I don't know anything about boost.units, but radians are
dimensionless. So if you are tracking radian as a dimensionless unit
it makes sense to have it in the units for torque. Not for
newton-meter in general though, which is not necessarily used for
torque.
On Thu, Aug 25, 2022 at 8:50 PM Albert Dvornik via Boost
It looks like my previous message wasn't visible by non-HTML readers (including the list archive) because I didn't configure my so-called mailer correctly. Sorry for the duplication and the hassle.
bert
-------------------- Hi, all.
The definitions of boost::units::torque_dimension, boost::units::si::torque, boost::units::si::newton_meter and boost::units::si::newton_meters seem broken. The most obvious symptom of this is that boost::units::si::newton_meter isn't the same thing as boost::units::si::newton * boost::units::si::meter (i.e. an actual N m), but is instead defined to be equivalent to *N m / rad* instead.
I'm attaching a source file that shows the problem. I've compiled it with Boost 1.80, g++ 10.2.1.
Am I confused in some way? If not, then I believe that the fix would be to modify boost/units/physical_dimensions/torque.hpp and remove the ", plane_angle_base_dimension,-1" in the current definition of torque_dimension.
Regards, bert Dvornik
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
-- Frank