Re: Proposed change to BC iterator parameters

[ada_wizard@toadmail.com]

MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
User-Agent: Internet Messaging Program (IMP) 3.2.1

"Steve" <nospam_steved94@comcast.net> writes:

> "Robert A Duff" <bobduff@shell01.TheWorld.com> wrote in message
> news:wccekvc2rol.fsf@shell01.TheWorld.com...
> [snip]
> >
> > I don't understand the point of Param_Type and Param above.  If you want
> > to pass extra information to Apply, nest the instantiation in a place
> > where it can see that data.
> 
> The reason for Param_Type and Param is to avoid using data outside of the
> scope of the Apply routine except for the arguments to the Apply machine.
> It comes from that old school of thought:
>   Avoid global data
>   All inputs and outputs of a procedure should appear in their parameter
> lists

Just for the record, I agree with Robert Duff here :).

While the rules you state are in general reasonable, no rule should be
treated as an absolute. A local instantiation of a generic iterator
that accesses data in the local scope that is not in the parameter
list is _not_ using global data. SAL takes this approach.

In addition, it is useful to have a version of the iterator that does
_not_ take Param_Type. I haven't looked at the booch components (yes,
that is a failing on my part :); perhaps it offers both flavors of
iterators. For me, that is an unnecessary complication, requiring
extra testing and maintenance.


On a slightly related topic:

There was an interesting paper at the SigAda conference yesterday (I'm
posting from my hotel) on "shortcuts" as an efficient way to implement
"safe" containers (no dangling links, even when iterators are used to
delete items). Looked like a good idea; I recommend checking it out. I
think the proceedings will be in a future issue of Ada Letters.

-- 
-- Stephe

___________________________________________________________
This mail sent using ToadMail -- Web based e-mail @ ToadNet