[
https://issues.apache.org/jira/browse/IGNITE-18556?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Maksim Zhuravkov updated IGNITE-18556:
--------------------------------------
Description:
Default implementation of TypeSystem::getDefaultPrecision, provided by Calcite,
returns the same value for FLOAT and DOUBLE types. Such behaviour causes
TypeFactory::leastRestrictiveType to return different results for (FLOAT,
DOUBLE) and (DOUBLE, FLOAT).
We fixed getDefaultPrecision to return different values to resolve the problem
with leastRestrictiveType in IGNITE-18163.
1) Investigate how these change affect behaviour of other operators
2) Choose the appropriate value for default precision in IgniteTypeSystem for
FLOAT and DOUBLE if necessary.
was:
Default implementation of TypeSystem::getDefaultPrecision, provided by Calcite,
returns the same value for FLOAT and DOUBLE types. Such behaviour causes
TypeFactory::leastRestrictiveType to return different results for (FLOAT,
DOUBLE) and (DOUBLE, FLOAT).
We fixed getDefaultPrecision to return different values to resolve the problem
with leastRestrictiveType in IGNITE-18163.
1) Investigate how these change affect behaviour of other operators
2) Choose appropriate value for default precision.
> Sql. TypeSystem. Default implementation of getDefaultPrecision FLOAT and
> DECIMAL returns the same value.
> --------------------------------------------------------------------------------------------------------
>
> Key: IGNITE-18556
> URL: https://issues.apache.org/jira/browse/IGNITE-18556
> Project: Ignite
> Issue Type: Improvement
> Components: sql
> Reporter: Maksim Zhuravkov
> Priority: Minor
> Labels: ignite-3
> Fix For: 3.0.0-beta1
>
>
> Default implementation of TypeSystem::getDefaultPrecision, provided by
> Calcite, returns the same value for FLOAT and DOUBLE types. Such behaviour
> causes TypeFactory::leastRestrictiveType to return different results for
> (FLOAT, DOUBLE) and (DOUBLE, FLOAT).
> We fixed getDefaultPrecision to return different values to resolve the
> problem with leastRestrictiveType in IGNITE-18163.
> 1) Investigate how these change affect behaviour of other operators
> 2) Choose the appropriate value for default precision in IgniteTypeSystem for
> FLOAT and DOUBLE if necessary.
>
>
>
>
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
