On 29 Jun 2011, at 2:26am, Lazarus 101 wrote: > the select statement is > SELECT * from files left join data on files.file_id=data.file_id;
So you read all the records for the correct file_id, and deal with each one as you find it, ignoring those you don't want. Hmm. I don't see why your app isn't far faster with the index. Can an SQLite expert explain it ? By the way, when SQLite does create the index automatically it deletes it when the SELECT finishes. So in your real application it'll keep recreating the index for every SELECT. Simon. _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
