dnet-hadoop/dhp-workflows/dhp-usage-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/usagestatsupdate/oozie_app/scripts/Step6.sql

14 lines
583 B
SQL

--Episciences log
DROP TABLE IF EXISTS ${usageStatsDB}.episcienceslogdistinct;
CREATE TABLE IF NOT EXISTS
${usageStatsDB}.episcienceslogdistinct(source INT, id_visit STRING,
country STRING, action STRING, url STRING,
entity_id STRING, source_item_type STRING, timestamp STRING, referrer_name STRING, agent STRING)
clustered by (source, id_visit, action, timestamp, entity_id)
into 100 buckets stored as orc tblproperties('transactional'='true');
INSERT INTO ${usageStatsDB}.episcienceslogdistinct
SELECT DISTINCT * FROM ${usageStatsDB}.episcienceslog WHERE entity_id is not null;