_DB}.concept"
echo "Cleaning up"
hdfs dfs -rm -f -r -skipTrash ${TMP}
rm concepts.csv
rm categories.csv
rm contexts.csv
echo "Finito!"
echo "Finito!"

View File

@ -0,0 +1,97 @@
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="dimitris.pierrakos"
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/
# 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/$FILE/
# change ownership to impala
hdfs dfs -conf /etc/impala_cluster/hdfs-site.xml -chmod -R 777 /tmp/$FILE/${db}.db
# drop tables from db
for i in `impala-shell --user $HADOOP_USER_NAME -i impala-cluster-dn1.openaire.eu -d ${db} --delimited -q "show tables"`;
do
`impala-shell -i impala-cluster-dn1.openaire.eu -d -d ${db} -q "drop table $i;"`;
done
# drop views from db
for i in `impala-shell --user $HADOOP_USER_NAME -i impala-cluster-dn1.openaire.eu -d ${db} --delimited -q "show tables"`;
do
`impala-shell -i impala-cluster-dn1.openaire.eu -d -d ${db} -q "drop view $i;"`;
done
# delete the database
impala-shell --user $HADOOP_USER_NAME -i impala-cluster-dn1.openaire.eu -q "drop database if exists ${db} cascade";
# create the databases
impala-shell --user $HADOOP_USER_NAME -i impala-cluster-dn1.openaire.eu -q "create database ${db}";
impala-shell --user $HADOOP_USER_NAME -q "INVALIDATE METADATA"
echo "creating schema for ${db}"
for (( k = 0; k < 5; k ++ )); do
for i in `impala-shell --user $HADOOP_USER_NAME -d ${db} --delimited -q "show tables"`;
do
impala-shell --user $HADOOP_USER_NAME -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
# for i in `impala-shell --user $HADOOP_USER_NAME -d ${db} --delimited -q "show tables"`;
# do
# impala-shell --user $HADOOP_USER_NAME -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 -
#
# # run the same command twice because we may have failures in the first run (due to views pointing to the same db)
# for i in `impala-shell --user $HADOOP_USER_NAME -d ${db} --delimited -q "show tables"`;
# do
# impala-shell --user $HADOOP_USER_NAME -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 -
# load the data from /tmp in the respective tables
echo "copying data in tables and computing stats"
for i in `impala-shell --user $HADOOP_USER_NAME -i impala-cluster-dn1.openaire.eu -d ${db} --delimited -q "show tables"`;
do
impala-shell --user $HADOOP_USER_NAME -i impala-cluster-dn1.openaire.eu -d ${db} -q "load data inpath '/tmp/$FILE/${db}.db/$i' into table $i";
impala-shell --user $HADOOP_USER_NAME -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
}
STATS_DB=$1
MONITOR_DB=$2
OBSERVATORY_DB=$3
EXT_DB=$4
USAGE_STATS_DB=$5
HADOOP_USER_NAME=$6
copydb $USAGE_STATS_DB
copydb $PROD_USAGE_STATS_DB
copydb $EXT_DB
copydb $STATS_DB
#copydb $MONITOR_DB
copydb $OBSERVATORY_DB
copydb $MONITOR_DB'_funded'
copydb $MONITOR_DB'_institutions'
copydb $MONITOR_DB'_RIs_tail'
contexts="knowmad::other dh-ch::other enermaps::other gotriple::other neanias-atmospheric::other rural-digital-europe::other covid-19::other aurora::other neanias-space::other north-america-studies::other north-american-studies::other eutopia::other"
for i in ${contexts}
do
tmp=`echo "$i" | sed 's/'-'/'_'/g' | sed 's/'::'/'_'/g'`
copydb ${MONITOR_DB}'_'${tmp}
done

View File

@ -0,0 +1,48 @@
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 -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 -
}
STATS_DB=$1
STATS_DB_SHADOW=$2
MONITOR_DB=$3
MONITOR_DB_SHADOW=$4
OBSERVATORY_DB=$5
OBSERVATORY_DB_SHADOW=$6
USAGE_STATS_DB=$7
USAGE_STATS_DB_SHADOW=$8
createShadowDB $STATS_DB $STATS_DB_SHADOW
createShadowDB $MONITOR_DB $MONITOR_DB_SHADOW
createShadowDB $OBSERVATORY_DB $OBSERVATORY_DB_SHADOW
createShadowDB USAGE_STATS_DB USAGE_STATS_DB_SHADOW
createShadowDB $MONITOR_DB'_funded' $MONITOR_DB'_funded_shadow'
createShadowDB $MONITOR_DB'_institutions' $MONITOR_DB'_institutions_shadow'
createShadowDB $MONITOR_DB'_RIs_tail' $MONITOR_DB'_RIs_tail_shadow'
contexts="knowmad::other dh-ch::other enermaps::other gotriple::other neanias-atmospheric::other rural-digital-europe::other covid-19::other aurora::other neanias-space::other north-america-studies::other north-american-studies::other eutopia::other"
for i in ${contexts}
do
tmp=`echo "$i" | sed 's/'-'/'_'/g' | sed 's/'::'/'_'/g'`
createShadowDB ${MONITOR_DB}'_'${tmp} ${MONITOR_DB}'_'${tmp}'_shadow'
done

