You don't need to define the PRIMARY KEY as NOT NULL - it's implied. The column constraint flow diagram in the documentation in fact doesn't allow it: http://www.sqlite.org/syntaxdiagrams.html#column-constraint
Maybe that's the problem? Try recreating the table without the NOT NULL constraint on the "id" column and see if it works after that. Thanks, Nick. -----Original Message----- From: sqlite-users-boun...@sqlite.org [mailto:sqlite-users-boun...@sqlite.org] On Behalf Of Brandon Wang Sent: 01 December 2009 17:06 To: sqlite-users@sqlite.org Subject: [sqlite] Possibly a bug in SQLite? Hello, I've come upon a interesting scenerio. .sqlite> .schema rg_configuration CREATE TABLE 'rg_configuration' ( "id" INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL, "configurationName" TEXT NOT NULL, "host" TEXT, "user" TEXT, "parentArch" INTEGER NOT NULL, "parentJob" INTEGER NOT NULL, "parentSubblock" INTEGER NOT NULL, "parentBlock" INTEGER NOT NULL, "canBeRun" INTEGER DEFAULT (1) ); [Addititonal indices, triggers, etc. here] One of my scripts attempts to execute the following: INSERT INTO main.rg_configuration (configurationName, parentArch, parentJob, parentSubblock, parentBlock, canBeRun) VALUES ('full_chip.nofeedthru', 9565, 3014, 33, 8, 1); Upon which I get the error: SQL error: PRIMARY KEY must be unique I'm not specifying the primary key, id. Is there some error on my part? Thanks! -Brandon _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
