Re: [HACKERS] Error creating index

From: "Thomas G(dot) Lockhart" <lockhart(at)alumni(dot)caltech(dot)edu>
To: Vince Vielhaber <vev(at)michvhf(dot)com>
Cc: hackers(at)postgreSQL(dot)org
Subject: Re: [HACKERS] Error creating index
Date: 1999-01-04 06:22:33
Message-ID: 36905E29.E24F2301@alumni.caltech.edu
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

> Is this telling me that I can't have an index of type int8? Why not?
> FreeBSD 2.2.8, PostgreSQL cvsup'd less than a week ago.

Because I added the int8 data type to the backend without knowing which
platforms would actually support it. (As an aside, I had tried to
solicit results of a small test program for various platforms on the
hackers list, but can't recall getting any responses.)

Without knowing that, it wasn't clear to me whether the int8 would be
able to stay, and I didn't want to embed these changes in the index
support system tables if they would need to be removed. These tables are
the most difficult to configure and support, and it isn't as easy to
trace which entries support which types.

Anyway, it seems int8 indices would be a good feature to have for v6.5,
eh? One could write some user-defined index operators as an interim
solution.

- Tom

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Thomas G. Lockhart 1999-01-04 07:12:29 Re: Date/time fixes for HAVE_TM_ZONE platforms
Previous Message Bruce Momjian 1999-01-04 04:44:13 Re: [HACKERS] Reloading a 6.3db into 6.4.1t