[
https://issues.apache.org/jira/browse/HIVE-5525?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13794526#comment-13794526
]
Gunther Hagleitner commented on HIVE-5525:
------------------------------------------
It'd be good to add some tests wrt to virtual columns and schema evolution
(since you're changing the MapOperator when vectorization kicks in):
VC:
select INPUT__FILE__NAME, key, BLOCK__OFFSET__INSIDE__FILE from src;
select key, count(INPUT__FILE__NAME) from src group by key order by key;
select count(key), INPUT__FILE__NAME from src group by INPUT__FILE_NAME;
select * from src where BLOCK__OFFSET__INSIDE__FILE > 12000 order by key;
Schema:
Test with multiple partitions, where:
- different partitions use different serdes
- have different + renamed columns
- have partition columns with different types/names
Otherwise patch looks good. I think we should consider moving the
non-vectorized eventually to a similar model where we handle VC+part at
planning time rather than at runtime.
> Vectorized query failing for partitioned tables.
> ------------------------------------------------
>
> Key: HIVE-5525
> URL: https://issues.apache.org/jira/browse/HIVE-5525
> Project: Hive
> Issue Type: Sub-task
> Components: Vectorization
> Reporter: Jitendra Nath Pandey
> Assignee: Jitendra Nath Pandey
> Attachments: HIVE-5525.1.patch
>
>
> {code}
> Caused by: java.lang.NullPointerException
> at
> org.apache.hadoop.hive.ql.exec.vector.VectorizationContext.getInputColumnIndex(VectorizationContext.java:154)
> at
> org.apache.hadoop.hive.ql.exec.vector.VectorizationContext.getVectorBinaryComparisonFilterExpression(VectorizationContext.java:1163)
> at
> org.apache.hadoop.hive.ql.exec.vector.VectorizationContext.getVectorExpression(VectorizationContext.java:436)
> at
> org.apache.hadoop.hive.ql.exec.vector.VectorizationContext.getVectorExpression(VectorizationContext.java:274)
> at
> org.apache.hadoop.hive.ql.exec.vector.VectorFilterOperator.<init>(VectorFilterOperator.java:50)
> {code}
--
This message was sent by Atlassian JIRA
(v6.1#6144)
