Thanks very much to all for answers! Sorry, I had some troubles with my internet connection, so I've read them just now. I need exactly unique value by set of this three columns in my table, that is correct. Application do not write temporary data to the disc, because it works in one transaction, which is started at begin and commited only after all queries. I have two general types of selects for this table. First one is "SELECT word,wform,id_norm FROM mgWords WHERE id=<some_value>" and second is "SELECT id FROM mgWords WHERE (word='<some_val1>') AND (wform='<some_val2>') AND (id_norm=<some_val3>)". So I tried to add indexes to all fileds in my table: CREATE INDEX id_norm ON mgWords (id_norm) CREATE INDEX word ON mgWords (word) CREATE INDEX wform ON mgWords (wform) But that didn't make desired result. Thanks for offering delete autoincrement value and operate by rowid, i'll try to do it. Sorry again for my bad English and thank you for attention! -- View this message in context: http://www.nabble.com/Slow-SELECTs-in-application-tp25992880p26073879.html Sent from the SQLite mailing list archive at Nabble.com.
_______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
