D-Net
/
dnet-hadoop
13
4
Fork 10
Code Issues 14 Pull Requests 8 Projects 2 Releases 1 Wiki Activity

Changes to execute the stats wf only in hive

This commit is contained in:
dimitrispie 2023-01-04 11:39:01 +02:00
parent 2a4bf32d4c
commit dcb958e146
11 changed files with 472 additions and 554 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/contexts.sh
View File

@ -31,8 +31,8 @@ hdfs dfs -copyFromLocal categories.csv ${TMP}
hdfs dfs -copyFromLocal concepts.csv ${TMP}
hdfs dfs -chmod -R 777 ${TMP}
export HADOOP_USER="antonis.lempesis"
export HADOOP_USER_NAME="antonis.lempesis"
export HADOOP_USER="dimitris.pierrakos"
export HADOOP_USER_NAME="dimitris.pierrakos"
echo "Creating and populating impala tables"
hive $HIVE_OPTS -e "create table ${TARGET_DB}.context (id string, name string) row format delimited fields terminated by ','"

6
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/finalizedb.sh
View File

@ -8,7 +8,9 @@ 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 --database ${SOURCE} -e "show tables" | grep -v WARN | sed "s/^\(.*\)/analyze table ${SOURCE}.\1 compute statistics;/" > foo
hive -f foo
hive $HIVE_OPTS --database ${SOURCE} -e "show tables" | grep -v WARN | sed "s/^\(.*\)/analyze table ${SOURCE}.\1 compute statistics;/" > foo
hive $HIVE_OPTS -f foo

4
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/indicators.sh
View File

@ -8,8 +8,8 @@ fi
export TARGET=$1
export SCRIPT_PATH=$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"
export HADOOP_USER="antonis.lempesis"
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 " $SCRIPT_PATH
hdfs dfs -copyToLocal $SCRIPT_PATH

15
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/monitor.sh
View File

@ -11,10 +11,15 @@ export TARGET=$2
export SHADOW=$3
export SCRIPT_PATH=$4
echo "Getting file from " $4
hdfs dfs -copyToLocal $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 " $SCRIPT_PATH
hdfs dfs -copyToLocal $SCRIPT_PATH
echo "Creating monitor database"
cat step20-createMonitorDB.sql | sed s/SOURCE/$1/g | sed s/TARGET/$2/g1 > foo
hive -f foo
echo "Impala shell finished"
#cat step20-createMonitorDB.sql | sed s/SOURCE/$1/g | sed s/TARGET/$2/g1 > foo
cat step20-createMonitorDB.sql | sed "s/TARGET/${TARGET}/g" | sed "s/SOURCE/${SOURCE}/g" > foo
hive $HIVE_OPTS -f foo
echo "Hive shell finished"

2
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/observatory-post.sh
View File

@ -12,4 +12,4 @@ export SHADOW=$3
hive --database ${TARGET} -e "show tables" | grep -v WARN | sed "s/\(.*\)/analyze table ${TARGET}.\1 compute statistics;/" > foo
hive -f foo
echo "Impala shell finished"
echo "Hive shell finished"

11
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step15_5.sql
View File

@ -29,6 +29,13 @@ 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 view ${stats_db_name}.rndgdpexpenditure as select * from stats_ext.rndgdpexpenditure;
create view ${stats_db_name}.doctoratestudents as select * from stats_ext.doctoratestudents;
create view ${stats_db_name}.totalresearchers as select * from stats_ext.totalresearchers;
create view ${stats_db_name}.totalresearchersft as select * from stats_ext.totalresearchersft;
create view ${stats_db_name}.hrrst as select * from stats_ext.hrrst;
create table ${stats_db_name}.result_instance stored as parquet as
select distinct r.*
from (
@ -43,4 +50,6 @@ 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
join ${stats_db_name}.result res on res.id=r.id
where r.amount is not null;
where r.amount is not null;
create view ${stats_db_name}.issn_gold_oa_dataset as select * from stats_ext.issn_gold_oa_dataset;

693
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16-createIndicatorsTables.sql
View File

File diff suppressed because it is too large Load Diff

105
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step20-createMonitorDB.sql
View File

