D-Net

<?xml version="1.0"?>

<!-- Note that documentation placed in comments in this file uses the 

"markdown" syntax (along with its way of dividing text into sections). -->

<workflow-app xmlns="uri:oozie:workflow:0.3" name="test-core_examples_hadoopstreaming_cloner_without_reducer_with_explicit_schema_file">

	<start to="data_producer" />

	<action name="data_producer">

		<java>

			<job-tracker>${jobTracker}</job-tracker>

			<name-node>${nameNode}</name-node>

			<!-- The data generated by this node is deleted in this section -->

			<prepare>

				<delete path="${nameNode}${workingDir}/data_producer" />

				<mkdir path="${nameNode}${workingDir}/data_producer" />

			</prepare>

			<configuration>

				<property>

					<name>mapred.job.queue.name</name>

					<value>${queueName}</value>

				</property>

			</configuration>

			<!-- This is simple wrapper for the Java code -->

			<main-class>eu.dnetlib.iis.core.java.ProcessWrapper</main-class>

			<!-- The business Java code that gets to be executed -->

			<arg>eu.dnetlib.iis.core.examples.java.SampleDataProducer</arg>

			<!-- All input and output ports have to be bound to paths in HDFS -->

			<arg>-Operson=${workingDir}/data_producer/person</arg>

			<arg>-Odocument=${workingDir}/data_producer/document</arg>

		</java>

		<ok to="python_cloner" />

		<error to="fail" />

	</action>

    <action name="python_cloner">

        <map-reduce>

            <job-tracker>${jobTracker}</job-tracker>

            <name-node>${nameNode}</name-node>

            <prepare>

                <delete path="${nameNode}${workingDir}/python_cloner"/>

                <mkdir path="${nameNode}${workingDir}/python_cloner"/>

            </prepare>

            <streaming>

            	<!-- Here, we give the relative path to the script and pass it

            	the parameters of the workflow node. The script is held

            	in a directory having the same name as the workflow node.

            	The parameters should be passed as **named** arguments. This

            	convention of passing them as named arguments makes the code

            	more readable/maintainable.

            	 -->

                <mapper>scripts/python_cloner/cloner.py --copies 3</mapper>

            </streaming>

            <configuration>

            	<!-- # Standard settings for our framework -->

                <property>

                    <name>mapred.output.format.class</name>

                    <value>com.cloudera.science.avro.streaming.AvroAsJSONOutputFormat</value>

                </property>

                <property>

                    <name>mapred.input.format.class</name>

                    <value>com.cloudera.science.avro.streaming.AvroAsJSONInputFormat</value>

                </property>

                <!-- # Custom settings for this workflow node -->

                <!-- We do not use any reducers, so we set their number to 0 -->

                <property>

                    <name>mapred.reduce.tasks</name>

                    <value>0</value>

                </property>

                <property>

                    <name>mapred.input.dir</name>

                    <value>${workingDir}/data_producer/person</value>

                </property>

                <!-- Path to the input schema. This is held in the same 

                directory as the script. -->

                <property>

                    <name>input.schema.url</name>

                    <value>${wf:appPath()}/lib/scripts/python_cloner/Person.avsc</value>

                </property>

                <property>

                    <name>mapred.output.dir</name>

                    <value>${workingDir}/python_cloner/output</value>

                </property>

                <!-- Path to the output schema. This is held in the same 

                directory as the script. -->

                <property>

                    <name>output.schema.url</name>

                    <value>${wf:appPath()}/lib/scripts/python_cloner/Person.avsc</value>

                </property>

            </configuration>

        </map-reduce>

        <ok to="cloner"/>

        <error to="fail"/>

    </action>

	<action name="cloner">

		<java>

			<job-tracker>${jobTracker}</job-tracker>

			<name-node>${nameNode}</name-node>

			<!-- The data generated by this node is deleted in this section -->

			<prepare>

				<delete path="${nameNode}${workingDir}/cloner" />

				<mkdir path="${nameNode}${workingDir}/cloner" />

			</prepare>

			<configuration>

				<property>

					<name>mapred.job.queue.name</name>

					<value>${queueName}</value>

				</property>

			</configuration>

			<!-- This is simple wrapper for the Java code -->

			<main-class>eu.dnetlib.iis.core.java.ProcessWrapper</main-class>

			<!-- The business Java code that gets to be executed -->

			<arg>eu.dnetlib.iis.core.examples.java.PersonCloner</arg>

			<!-- All input and output ports have to be bound to paths in HDFS -->

			<arg>-Iperson=${workingDir}/python_cloner/output</arg>

			<arg>-Operson=${workingDir}/cloner/person</arg>

		</java>

		<ok to="end" />

		<error to="fail" />

	</action>

    <kill name="fail">

        <message>Unfortunately, the process failed -- error message:

        			[${wf:errorMessage(wf:lastErrorNode())}]

        		</message>

    </kill>

    <end name="end"/>

</workflow-app>