ramesh0201 commented on a change in pull request #880: HIVE-22515 Support cast
to decimal64 in Vectorization
URL: https://github.com/apache/hive/pull/880#discussion_r371567891
##########
File path:
ql/src/test/results/clientpositive/llap/vector_reuse_scratchcols.q.out
##########
@@ -369,7 +369,7 @@ STAGE PLANS:
includeColumns: [0, 1, 2, 3, 4, 5, 6, 7, 9, 10, 11]
dataColumns: ctinyint:tinyint, csmallint:smallint,
cint:int, cbigint:bigint, cfloat:float, cdouble:double, cstring1:string,
cstring2:string, ctimestamp1:timestamp, ctimestamp2:timestamp,
cboolean1:boolean, cboolean2:boolean
partitionColumnCount: 0
- scratchColumnTypeNames: [double, double, double,
decimal(22,3), decimal(13,3), double, double, double, double, double, double,
double, double]
+ scratchColumnTypeNames: [double, double, double,
decimal(22,3), decimal(13,3)/DECIMAL_64, double, double, double, double,
double, double, double, double]
Review comment:
Yes, because we can safely store a decimal with (13,3) by scaling them up
and storing in long vectors and tracking the scale. We do this until
precision+scale is <=18.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]