Radovan Antloga <radovan.antloga@...> writes: > http://www.sqlite.org/releaselog/3_7_6.html
Ok, I see it's mentioned there. Given that it changes semantics I would have expected it to be mentioned more prominently though. Now may I ask why it was done? Are there some known real world cases where it is better (I have one real-world case where it's worse and generally think that if application expects reals, it uses reals already anyway, so it does not end up causing integer overflow). > > * Up to 3.7.5, integer operations are always carried out in 64-bit > > integers > > and overflow as expected for that type. > > * From 3.7.6 on, integer operands of *some operations* (left shift being > > notable exception) are converted to real if the operation would > > overflow. _______________________________________________ sqlite-users mailing list [email protected] http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
