Re: AW: AW: [HACKERS] attlen weirdness? - Mailing list pgsql-hackers

From Bruce Momjian
Subject Re: AW: AW: [HACKERS] attlen weirdness?
Date
Msg-id 199803121348.IAA15190@candle.pha.pa.us
Whole thread Raw
In response to AW: AW: [HACKERS] attlen weirdness?  (Zeugswetter Andreas <andreas.zeugswetter@telecom.at>)
List pgsql-hackers
>
> > Right now: yes, but I would like to see the char() type changed to a fixed length type (which it really is)
> > then it would have the length in attlen. Right now the length for char and varchar is in atttypmod.
> >
>
> Again, we can do this, but we then have to have atttypmod available in
> all places where we check for length.  I will add it to the TODO list.
>
> I am starting to see the problem, only atttypmod is passed around not attlen.
> Therefore even if we do this, the char() length of a column still has to be in atttypmod.
> So it will stay the same for Julie, whether it is taken out or not. But since it is really
> hard for a generic application like ODBC, maybe we could change atttypmod to not
> include the VARHDRSZ, most of the time it is needed without VARHDRSZ anyway ?
> I am starting to think that this would be more important than to remove the 4 bytes.

Those four bytes have always been there.  Check psql or pg_dump.  See
them subtracting VARHDRSZ all the time.

--
Bruce Momjian                          |  830 Blythe Avenue
maillist@candle.pha.pa.us              |  Drexel Hill, Pennsylvania 19026
  +  If your life is a hard drive,     |  (610) 353-9879(w)
  +  Christ can be your backup.        |  (610) 853-3000(h)

pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: [HACKERS] port/getrusage.c?]
Next
From: Bruce Momjian
Date:
Subject: Re: [HACKERS] Re: [QUESTIONS] Does Storage Manager support >2GB tables?