www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.learn - lvalue <-> rvalue problem

reply "Namespace" <rswhite4 googlemail.com> writes:
The following code prints:

/home/c494/c719.d(27): Error: (Vec2!(float) __ctmp1173 = 
_D4c71911__T4Vec2TfZ4Vec26__initZ;
, __ctmp1173).this(4F, 2F) is not an lvalue

[code]
import std.stdio;

struct Vec2(T) {
public:
	T x;
	T y;
	
	this(T x, T y) {
	
	}
}

alias Vec2f = Vec2!(float);

class Foo {
public:
	static Foo make(float i, ref const Vec2f a) {
		return new Foo();
	}
	
	static Foo make(float i, const Vec2f a) {
		return Foo.make(i, a);
	}
}

void main() {
	Foo f = Foo.make(42, Vec2f(4, 2));
}
[/code]

If I change i from float to int it works. Also if I write 42f 
instead of 42. That is strange. Could someone explain me that?

Same with this code: http://dpaste.1azy.net/2c98fe95
But there I found no workaround.
Jan 27 2013
next sibling parent reply "Dicebot" <m.strashun gmail.com> writes:
Looks like a bug in function overload selection.
Jan 27 2013
next sibling parent "Namespace" <rswhite4 googlemail.com> writes:
On Sunday, 27 January 2013 at 11:42:29 UTC, Dicebot wrote:

 Looks like a bug in function overload selection.


I hope not!
That would be really bad. That is part of my solution as long as 
auto ref isn't there. :D
Jan 27 2013
prev sibling parent reply Jonathan M Davis <jmdavisProg gmx.com> writes:
On Sunday, January 27, 2013 12:42:28 Dicebot wrote:

 Looks like a bug in function overload selection.


Definitely.

- Jonathan M Davis
Jan 27 2013
parent reply "Namespace" <rswhite4 googlemail.com> writes:
On Sunday, 27 January 2013 at 23:05:16 UTC, Jonathan M Davis 
wrote:

 On Sunday, January 27, 2013 12:42:28 Dicebot wrote:

 Looks like a bug in function overload selection.


 Definitely.

 - Jonathan M Davis


And that prevents a workaround for the missing auto ref. :o)
It seems dmd 2.060 is far more useable than 2.061 (for users 
which use structs, of course).
I will switch back.
Jan 27 2013
parent reply "Namespace" <rswhite4 googlemail.com> writes:
 And that prevents a workaround for the missing auto ref. :o)
 It seems dmd 2.060 is far more useable than 2.061 (for users 
 which use structs, of course).
 I will switch back.


I'm still interested in this problem. But how can I identify the 
specific compiler code where this error comes from?
I know where the error is thrown and where the matching fail 
comes from (expression.c) but not _why_ the matching fails happen.
Could you give me a quick explanation?
Jan 28 2013
parent reply "Jonathan M Davis" <jmdavisProg gmx.com> writes:
On Monday, January 28, 2013 11:16:26 Namespace wrote:

 And that prevents a workaround for the missing auto ref. :o)
 It seems dmd 2.060 is far more useable than 2.061 (for users
 which use structs, of course).
 I will switch back.


 
 I'm still interested in this problem. But how can I identify the
 specific compiler code where this error comes from?
 I know where the error is thrown and where the matching fail
 comes from (expression.c) but not _why_ the matching fails happen.
 Could you give me a quick explanation?


I have no idea why it's failing. That would require debugging the compiler, 
and the only part of the compiler that I know much about is the lexer.

- Jonathan M Davis
Jan 28 2013
parent "Namespace" <rswhite4 googlemail.com> writes:
It's fixed already. But I searched at the wrong place. The 
compiler is too huge for me. o.O
Jan 28 2013
prev sibling parent reply "Namespace" <rswhite4 googlemail.com> writes:
 Same with this code: http://dpaste.1azy.net/2c98fe95
 But there I found no workaround.


This works: new C(cast(A) new B(), FloatRect(0, 1, 2, 3));
But that is ugly. o.O

I think you are right and it is a bug in function overload 
selection. :(
Jan 27 2013
parent "Namespace" <rswhite4 googlemail.com> writes:
Ok, I will open a bug report for this.
Jan 27 2013