Documentation of "scope" function parameters

[Jason House]

Michel Fortin wrote:

> I'm reading the Functions section of the D 2.0 documentation.
> <http://www.digitalmars.com/d/2.0/function.html#parameters>.
> Under Function Parameters, I read this:
> 
> Parameter storage classes are in, out, ref, lazy, final, const,
> invariant, or scope.
> 
> then this:
> 
> The in storage class is equivalent to const scope.
> 
> But there is no explaination how scope affect things.
> 
> If I go find scope in the Attributes section
> <http://www.digitalmars.com/d/2.0/attribute.html#scope>
> I can find this which may be relevant to the case of a funciton parameter:
> 
> For local declarations, scope implements the RAII (Resource Acquisition
> Is Initialization) protocol. This means that the destructor for an object
> is automatically called when the reference to it goes out of scope.
> 
> Is this really what it does? It sounds pretty silly that the function
> destroys the object the caller has passed to it, so I guess it doesn't.
> 
> However, this part may apply when talking about function parameters:
> 
> Assignment to a scope, other than initialization, is not allowed.
> Rationale: These restrictions may get relaxed in the future if a
> compelling reason to appears.
> 
> which would mean you just can't reassign a new value to a scope parameter
> inside the function. That's just a guess of what it does, and to me it
> sounds pointless and misnamed.
> 
> What is it in reality?
> 
> Anyhow, I think what scope does with a function parameters should be
> better documented in the Functions section.
> 

My understanding is that the scope part of in hasn't been implemented yet :( 
My understanding is that the scope keyword should imply that the data is
not available after the function call is over.  I personally like the idea
of (non-invariant) data being able to become scope invariant (for the
duration of a function call), but I haven't found anyone that shares my
ideas.  To me, it's a logical extension of the const system...