Class implementation such that assignment x=MyObj does in fact x=MyObj.prop1

Let's say I have a class

classdef TestClass < handle
    properties
        prop1 = 7
        prop2
    end
    methods
        function method1(obj)
            fprintf('prop1 is %g.', obj.prop1);
        end
    end
end

Now, when I do

Obj = TestClass;
x = Obj;

I want that the value of property prop1 is assigned to x, instead of the reference to Obj. So this assignment results to x=7.

And I would prefer to keep the class to be inherited from handle. But value class will also do.

I still would like to access prop2 as Obj.prop2 and call all the methods.

How this can be implemented in a custom class?

Thanks!

[Matt J]

The similar behaviour the MATLAB's class string has.

I don't think so. You should probably demonstrate what you mean.

[Oleg Iupikov]

I meant a case like

s = "abc";  s1 = s;  s1

s1 = "abc"

so it displays the original string. But now I see that I was wrong. Perhaps it just overloads disp, or does a similar trick :) I removed that part from the question to not confuse people.

Answer 1

[Matt J]

You can't have precisely what you have asked for, but you can do this,

classdef TestClass < handle
    properties
        prop1 = 7
        prop2
    end
    methods
        function method1(obj)
            fprintf('prop1 is %g.', obj.prop1);
        end
        
        function out=double(obj) %converter
            out=double(obj.prop1);         
        end
        
        function out=single(obj) %converter          
            out=single(obj.prop1);           
        end
        
    end
end

and then,

>> Obj=TestClass

Obj =

TestClass with properties:

prop1: 7

prop2: []

>> x=zeros(1,3)

x =

0 0 0

>> x(3)=Obj

x =

0 0 7

[Oleg Iupikov]

Unfortunatelly, for this we already need to have a variable of class double or single, and do an indexed assignment.

In any case, thanks for your time and answer!