1) Resolved issue related to reduce operator closing correctly. 
 Reducer.close is not called and hence when running hive on hyracks, the query does not complete with job running indefinitely.
 
 2) Dynamic Record Descriptors
    Hadoop read and map operators needed to have record descriptors for each partition as the record descriptors may vary for every partition. 
    
     
 

git-svn-id: https://hyracks.googlecode.com/svn/trunk/hyracks@152 123451ca-8445-de46-9d55-352943316053
8 files changed