antonis.lempesis
/
dnet-hadoop
forked from D-Net/dnet-hadoop
1
1
Fork 1
Code Issues Pull Requests Projects Releases Wiki Activity

Merge branch 'hive' of https://code-repo.d4science.org/antonis.lempesis/dnet-hadoop into hive 

# Conflicts:
#	dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step10.sql
#	dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step13.sql
#	dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step14.sql
#	dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step16_1-definitions.sql
#	dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step7.sql
This commit is contained in:
dimitrispie 2022-12-22 10:22:46 +02:00
parent c8309fe18e 6fc9ef53f6
commit 2a4bf32d4c
1 changed files with 7 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
dhp-workflows/dhp-stats-update/src/main/resources/eu/dnetlib/dhp/oa/graph/stats/oozie_app/scripts/step8.sql
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. -- 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 -- Creating a temporary dual table that will be removed after the following insert
CREATE TABLE ${stats_db_name}.dual CREATE TABLE ${stats_db_name}.dual ( dummy CHAR(1));
(
dummy CHAR(1) INSERT INTO ${stats_db_name}.dual VALUES ('X');
);
INSERT INTO ${stats_db_name}.dual
VALUES ('X');
INSERT INTO ${stats_db_name}.datasource_tmp (`id`, `name`, `type`, `dateofvalidation`, `yearofvalidation`, `harvested`, INSERT INTO ${stats_db_name}.datasource_tmp (`id`, `name`, `type`, `dateofvalidation`, `yearofvalidation`, `harvested`,
`piwik_id`, `latitude`, `longitude`, `websiteurl`, `compatibility`, `issn_printed`, `issn_online`) `piwik_id`, `latitude`, `longitude`, `websiteurl`, `compatibility`, `issn_printed`, `issn_online`)
SELECT 'other', 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'); WHERE 'other' not in (SELECT id FROM ${stats_db_name}.datasource_tmp WHERE name = 'Unknown Repository');
DROP TABLE ${stats_db_name}.dual; DROP TABLE ${stats_db_name}.dual;
UPDATE ${stats_db_name}.datasource_tmp UPDATE ${stats_db_name}.datasource_tmp SET name='Other' WHERE name = 'Unknown Repository';
SET name='Other' UPDATE ${stats_db_name}.datasource_tmp SET yearofvalidation=null WHERE yearofvalidation = '-1';
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 CREATE TABLE ${stats_db_name}.datasource_languages STORED AS PARQUET AS
SELECT substr(d.id, 4) AS id, langs.languages AS language SELECT substr(d.id, 4) AS id, langs.languages AS language