Re: Poor performance using CTE - Mailing list pgsql-performance

From Claudio Freire
Subject Re: Poor performance using CTE
Date
Msg-id CAGTBQpZCPOpt4azPyUP-Nek94qapYFNfRX-cY=4DD08XM_G2Cw@mail.gmail.com
Whole thread Raw
In response to Re: Poor performance using CTE  (Andrew Dunstan <andrew@dunslane.net>)
List pgsql-performance
On Wed, Nov 21, 2012 at 1:35 PM, Andrew Dunstan <andrew@dunslane.net> wrote:
>>>> Why syntax? What about a guc?
>>>>
>>>> collapse_cte_limit?
>>>
>>> Because there are very good reasons to want to current behaviour. A guc
>>> is a global either/or so I don't see it helping much.
>>
>> set collapse_cte_limit=8;
>> with blah as (blah) select blah;
>>
>> Not global at all.
>>
>
> Then you have to unset it again, which is ugly. You might even want it
> applying to *part* of a query, not the whole thing, so this strikes me as a
> dead end.

Really?

Because I've seen here people that want it generally (because
Oracle/MSSQL/your favourite db does it), and people that don't want it
(generally because they need it). I haven't seen any mention to mixing
fenced and unfenced usage.


pgsql-performance by date:

Previous
From: Andrew Dunstan
Date:
Subject: Re: Poor performance using CTE
Next
From: Tom Lane
Date:
Subject: Re: Poor performance using CTE