[
https://issues.apache.org/jira/browse/HIVE-9978?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14363475#comment-14363475
]
Gopal V commented on HIVE-9978:
-------------------------------
I see comments about this in code from [~sershe].
{code}
try {
List<Integer> includedCols = ColumnProjectionUtils.isReadAllColumns(job)
? null : ColumnProjectionUtils.getReadColumnIDs(job);
if (includedCols.isEmpty()) {
includedCols = null; // Also means read all columns? WTF?
}
{code}
> LLAP: OrcColumnVectorProducer should handle reading isPresent columns only
> --------------------------------------------------------------------------
>
> Key: HIVE-9978
> URL: https://issues.apache.org/jira/browse/HIVE-9978
> Project: Hive
> Issue Type: Sub-task
> Components: File Formats
> Affects Versions: llap
> Reporter: Gopal V
> Assignee: Sergey Shelukhin
>
> LlapInputFormat does not understand the difference between empty columns list
> and null columns list.
> The empty columns list indicates no columns read except the root struct
> isPresent column, while the null columns list indicates that all columns are
> being read.
> {code}
> select count(1) from store_sales join date_dim on ss_sold_date_sk = d_date_sk
> where d_date = '1998-01-01';
> ...
> Caused by: java.lang.NullPointerException
> at
> org.apache.hadoop.hive.llap.io.decode.OrcColumnVectorProducer.createReadPipeline(OrcColumnVectorProducer.java:72)
> at
> org.apache.hadoop.hive.llap.io.api.impl.LlapInputFormat$LlapRecordReader.startRead(LlapInputFormat.java:181)
> at
> org.apache.hadoop.hive.llap.io.api.impl.LlapInputFormat$LlapRecordReader.next(LlapInputFormat.java:140)
> at
> org.apache.hadoop.hive.llap.io.api.impl.LlapInputFormat$LlapRecordReader.next(LlapInputFormat.java:99)
> at
> org.apache.hadoop.hive.ql.io.HiveContextAwareRecordReader.doNext(HiveContextAwareRecordReader.java:350)
> ... 22 more
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
