Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Threaded View
Pig >> mail # dev >> Pig standalone jar


Copy link to this message
-
Re: Pig standalone jar
We already generate core pig jar even though it's not published on Maven
repo. If you run "ant clean jar", you will find two jars under the build
dir:

-rw-r--r--  1 cheolsoo cheolsoo  3411312 Feb 14 09:41
pig-0.12.0-SNAPSHOT.jar
-rw-r--r--  1 cheolsoo cheolsoo 16225627 Feb 14 09:41
pig-0.12.0-SNAPSHOT-withdependencies.jar

"pig-0.12.0-SNAPSHOT.jar" contains only Pig but nothing else, whereas
"pig-0.12.0-SNAPSHOT-withdependencies.jar" contains minimal dependencies.
Please look at build.xml regarding what dependencies are packaged into
these jar files:

    <fileset dir="${ivy.lib.dir}"
id="runtime.dependencies-withouthadoop.jar">
        <patternset id="pattern.runtime.dependencies-withouthadoop.jar">
            <include name="antlr-runtime-${antlr.version}.jar"/>
            <include name="ST4-${stringtemplate.version}.jar"/>
            <include name="jline-${jline.version}.jar"/>
            <include name="jackson-mapper-asl-${jackson.version}.jar"/>
            <include name="jackson-core-asl-${jackson.version}.jar"/>
            <include name="joda-time-${joda-time.version}.jar"/>
            <include name="guava-${guava.version}.jar"/>
            <include name="automaton-${automaton.version}.jar"/>
            <include name="jansi-${jansi.version}.jar"/>
        </patternset>
    </fileset>

    <fileset dir="${ivy.lib.dir}" id="runtime.dependencies.jar">
        <patternset id="pattern.runtime.dependencies.jar">
            <patternset
refid="pattern.runtime.dependencies-withouthadoop.jar"/>
            <include name="hadoop-core-${hadoop-core.version}.jar"/>
            <include name="hadoop-*-${hadoop-common.version}.jar"/>
            <exclude name="hadoop-hdfs-${hadoop-hdfs.version}.jar"/>
            <include name="junit-${junit.version}.jar"/>
            <include name="jsch-${jsch.version}.jar"/>
            <include name="protobuf-java-${protobuf-java.version}.jar"/>
            <include name="avro-${avro.version}.jar"/>
            <include name="commons*.jar"/>
            <include name="log4j*.jar"/>
            <include name="slf4j*.jar"/>
            <include name="jsp-api*.jar"/>
        </patternset>
    </fileset>

    <fileset dir="${ivy.lib.dir}" id="core.dependencies.jar">
        <exclude name="**.*jar"/>
    </fileset>

Thanks,
Cheolsoo

On Thu, Feb 14, 2013 at 9:32 AM, Bill Graham <[EMAIL PROTECTED]> wrote:

> How would this be different than pig-*-withdependencies.jar?
>
>
> On Wed, Feb 13, 2013 at 9:59 PM, Prashant Kommireddi <[EMAIL PROTECTED]
> >wrote:
>
> > Should pig have a standalone jar? We have a case when building pig with
> > external code produces jar conflicts. Just like we have
> "pig-withouthadoop"
> > as a target, does it make sense to do "pig-standalone.jar" too?
> >
> > -Prashant
> >
>
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB