Simon, On Mon, Apr 29, 2013 at 2:10 PM, Simon Slavin <[email protected]> wrote:
> > On 29 Apr 2013, at 8:56pm, Igor Korot <[email protected]> wrote: > > > Yes, I read this link. But I don't understand how it applies here. > > Can you make some pseudo code, please? > > I'm sorry, but if you're using a class someone else wrote you may not be > able to use sqlite3_reset(). It depends on what the class assumes about > _step(). > No, this is my class. > > The idea is that instead of > > _prepare(statement1) > _step(statement1), _step(statement1) ... > _finalize(statement1) > > _prepare(statement2) > _step(statement2), _step(statement2) ... > _finalize(statement2) > > you can do this > > _prepare(statement1) > _step(statement1), _step(statement1) ... > _reset(statement1) > _bind(statement1) > _step(statement1), _step(statement1) ... > _finalize(statement1) > I think you are missing "statement2" somewhere here... Now, statement1 is an update and statement2 is insert. It shouldn't change anything, though. Thank you. > > As you correctly noted, it's the _prepare() step that takes a lot of time. > This second form conforms to your original request > > > Problem is sqlite3_prepare_v2() wastes time by compiling SQL statements > and > > I'd like to avoid it by making this call only once and just > > bind the appropriate parameters on each loop iteration. > > Simon. > _______________________________________________ > sqlite-users mailing list > [email protected] > http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users > _______________________________________________ sqlite-users mailing list [email protected] http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
