Horton

Thread

Commits

  1. Remove pgbench's restriction on placement of -M switch.

  1. Confusing error message in pgbench

    Tatsuo Ishii <ishii@sraoss.co.jp> — 2017-08-02T02:03:28Z

    I found an error message in pgbench is quite confusing.

pgbench -S -M extended -c 1 -T 30 test
query mode (-M) should be specified before any transaction scripts (-f or -b)

Since there's no -f or -b option is specified, users will be
confused. Actually the error occurs because pgbench implicitly
introduces a built in script for -S. To eliminate the confusion, I
think the error message should be fixed like this:

query mode (-M) should be specified before transaction type (-S or -N) or any transaction scripts (-f or -b)

Patch attached.

Best regards,
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese:http://www.sraoss.co.jp

  2. Re: Confusing error message in pgbench

    Robert Haas <robertmhaas@gmail.com> — 2017-08-02T12:22:02Z

    On Tue, Aug 1, 2017 at 10:03 PM, Tatsuo Ishii <ishii@sraoss.co.jp> wrote:
> I found an error message in pgbench is quite confusing.
>
> pgbench -S -M extended -c 1 -T 30 test
> query mode (-M) should be specified before any transaction scripts (-f or -b)
>
> Since there's no -f or -b option is specified, users will be
> confused. Actually the error occurs because pgbench implicitly
> introduces a built in script for -S. To eliminate the confusion, I
> think the error message should be fixed like this:
>
> query mode (-M) should be specified before transaction type (-S or -N) or any transaction scripts (-f or -b)
>
> Patch attached.

Not really objecting, but an even better fix might be to remove the
restriction on the order in which the options can be specified.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

  3. Re: Confusing error message in pgbench

    Fabien COELHO <coelho@cri.ensmp.fr> — 2017-08-02T15:26:24Z

    Hello Tatsuo-san,

> I found an error message in pgbench is quite confusing.
>
> pgbench -S -M extended -c 1 -T 30 test
> query mode (-M) should be specified before any transaction scripts (-f or -b)
>
> Since there's no -f or -b option is specified, users will be
> confused.

Indeed.

> Actually the error occurs because pgbench implicitly introduces a built 
> in script for -S. To eliminate the confusion, I think the error message 
> should be fixed like this:

The idea is that -S/-N documentations say that it is just a shortcut for 
-b, but the explanation (eg --help) is too far away.

> query mode (-M) should be specified before transaction type (-S or -N) 
> or any transaction scripts (-f or -b)

I would suggest to make it even shorter, see attached:

query mode (-M) should be specified before any transaction scripts (-f, 
-b, -S or -N).

I'm wondering whether it could/should be "any transaction script". My 
English level does not allow to decide.

-- 
Fabien.

  4. Re: Confusing error message in pgbench

    Tom Lane <tgl@sss.pgh.pa.us> — 2017-08-02T19:45:05Z

    Robert Haas <robertmhaas@gmail.com> writes:
> On Tue, Aug 1, 2017 at 10:03 PM, Tatsuo Ishii <ishii@sraoss.co.jp> wrote:
>> I found an error message in pgbench is quite confusing.

> Not really objecting, but an even better fix might be to remove the
> restriction on the order in which the options can be specified.

Indeed.  It doesn't look that hard: AFAICS the problem is just that
process_sql_command() is making premature decisions about whether to
extract parameters from the SQL commands.  Proposed patch attached.

			regards, tom lane

  5. Re: Confusing error message in pgbench

    Tatsuo Ishii <ishii@sraoss.co.jp> — 2017-08-02T23:01:35Z

    > Not really objecting, but an even better fix might be to remove the
> restriction on the order in which the options can be specified.

+100 :-)

Best regards,
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese:http://www.sraoss.co.jp

  6. Re: Confusing error message in pgbench

    Tatsuo Ishii <ishii@sraoss.co.jp> — 2017-08-03T00:53:46Z

    >> Not really objecting, but an even better fix might be to remove the
>> restriction on the order in which the options can be specified.
> 
> Indeed.  It doesn't look that hard: AFAICS the problem is just that
> process_sql_command() is making premature decisions about whether to
> extract parameters from the SQL commands.  Proposed patch attached.

Great. Patch looks good to me.

Best regards,
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese:http://www.sraoss.co.jp

  7. Re: Confusing error message in pgbench

    Fabien COELHO <coelho@cri.ensmp.fr> — 2017-08-03T09:09:50Z

    >> Indeed.  It doesn't look that hard: AFAICS the problem is just that
>> process_sql_command() is making premature decisions about whether to
>> extract parameters from the SQL commands.  Proposed patch attached.
>
> Great. Patch looks good to me.

Too me as well: code looks ok, patch applies, compiles, make check 
ok, manual tests with pgbench ok.

That is one more patch about pgbench in the queue.

-- 
Fabien.