--- Scott Marlowe <smarlowe@xxxxxxxxxxxxxxxxx> wrote: > On Wed, 2005-09-21 at 15:02, CSN wrote: > > This appears related to my previous post: > > > http://archives.postgresql.org/pgsql-general/2005-09/msg00809.php > > > > I setup a unique index using the title, yield, and > > directions fields. Some inserts are causing this > > error: > > > > DBD::Pg::st execute failed: ERROR: index row size > > 2832 exceeds btree maximum, 2713 > > CONTEXT: SQL statement "insert into stuff (title, > > yield, directions) values ( $1 , $2 , $3 )" > > > > What do I do? > > Don't insert such big values? :) > > Actually, the standard solution is to use an md5 of > the three fields: > > create unique index threefieldindex on table1 > (md5(field1||field2||field3)); > Ah, cool! Looks like using tsearch2 would be another option, but I don't plan on searching through the yield or directions fields (except at insert time). http://joseph.randomnetworks.com/archives/2005/08/05/postgresql-index-limitation-index-row-size-xxxxx-exceeds-btree-maximum-2713/ CSN __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com ---------------------------(end of broadcast)--------------------------- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
