Re: index on calculated column

On Wed, 17 Jun 1998 14:35:52 -0700, Richard Caldera <richardc@skechers.com> wrote: > >Joseph: > >An index is placed on table column(s) to provide faster searching/sorting >on that table. Having said this, I don't see the possibility of building >any sort of useful index based on the difference between two column values. An example: select ... from ... where started - finished = 5 would be much faster if there where an index on started - finished. The UDO and possibly some of the 8.x engines can do this sort of thing, but not the standard IDS (from 4.x to 7.x) engines. > Perhaps you can provide some background and more information on what you >are trying to do. > >If you really do want to build an index based on the difference between two >column values, one suggestion is to add a third column for the result of >your calculation and create your index on it. > >-- Richard > >At 12:58 6/17/98 -0500, Joseph Cullipher wrote: >>I am trying to create an index on a calculated column but keep getting >>a syntax error. >> >>create temp table t1 (started smallint,finished smallint); >>create index t1 on t1(started-finished)>> >>I couldn't find anything in the manuals that say I cannot do this but >>it still won't let me. Any suggestions on how I can accomplish this task? >> >>------------------------------------------------------------------------- >>Joseph Cullipher <joseph@cannonexpress.com> Standard Disclaimers Apply >>------------------------------------------------------------------------- >> >> >> >-- > >Richard Caldera >SKECHERS USA, Inc. >MIS Department >richardc@skechers.com >(310) 318-3100 x 1228 Nils Myklebust NM Data AS Norway E-mail: Nils.Myklebust@nmdata.com FAQ at: http://www.iiug.org/techinfo/faq/faq_top.html (Now with ODBC info under "Third party products".)