Re: pg_get_functiondef forgets about most GUC_LIST_INPUT GUCs

Kyotaro HORIGUCHI <horiguchi.kyotaro@lab.ntt.co.jp> writes:
> At Fri, 16 Mar 2018 21:15:54 +0900, Michael Paquier <michael@paquier.xyz> wrote in
<20180316121554.GA2552@paquier.xyz>
>> Let's be clear.  I have listed all the variables in the patch to gather
>> more easily opinions, and because it is easier to review the whole stack
>> this way. I personally think that the only variables where the patch
>> makes sense are:
>> - DateStyle
>> - search_path
>> - plpgsql.extra_errors
>> - plpgsql.extra_warnings
>> - wal_consistency_checking
>> So I would be incline to drop the rest from the patch.  If there are
>> authors of popular extensions willing to get this support, let's update
>> the list once they argue about it and only if it makes sense.  However,
>> as far as I can see, there are no real candidates.  So let's keep the
>> list simple.

> FWIW +1 from me. It seems reasonable as the amendment to the
> current status.

It suddenly struck me that the scope of the patch is wider than it needs
to be.  We don't need special behavior for all GUC_LIST variables, only
for GUC_LIST_QUOTE variables.  (For example, SET datestyle = 'iso, mdy'
works just as well as SET datestyle = iso, mdy.)

This is a good thing not least because all the GUC_LIST_QUOTE variables
are in core.  I've been trying to think of a way that we could have
correct behavior for variables that the core backend doesn't know about
and whose extension shlibs aren't currently loaded, and I can't come up
with one.  So I think that in practice, we have to document that
GUC_LIST_QUOTE is unsupported for extension variables (and, perhaps,
enforce this in DefineCustomStringVariable? or is that cure worse than
the disease?)

            regards, tom lane