Thank you, I was missing the parens.
If I do an explain I see:
-> Index Scan using m_u_and_p_key on m (cost=0.00..3035.22 rows=1363 width=12)
even if I put a limit 1 on the select. Why is that?
Stephan Szabo wrote:
> On Mon, 20 Aug 2001, Joseph Shraibman wrote:
>
>
>>I want to select a boolean if there exists a row in another table that matches this one.
>>So I did select ..., (select count(*) from table2 where ...) > 0 ...
>>but that count(*) was taking forever. I know there is a better way to do it, but whenever
>>I try to use EXISTS I get a syntax error. What is the proper way?
>>
>
> Hmm, on current sources I can do:
> select ..., exists (select * from table2 where ...) from table1;
>
> I don't know if that's new though...
>
--
Joseph Shraibman
jks@selectacast.net
Increase signal to noise ratio. http://www.targabot.com