I see, great. I was tricked by "greater than 1" in "where s if is an integer greater than 1". Perhaps that should go then.
No pressure, just to have an idea, will this be usable soon as a full SQL functionality? Just today I happen to need it :) Cheers, Roberto On Thu, Sep 22, 2011 at 14:34, Lefteris <[email protected]> wrote: > > SAMPLE 1 will mean 1 tuple > > and > > SAMPLE 1.0 will mean 100% > > it it is an integer is number of tuples and if it is a double is the > fraction of the total size of results > > On Thu, Sep 22, 2011 at 1:37 PM, Roberto Cornacchia < > [email protected]> wrote: > >> + * A new SQL operator has been added to support sampling the result of a >>> query. >>> + * The syntax for sampling is: >>> + * SELECT ... FROM ... WHERE ... SAMPLE s >>> + * >>> + * where s if is an integer greater than 1, it defines the number of >>> rows to be >>> + * in the sample. If s is a double between [0.0,1.0] the it refers to >>> the >>> + * percentage of the result to be sampled. That is if s=0.3 then the >>> sample >>> + * will be 30% the size of the query result. >>> >> >> Does this mean that it is not possible to select 1 random tuple? ("SAMPLE >> 1" would mean 100%) >> >> Roberto >> >> _______________________________________________ >> Checkin-list mailing list >> [email protected] >> http://mail.monetdb.org/mailman/listinfo/checkin-list >> >> > > _______________________________________________ > Checkin-list mailing list > [email protected] > http://mail.monetdb.org/mailman/listinfo/checkin-list > >
_______________________________________________ Checkin-list mailing list [email protected] http://mail.monetdb.org/mailman/listinfo/checkin-list
