Quanlong Huang created IMPALA-10882:
---------------------------------------
Summary: Push down Min-Max predicates of CHAR/VARCHAR to ORC reader
Key: IMPALA-10882
URL: https://issues.apache.org/jira/browse/IMPALA-10882
Project: IMPALA
Issue Type: Improvement
Components: Backend
Reporter: Quanlong Huang
This is a follow-up work of IMPALA-6505. Due to padding/truncation issues of
CHAR/VARCHAR types when comparing to string literals, we might get wrong
results if we simply push down CHAR/VARCHAR predicates into the ORC reader,
especially when the file schema differs from the table schema.
See more discussion in the Gerrit review:
https://gerrit.cloudera.org/c/15403/1/be/src/exec/hdfs-orc-scanner.cc#889
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
