Re: pgbench - extend initialization phase control

Hello Masao-san,

>> I do not think that this is desirable. It would be a regression, and
>> allowing a no-op is not an issue in anyway.
>
> Why is that regression, you think?

Because "pgbench -I ' d'" currently works and it would cease to work after
the patch.

> I think that's an oversight. If I'm missing something and accepting a
> blank character as no-op in also checkInitSteps() is really necessary
> for some reasons, which should be documented. But, if so, another
> question is; why should only blank character be treated as no-op, in
> checkInitSteps()?

The idea is to have one character that can be substituted to remove any
operation.

On principle, allowing a no-op character, whatever the choice, is a good
idea, because it means that the caller can take advantage of that if need
be.

I think that the actual oversight is that the checkInitSteps should be
called at the beginning of processing initialization steps rather than
while processing -I, because currently other places modify the
initialization string (no-vacuum, foreign key) and thus are not checked.

I agree that it should be documented.

Attached patch adds a doc and moves the check where it should be, and
modifies a test with an explicit no-op space initialization step.

--
Fabien.