There are some experimental features of 0.12 that doesn't work with Hadoop 2.* (such as new UDP for Python extension)
Probably it just won't compile together at all because of that. Sent from my iPhone > On Jan 27, 2014, at 11:15 AM, Jay Vyas <[email protected]> wrote: > > Hi, looking in > > http://mvnrepository.com/artifact/org.apache.pig/pig/0.12.0 > > It seems that pig doesnt have an explicit dependency on hadoop-common. Im > trying to decide which is the "Right" version of hadoop to have on my > classpath when running pig 0.12 in my IDE, and hadoop 2.2 seems to cause a > cryptic exception: > > > org.apache.pig.impl.logicalLayer.FrontendException: ERROR 1066: Unable to > open iterator for alias uniqcnt > at org.apache.pig.PigServer.openIterator(PigServer.java:880) > at > org.bigtop.bigpetstore.etl.PigETL.numberOfProductsByProduct(PigETL.java:99) > at org.bigtop.bigpetstore.pigtest.TestPig.runPig(TestPig.java:61) > at org.bigtop.bigpetstore.pigtest.TestPig.testPigETL(TestPig.java:51) > at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) > > So, any help would be appreciated in understanding the compile time > integration of pig and hadoop . > > -- > Jay Vyas > http://jayunit100.blogspot.com
