Re: status of Ada STL?

[Brian Rogoff <bpr@bpr.best.vwh.net>]

On Wed, 19 Jun 2002, Robert A Duff wrote:
> Brian Rogoff <bpr@bpr.best.vwh.net> writes:
>
> > I don't know why anyone would hate garbage collection. That seems like a
> > rather odd sentiment. It also seems that if this is true he'll never be
> > satisfied, since GC is just the extreme on the automatic end for memory
> > management; all non-GC mem managers are "by hand" at some level.
>
> By "by hand", I mean the malloc/free or new/Unchecked_Deallocate style
> of C, Ada, Pascal, etc.  Cyclone's regions are not "by hand", and can
> replace GC safely and efficiently -- if your program matches the
> necessary restrictions.

Well, I think that there will always (as long as I live, say ;) be places
where doing everything "by hand" will be cost effective, so there will
always be a place for a language which facilitates this. And I agree that
Cyclone currently does not provide all of the forms of memort management
one could reasonably need. It's a research project.

> > Well, it ain't derived from Ada!
>
> I meant regions, specifically.

Got it, my misinterpretation. For anyone reading who is unaware, regions
are roughly what some people call arenas, with some extensions.

> >... Anyways, pattern matching, like all
> > modern functional languages have, would be a worthy addition to an
> > Ada like language.
>
> Feel free to sketch out your ideas.  I'd be interested.

I believe the Cyclone work could be extended directly to an Ada like
language. Conceptually, we want to add let (like destructuring-bind for
you Lispers) and enhance Ada's case statement with pattern matching forms.
In Cyclone, they add a "$" to the C syntax to handle this, and uses
C's && for pattern guards. We'd use "and".

Probably it would be more interesting if such a syntax could be entirely
upward compatible with Ada 95, but I'm not sure how important that is.

I'll put together some examples, and post them under a more relevant
subject line.

Just so everyone knows where I'm coming from, rather than being some
disgruntled C programmer  who wants to make Ada look like C, I'm a
disgruntled ML programmer who'd like to see an Ada like language (meaning
low level) with some MLish features. As far as high level languages go,
I think some ML variant (strict FPL with imperative features, modules)
extended with support for overloading is about right for me. Even the
current MLs are better than the amazingly bad gunk (like Java) that gets
foisted on the programming community.

> > > As you say, it's worthy of study; Tucker has mentioned some ideas for
> > > making it more powerful.
> >
> > I must have missed this. Did the discussion happen here, or elsewhere?
>
> No, sorry, I didn't mean here.  These discussions happened in my
> office and Tucker's office.  ;-)

Are you preparing a paper on it?

-- Brian