zlzhang0122 created FLINK-24306: ----------------------------------- Summary: group by index throw SqlValidatorException Key: FLINK-24306 URL: https://issues.apache.org/jira/browse/FLINK-24306 Project: Flink Issue Type: Bug Components: Table SQL / API Affects Versions: 1.13.1, 1.12.2 Reporter: zlzhang0122 Fix For: 1.15.0 Attachments: sql_exception.png
We create a table using following DDL: {code:java} create table if not exists datagen_source ( id int, name string, sex string, age int, birthday string, proc_time as proctime() ) with ( 'connector' = 'datagen', 'rows-per-second' = '10000', 'fields.id.kind' = 'random', 'fields.id.min' = '1', 'fields.id.max' = '2000000');{code} When we running {code:java} select id, count(*) from datagen_source group by id;{code} Everything will be fine.But if we running {code:java} select id, count(*) from datagen_source group by 1;{code} We will get a SqlValidatorException like this: !sql_exception.png! >From the error message we know that Flink SQL also can reconize that the '1' >index reprent the 'id'.Since MySql\Hive\Spark SQL\etc. all support group by >index, I think Flink also should support this syntax too. -- This message was sent by Atlassian Jira (v8.3.4#803005)