View File

@ -8,12 +8,12 @@ fi
export SOURCE=$1
export SHADOW=$2
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 "Updating shadow database"
impala-shell -q "invalidate metadata"
impala-shell -d ${SOURCE} -q "invalidate metadata"
impala-shell -d ${SOURCE} -q "show tables" --delimited | sed "s/^\(.*\)/compute stats ${SOURCE}.\1;/" | impala-shell -c -f -
impala-shell -q "create database if not exists ${SHADOW}"
impala-shell -d ${SHADOW} -q "show tables" --delimited | sed "s/^/drop view if exists ${SHADOW}./" | sed "s/$/;/" | impala-shell -c -f -
impala-shell -d ${SOURCE} -q "show tables" --delimited | sed "s/\(.*\)/create view ${SHADOW}.\1 as select * from ${SOURCE}.\1;/" | impala-shell -c -f -
echo "Shadow db ready!"
hive -e "drop database if exists ${SHADOW} cascade"
hive -e "create database if not exists ${SHADOW}"
hive $HIVE_OPTS --database ${SOURCE} -e "show tables" | grep -v WARN | sed "s/\(.*\)/create view ${SHADOW}.\1 as select * from ${SOURCE}.\1;/" > foo
hive -f foo
echo "Updated shadow database"

View File

@ -7,13 +7,17 @@ then
fi
export TARGET=$1
export SCRIPT_PATH=$2
export STATS_EXT=$2
export SCRIPT_PATH=$3
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 -hiveconf hive.auto.convert.join=false"
export HADOOP_USER_NAME="oozie"
echo "Getting file from " $SCRIPT_PATH
hdfs dfs -copyToLocal $SCRIPT_PATH
echo "Creating indicators"
impala-shell -q "invalidate metadata"
impala-shell -d ${TARGET} -q "show tables" --delimited | sed "s/^\(.*\)/compute stats ${TARGET}.\1;/" | impala-shell -c -f -
cat step16-createIndicatorsTables.sql | impala-shell -d $TARGET -f -
echo "Indicators created"
hive $HIVE_OPTS --database ${TARGET} -e "show tables" | grep -v WARN | sed "s/STATS_EXT/${STATS_EXT}/g" |sed "s/^\(.*\)/analyze table ${TARGET}.\1 compute statistics;/" > foo
hive $HIVE_OPTS -f foo
hive $HIVE_OPTS --database ${TARGET} -f step16-createIndicatorsTables.sql
echo "Indicators created"

View File

@ -0,0 +1,19 @@
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 SHADOW=$2
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 "Updating shadow database"
hive -e "drop database if exists ${SHADOW} cascade"
hive -e "create database if not exists ${SHADOW}"
hive $HIVE_OPTS --database ${SOURCE} -e "show tables" | grep -v WARN | sed "s/\(.*\)/create view ${SHADOW}.\1 as select * from ${SOURCE}.\1;/" > foo
hive -f foo
echo "Updated shadow database"

