Hi, did you add the dependency of parquet[1]?
[1] https://mvnrepository.com/artifact/org.apache.flink/flink-sql-parquet
在 2022-07-11 18:33:38，pod...@gmx.com 写道：
This example?:
CREATETABLEkafka_table(user_idSTRING,order_amountDOUBLE,log_tsTIMESTAMP(3),WATERMARKFORlog_tsASlog_ts-INTERVAL'5'SECOND)WITH(...);CREATETABLEfs_table(user_idSTRING,order_amountDOUBLE,dtSTRING,`hour`STRING)PARTITIONEDBY(dt,`hour`)WITH('connector'='filesystem','path'='...','format'='parquet','sink.partition-commit.delay'='1
h','sink.partition-commit.policy.kind'='success-file');-- streaming sql,
insert into file system table
INSERTINTOfs_tableSELECTuser_id,order_amount,DATE_FORMAT(log_ts,'yyyy-MM-dd'),DATE_FORMAT(log_ts,'HH')FROMkafka_table;--
batch sql, select with partition pruning
SELECT*FROMfs_tableWHEREdt='2020-05-20'and`hour`='12';
CREATETABLEkafka_table - creates table, not file
CREATETABLEfs_table - creates table, not file
INSERTINTOfs_table - inserts into table from another table
I do not see any create file section.
Anyway:
Caused by: org.apache.flink.table.api.ValidationException: Could not find any
format factory for identifier 'parquet' in the classpath.
Sent: Monday, July 11, 2022 at 10:31 AM
From: "Lijie Wang" <wangdachui9...@gmail.com>
To: pod...@gmx.com
Cc: user@flink.apache.org
Subject: Re: Is it possible to save Table to CSV?
You can use the FileSink and set the format to csv. An example of FileSink:
https://nightlies.apache.org/flink/flink-docs-master/docs/connectors/table/filesystem/#full-example
Best,
Lijie
<pod...@gmx.com> 于2022年7月11日周一 16:16写道：
If I create dynamic table with:
CREATETABLEsome_table(nameSTRING,scoreINT)WITH('format'='csv','...');
//do some other stuff here
Then how to save table result to CSV file?
Best,
Mike
