You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Danny Chen (Jira)" <ji...@apache.org> on 2020/05/20 08:48:00 UTC
[jira] [Updated] (FLINK-15397) Streaming and batch has different
value in the case of count function
[ https://issues.apache.org/jira/browse/FLINK-15397?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Danny Chen updated FLINK-15397:
-------------------------------
Fix Version/s: (was: 1.11.0)
> Streaming and batch has different value in the case of count function
> ---------------------------------------------------------------------
>
> Key: FLINK-15397
> URL: https://issues.apache.org/jira/browse/FLINK-15397
> Project: Flink
> Issue Type: Bug
> Components: Table SQL / Client
> Affects Versions: 1.10.0
> Reporter: xiaojin.wy
> Priority: Major
>
> *The sql is:*
> CREATE TABLE `testdata` (
> a INT,
> b INT
> ) WITH (
> 'connector.path'='/defender_test_data/daily_regression_batch_spark_1.10/test_group_agg/sources/testdata.csv',
> 'format.empty-column-as-null'='true',
> 'format.field-delimiter'='|',
> 'connector.type'='filesystem',
> 'format.derive-schema'='true',
> 'format.type'='csv'
> );
> SELECT COUNT(1) FROM testdata WHERE false;
> If the configuration's type is batch ,the result will be 0, but if the configuration is streaming, there will be no value;
> *The configuration is:*
> execution:
> planner: blink
> type: streaming
> *The input data is:*
> {code:java}
> 1|1
> 1|2
> 2|1
> 2|2
> 3|1
> 3|2
> |1
> 3|
> |
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)