View File

@ -10,16 +10,88 @@ export SOURCE=$1
export TARGET=$2
export SHADOW=$3
export SCRIPT_PATH=$4
export SCRIPT_PATH2=$5
export SCRIPT_PATH3=$6
export SCRIPT_PATH4=$7
export SCRIPT_PATH5=$8
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 "Creating monitor database"
cat step20-createMonitorDB.sql | sed s/SOURCE/$1/g | sed s/TARGET/$2/g1 | impala-shell -f -
echo "Impala shell finished"
echo "Getting file from " $5
hdfs dfs -copyToLocal $5
echo "Updating shadow monitor database"
impala-shell -q "create database if not exists ${SHADOW}"
impala-shell -d ${SHADOW} -q "show tables" --delimited | sed "s/^/drop view if exists ${SHADOW}./" | sed "s/$/;/" | impala-shell -f -
impala-shell -d ${TARGET} -q "show tables" --delimited | sed "s/\(.*\)/create view ${SHADOW}.\1 as select * from ${TARGET}.\1;/" | impala-shell -f -
echo "Shadow db ready!"
echo "Getting file from " $6
hdfs dfs -copyToLocal $6
echo "Getting file from " $7
hdfs dfs -copyToLocal $7
echo "Getting file from " $8
hdfs dfs -copyToLocal $8
echo "Creating monitor database"
cat step20-createMonitorDB_funded.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2_funded/g1" > foo
hive $HIVE_OPTS -f foo
cat step20-createMonitorDB.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2_funded/g1" > foo
hive $HIVE_OPTS -f foo
#
cat step20-createMonitorDB_institutions.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2_institutions/g1" > foo
hive $HIVE_OPTS -f foo
cat step20-createMonitorDB.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2_institutions/g1" > foo
hive $HIVE_OPTS -f foo
contexts="knowmad::other dh-ch::other enermaps::other gotriple::other neanias-atmospheric::other rural-digital-europe::other covid-19::other aurora::other neanias-space::other north-america-studies::other north-american-studies::other eutopia::other"
for i in ${contexts}
do
tmp=`echo "$i" | sed 's/'-'/'_'/g' | sed 's/'::'/'_'/g'`
tmp2=`echo "$i" |sed 's/:.*//' `
cat step20-createMonitorDB_RIs.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2_$tmp/g1" | sed "s/CONTEXT/\'%$tmp2%\'/g" > foo
hive $HIVE_OPTS -f foo
cat step20-createMonitorDB.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2_$tmp/g1" > foo
hive $HIVE_OPTS -f foo
done
cat step20-createMonitorDB_RIs_tail.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2_RIs_tail/g1" | sed "s/CONTEXTS/\"'knowmad::other','dh-ch::other', 'enermaps::other', 'gotriple::other', 'neanias-atmospheric::other', 'rural-digital-europe::other', 'covid-19::other', 'aurora::other', 'neanias-space::other', 'north-america-studies::other', 'north-american-studies::other', 'eutopia::other'\"/g" > foo
hive $HIVE_OPTS -f foo
cat step20-createMonitorDB.sql | sed "s/SOURCE/$1/g" | sed "s/TARGET/$2_RIs_tail/g1" > foo
hive $HIVE_OPTS -f foo
echo "Hive shell finished"
echo "Updating shadow monitor funded database"
hive -e "drop database if exists ${SHADOW}_funded cascade"
hive -e "create database if not exists ${SHADOW}_funded"
hive $HIVE_OPTS --database ${2}_funded -e "show tables" | grep -v WARN | sed "s/\(.*\)/create view ${SHADOW}_funded.\1 as select * from ${2}_funded.\1;/" > foo
hive -f foo
echo "Updated shadow monitor funded database"
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!"
echo "Updating shadow monitor RIs database"
for i in $contexts
do
tmp=`echo "$i" | sed 's/'-'/'_'/g' | sed 's/'::'/'_'/g'`
hive -e "drop database if exists ${SHADOW}_${tmp} cascade"
hive -e "create database if not exists ${SHADOW}_${tmp}"
hive $HIVE_OPTS --database ${2}_${tmp} -e "show tables" | grep -v WARN | sed "s/\(.*\)/create view ${SHADOW}_${tmp}.\1 as select * from ${2}_${tmp}.\1;/" > foo
hive -f foo
done
echo "Shadow db monitor RIs ready!"
echo "Updating shadow monitor RIs tail database"
hive -e "drop database if exists ${SHADOW}_ris_tail cascade"
hive -e "create database if not exists ${SHADOW}_ris_tail"
hive $HIVE_OPTS --database ${2}_ris_tail -e "show tables" | grep -v WARN | sed "s/\(.*\)/create view ${SHADOW}_ris_tail.\1 as select * from ${2}_ris_tail.\1;/" > foo
hive -f foo
echo "Shadow db monitor RIs tail ready!"

View File

@ -7,15 +7,13 @@ then
fi
export SOURCE=$1
export TARGET=$2
export SHADOW=$3
export SHADOW=$2
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"
impala-shell -q "invalidate metadata;"
impala-shell -d ${TARGET} -q "show tables" --delimited | sed "s/\(.*\)/compute stats ${TARGET}.\1;/" | impala-shell -f -
echo "Impala shell finished"
echo "Updating shadow observatory database"
impala-shell -q "create database if not exists ${SHADOW}"
impala-shell -d ${SHADOW} -q "show tables" --delimited | sed "s/^/drop view if exists ${SHADOW}./" | sed "s/$/;/" | impala-shell -f -
impala-shell -d ${TARGET} -q "show tables" --delimited | sed "s/\(.*\)/create view ${SHADOW}.\1 as select * from ${TARGET}.\1;/" | impala-shell -f -
echo "Shadow db ready!"
echo "Updating shadow database"
hive -e "drop database if exists ${SHADOW} cascade"
hive -e "create database if not exists ${SHADOW}"
hive $HIVE_OPTS --database ${SOURCE} -e "show tables" | grep -v WARN | sed "s/\(.*\)/create view ${SHADOW}.\1 as select * from ${SOURCE}.\1;/" > foo
hive -f foo
echo "Updated shadow database"

View File

@ -10,7 +10,11 @@ export SOURCE=$1
export TARGET=$2
export SHADOW=$3
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 "Creating observatory database"
impala-shell -q "drop database if exists ${TARGET} cascade"
impala-shell -q "create database if not exists ${TARGET}"
impala-shell -d ${SOURCE} -q "show tables" --delimited | grep -iv roar | sed "s/\(.*\)/create view ${TARGET}.\1 as select * from ${SOURCE}.\1;/" | impala-shell -f -
hive -e "drop database if exists ${TARGET} cascade"
hive -e "create database if not exists ${TARGET}"
hive $HIVE_OPTS --database ${SOURCE} -e "show tables" | grep -v WARN | grep -iv roar | sed "s/\(.*\)/create view ${TARGET}.\1 as select * from ${SOURCE}.\1;/" > foo
hive -f foo

View File

@ -46,4 +46,8 @@ FROM (
LEFT OUTER JOIN (
SELECT substr(d.id, 4) id
from ${openaire_db_name}.datasource d
WHERE d.datainfo.deletedbyinference=false and d.datainfo.invisible = FALSE) d on o.datasource = d.id;
WHERE d.datainfo.deletedbyinference=false and d.datainfo.invisible = FALSE) d on o.datasource = d.id;
CREATE TABLE IF NOT EXISTS ${stats_db_name}.result_accessroute STORED AS PARQUET as
select distinct substr(id,4) as id, accessroute from ${openaire_db_name}.result
lateral view explode (instance.accessright.openaccessroute) openaccessroute as accessroute;

