Re: const mapper functions

[John W. Eaton]

On 1/8/19 3:24 PM, Rik wrote:
> jwe,
>
> Is there any reason not to make the function reference passed in to
> do_rc_map constant?
>
> ov-flt-re-mat.cc (738)
>      V669     The 'fcn' argument is a non-constant reference. The analyzer
> is unable to determine the position at which this argument is being
> modified. It is possible that the function contains an error.
>
> Line 737-738 of ov-flt-re-mat.cc is
>
> static octave_value
> do_rc_map (const FloatNDArray& a, FloatComplex (&fcn) (float))
>
> The other *-re-mat.cc have the same question.  The function passed in is a
> mapper like sin, cos, etc. and I can't imagine we would change it mid-stream.

I'm in favor of declaring any reference or pointer arguments and member 
functions const if we can.  So yes, if you want to do i and have the 
time, please fix them.

I generally don't see a need for doing that for local values that are 
never modified after initialization, but I can also see the argument for 
that in some cases.

jwe