[
https://issues.apache.org/jira/browse/FLINK-29558?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17678042#comment-17678042
]
lincoln lee commented on FLINK-29558:
-------------------------------------
[~godfreyhe] Could you also take a look at this when you have time? There're
several existing cases already tested the case which select nothing from
source, e.g., "SELECT COUNT(1) FROM T" in
`org.apache.flink.table.planner.plan.stream.sql.TableSourceTest#testProjectWithoutInputRef`
and the expected plan is:
{code}
<TestCase name="testProjectWithoutInputRef">
<Resource name="sql">
<![CDATA[SELECT COUNT(1) FROM T]]>
</Resource>
<Resource name="ast">
<![CDATA[
LogicalAggregate(group=[{}], EXPR$0=[COUNT()])
+- LogicalProject($f0=[0])
+- LogicalTableScan(table=[[default_catalog, default_database, T]])
]]>
</Resource>
<Resource name="optimized exec plan">
<![CDATA[
GroupAggregate(select=[COUNT(*) AS EXPR$0])
+- Exchange(distribution=[single])
+- TableSourceScan(table=[[default_catalog, default_database, T, project=[],
metadata=[]]], fields=[])
]]>
</Resource>
</TestCase>
{code}
so is there any contract that connectors should follow in such cases？
> Use select count(*) from xxx; and get SQL syntax
> ------------------------------------------------
>
> Key: FLINK-29558
> URL: https://issues.apache.org/jira/browse/FLINK-29558
> Project: Flink
> Issue Type: Bug
> Components: Connectors / JDBC
> Affects Versions: 1.15.3
> Environment: flink 1.15.2
> CentOS Linux release 7.9.2009 (Core)
> 5.7.32-log MySQL Community Server (GPL)
> Reporter: StarBoy1005
> Assignee: lincoln lee
> Priority: Major
> Labels: pull-request-available
> Fix For: 1.17.0
>
> Attachments: image-2022-10-10-15-31-34-341.png,
> image-2022-10-19-17-55-14-700.png, image-2022-11-03-18-16-12-127.png,
> screenshot-1.png, screenshot-2.png
>
>
> Hi, I use flink sql to make kafka records to mysql.
> so I create these 2 tables in flink sql,here is the mysql ,and I created the
> table in mysql before I did the insert action in flink sql.
> CREATE TABLE mysql_MyUserTable (
> id STRING,
> name STRING,
> age STRING,
> status STRING,
> PRIMARY KEY (id) NOT ENFORCED
> ) WITH (
> 'connector' = 'jdbc',
> 'url' = 'jdbc:mysql://10.19.29.170:3306/fromflink152',
> 'table-name' = 'users',
> 'username' = 'root',
> 'password' = '******'
> );
> In mysql, I created database "fromflink152" then created the table like this
> way
> CREATE TABLE `users` (
> `id` varchar(64) NOT NULL DEFAULT '',
> `name` varchar(255) DEFAULT NULL,
> `age` varchar(255) DEFAULT NULL,
> `status` varchar(255) DEFAULT NULL,
> PRIMARY KEY (`id`)
> )
> After executed insert sql,I found 'select * from mysql_MyUserTable' can get
> correct result,but ’select count(\*) from mysql_MyUserTable‘ or ’select
> count(id) from mysql_MyUserTable‘ ,the collect job in flink app keep
> restarting again and again.The exception is:
> !image-2022-10-10-15-31-34-341.png!
> So I wonder which config that I missed about the table in flink or mysql side
> :(
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
