[DISCUSS] add a '--filename' parameter for sql-client

[Jun Zhang]

I want to execute some flink sql batch jobs regularly, such as 'insert into
select .....', but I can't find a suitable method so far, so reference
 hive, I changed the source code and add a  '--filename'  parameter  so
that we can execute a sql file.

like this:

/home/flink/bin/sql-client.sh embedded -f flink.sql

what about any ideas or plans for this feature community?

[godfreyhe]

hi Jun,

Currently, sql client has supported -u option, just like:
 ./bin/sql-client.sh embedded -u "insert_statement".

There is already a JIRA [1] that wants to support -f option

[1] https://issues.apache.org/jira/browse/FLINK-12828

Best,
Godfrey

Jun Zhang <[hidden email]> 于2020年7月29日周三 上午9:22写道：

> I want to execute some flink sql batch jobs regularly, such as 'insert into
> select .....', but I can't find a suitable method so far, so reference
>  hive, I changed the source code and add a  '--filename'  parameter  so
> that we can execute a sql file.
>
> like this:
>
> /home/flink/bin/sql-client.sh embedded -f flink.sql
>
> what about any ideas or plans for this feature community?
>

[Jun Zhang]

hi，godfrey：
Thanks for your reply

1. I have seen the -u parameter, but my sql file may not only include
'insert into select ....', but also SET, DDL, etc.

2. I may not have noticed this issue. I took a look at this issue. I think
this issue may have some problems. For example, he finally called the
CliClient.callCommand method.
But I think that many options in callCommand are not completely suitable
for sql files, such as HELP, CLEAR, SELECT, etc. The select operation opens
a window to display the results, obviously this is not suitable for
executing sql files

godfrey he <[hidden email]> 于2020年7月29日周三 上午9:56写道：

> hi Jun,
>
> Currently, sql client has supported -u option, just like:
>  ./bin/sql-client.sh embedded -u "insert_statement".
>
> There is already a JIRA [1] that wants to support -f option
>
> [1] https://issues.apache.org/jira/browse/FLINK-12828
>
> Best,
> Godfrey
>
> Jun Zhang <[hidden email]> 于2020年7月29日周三 上午9:22写道：
>
>> I want to execute some flink sql batch jobs regularly, such as 'insert
>> into
>> select .....', but I can't find a suitable method so far, so reference
>>  hive, I changed the source code and add a  '--filename'  parameter  so
>> that we can execute a sql file.
>>
>> like this:
>>
>> /home/flink/bin/sql-client.sh embedded -f flink.sql
>>
>> what about any ideas or plans for this feature community?
>>
>

[godfreyhe]

Yes, The pr still needs to be improved.
In most cases, there are more than one statement in the sql file,
so -f option should support multiple statements.
however, a related PR [1] has not completed yet.

[1] https://github.com/apache/flink/pull/8738

Best,
Godfrey

Jun Zhang <[hidden email]> 于2020年7月29日周三 上午10:17写道：

> hi，godfrey：
> Thanks for your reply
>
> 1. I have seen the -u parameter, but my sql file may not only include
> 'insert into select ....', but also SET, DDL, etc.
>
> 2. I may not have noticed this issue. I took a look at this issue. I think
> this issue may have some problems. For example, he finally called the
> CliClient.callCommand method.
> But I think that many options in callCommand are not completely suitable
> for sql files, such as HELP, CLEAR, SELECT, etc. The select operation opens
> a window to display the results, obviously this is not suitable for
> executing sql files
>
> godfrey he <[hidden email]> 于2020年7月29日周三 上午9:56写道：
>
>> hi Jun,
>>
>> Currently, sql client has supported -u option, just like:
>>  ./bin/sql-client.sh embedded -u "insert_statement".
>>
>> There is already a JIRA [1] that wants to support -f option
>>
>> [1] https://issues.apache.org/jira/browse/FLINK-12828
>>
>> Best,
>> Godfrey
>>
>> Jun Zhang <[hidden email]> 于2020年7月29日周三 上午9:22写道：
>>
>>> I want to execute some flink sql batch jobs regularly, such as 'insert
>>> into
>>> select .....', but I can't find a suitable method so far, so reference
>>>  hive, I changed the source code and add a  '--filename'  parameter  so
>>> that we can execute a sql file.
>>>
>>> like this:
>>>
>>> /home/flink/bin/sql-client.sh embedded -f flink.sql
>>>
>>> what about any ideas or plans for this feature community?
>>>
>>