Floating Point Lib And Member Variables?

[Neil 0]

Hi All

 

I think I have found a problem using the floating point lib with member c++ member variables.

 

The complier will not let you

 

void mycalss::func()

{

float localvar = var;

}

 

where var is float myclass::_var;

A public member that is a float

 

Is this a bug?

 

 

Neil

[Dave 0]

Neil,

 

Is this a bug?

Looks like it is a compiler bug

 

Regards

Dave

[Neil 0]

Hi Dave

 

Thanks. Im sure you will fix it.

 

I have noticed some other limitations of the compiler as a C++ complier. Are these on a todo list?

 

You cant use operator type functions, e.g. operator[]

You cant return a reference from a member function int& foo()

You don't seem to be able to define a copy constructor foo::foo(const foo& that)

You can not do int foo(0), you have to do int foo = 0; which is not the same.

 

It also seems that if you create an object that is bigger than 256 bytes it does not fit into memory even if the individual member variables are all smaller than 256 bytes

 

I dont want to pick faults, as Ive said before, I think your product is good. After all I am using it to develop a product and I have paid for a license!!

 

Best regards

 

Neil

 

PS any chance that the float lib will have a proper C++ interface so you can do float foo = bar + poo rather than the convoluted C function calles?

[Pavel 0]

...The complier will not let you

 

void mycalss::func()

{

float localvar = var;

}

 

where var is float myclass::_var;

A public member that is a float...

 

This problem has been fixed and fix will available in the next release.

 

Regards,

Pavel