On Mar 24, 2009, at 8:46 AM, Carl Witty wrote: > On Mon, Mar 23, 2009 at 11:52 PM, Robert Bradshaw > <[email protected]> wrote: >>> A warning will be somewhat cumbersome to users. The change won't >>> affect, >>> say, 90% of the code using "%", but it will require everyone to >>> care about >>> it. I'm not saying we shouldn't do that, just that it's a heavy >>> thing to do >>> (that, is: if we don't kill performance entirely and check the >>> value at >>> runtime to emit a warning only when it's necessary). >> >> I'm thinking a single (compile time) warning if % is used on cdef >> ints. > > It would be nice if the "kill performance entirely and check the > value at > runtime to emit a warning only when it's necessary" option were also > available (under the control of a compiler flag, probably, and > definitely not by default). Then somebody could compile Sage with > that flag and see if doctesting reveals any problems.
Sure. We have cython.cdivision(True), cython.cdivision(False), and ??? - Robert _______________________________________________ Cython-dev mailing list [email protected] http://codespeak.net/mailman/listinfo/cython-dev
