On Thu, Oct 28, 2010 at 1:05 PM, Robert Haas <robertmhaas@xxxxxxxxx> wrote: > On Tue, Oct 19, 2010 at 7:56 PM, Merlin Moncure <mmoncure@xxxxxxxxx> wrote: >> On Mon, Oct 18, 2010 at 1:16 AM, AI Rumman <rummandba@xxxxxxxxx> wrote: >>> At present for reporting I use following types of query: >>> select crm.*, crm_cnt.cnt >>> from crm, >>> (select count(*) as cnt from crm) crm_cnt; >>> Here count query is used to find the total number of records. >>> Same FROM clause is copied in both the part of the query. >>> Is there any other good alternative way to get this similar value? >> >> Probably the best way to do this type of thing is handle it on the >> client. However, if you want to do it this way and your from clause >> is more complex than 'from table', you can possibly improve on this >> with a CTE: >> >> with q as (select * from <something expensive>) >> select q.* q_cnt.cnt from q, (select count(*) as cnt from q) q_cnt; >> >> The advantage here is that the CTE is materialized without having to >> do the whole query again. This can be win or loss depending on the >> query. > > What about > > select crm.*, sum(1) over () as crm_count from crm limit 10; Hm, after a few quick tests it seems your approach is better in just about every respect :-). merlin -- Sent via pgsql-performance mailing list (pgsql-performance@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-performance
