On Mon, 6 May 2013 20:55:01 +0300 Mikael <mikael.tr...@gmail.com> wrote:
> Currently in the absence of a sqlite3_reset_last_insert_rowid(), the > closest you can do to figure out what ID was actually inserted by the > most recent SQLite query is: > > * Lock mutex > * Get sqlite*'s last insert row ID > * Perform the query > * Get number of changed rows > * Get sqlite*'s last insert row ID > * Unlock mutex I don't understand. The process doing the insert presumably knows which data it inserted. Why not turn around and SELECT the ids for the inserted values? No mutex and no guesswork. Your process seems to imply that the the inserted values don't constitute a key, or that you want to compute the row IDs for performance. Those don't strike me as sufficient reason to perturb the API. --jkl _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
