On Mon, Aug 13, 2012 at 03:43:46PM +0200, Bishwa Shrestha scratched on the wall: > Hi, > > I've recently started using sqlite3 c-library. I'm using the > in-memory databases which are pretty fast since I'm loading data > dynamically into sqlite3 from a different backend (had to do this as > we wanted to provide an SQL interface to our users).
If you want to provide an SQL interface to an existing data source, I would strongly recommend you look at Virtual Tables. They look and smell like normal tables, but all of their access and data generation is done via application level code. You can write a series of functions that essentially slip in between the SQL engine and the data access layer, and can redirect requests for table data to an external source. Virtual Tables can be read-write or read-only. Read-only tables are pretty fast to implement, while read-write tables would allow external applications to modify the backend (no idea if that is desirable or not in your case). Virtual Tables are an extremely powerful and often overlooked part of SQLite. You can do some pretty amazing and powerful things with them, as evident by the Full Text Search engine in SQLite, as well as the R-Trees extension, both of which are built on top of Virtual Tables. SQLite docs: http://www.sqlite.org/vtab.html http://www.sqlite.org/c3ref/create_module.html Virtual Tables are also covered fairly well in "Using SQLite": http://shop.oreilly.com/product/9780596521196.do -j -- Jay A. Kreibich < J A Y @ K R E I B I.C H > "Intelligence is like underwear: it is important that you have it, but showing it to the wrong people has the tendency to make them feel uncomfortable." -- Angela Johnson _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
