Re: Oct-file version of x>0

[Keith Goodman]

On 5/23/06, John W. Eaton <address@hidden> wrote:
> On 23-May-2006, Keith Goodman wrote:
>
> | While porting a m-file to an oct-file I got stuck on the line
> |
> | y = x'*(x > 0);
> |
> | where x is Nx1.
> |
> | How do I port x>0?
> |
> | I figured out how not to do it all by myself:
> |
> | #include <oct.h>
> | DEFUN_DLD(abc, args, nargout,"y = x'*(x > 0) where x is Nx1")
> | {
> |   octave_value retval;
> |   Matrix x = args(0).matrix_value();
> |   retval = x.transpose() * (x > 0);
> |   return retval;
> | }
> |
> | $ mkoctfile abc.cc
> | abc.cc: In function 'octave_value_list Fabc(const octave_value_list&, int)':
> | abc.cc:6: error: ambiguous overload for 'operator>' in 'x > 0'
>
> #include <oct.h>
> DEFUN_DLD(abc, args, nargout,"y = x'*(x > 0) where x is Nx1")
> {
>   octave_value retval;
>   Matrix x = args(0).matrix_value();
>   retval = x.transpose() * mx_el_gt (x, 0);
>   return retval;
> }
>
> If your next question is "Where is this documented?" the only answer I
> has is that the internals of Octave are not well documented, so use
> the Source, Luke.  Then you'll probably say, "I could have looked all
> day/week/month/year/decade and not found this function!" and I'd have
> to say, "Me too, when is someone going to document this mess?"

Thank you. Thank you. Thank you. You are the documentation!

I think the best way to start is with a table that translates m-file
commands into oct-file commands.

The first column is the m-file command; the second column is the
corresponding oct-file command; maybe a third column for comments.

One table could be for matrix manipulation:

x > 2
mx_el_gt (x, 2)

x'
x.transpose()

any(~isfinite(x))
x.any_element_is_inf_or_nan()

abs(x)
x.abs()

y = [x; z]
x.stack(z)

inv(x)
x.inverse()

x*y
x*y

x.*y
product(x, y)

x./y
quotient(x,y)

x(1:2,5:6)
x.extract(0,4,1,5)

randn(3,4)
octave_rand::distribution("normal")
octrandn = octave_rand::matrix(3,4)

sumsq(x)
x.sumsq()

x(3)
x(2)

sum(x,1)
x.sum(0)

sum(x,2)
x.sum(1)

size(x,1)
x.rows()

size(x,2)
x.cols()

etc.