hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mohammad Kamrul Islam (JIRA)" <>
Subject [jira] [Created] (HIVE-12475) Parquet schema evolution within array<struct<>> doesn't work
Date Fri, 20 Nov 2015 00:26:11 GMT
Mohammad Kamrul Islam created HIVE-12475:

             Summary: Parquet schema evolution within array<struct<>> doesn't
                 Key: HIVE-12475
             Project: Hive
          Issue Type: Bug
          Components: File Formats
    Affects Versions: 1.1.0
            Reporter: Mohammad Kamrul Islam
            Assignee: Mohammad Kamrul Islam

If we create a table with type array<struct<>>, and later added a field in the
struct, we got the following exception.

The following SQL statements would recreate the error:

CREATE TABLE pq_test (f1 array<struct<c1:int,c2:int>>) STORED AS  PARQUET;
INSERT INTO TABLE pq_test select array(named_struct("c1",1,"c2",2)) FROM tmp LIMIT 2;

SELECT * from pq_test;

ALTER TABLE pq_test REPLACE COLUMNS (f1 array<struct<c1:int,c2:int,cccccc:int>>);
//***** cccccc
SELECT * from pq_test;

Caused by: java.lang.ArrayIndexOutOfBoundsException: 2
        at org.apache.hadoop.hive.serde2.SerDeUtils.buildJSONString(
        at org.apache.hadoop.hive.serde2.SerDeUtils.buildJSONString(
        at org.apache.hadoop.hive.serde2.SerDeUtils.getJSONString(
        at org.apache.hadoop.hive.serde2.DelimitedJSONSerDe.serializeField(
        at org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe.doSerialize(
        at org.apache.hadoop.hive.serde2.AbstractEncodingAwareSerDe.serialize(
        at org.apache.hadoop.hive.ql.exec.DefaultFetchFormatter.convert(
        at org.apache.hadoop.hive.ql.exec.DefaultFetchFormatter.convert(
        at org.apache.hadoop.hive.ql.exec.ListSinkOperator.process(

This message was sent by Atlassian JIRA

View raw message