[jira] [Updated] (FLINK-16922) DecimalData.toUnscaledBytes should be consistent with BigDecimla.unscaledValue.toByteArray

["Danny Chen (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/FLINK-16922?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Danny Chen updated FLINK-16922:
-------------------------------
    Summary: DecimalData.toUnscaledBytes should be consistent with BigDecimla.unscaledValue.toByteArray  (was: Decimal.toUnscaledBytes should be consistent with BigDecimla.unscaledValue.toByteArray)

> DecimalData.toUnscaledBytes should be consistent with BigDecimla.unscaledValue.toByteArray
> ------------------------------------------------------------------------------------------
>
>                 Key: FLINK-16922
>                 URL: https://issues.apache.org/jira/browse/FLINK-16922
>             Project: Flink
>          Issue Type: Bug
>          Components: Table SQL / Runtime
>            Reporter: Jingsong Lee
>            Assignee: Jark Wu
>            Priority: Critical
>             Fix For: 1.11.0
>
>
> In Decimal:
> {code:java}
> public byte[] toUnscaledBytes() {
>    if (!isCompact()) {
>       return toBigDecimal().unscaledValue().toByteArray();
>    }
>    // big endian; consistent with BigInteger.toByteArray()
>    byte[] bytes = new byte[8];
>    long l = longVal;
>    for (int i = 0; i < 8; i++) {
>       bytes[7 - i] = (byte) l;
>       l >>>= 8;
>    }
>    return bytes;
> }
> {code}
> When is compact, it will return fix 8 length byte array.
> This should not happen, it brings an incompatible byte array.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)