Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Plain View
Pig >> mail # user >> ~/.pigrc or ~/.pigbootup? Impossible to parametize anything in Grunt?


Copy link to this message
-
~/.pigrc or ~/.pigbootup? Impossible to parametize anything in Grunt?
I need to run these things as Pig boots, and have them work in Grunt:

/* Setup for Piggybank */
%default PIGGYBANK_LIB '/me/Software/pig/contrib/piggybank/java'
REGISTER $PIBBYBANK_LIB/piggybank.jar

/* Setup for Avro */
%default PIG_LIB '/me/Software/pig/build/ivy/lib/Pig';
REGISTER $PIG_LIB/avro-1.5.3.jar
REGISTER $PIG_LIB/json-simple-1.1.jar
DEFINE AvroStorage org.apache.pig.piggybank.storage.avro.AvroStorage();

/* Setup for MongoDB */
$default MONGO_LIB '/me/Software/mongo-hadoop/'
REGISTER $MONGO_LIB/mongo-2.10.1.jar
REGISTER $MONGO_LIB/core/target/mongo-hadoop-core-1.1.0-SNAPSHOT.jar
REGISTER $MONGO_LIB/pig/target/mongo-hadoop-pig-1.1.0-SNAPSHOT.jar
DEFINE MongoStorage com.mongodb.hadoop.pig.MongoStorage();
However, when I place them in ~/.pigrc and start Pig 0.10... nothing
happens. When I run Pig 0.11, nothing happens.

I have gone through half a dozen JIRAs, and not one specifies how this
interface works, or how to load libraries when Grunt starts. Should I be
using .pigbootup - and Pig 0.11? That throws tons of errors.

How can I write a script that doesn't require a Grunt user to customize the
path to his pig installation EACH AND EVERY TIME they use grunt?
-----------------------

--
Russell Jurney twitter.com/rjurney
[EMAIL PROTECTED]<https://mail.google.com/mail/?view=cm&fs=1&tf=1&[EMAIL PROTECTED]>
 datasyndrome.com
+
Prashant Kommireddi 2013-01-05, 21:51
+
Russell Jurney 2013-01-05, 21:53