Re: regclass and format('%I')

2015-03-15 3:09 GMT+01:00 Jason Dusek <jason(dot)dusek(at)gmail(dot)com>:

> On 14 March 2015 at 09:17, David G. Johnston <david(dot)g(dot)johnston(at)gmail(dot)com>
> wrote:
> > On Saturday, March 14, 2015, Jason Dusek <jason(dot)dusek(at)gmail(dot)com> wrote:
> >> It honestly seems far more reasonable to me that %s and %I should do
> >> the exact same thing with regclass. My reasoning is as follows:
> >>
> >> ‘%I’ formats a something such that it is a valid identifier,
> >>
> >> regclass is already a valid identifier,
> >>
> >> therefore, do nothing.
> >>
> >> Another line of reasoning:
> >>
> >> If you format with ‘%s’ you are saying: I don’t care whether it’s a
> >> valid identifier or literal or whatever, just put the string there,
> >>
> >> but when we sub a regclass into a string, we want it to be a valid
> >> identifier,
> >>
> >> therefore we should write ‘%I’ when subbing it, so as not to confuse
> >> our readers,
> >>
> >> therefore ‘%I’ should do nothing.
> >>
> >
> > I agree with the theory but adding type specific logic to format is
> going to
> > be difficult. The first thing the system does is convert all of the
> inputs
> > to text. Inside format() everything is text and so it has no way to know
> > that the type was regclass and should not be quoted again.
>
> Could it work to add type-specific logic for the cast from `regclass`
> to `name`? It would be nice to have something formulaic: always format
> identifiers with `%I`, always cast to `name` before formatting.
>

I don't think, so it can help - first, it is workaround and it doesn't help
for somebody who doesn't read a documentation. Same effect you can get if
you write "doesn't use %I for regclass, regtype types", although this
sentence is strange. I agree with you so it is bug or minimally not user
friendly design.

A some good solution should be safe function quote_identif, that protect us
against double quoting. This logic should be elsewhere than inside "format"
function. I am thinking so we can do it. It breaks nothing. Implementation
should not be too much complex, because "new" function quote_identif can do
nothing for safe types, so it can take string as input parameter and typid
as second parameter.

Pavel

>
> Kind Regards,
> Jason Dusek
>
>
> --
> Sent via pgsql-general mailing list (pgsql-general(at)postgresql(dot)org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-general
>