mtlb_realmax

Description

Scilab equivalent for Matlab realmax is number_properties but not all cases are implemented:

• Scilab equivalent for Matlab realmax or realmax('double') is number_properties("huge").

• There is no Scilab equivalent for Matlab realmax('single')

The function mtlb_realmax(prec) is used by mfile2sci to replace realmax(prec) when it was not possible to know what was the input while porting Matlab code to Scilab. This function will determine the correct semantic at run time. If you want to have a more efficient code it is possible to replace mtlb_realmax calls:

• If prec is equal to "double" mtlb_realmax(prec) may be replaced by number_properties("huge")

Caution: mtlb_realmax has not to be used for hand coded functions.

See also

• number_properties — determine floating-point parameters