View File

@ -33,4 +33,12 @@ select * from ${stats_db_name}.dataset_refereed
union all
select * from ${stats_db_name}.software_refereed
union all
select * from ${stats_db_name}.otherresearchproduct_refereed;
select * from ${stats_db_name}.otherresearchproduct_refereed;
create table if not exists ${stats_db_name}.indi_impact_measures STORED AS PARQUET as
select substr(id, 4) as id, measures_ids.id impactmetric, cast(measures_ids.unit.value[0] as double) score,
cast(measures_ids.unit.value[0] as decimal(6,3)) score_dec, measures_ids.unit.value[1] class
from ${openaire_db_name}.result lateral view explode(measures) measures as measures_ids
where measures_ids.id!='views' and measures_ids.id!='downloads';
ANALYZE TABLE indi_impact_measures COMPUTE STATISTICS;

View File

@ -1,21 +1,21 @@
-------------------------------------------
--- Extra tables, mostly used by indicators
create table ${stats_db_name}.result_projectcount STORED AS PARQUET as
create table if not exists ${stats_db_name}.result_projectcount STORED AS PARQUET as
select r.id, count(distinct p.id) as count
from ${stats_db_name}.result r
left outer join ${stats_db_name}.result_projects rp on rp.id=r.id
left outer join ${stats_db_name}.project p on p.id=rp.project
group by r.id;
create table ${stats_db_name}.result_fundercount STORED AS PARQUET as
create table if not exists ${stats_db_name}.result_fundercount STORED AS PARQUET as
select r.id, count(distinct p.funder) as count
from ${stats_db_name}.result r
left outer join ${stats_db_name}.result_projects rp on rp.id=r.id
left outer join ${stats_db_name}.project p on p.id=rp.project
group by r.id;
create table ${stats_db_name}.project_resultcount STORED AS PARQUET as
create table if not exists ${stats_db_name}.project_resultcount STORED AS PARQUET as
with rcount as (
select p.id as pid, count(distinct r.id) as `count`, r.type as type
from ${stats_db_name}.project p
@ -29,17 +29,22 @@ select rcount.pid, sum(case when rcount.type='publication' then rcount.count els
from rcount
group by rcount.pid;
create view ${stats_db_name}.rndexpenditure as select * from stats_ext.rndexpediture;
create or replace view ${stats_db_name}.rndexpenditure as select * from stats_ext.rndexpediture;
create or replace view ${stats_db_name}.rndgdpexpenditure as select * from stats_ext.rndgdpexpenditure;
create or replace view ${stats_db_name}.doctoratestudents as select * from stats_ext.doctoratestudents;
create or replace view ${stats_db_name}.totalresearchers as select * from stats_ext.totalresearchers;
create or replace view ${stats_db_name}.totalresearchersft as select * from stats_ext.totalresearchersft;
create or replace view ${stats_db_name}.hrrst as select * from stats_ext.hrrst;
create table ${stats_db_name}.result_instance stored as parquet as
create table if not exists ${stats_db_name}.result_instance stored as parquet as
select distinct r.*
from (
select substr(r.id, 4) as id, inst.accessright.classname as accessright, substr(inst.collectedfrom.key, 4) as collectedfrom,
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
join ${stats_db_name}.result res on res.id=r.id;
create table ${stats_db_name}.result_apc as
create table if not exists ${stats_db_name}.result_apc STORED AS PARQUET as
select 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
@ -47,4 +52,4 @@ from (
join ${stats_db_name}.result res on res.id=r.id
where r.amount is not null;
create view ${stats_db_name}.issn_gold_oa_dataset as select * from stats_ext.issn_gold_oa_dataset;
create or replace view ${stats_db_name}.issn_gold_oa_dataset as select * from ${external_stats_db_name}.issn_gold_oa_dataset;

