From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | "Josh Berkus" <josh(at)agliodbs(dot)com> |
Cc: | pgsql-sql(at)postgresql(dot)org |
Subject: | Re: creating variable views |
Date: | 2001-07-08 03:49:36 |
Message-ID: | 5832.994564176@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-sql |
"Josh Berkus" <josh(at)agliodbs(dot)com> writes:
>> I don't understand the distinction you're trying to make here. In
>> general, a view column that is implemented as a function will give
>> the
>> optimizer headaches if you refer to it in WHERE --- the simplicity or
>> complexity of the function has got nothing to do with that AFAICS.
> OK. I thought that I noticed a difference between simple SQL functions
> and PL/pgSQL functions in this respect, but I could easily be mistaken.
AFAIR, the optimizer doesn't pay any attention at all to the
implementation language of a function. The only thing that comes to
mind here is that some care has been taken to mark all the built-in
functions as "iscachable" (or not, as appropriate); but user-created
functions may not be so marked when they should be, leading to loss of
performance in some cases.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Sharmad Naik | 2001-07-08 04:25:22 | distributed database |
Previous Message | Josh Berkus | 2001-07-07 23:12:49 | Re: creating variable views |