On 7 Jun 2011, at 12:01am, Nico Williams wrote: > On Mon, Jun 6, 2011 at 4:58 PM, Jean-Christophe Deschamps > <j...@antichoc.net> wrote: >> Look at a FP-intensive product like Spatialite (SQLite-based). You'd >> probably agree it performs much more complex tasks than average, mean >> squares and such. >> I'd be very surprised if it used NaN representations! > > Sure, if you're just computing average() then you'll not get any NaNs.
You will if one of the numbers you're taking an average of is a NaN. I've rethought my earlier position. This re-think is the result of the SQL standard being incompatible with the IEEE standard. If you want to do IEEE arithmetic, do it in your own software, and use SQL just for retrieving the numbers you're going to handle. You can store those numbers as numbers, text or BLOBs, whatever works for your purposes. If SQLite is to be changed at all, it should be changed to conform to the SQL standard, including treating division by zero as an error. I regard this change as one worth putting on the bug list, but not one which needs immediate attention. Simon. _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
