[
https://issues.apache.org/jira/browse/CALCITE-7368?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18051990#comment-18051990
]
Mihai Budiu commented on CALCITE-7368:
--------------------------------------
Here are some examples which describe how this would work, in a quidem-like
format.
{code:java}
SELECT CAST(10 AS BINARY(4)) AS r;
0000000A
SELECT CAST(10 AS BINARY(2)) AS r;
000A
SELECT CAST(10 AS VARBINARY(4)) AS r;
0000000A
SELECT CAST(10 AS VARBINARY(8)) AS r;
0000000A
SELECT CAST(10 AS VARBINARY(2)) AS r;
000A
SELECT CAST(CAST(10 AS INT UNSIGNED) AS BINARY(4)) AS r;
0000000A
SELECT CAST(-1 AS BINARY(4)) AS r;
FFFFFFFF
SELECT CAST(-1 AS VARBINARY(8)) AS r;
FFFFFFFF
SELECT CAST(CAST(10 AS TINYINT) AS VARBINARY);
0A
SELECT CAST(CAST(-1 AS TINYINT) AS VARBINARY);
FF
SELECT CAST(CAST(-1 AS TINYINT) AS BINARY(4));
000000FF{code}
> The validator accepts CAST(INT TO BINARY), but the runtime does not implement
> them
> ----------------------------------------------------------------------------------
>
> Key: CALCITE-7368
> URL: https://issues.apache.org/jira/browse/CALCITE-7368
> Project: Calcite
> Issue Type: New Feature
> Components: core
> Affects Versions: 1.41.0
> Reporter: Mihai Budiu
> Priority: Minor
>
> Most SQL dialects which support this feature seem to convert integers to big
> endian values, and then truncate or pad on the left when the size of the
> target BINARY does not exactly match the integer's size.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
