cv-qualification

[Jim Dehnert]

> From: Alain Miniussi <alainm at cup.hp.com>
> 
> <qualified-name> ::= <CV-qualifier> <qualified-name>
>                  ::= <substitution>
> 
> As explained in the preceding paragraph, the only purpose of 
> the CV-qualifier is to encode the cv qualification of a
> method. I think it is a way to encode the cv-qualification of
> this (since this's type is not represented in the signature).
> 
> then, I have two questions:
> 	1) We don't need to encode the type of this, only it's
> pointed value cv-qualification. Still:
> 	<CV-qualifier> ::= K <type> 
> Is the intent to encode the <type> ?

No.

> If the answer is no, maybe we should have something like:
>         <type> ::= <CV-qualifier> <type>
> and
>         <CV-qualifier> ::= K | V | r

I've fixed it, and also added a tentative move towards your earlier
suggestion of not substituting intermediate types (with a subset of the
qualifiers).

> 	2) if we have two functions with the same name used in a mangling,
> do we want to substitute the second occurence because it's the same name
> or to keed two different encoding since they do not represent the same
> entity ?
> (I am not sure how to interpret:
> "Note that substitutable components are the represented symbolic
> constructs, not their associated mangling character strings."
> in that situation.)

It means that two functions are two distinct names for mangling
purposes, even if they are spelled the same.

Jim

-	    Jim Dehnert		dehnert at sgi.com
				(650)933-4272