Re: array of string

[Adam Beneschan <adambeneschan@gmail.com>]

On Saturday, October 4, 2014 8:20:36 AM UTC-7, Dmitry A. Kazakov wrote:

> > On Saturday, October 4, 2014 10:32:44 AM UTC+2, Dmitry A. Kazakov wrote: 
> >> this case either. Assignment is not an operation in Ada.
> 
> > Just being picky on nomenclature: There is an *assignment statement* in
> > Ada, a part of that is the *assignment operation*.
> 
> > What you mean, is: Assignment is not an *operator* in Ada.
> 
> Not quite.
> 
> Though, I prefer "operator" to mean a lexical element of an expression
> (e.g. +) and "operation" to mean a subprogram taking the type as an
> argument or result (e.g. a primitive operation).

Actually, RM 3.2.3(1) considers both predefined operators (which are subprograms) and "language-defined operations such as assignment or a membership test" (which are not subprograms) to be "predefined operations", and it says those are included in "primitive operations".  

The AARM has this note (3.2.3(1.c)):

# Discussion: We use the term "primitive subprogram" in most of the
# rest of the manual. The term "primitive operation" is used mostly in
# conceptual discussions.

Whether this is actually true is debatable.  I tried searching for the term "primitive operation" in the AARM and found some places in 3.9 where it looks like the term was used where "primitive subprogram" would have been more appropriate, but those where only in the AARM, not the RM.  I didn't search really hard, though.  In any case, I can understand why there would be confusion.

                             -- Adam