@ -10,6 +10,11 @@ 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 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 (
@ -54,84 +59,87 @@ create table TARGET.result stored as parquet as
'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____::db7686f30f22cbe73a4fde872ce812a6', -- University of Milan
'openorgs____::b8b8ca674452579f3f593d9f5e557483' -- University College Cork
) )) foo;
compute stats TARGET.result;
ANALYZE TABLE TARGET.result COMPUTE STATISTICS;
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 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 +148,7 @@ 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;
@ -157,28 +165,28 @@ create view if not exists TARGET.project_resultcount as select * from SOURCE.pro
create view if not exists TARGET.project_classification as select * from SOURCE.project_classification;
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 +195,30 @@ 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_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 +229,12 @@ 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);
ANALYZE TABLE TARGET.indi_result_with_pid COMPUTE STATISTICS;
--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;

40
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step21-createObservatoryDB.sql
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;

2
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step8.sql
View File

@ -96,6 +96,6 @@ select substr(d.id, 4) as id, substr(cf.key, 4) as datasource
from ${openaire_db_name}.datasource d lateral view explode(d.collectedfrom) cfrom as cf
where d.datainfo.deletedbyinference = false and d.datainfo.invisible=false;
CREATE OR REPLACE VIEW ${stats_db_name}.datasource_results STORED AS PARQUET AS
CREATE OR REPLACE VIEW ${stats_db_name}.datasource_results AS
SELECT datasource AS id, id AS result
FROM ${stats_db_name}.result_datasources;

144
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/workflow.xml
View File

@ -74,7 +74,7 @@
</configuration>
</global>
<start to="Contexts"/>
<start to="Step1"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
@ -302,22 +302,22 @@
<param>stats_db_name=${stats_db_name}</param>
<param>openaire_db_name=${openaire_db_name}</param>
</hive2>
<ok to="Step19-finalize"/>
<error to="Kill"/>
</action>
<action name="Step19-finalize">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<exec>finalizedb.sh</exec>
<argument>${stats_db_name}</argument>
<argument>${stats_db_shadow_name}</argument>
<file>finalizedb.sh</file>
</shell>
<ok to="step20-createMonitorDB"/>
<error to="Kill"/>
</action>
<!-- <action name="Step19-finalize">-->
<!-- <shell xmlns="uri:oozie:shell-action:0.1">-->
<!-- <job-tracker>${jobTracker}</job-tracker>-->
<!-- <name-node>${nameNode}</name-node>-->
<!-- <exec>finalizedb.sh</exec>-->
<!-- <argument>${stats_db_name}</argument>-->
<!-- <argument>${stats_db_shadow_name}</argument>-->
<!-- <file>finalizedb.sh</file>-->
<!-- </shell>-->
<!-- <ok to="step20-createMonitorDB"/>-->
<!-- <error to="Kill"/>-->
<!-- </action>-->
<action name="step20-createMonitorDB">
<shell xmlns="uri:oozie:shell-action:0.1">
@ -355,67 +355,67 @@
<param>stats_db_name=${stats_db_name}</param>
<param>observatory_db_name=${observatory_db_name}</param>
</hive2>
<ok to="step21-createObservatoryDB-post"/>
<error to="Kill"/>
</action>
<action name="step21-createObservatoryDB-post">
<shell xmlns="uri:oozie:shell-action:0.1">
<job-tracker>${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-copyDataToImpalaCluster"/>
<error to="Kill"/>
</action>
<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>
<argument>${external_stats_db_name}</argument>
<argument>${stats_db_name}</argument>
<argument>${monitor_db_name}</argument>
<argument>${observatory_db_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>
<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>
<exec>updateCache.sh</exec>
<argument>${stats_tool_api_url}</argument>
<file>updateCache.sh</file>
</shell>
<ok to="End"/>
<error to="Kill"/>
</action>
<!-- <action name="step21-createObservatoryDB-post">-->
<!-- <shell xmlns="uri:oozie:shell-action:0.1">-->
<!-- <job-tracker>${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-copyDataToImpalaCluster"/>-->
<!-- <error to="Kill"/>-->
<!-- </action>-->
<!-- <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>-->
<!-- <argument>${external_stats_db_name}</argument>-->
<!-- <argument>${stats_db_name}</argument>-->
<!-- <argument>${monitor_db_name}</argument>-->
<!-- <argument>${observatory_db_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>-->
<!-- <file>finalizeImpalaCluster.sh</file>-->
<!-- </shell>-->
<!-- <ok to="End"/>-->
<!-- <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>-->
<!-- <exec>updateCache.sh</exec>-->
<!-- <argument>${stats_tool_api_url}</argument>-->
<!-- <file>updateCache.sh</file>-->
<!-- </shell>-->
<!-- <ok to="End"/>-->
<!-- <error to="Kill"/>-->
<!-- </action>-->
<end name="End"/>
</workflow-app>