D-Net
/
dnet-hadoop
14
5
Fork 11
Code Issues 14 Pull Requests 13 Projects 2 Releases 2 Wiki Activity

fix a bug with organization countries

This commit is contained in:
Michele Artini 2020-03-24 08:43:56 +01:00
parent 15160032bd
commit e3760c7f39
4 changed files with 83 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/migration/sql/queryOrganizations.sql
View File

@ -22,8 +22,7 @@ SELECT
'' AS inferenceprovenance, '' AS inferenceprovenance,
d.id AS collectedfromid, d.id AS collectedfromid,
d.officialname AS collectedfromname, d.officialname AS collectedfromname,
o.country || '@@@' || o.country || '@@@dnet:countries@@@dnet:countries' AS country,
o.country || '@@@dnet:countries' AS country,
'sysimport:crosswalk:entityregistry@@@sysimport:crosswalk:entityregistry@@@dnet:provenance_actions@@@dnet:provenance_actions' AS provenanceaction, 'sysimport:crosswalk:entityregistry@@@sysimport:crosswalk:entityregistry@@@dnet:provenance_actions@@@dnet:provenance_actions' AS provenanceaction,
ARRAY[]::text[] AS pid ARRAY[]::text[] AS pid

4
dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/migration/sql/queryOrganizationsFromOpenOrgsDB.sql
View File

@ -11,7 +11,7 @@ SELECT
'' AS inferenceprovenance, '' AS inferenceprovenance,
'openaire____::openorgs' AS collectedfromid, 'openaire____::openorgs' AS collectedfromid,
'OpenOrgs Database' AS collectedfromname, 'OpenOrgs Database' AS collectedfromname,
o.country || '@@@dnet:countries' AS country, o.country || '@@@' || o.country || '@@@dnet:countries@@@dnet:countries' AS country,
'sysimport:crosswalk:entityregistry@@@sysimport:crosswalk:entityregistry@@@dnet:provenance_actions@@@dnet:provenance_actions' AS provenanceaction, 'sysimport:crosswalk:entityregistry@@@sysimport:crosswalk:entityregistry@@@dnet:provenance_actions@@@dnet:provenance_actions' AS provenanceaction,
array_agg(DISTINCT i.otherid || '###' || i.type || '@@@dnet:pid_types') AS pid array_agg(DISTINCT i.otherid || '###' || i.type || '@@@dnet:pid_types') AS pid
FROM organizations o FROM organizations o
@ -40,7 +40,7 @@ SELECT
'' AS inferenceprovenance, '' AS inferenceprovenance,
'openaire____::openorgs' AS collectedfromid, 'openaire____::openorgs' AS collectedfromid,
'OpenOrgs Database' AS collectedfromname, 'OpenOrgs Database' AS collectedfromname,
o.country || '@@@dnet:countries' AS country, o.country || '@@@' || o.country || '@@@dnet:countries@@@dnet:countries' AS country,
'sysimport:crosswalk:entityregistry@@@sysimport:crosswalk:entityregistry@@@dnet:provenance_actions@@@dnet:provenance_actions' AS provenanceaction, 'sysimport:crosswalk:entityregistry@@@sysimport:crosswalk:entityregistry@@@dnet:provenance_actions@@@dnet:provenance_actions' AS provenanceaction,
array_agg(DISTINCT i.otherid || '###' || i.type || '@@@dnet:pid_types') AS pid array_agg(DISTINCT i.otherid || '###' || i.type || '@@@dnet:pid_types') AS pid
FROM other_names n FROM other_names n

18
dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/migration/wfs/regular_step1_onlydb/oozie_app/config-default.xml Normal file
View File

@ -0,0 +1,18 @@
<configuration>
<property>
<name>jobTracker</name>
<value>yarnRM</value>
</property>
<property>
<name>nameNode</name>
<value>hdfs://nameservice1</value>
</property>
<property>
<name>oozie.use.system.libpath</name>
<value>true</value>
</property>
<property>
<name>oozie.action.sharelib.for.spark</name>
<value>spark2</value>
</property>
</configuration>

62
dhp-workflows/dhp-aggregation/src/main/resources/eu/dnetlib/dhp/migration/wfs/regular_step1_onlydb/oozie_app/workflow.xml Normal file
View File

@ -0,0 +1,62 @@
<workflow-app name="import regular entities as Graph (step 1)" xmlns="uri:oozie:workflow:0.5">
<parameters>
<property>
<name>migrationPathStep1</name>
<description>the base path to store hdfs file</description>
</property>
<property>
<name>postgresURL</name>
<description>the postgres URL to access to the database</description>
</property>
<property>
<name>postgresUser</name>
<description>the user postgres</description>
</property>
<property>
<name>postgresPassword</name>
<description>the password postgres</description>
</property>
<property>
<name>sparkDriverMemory</name>
<description>memory for driver process</description>
</property>
<property>
<name>sparkExecutorMemory</name>
<description>memory for individual executor</description>
</property>
<property>
<name>sparkExecutorCores</name>
<description>number of cores used by single executor</description>
</property>
</parameters>
<start to="ResetWorkingPath"/>
<kill name="Kill">
<message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message>
</kill>
<action name="ResetWorkingPath">
<fs>
<delete path='${migrationPathStep1}/db_records'/>
</fs>
<ok to="ImportDB"/>
<error to="Kill"/>
</action>
<action name="ImportDB">
<java>
<job-tracker>${jobTracker}</job-tracker>
<name-node>${nameNode}</name-node>
<main-class>eu.dnetlib.dhp.migration.step1.MigrateDbEntitiesApplication</main-class>
<arg>-p</arg><arg>${migrationPathStep1}/db_records</arg>
<arg>-pgurl</arg><arg>${postgresURL}</arg>
<arg>-pguser</arg><arg>${postgresUser}</arg>
<arg>-pgpasswd</arg><arg>${postgresPassword}</arg>
</java>
<ok to="End"/>
<error to="Kill"/>
</action>
<end name="End"/>
</workflow-app>