Re: CIDR output format

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Larry Rosenman <ler(at)lerctr(dot)org>
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: CIDR output format
Date: 2000-12-21 15:48:51
Message-ID: 16520.977413731@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Larry Rosenman <ler(at)lerctr(dot)org> writes:
>> It would seem that the appropriate behavior would be to make the default
>> display format for CIDR be like "10.0.0.0/8". Now the text() conversion
>> function already produces this same format. I'd be inclined to leave
>> text() as-is and add a new conversion function with some other name
>> (suggestions anyone?) that produces the shorthand form "10/8" as text,
>> for those who prefer it.

> I would call it cidrshort().

I was thinking something like abbrev(). There is no need to put the
type name in the function; that's what function overloading is for.

> I assume this also is true for INET?

INET doesn't use abbreviation of the address part anyway. The only
display shortcut it has is to suppress "/32" when the netmask is 32.
I figured that text() could produce an un-abbreviated result for an
INET input (as it does now), and abbrev() could produce one with
/32 suppression. In short:

Value Default output text() abbrev()

'127.0.0.1/32'::inet 127.0.0.1 127.0.0.1/32 127.0.0.1
'127.0.0.1/32'::cidr 127.0.0.1/32 127.0.0.1/32 127.0.0.1/32
'127/8'::cidr 127.0.0.0/8 127.0.0.0/8 127/8

This would be a little bit inconsistent, because the default output
format would match text() for CIDR values but abbrev() for INET values.
But that seems like the most useful behavior to me. Possibly others
will disagree ;-)

regards, tom lane

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Michael Fork 2000-12-21 15:56:25 Re: AW: Three types of functions, ala function redux.
Previous Message Matthew Kirkwood 2000-12-21 15:48:46 Re: SSL Connections