From e5abbec2ba6e312bb2ef790c6f2e196774d6ebe5 Mon Sep 17 00:00:00 2001
From: Claudio Atzori <claudio.atzori@isti.cnr.it>
Date: Thu, 22 Apr 2021 11:22:10 +0200
Subject: [PATCH] [orcid] download of the lambda file defined in a script

---
 .../oozie_app/get_orcid_lambda_file.sh                |  3 +++
 .../orcid_updates_download/oozie_app/workflow.xml     | 11 ++---------
 2 files changed, 5 insertions(+), 9 deletions(-)
 create mode 100644 dhp-workflows/dhp-doiboost/src/main/resources/eu/dnetlib/dhp/doiboost/orcid_updates_download/oozie_app/get_orcid_lambda_file.sh

diff --git a/dhp-workflows/dhp-doiboost/src/main/resources/eu/dnetlib/dhp/doiboost/orcid_updates_download/oozie_app/get_orcid_lambda_file.sh b/dhp-workflows/dhp-doiboost/src/main/resources/eu/dnetlib/dhp/doiboost/orcid_updates_download/oozie_app/get_orcid_lambda_file.sh
new file mode 100644
index 000000000..4f6bb8610
--- /dev/null
+++ b/dhp-workflows/dhp-doiboost/src/main/resources/eu/dnetlib/dhp/doiboost/orcid_updates_download/oozie_app/get_orcid_lambda_file.sh
@@ -0,0 +1,3 @@
+wget -O /tmp/last_modified.csv.tar http://74804fb637bd8e2fba5b-e0a029c2f87486cddec3b416996a6057.r3.cf1.rackcdn.com/last_modified.csv.tar
+hdfs dfs -copyFromLocal /tmp/last_modified.csv.tar /data/orcid_activities_2020/last_modified.csv.tar
+rm -f /tmp/last_modified.csv.tar
\ No newline at end of file
diff --git a/dhp-workflows/dhp-doiboost/src/main/resources/eu/dnetlib/dhp/doiboost/orcid_updates_download/oozie_app/workflow.xml b/dhp-workflows/dhp-doiboost/src/main/resources/eu/dnetlib/dhp/doiboost/orcid_updates_download/oozie_app/workflow.xml
index fa161ad35..72a220ef1 100644
--- a/dhp-workflows/dhp-doiboost/src/main/resources/eu/dnetlib/dhp/doiboost/orcid_updates_download/oozie_app/workflow.xml
+++ b/dhp-workflows/dhp-doiboost/src/main/resources/eu/dnetlib/dhp/doiboost/orcid_updates_download/oozie_app/workflow.xml
@@ -24,12 +24,6 @@
             <name>token</name>
             <description>access token</description>
         </property>
-        <property>
-            <name>shell_cmd</name>
-            <value>wget -O /tmp/last_modified.csv.tar http://74804fb637bd8e2fba5b-e0a029c2f87486cddec3b416996a6057.r3.cf1.rackcdn.com/last_modified.csv.tar ; hdfs dfs -copyFromLocal /tmp/last_modified.csv.tar /data/orcid_activities_2020/last_modified.csv.tar ; rm -f /tmp/last_modified.csv.tar
-            </value>
-            <description>the shell command that downloads the lambda file from orcid containing last orcid update informations</description>
-        </property>
         <property>
             <name>sparkDriverMemory</name>
             <value>7G</value>
@@ -103,9 +97,8 @@
         <shell xmlns="uri:oozie:shell-action:0.1">
             <job-tracker>${jobTracker}</job-tracker>
             <name-node>${nameNode}</name-node>
-            <exec>bash</exec>
-            <argument>-c</argument>
-            <argument>${shell_cmd}</argument>
+            <exec>get_orcid_lambda_file.sh</exec>
+            <file>get_orcid_lambda_file.sh</file>
             <capture-output/>
         </shell>
         <ok to="GenLastModifiedSeq"/>