I found the problem. I had changed the table definition of t3.
The stored as inputformat 'org.apache.hadoop.mapred.SequenceFileInputFormat' does cause it to read the sequence file, and I had different "stored as" clauses as I experimented. The key/value of the sequence file does not need to be specified. From: Rodriguez, John [mailto:jrodrig...@verisign.com] Sent: Tuesday, September 14, 2010 11:08 PM To: hive-user@hadoop.apache.org Subject: question about reading a sequencefile I have a sequence file that has custom Writables in it. For illustration, assume the value in the SequenceFile is like this: public class DigestWritable implements org.apache.hadoop.io.Writable { String str; int I; } I wrote a SerDe, expecting that the value passed into deserialize would be a DigestWritable but I get a class cast exception in public Object deserialize(Writable field) throws SerDeException { DigestWritable digest = (DigestWritable)writable; // class cast exception here Failed with exception java.io.IOException:java.lang.ClassCastException: org.apache.hadoop.io.Text cannot be cast to DigestWritable Is my create table statement wrong? Does the "stored as inputformat" part cause Hive to read a SequenceFile? Do I need to specify the key and value of the SeuenceFile? create external table if not exists t3 ( str string ) row format serde 'com.utils.DigestSerde' stored as inputformat 'org.apache.hadoop.mapred.SequenceFileInputFormat' outputformat 'org.apache.hadoop.mapred.SequenceFileOutputFormat' location '/user/jr/hiveTables';
