Skip site navigation (1) Skip section navigation (2)

Re: [HACKERS] pg_type populated incorrectly in some cases?

From: Bruce Momjian <maillist(at)candle(dot)pha(dot)pa(dot)us>
To: mgittens(at)david(dot)gits(dot)nl (Maurice Gittens)
Cc: hackers(at)postgreSQL(dot)org
Subject: Re: [HACKERS] pg_type populated incorrectly in some cases?
Date: 1998-04-27 03:17:24
Message-ID: 199804270317.XAA19761@candle.pha.pa.us (view raw or flat)
Thread:
Lists: pgsql-hackers
> 
> Hi,
> 
> In include/catalog/pg_type.h I've noticed the following code.
> 
> <Code>
> CATALOG(pg_type) BOOTSTRAP
> {
> 	NameData	typname;
> 	Oid			typowner;
> 	int2		typlen;
> 
> 	/*
> 	 * typlen is the number of bytes we use to represent a value of this
> 	 * type, e.g. 4 for an int4.  But for a variable length type, typlen
> 	 * is -1.
> 
> 	...
> </Code>
> 
> The pg_type catalog is then populated with lines like the following.
> 
> <Code>
> DATA(insert OID = 71 (	pg_type		 PGUID 1 1 t b t \054 1247 0 foo bar foo bar c _null_));
> DATA(insert OID = 75 (	pg_attribute PGUID 1 1 t b t \054 1249 0 foo bar foo bar c _null_));
> DATA(insert OID = 81 (	pg_proc		 PGUID 1 1 t b t \054 1255 0 foo bar foo bar c _null_));
> DATA(insert OID = 83 (	pg_class	 PGUID 1 1 t b t \054 1259 0 foo bar foo bar c _null_));
> 
> </Code>
> 
> Notice that the type length for types like pg_class have the value of one (1)?
> I would have expected them to the same length as an oid.
> Am I just seeing things?
> 
> Thanks with regards from Maurice
> 
> 
> 


I believe these should be -1, and I have set them to that.  typprtlen is
not used for any purpose currently, but it pads the other int2 field, so
it is not taking any space, and someday may be useful.


-- 
Bruce Momjian                          |  830 Blythe Avenue
maillist(at)candle(dot)pha(dot)pa(dot)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)

In response to

pgsql-hackers by date

Next:From: Bruce MomjianDate: 1998-04-27 03:21:09
Subject: Re: [HACKERS] Agregates in update?
Previous:From: Bruce MomjianDate: 1998-04-27 03:05:02
Subject: Re: [HACKERS] postgres init script things solved

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group