From: Iban Hatchond
Subject: STREAM problems
Date: 
Message-ID: <3AACA71C.5AE88195@emi.u-bordeaux.fr>
Hi,
Does anybody know the way to create a binary-stream who's not attach to
a file ???
It may be done with the gray-streams but I cant figure how.

thanks.

From: Kent M Pitman
Subject: Re: STREAM problems
Date: 
Message-ID: <sfw7l1vyyma.fsf@world.std.com>
Iban Hatchond <········@emi.u-bordeaux.fr> writes:

> Does anybody know the way to create a binary-stream who's not attach to
> a file ???
> It may be done with the gray-streams but I cant figure how.

Is this because you don't have any documentation of "gray streams" or
because you didn't understand the documentation?

The original proposal, which is failed (well, tabled) x3j13 issue
STREAM-DEFINITION-BY-USER (by David Gray, hence the informal name
"gray streams") looks pretty clear on how you would do this.  It says,
really very plainly, that you should make a class that is a subclass of
FUNDAMENTAL-BINARY-INPUT-STREAM and that implements STREAM-ELEMENT-TYPE
and STREAM-READ-BYTE.  You probably have  to find out if your vendor
implements gray streams, and also what package they put these various
symbols in.  But other than that, it looks dead easy.

For those who want to read it, I've posted a copy of the issue
STREAM-DEFINITION-BY-USER at my web site.

 http://world.std.com/~pitman/CL/Issues/stream-definition-by-user.html
From: Christophe Rhodes
Subject: Re: STREAM problems
Date: 
Message-ID: <sqlmqbcgeb.fsf@lambda.jesus.cam.ac.uk>
Kent M Pitman <······@world.std.com> writes:

> For those who want to read it, I've posted a copy of the issue
> STREAM-DEFINITION-BY-USER at my web site.
> 
>  http://world.std.com/~pitman/CL/Issues/stream-definition-by-user.html

I presume that the lack of reference to read-sequence and
write-sequence in that proposal is because those functions hadn't yet
been specified; is that the case, or is there some other reason that
they weren't included in that issue, as far as you know?

I mention this because stream-{read,write}-sequence appear to exist in
some implementations and not others...

Christophe
-- 
Jesus College, Cambridge, CB5 8BL                           +44 1223 524 842
(FORMAT T "(·@{~w ········@{~w~^ ~})" 'FORMAT T "(·@{~w ········@{~w~^ ~})")
From: Iban Hatchond
Subject: Re: STREAM problems
Date: 
Message-ID: <3AACE347.FA454449@emi.u-bordeaux.fr>
Christophe Rhodes wrote:

> Kent M Pitman <······@world.std.com> writes:
>
> > For those who want to read it, I've posted a copy of the issue
> > STREAM-DEFINITION-BY-USER at my web site.
> >
> >  http://world.std.com/~pitman/CL/Issues/stream-definition-by-user.html
>
> I presume that the lack of reference to read-sequence and
> write-sequence in that proposal is because those functions hadn't yet
> been specified; is that the case, or is there some other reason that
> they weren't included in that issue, as far as you know?
>
> I mention this because stream-{read,write}-sequence appear to exist in
> some implementations and not others...

It's true and this was a point of my question. Because I want a
two-way-binary-stream, and use it for read-seq and write-seq in it. Under
CMUCL, I still do not find how to do that with the gray-streams extensions.