Right. I meant what would you propose for checking inequality for floating-point types? Similar to the original poster asking for BOOST_CHECK_NOT_EQUAL(), what I'm asking about is a BOOST_CHECK_NOT_CLOSE() or BOOST_CHECK_NOT_SMALL() which would use the appropriate tolerance algorithm. Is there a way of getting this kind of result with the current tools? On 6/8/06, Gennadiy Rozental <gennadiy.rozental@thomson.com> wrote:
"Pierre-Jules Tremblay" <iamtrep@gmail.com> wrote in message news:4590bfef0606080504m24a76fffi9a7b2883f56c37b0@mail.gmail.com...
What would you propose for floating-point comparisons?
BOOST_CHECK_CLOSE - checks that two FP values relative difference doesn't exceed specified percent. BOOST_CHECK_CLOSE_FRACTION - checks that two FP values relative difference doesn't exceed specified value BOOST_CHECK_SMALL - checks that module of particular value doesn't exceed tolerance value
Gennadiy
_______________________________________________ Boost-users mailing list Boost-users@lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users