Re: [HACKERS] union and LIMIT problem - Mailing list pgsql-hackers

From Tom Lane
Subject Re: [HACKERS] union and LIMIT problem
Date
Msg-id 29687.939317382@sss.pgh.pa.us
Whole thread Raw
In response to Re: [HACKERS] union and LIMIT problem  (Bruce Momjian <maillist@candle.pha.pa.us>)
Responses Re: [HACKERS] union and LIMIT problem  (Bruce Momjian <maillist@candle.pha.pa.us>)
List pgsql-hackers
>>>> * UNION with LIMIT fails
>> 
>> and must fail by it's implementation. LIMIT is handled by the
>> outermost executor loop, suppressing OFFSET result tuples and
>> stopping execution after LIMIT results sent to the client.

> Ah, but it works sometimes:

Well, the real question is what do you mean by "works" or "fails".
In particular, do you think that LIMIT applies to the overall result
of the whole query, or to any one sub-select?

IIRC, ORDER BY is supposed to apply to the end result (and you can
only write it at the very end of the query, not after a sub-select),
and I'd vote for making LIMIT work the same.  In which case the
executor should be fine, and we probably just have a problem with
the parser hanging the info on the wrong node of the querytree...
        regards, tom lane


pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: [HACKERS] Re: psql and comments
Next
From: Tom Lane
Date:
Subject: Re: [HACKERS] psql and comments