Re: oct file ; problem with argument checking

[Mike Miller]

On Fri, Sep 22, 2017 at 13:52:20 +0200, arnaud Lejeune wrote:
> it works with int8(2) but isn't any member of "octave_value" that recognizes
> "2" as an integer when the user writes
> [x,y] = toto( 2 )

It sounds like want your function to accept floating point values that
happen to have integer values, is that correct? And you want to issue an
error when the floating point value is not exacty an integer?

If that's the case then you want to do something like the following

    double a = args(0).scalar_value ();
    if (a != std::floor (a))
      error ("myfunc: A must be an integer scalar value");
    int value = static_cast<int> (a);

If you are working with an older version of Octave, I think you'll need
a return statement after the error call.

-- 
mike

signature.asc
Description: PGP signature