Hi, I have a table of unique values in the following format:
CREATE TABLE categories (id INTEGER PRIMARY KEY, category UNIQUE TEXT) I want inserts into this table to succeed, even though the corresponding entry already exists. So I use inserts in the following format: INSERT OR IGNORE INTO categories VALUES (NULL, ?) However, if I follow this successful execution with a call to sqlite3_last_insert_rowid() I don't get the rowid of row, which caused the insert to be ignored, but one I preformed previously (which doesn't necessarily have anything to do with this one.) This causes some relations in my database model to break. I know I could use INSERT OR FAIL and a subsequent SELECT, but that seems awkward and like unnecessary code bloat to me. Additionally I kind of think, this breaks the description and sense of sqlite3_last_insert_rowid(). SQlite version used is 3.3.16. Is this intentional? Any suggestions or should I file a ticket for this? Thanks! Mike ----------------------------------------------------------------------------- To unsubscribe, send email to [EMAIL PROTECTED] -----------------------------------------------------------------------------