View File

@ -1,5 +1,78 @@
drop database if exists TARGET cascade;
create database if not exists TARGET;
--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;
@ -10,128 +83,90 @@ create view if not exists TARGET.creation_date as select * from SOURCE.creation_
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 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
) )) foo;
compute stats TARGET.result;
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_citations stored as parquet as select * from SOURCE.result_citations orig where exists (select 1 from TARGET.result r where r.id=orig.id);
compute stats TARGET.result_citations;
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);
compute stats TARGET.result_references_oc;
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);
compute stats TARGET.result_citations_oc;
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);
compute stats TARGET.result_classifications;
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);
compute stats TARGET.result_apc;
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);
compute stats TARGET.result_concepts;
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);
compute stats TARGET.result_datasources;
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);
compute stats TARGET.result_fundercount;
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);
compute stats TARGET.result_gold;
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);
compute stats TARGET.result_greenoa;
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);
compute stats TARGET.result_languages;
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);
compute stats TARGET.result_licenses;
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);
compute stats TARGET.result_oids;
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);
compute stats TARGET.result_organization;
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);
compute stats TARGET.result_peerreviewed;
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);
compute stats TARGET.result_pids;
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);
compute stats TARGET.result_projectcount;
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);
compute stats TARGET.result_projects;
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);
compute stats TARGET.result_refereed;
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);
compute stats TARGET.result_sources;
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);
compute stats TARGET.result_topics;
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);
compute stats TARGET.result_fos;
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 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;
compute stats TARGET.result_result;
ANALYZE TABLE TARGET.result_result COMPUTE STATISTICS;
-- datasources
create view if not exists TARGET.datasource as select * from SOURCE.datasource;
@ -140,7 +175,7 @@ create view if not exists TARGET.datasource_organizations as select * from SOURC
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;
compute stats TARGET.datasource_results;
ANALYZE TABLE TARGET.datasource_results COMPUTE STATISTICS;
-- organizations
create view if not exists TARGET.organization as select * from SOURCE.organization;
@ -155,30 +190,31 @@ create view if not exists TARGET.project_oids as select * from SOURCE.project_oi
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;
compute stats TARGET.project_results;
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);
compute stats TARGET.indi_pub_green_oa;
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);
compute stats TARGET.indi_pub_grey_lit;
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);
compute stats TARGET.indi_pub_doi_from_crossref;
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);
compute stats TARGET.indi_result_has_cc_licence;
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);
compute stats TARGET.indi_result_has_cc_licence_url;
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);
compute stats TARGET.indi_pub_has_abstract;
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);
compute stats TARGET.indi_result_with_orcid;
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);
compute stats TARGET.indi_funded_result_with_fundref;
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;
@ -187,30 +223,32 @@ create view TARGET.indi_funder_country_collab as select * from SOURCE.indi_funde
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);
compute stats TARGET.indi_pub_diamond;
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);
compute stats TARGET.indi_pub_in_transformative;
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);
compute stats TARGET.indi_pub_closed_other_open;
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);
compute stats TARGET.indi_result_no_of_copies;
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);
compute stats TARGET.indi_pub_hybrid_oa_with_cc;
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);
compute stats TARGET.indi_pub_downloads;
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);
compute stats TARGET.indi_pub_downloads_datasource;
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);
compute stats TARGET.indi_pub_downloads_year;
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);
compute stats TARGET.indi_pub_downloads_datasource_year;
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);
compute stats TARGET.indi_pub_gold_oa;
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);
compute stats TARGET.indi_pub_hybrid;
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;
@ -221,11 +259,12 @@ 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);
--create table TARGET.indi_datasets_gold_oa stored as parquet as select * from SOURCE.indi_datasets_gold_oa orig where exists (select 1 from TARGET.result r where r.id=orig.id);
--compute stats TARGET.indi_datasets_gold_oa;
--create table TARGET.indi_software_gold_oa stored as parquet as select * from SOURCE.indi_software_gold_oa orig where exists (select 1 from TARGET.result r where r.id=orig.id);
--compute stats TARGET.indi_software_gold_oa;
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;

View File

@ -0,0 +1,15 @@
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_concepts rc
join SOURCE.concept conc on conc.id=rc.concept
join SOURCE.category cat on cat.id=conc.category
join SOURCE.context cont on cont.id=cat.context
-- join SOURCE.result
where rc.id=r.id and conc.category like CONTEXT)
) foo;
ANALYZE TABLE TARGET.result COMPUTE STATISTICS;

View File

@ -0,0 +1,15 @@
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_concepts rc
join SOURCE.concept conc on conc.id=rc.concept
join SOURCE.category cat on cat.id=conc.category
join SOURCE.context cont on cont.id=cat.context
-- join SOURCE.result
where rc.id=r.id and conc.category not in (CONTEXTS))
) foo;
ANALYZE TABLE TARGET.result COMPUTE STATISTICS;

View File

@ -0,0 +1,9 @@
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_projects rp join SOURCE.project p on rp.project=p.id where rp.id=r.id)
) foo;
ANALYZE TABLE TARGET.result COMPUTE STATISTICS;

View File

@ -0,0 +1,58 @@
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
))) foo;
ANALYZE TABLE TARGET.result COMPUTE STATISTICS;

