www.digitalmars.com         C & C++   DMDScript  

digitalmars.D - ? valid opAssign

reply Thomas Kuehne <thomas-dloop kuehne.thisisspam.cn> writes:
Is the code below valid? 







The reason I ask is that this code passes the compiler,
but I can't find and documentation on it's behaviour.

Thomas
Nov 16 2004
parent reply Daniel Keep <daniel.keep dummy.gmail.com> writes:
Ok, I could be wrong, BUT, I think that it's valid.

The trick here is that (unless Walter changed it from C/C++), assignment 
can be used in an expression.  Specifically, (a=b) first assigns the 
value of b to a, then returns a.

So `return b=b=a;' is assigning a to b, b to itself, then returning the 
value of b.

Maybe :P

	-- Daniel

Thomas Kuehne wrote:
 Is the code below valid? 
 





 
 The reason I ask is that this code passes the compiler,
 but I can't find and documentation on it's behaviour.
 
 Thomas
Nov 16 2004
parent reply Regan Heath <regan netwin.co.nz> writes:
On Tue, 16 Nov 2004 19:11:30 +1100, Daniel Keep 
<daniel.keep dummy.gmail.com> wrote:

 Ok, I could be wrong, BUT, I think that it's valid.

 The trick here is that (unless Walter changed it from C/C++), assignment 
 can be used in an expression.  Specifically, (a=b) first assigns the 
 value of b to a, then returns a.

 So `return b=b=a;' is assigning a to b, b to itself, then returning the 
 value of b.
Or, is the order reversed, i.e. assigns b to b, then a to b. Of course in both cases the result is the same, b=a. The real question is, is the order defined? If so, what is it? Regan
 Thomas Kuehne wrote:






 The reason I ask is that this code passes the compiler,
 but I can't find and documentation on it's behaviour.

 Thomas
-- Using M2, Opera's revolutionary e-mail client: http://www.opera.com/m2/
Nov 16 2004
parent Daniel Keep <daniel.keep dummy.gmail.com> writes:
Hmm... my intuition tells me it's right to left.  Assignment *should* be 
right associative--it computes the right hand side first (thus, 
performing all the assignments right to left).

But then again, IANALL

Regan Heath wrote:
 On Tue, 16 Nov 2004 19:11:30 +1100, Daniel Keep 
 <daniel.keep dummy.gmail.com> wrote:
 
 Ok, I could be wrong, BUT, I think that it's valid.

 The trick here is that (unless Walter changed it from C/C++), 
 assignment can be used in an expression.  Specifically, (a=b) first 
 assigns the value of b to a, then returns a.

 So `return b=b=a;' is assigning a to b, b to itself, then returning 
 the value of b.
Or, is the order reversed, i.e. assigns b to b, then a to b. Of course in both cases the result is the same, b=a. The real question is, is the order defined? If so, what is it? Regan
 Thomas Kuehne wrote:







 The reason I ask is that this code passes the compiler,
 but I can't find and documentation on it's behaviour.

 Thomas
Nov 17 2004