[ 
https://issues.apache.org/jira/browse/FLINK-36754?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

MOBIN updated FLINK-36754:
--------------------------
    Description: 
When projection is empty or NULL, the projection should be treated as an 
asterisk. The following transform rule should throw an exception [different 
column counts], but the actual schema columns are id and name, which is not 
expected.
{code:java}
transform:
  - source-table: testdb.test_table1
    filter: age> 8 and age < 18
  - source-table: testdb.test_table1
    projection: id,UPPER(name) AS name
    filter: age >= 18 {code}
 

The following transform rule will throw an exception [different column counts]
{code:java}
transform:
  - source-table: testdb.test_table1
    projection: *
    filter: age> 8 and age < 18
  - source-table: testdb.test_table1
    projection: id,UPPER(name) AS name
    filter: age >= 18 {code}
We should treat projection as asterisk when it is empty or NULL, to be 
consistent

  was:
When projection is empty or NULL, the projection should be treated as an 
asterisk. The following transform rule should throw an exception [different 
column counts], but the actual schema columns are id and name, which is not 
expected.
{code:java}
transform:
  - source-table: testdb.test_table1
    filter: age> 8 and age < 18
  - source-table: testdb.test_table1
    projection: id,UPPER(name) AS name
    filter: age >= 18 {code}


> When projection is empty or null, it is treated as an asterisk
> --------------------------------------------------------------
>
>                 Key: FLINK-36754
>                 URL: https://issues.apache.org/jira/browse/FLINK-36754
>             Project: Flink
>          Issue Type: Improvement
>          Components: Flink CDC
>    Affects Versions: cdc-3.2.0
>            Reporter: MOBIN
>            Priority: Not a Priority
>
> When projection is empty or NULL, the projection should be treated as an 
> asterisk. The following transform rule should throw an exception [different 
> column counts], but the actual schema columns are id and name, which is not 
> expected.
> {code:java}
> transform:
>   - source-table: testdb.test_table1
>     filter: age> 8 and age < 18
>   - source-table: testdb.test_table1
>     projection: id,UPPER(name) AS name
>     filter: age >= 18 {code}
>  
> The following transform rule will throw an exception [different column counts]
> {code:java}
> transform:
>   - source-table: testdb.test_table1
>     projection: *
>     filter: age> 8 and age < 18
>   - source-table: testdb.test_table1
>     projection: id,UPPER(name) AS name
>     filter: age >= 18 {code}
> We should treat projection as asterisk when it is empty or NULL, to be 
> consistent



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to