View File

@ -8,6 +8,8 @@ from ${stats_db_name}.result r
group by rl.id
) rln on rln.id=r.id;
ANALYZE TABLE ${observatory_db_name}.result_cc_licence COMPUTE STATISTICS;
create table ${observatory_db_name}.result_affiliated_country stored as parquet as
select
count(distinct r.id) as total,
@ -37,6 +39,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, c.code, c.name;
ANALYZE TABLE ${observatory_db_name}.result_affiliated_country COMPUTE STATISTICS;
create table ${observatory_db_name}.result_affiliated_year stored as parquet as
select
count(distinct r.id) as total,
@ -66,6 +70,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, r.year;
ANALYZE TABLE ${observatory_db_name}.result_affiliated_year COMPUTE STATISTICS;
create table ${observatory_db_name}.result_affiliated_year_country stored as parquet as
select
count(distinct r.id) as total,
@ -95,6 +101,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, r.year, c.code, c.name;
ANALYZE TABLE ${observatory_db_name}.result_affiliated_year_country COMPUTE STATISTICS;
create table ${observatory_db_name}.result_affiliated_datasource stored as parquet as
select
count(distinct r.id) as total,
@ -126,6 +134,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, d.name;
ANALYZE TABLE ${observatory_db_name}.result_affiliated_datasource COMPUTE STATISTICS;
create table ${observatory_db_name}.result_affiliated_datasource_country stored as parquet as
select
count(distinct r.id) as total,
@ -157,6 +167,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, d.name, c.code, c.name;
ANALYZE TABLE ${observatory_db_name}.result_affiliated_datasource_country COMPUTE STATISTICS;
create table ${observatory_db_name}.result_affiliated_organization stored as parquet as
select
count(distinct r.id) as total,
@ -186,6 +198,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, o.name;
ANALYZE TABLE ${observatory_db_name}.result_affiliated_organization COMPUTE STATISTICS;
create table ${observatory_db_name}.result_affiliated_organization_country stored as parquet as
select
count(distinct r.id) as total,
@ -215,6 +229,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, o.name, c.code, c.name;
ANALYZE TABLE ${observatory_db_name}.result_affiliated_organization_country COMPUTE STATISTICS;
create table ${observatory_db_name}.result_affiliated_funder stored as parquet as
select
count(distinct r.id) as total,
@ -246,6 +262,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, p.funder;
ANALYZE TABLE ${observatory_db_name}.result_affiliated_funder COMPUTE STATISTICS;
create table ${observatory_db_name}.result_affiliated_funder_country stored as parquet as
select
count(distinct r.id) as total,
@ -277,6 +295,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, p.funder, c.code, c.name;
ANALYZE TABLE ${observatory_db_name}.result_affiliated_funder_country COMPUTE STATISTICS;
create table ${observatory_db_name}.result_deposited_country stored as parquet as
select
count(distinct r.id) as total,
@ -308,6 +328,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, c.code, c.name;
ANALYZE TABLE ${observatory_db_name}.result_deposited_country COMPUTE STATISTICS;
create table ${observatory_db_name}.result_deposited_year stored as parquet as
select
count(distinct r.id) as total,
@ -339,6 +361,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, r.year;
ANALYZE TABLE ${observatory_db_name}.result_deposited_year COMPUTE STATISTICS;
create table ${observatory_db_name}.result_deposited_year_country stored as parquet as
select
count(distinct r.id) as total,
@ -370,6 +394,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, r.year, c.code, c.name;
ANALYZE TABLE ${observatory_db_name}.result_deposited_year_country COMPUTE STATISTICS;
create table ${observatory_db_name}.result_deposited_datasource stored as parquet as
select
count(distinct r.id) as total,
@ -401,6 +427,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, d.name;
ANALYZE TABLE ${observatory_db_name}.result_deposited_datasource COMPUTE STATISTICS;
create table ${observatory_db_name}.result_deposited_datasource_country stored as parquet as
select
count(distinct r.id) as total,
@ -432,6 +460,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, d.name, c.code, c.name;
ANALYZE TABLE ${observatory_db_name}.result_deposited_datasource_country COMPUTE STATISTICS;
create table ${observatory_db_name}.result_deposited_organization stored as parquet as
select
count(distinct r.id) as total,
@ -463,6 +493,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, o.name;
ANALYZE TABLE ${observatory_db_name}.result_deposited_organization COMPUTE STATISTICS;
create table ${observatory_db_name}.result_deposited_organization_country stored as parquet as
select
count(distinct r.id) as total,
@ -494,6 +526,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, o.name, c.code, c.name;
ANALYZE TABLE ${observatory_db_name}.result_deposited_organization_country COMPUTE STATISTICS;
create table ${observatory_db_name}.result_deposited_funder stored as parquet as
select
count(distinct r.id) as total,
@ -527,6 +561,8 @@ group by r.green, r.gold, case when rl.type is not null then true else false end
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, p.funder;
ANALYZE TABLE ${observatory_db_name}.result_deposited_funder COMPUTE STATISTICS;
create table ${observatory_db_name}.result_deposited_funder_country stored as parquet as
select
count(distinct r.id) as total,
@ -558,4 +594,6 @@ from ${stats_db_name}.result r
left outer join ${stats_db_name}.result_fundercount rfc on rfc.id=r.id
group by r.green, r.gold, case when rl.type is not null then true else false end, case when pids.pid is not null then true else false end,
case when r.access_mode in ('Open Access', 'Open Source') then true else false end, r.peer_reviewed, r.type, abstract,
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, p.funder, c.code, c.name;
cc_licence, r.authors > 1, rpc.count > 1, rfc.count > 1, p.funder, c.code, c.name;
ANALYZE TABLE ${observatory_db_name}.result_deposited_funder_country COMPUTE STATISTICS;

View File

@ -84,4 +84,12 @@ 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
from ${openaire_db_name}.project p lateral view explode(p.fundingtree.value) fundingtree as fund;
from ${openaire_db_name}.project p lateral view explode(p.fundingtree.value) fundingtree as fund;
CREATE TABLE ${stats_db_name}.project_organization_contribution STORED AS PARQUET AS
SELECT distinct substr(r.source, 4) AS project, substr(r.target, 4) AS organization,
properties[0].value contribution, properties[1].value currency
from ${openaire_db_name}.relation r
LATERAL VIEW explode (r.properties) properties
where properties[0].key='contribution' and r.reltype = 'projectOrganization' and r.source like '40|%'
and properties[0].value>0.0 and r.datainfo.deletedbyinference = false and r.datainfo.invisible=false;

