| From: | "Robert Haas" <robertmhaas(at)gmail(dot)com> |
|---|---|
| To: | Decibel! <decibel(at)decibel(dot)org> |
| Cc: | "Tom Lane" <tgl(at)sss(dot)pgh(dot)pa(dot)us>, "Gregory Stark" <stark(at)enterprisedb(dot)com>, "Postgres Hackers" <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Regression in IN( field, field, field ) performance |
| Date: | 2008-10-25 18:31:06 |
| Message-ID: | 603c8f070810251131m4e9cf4efl8c6e877d1fbd2022@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
> There's 6 cases here, in a 2x3 array. In one dimension, the LHS can be
> either a Var or a fixed value. In the other dimension, the three
> possibilities are 1: everything on the RHS is a fixed value, 2: some fixed,
> some not, 3: everything on the RHS is a variable:
[...lengthy discussion of cases...]
It seems like you're saying that the only time the array wins here is
when you're comparing an expression against a whole bunch of
constants. Given that, would it make any sense to put all the
constants in an ARRAY and use OR for any variables? i.e. given
Whatever IN (Const1, Var1, Const2, Var2, Const3, Var3)
Generate:
Whatever = ANY (ARRAY[Const1, Const2, Const3]) OR Whatever = Var1 OR
Whatever = Var2 OR Whatever = Var3
...Robert
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Robert Haas | 2008-10-25 18:59:07 | Re: Handling NULL records in plpgsql |
| Previous Message | Tom Lane | 2008-10-25 16:36:24 | Re: Full socket send buffer prevents cancel, timeout |