Re: Standard Queue status

[Ted Dennison<dennison@telepath.com>]

In article <9u627b$42t$1@nh.pace.co.uk>, Marin David Condic says...
>
>"Stephen Leake" <stephen.a.leake.1@gsfc.nasa.gov> wrote in message
>news:uu1vdfn1c.fsf@gsfc.nasa.gov...
>>
>> > lists must not be a tagged type
>>
>> I could go either way. Making it tagged allows deep copy on a
>> non-limited list, so I lean that way.
>>
>I'd kind of favor tagged and inheriting from Controlled.

That's actually kind of required, I'm afraid. Any other option is going to run
horribly afowl of the transparent storage management fans (and probably require
removal of all the funtional calls).

>> "List elements must be limited private".
>>
>O.K. I can live with that.

Perhaps, but I'm pretty sure we already argued that one and decided against it.
It would require some kind of "copy" routine be supplied as a generic, which
would require even the 90% who *don't* want to deal with limited types to go
create themselves a copy routine that performs vanilla assignment just to make
the generic happy. Some suggested that we could create limited and non-limited
versions, but the general agreement we arrived at was to just drop the whole
thing and not deal with limited types. If it ends up being a huge problem,
someone can always create a parallel "Containers.Lists.Limited_Unbounded package
later.

---
T.E.D.    homepage   - http://www.telepath.com/dennison/Ted/TED.html

No trees were killed in the sending of this message. 
However a large number of electrons were terribly inconvenienced.