View File

@ -48,12 +48,10 @@ WHERE d1.datainfo.deletedbyinference = FALSE and d1.datainfo.invisible=false;
-- Updating temporary table with everything that is not based on results -> This is done with the following "dual" table.
-- Creating a temporary dual table that will be removed after the following insert
CREATE TABLE ${stats_db_name}.dual
(
dummy CHAR(1)
);
INSERT INTO ${stats_db_name}.dual
VALUES ('X');
CREATE TABLE ${stats_db_name}.dual ( dummy CHAR(1));
INSERT INTO ${stats_db_name}.dual VALUES ('X');
INSERT INTO ${stats_db_name}.datasource_tmp (`id`, `name`, `type`, `dateofvalidation`, `yearofvalidation`, `harvested`,
`piwik_id`, `latitude`, `longitude`, `websiteurl`, `compatibility`, `issn_printed`, `issn_online`)
SELECT 'other',
@ -73,12 +71,8 @@ FROM ${stats_db_name}.dual
WHERE 'other' not in (SELECT id FROM ${stats_db_name}.datasource_tmp WHERE name = 'Unknown Repository');
DROP TABLE ${stats_db_name}.dual;
UPDATE ${stats_db_name}.datasource_tmp
SET name='Other'
WHERE name = 'Unknown Repository';
UPDATE ${stats_db_name}.datasource_tmp
SET yearofvalidation=null
WHERE yearofvalidation = '-1';
UPDATE ${stats_db_name}.datasource_tmp SET name='Other' WHERE name = 'Unknown Repository';
UPDATE ${stats_db_name}.datasource_tmp SET yearofvalidation=null WHERE yearofvalidation = '-1';
CREATE TABLE ${stats_db_name}.datasource_languages STORED AS PARQUET AS
SELECT substr(d.id, 4) AS id, langs.languages AS language
@ -104,4 +98,4 @@ where d.datainfo.deletedbyinference = false and d.datainfo.invisible=false;
CREATE OR REPLACE VIEW ${stats_db_name}.datasource_results AS
SELECT datasource AS id, id AS result
FROM ${stats_db_name}.result_datasources;
FROM ${stats_db_name}.result_datasources;

