Re: pgbench - allow to create partitioned tables - Mailing list pgsql-hackers

From Amit Langote
Subject Re: pgbench - allow to create partitioned tables
Date
Msg-id CA+HiwqGKW0UJy5QyK64bMxHJCCuzu=qefC_Ftg9dd5ODDyrphA@mail.gmail.com
Whole thread Raw
In response to Re: pgbench - allow to create partitioned tables  (Fabien COELHO <coelho@cri.ensmp.fr>)
Responses Re: pgbench - allow to create partitioned tables
List pgsql-hackers
Hi Fabien,

On Wed, Sep 18, 2019 at 3:49 AM Fabien COELHO <coelho@cri.ensmp.fr> wrote:
> Attached v9:

Thanks.  This seems to work well.

Couple of nitpicks on parameter error messages.

+                    fprintf(stderr, "invalid partition type,
expecting \"range\" or \"hash\","

How about "partitioning method" instead of "partition type"?

+            fprintf(stderr, "--partition-method requires actual
partitioning with --partitions\n");

Assuming that this error message is to direct the user to fix a
mistake they might have inadvertently made in specifying --partitions,
I don't think the message is very clear.  How about:

"--partition-method requires --partitions to be greater than zero"

but this wording might suggest to some users that some partitioning
methods do allow zero partitions.  So, maybe:

"specifying --partition-method requires --partitions to be greater than zero"

Thanks,
Amit



pgsql-hackers by date:

Previous
From: David Fetter
Date:
Subject: Re: Efficient output for integer types
Next
From: "Smith, Peter"
Date:
Subject: Proposal: Add more compile-time asserts to expose inconsistencies.