[
https://issues.apache.org/jira/browse/FLINK-14848?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Flink Jira Bot updated FLINK-14848:
-----------------------------------
Labels: auto-deprioritized-major pull-request-available (was:
pull-request-available stale-major)
Priority: Minor (was: Major)
This issue was labeled "stale-major" 7 ago and has not received any updates so
it is being deprioritized. If this ticket is actually Major, please raise the
priority and ask a committer to assign you the issue or revive the public
discussion.
> BaseRowSerializer.toBinaryRow wrongly process null for non-compact decimal
> ---------------------------------------------------------------------------
>
> Key: FLINK-14848
> URL: https://issues.apache.org/jira/browse/FLINK-14848
> Project: Flink
> Issue Type: Bug
> Components: Table SQL / Runtime
> Affects Versions: 1.9.1
> Reporter: Zhenghua Gao
> Priority: Minor
> Labels: auto-deprioritized-major, pull-request-available
>
> Take non-compact decimal(precision > 18 null as an example, the writer not
> only need to set null bits, but also need to assign bytes of variable-length
> part for future update. The following test case can reproduce the problem.
>
>
> {code:java}
> @Test
> public void test() {
> Decimal dec = Decimal.fromBigDecimal(new BigDecimal(11), 38, 0);
> BinaryRow row1 = new BinaryRow(2);
> BinaryRowWriter writer = new BinaryRowWriter(row1);
> writer.writeDecimal(0, dec, 38);
> writer.writeDecimal(1, null, 38);
> writer.complete();
> BaseRowSerializer serializer = new BaseRowSerializer(null, RowType.of(new
> DecimalType(38, 0), new DecimalType(38, 0)));
> GenericRow row2 = new GenericRow(2);
> row2.setField(0, dec);
> row2.setField(1, null);
> BinaryRow row3 = serializer.toBinaryRow(row2);
> assertTrue(row1.equalsWithoutHeader(row3));
> }
> {code}
>
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