View File

@ -1,4 +1,4 @@
<workflow-app name="Graph Stats" xmlns="uri:oozie:workflow:0.5">
<workflow-app name="Graph Stats Hive" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>stats_db_name</name>
@ -10,9 +10,12 @@
</property>
<property>
<name>external_stats_db_name</name>
<value>stats_ext</value>
<description>the external stats that should be added since they are not included in the graph database</description>
</property>
<property>
<name>usage_stats_db_name</name>
<description>the usage statistics database name</description>
</property>
<property>
<name>stats_db_shadow_name</name>
<description>the name of the shadow schema</description>
@ -33,6 +36,10 @@
<name>observatory_db_shadow_name</name>
<description>the name of the shadow monitor db</description>
</property>
<property>
<name>usage_stats_db_shadow_name</name>
<description>the name of the shadow usage stats db</description>
</property>
<property>
<name>stats_tool_api_url</name>
<description>The url of the API of the stats tool. Is used to trigger the cache update.</description>
@ -53,6 +60,10 @@
<name>context_api_url</name>
<description>the base url of the context api (https://services.openaire.eu/openaire)</description>
</property>
<property>
<name>hadoop_user_name</name>
<description>user name of the wf owner</description>
</property>
</parameters>
<global>
@ -67,10 +78,47 @@
<name>hive.txn.timeout</name>
<value>${hive_timeout}</value>
</property>
<property>
<name>mapred.job.queue.name</name>
<value>analytics</value>
</property>
</configuration>
</global>
<start to="Step1"/>
<start to="resume_from"/>
<decision name="resume_from">
<switch>
<case to="Step1">${wf:conf('resumeFrom') eq 'Step1'}</case>
<case to="Step2">${wf:conf('resumeFrom') eq 'Step2'}</case>
<case to="Step3">${wf:conf('resumeFrom') eq 'Step3'}</case>
<case to="Step4">${wf:conf('resumeFrom') eq 'Step4'}</case>
<case to="Step5">${wf:conf('resumeFrom') eq 'Step5'}</case>
<case to="Step6">${wf:conf('resumeFrom') eq 'Step6'}</case>
<case to="Step7">${wf:conf('resumeFrom') eq 'Step7'}</case>
<case to="Step8">${wf:conf('resumeFrom') eq 'Step8'}</case>
<case to="Step9">${wf:conf('resumeFrom') eq 'Step9'}</case>
<case to="Step10">${wf:conf('resumeFrom') eq 'Step10'}</case>
<case to="Step11">${wf:conf('resumeFrom') eq 'Step11'}</case>
<case to="Step12">${wf:conf('resumeFrom') eq 'Step12'}</case>
<case to="Step13">${wf:conf('resumeFrom') eq 'Step13'}</case>
<case to="Step14">${wf:conf('resumeFrom') eq 'Step14'}</case>
<case to="Step15">${wf:conf('resumeFrom') eq 'Step15'}</case>
<case to="Step15_5">${wf:conf('resumeFrom') eq 'Step15_5'}</case>
<case to="Contexts">${wf:conf('resumeFrom') eq 'Contexts'}</case>
<case to="Step16-createIndicatorsTables">${wf:conf('resumeFrom') eq 'Step16-createIndicatorsTables'}</case>
<case to="Step16_1-definitions">${wf:conf('resumeFrom') eq 'Step16_1-definitions'}</case>
<case to="Step16_5">${wf:conf('resumeFrom') eq 'Step16_5'}</case>
<case to="Step19-finalize">${wf:conf('resumeFrom') eq 'Step19-finalize'}</case>
<case to="step20-createMonitorDB">${wf:conf('resumeFrom') eq 'step20-createMonitorDB'}</case>
<case to="step21-createObservatoryDB-pre">${wf:conf('resumeFrom') eq 'step21-createObservatoryDB-pre'}</case>
<case to="step21-createObservatoryDB">${wf:conf('resumeFrom') eq 'step21-createObservatoryDB'}</case>
<case to="step21-createObservatoryDB-post">${wf:conf('resumeFrom') eq 'step21-createObservatoryDB-post'}</case>
<case to="step22-copyDataToImpalaCluster">${wf:conf('resumeFrom') eq 'step22-copyDataToImpalaCluster'}</case>
<case to="step23-finalizeImpalaCluster">${wf:conf('resumeFrom') eq 'step23-finalizeImpalaCluster'}</case>
<case to="Step24-updateCache">${wf:conf('resumeFrom') eq 'Step24-updateCache'}</case>
<default to="Step1"/>
</switch>
</decision>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
@ -249,6 +297,7 @@
<script>scripts/step15_5.sql</script>
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
<param>external_stats_db_name=${external_stats_db_name}</param>
</hive2>
<ok to="Contexts"/>
<error to="Kill"/>
@ -273,6 +322,7 @@
<name-node>${nameNode}</name-node>
<exec>indicators.sh</exec>
<argument>${stats_db_name}</argument>
<argument>${external_stats_db_name}</argument>
<argument>${wf:appPath()}/scripts/step16-createIndicatorsTables.sql</argument>
<file>indicators.sh</file>
</shell>
@ -301,7 +351,7 @@
<ok to="Step19-finalize"/>
<error to="Kill"/>
</action>
<action name="Step19-finalize">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
@ -324,12 +374,29 @@
<argument>${monitor_db_name}</argument>
<argument>${monitor_db_shadow_name}</argument>
<argument>${wf:appPath()}/scripts/step20-createMonitorDB.sql</argument>
<argument>${wf:appPath()}/scripts/step20-createMonitorDB_funded.sql</argument>
<argument>${wf:appPath()}/scripts/step20-createMonitorDB_institutions.sql</argument>
<argument>${wf:appPath()}/scripts/step20-createMonitorDB_RIs.sql</argument>
<argument>${wf:appPath()}/scripts/step20-createMonitorDB_RIs_tail.sql</argument>
<file>monitor.sh</file>
</shell>
<ok to="step21-createObservatoryDB-pre"/>
<error to="Kill"/>
</action>
<!-- <action name="step20-createMonitorDB-post">-->
<!-- <shell xmlns="uri:oozie:shell-action:0.1">-->
<!-- <job-tracker>${jobTracker}</job-tracker>-->
<!-- <name-node>${nameNode}</name-node>-->
<!-- <exec>monitor-post.sh</exec>-->
<!-- <argument>${monitor_db_name}</argument>-->
<!-- <argument>${monitor_db_shadow_name}</argument>-->
<!-- <file>monitor-post.sh</file>-->
<!-- </shell>-->
<!-- <ok to="step21-createObservatoryDB-pre"/>-->
<!-- <error to="Kill"/>-->
<!-- </action>-->
<action name="step21-createObservatoryDB-pre">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
@ -360,16 +427,53 @@
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>observatory-post.sh</exec>
<argument>${stats_db_name}</argument>
<argument>${observatory_db_name}</argument>
<argument>${observatory_db_shadow_name}</argument>
<file>observatory-post.sh</file>
</shell>
<ok to="Step22"/>
<ok to="step22-copyDataToImpalaCluster"/>
<error to="Kill"/>
</action>
<action name="Step22">
<action name="step22-copyDataToImpalaCluster">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>copyDataToImpalaCluster.sh</exec>
<!-- <env-var>HADOOP_USER_NAME=${wf:user()}</env-var>-->
<!-- <argument>${external_stats_db_name}</argument>-->
<argument>${stats_db_name}</argument>
<argument>${monitor_db_name}</argument>
<argument>${observatory_db_name}</argument>
<argument>${external_stats_db_name}</argument>
<argument>${usage_stats_db_name}</argument>
<argument>${hadoop_user_name}</argument>
<file>copyDataToImpalaCluster.sh</file>
</shell>
<ok to="step23-finalizeImpalaCluster"/>
<error to="Kill"/>
</action>
<action name="step23-finalizeImpalaCluster">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>finalizeImpalaCluster.sh</exec>
<argument>${stats_db_name}</argument>
<argument>${stats_db_shadow_name}</argument>
<argument>${monitor_db_name}</argument>
<argument>${monitor_db_shadow_name}</argument>
<argument>${observatory_db_name}</argument>
<argument>${observatory_db_shadow_name}</argument>
<argument>${usage_stats_db_name}</argument>
<argument>${usage_stats_db_shadow_name}</argument>
<file>finalizeImpalaCluster.sh</file>
</shell>
<ok to="Step24-updateCache"/>
<error to="Kill"/>
</action>
<action name="Step24-updateCache">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
@ -382,4 +486,4 @@
</action>
<end name="End"/>
</workflow-app>
</workflow-app>