Taylor R Campbell | 1 Sep 2006 08:34
Favicon

Re: 64-bit Scheme48

Several hours after taking another stab at fixing bignums in my
patched Scheme48 for 64-bit environments, I've tracked down a number
of other hard-coded constants designed for 32-bit environments, and in
the process fixed flonums as well.  I've uploaded updated patches at
the same locations,

<http://people.csail.mit.edu/riastradh/tmp/s48-64-all.patch>
<http://people.csail.mit.edu/riastradh/tmp/s48-64.patch>;

with the same instructions for application, except that you also have
to touch `c/scheme48.h' before trying to build.  Also, if you do not
have an existing Scheme48 instance (for example, if you use a AMD64
machine but have not installed 32-bit compatibility libraries), you
can substitute the (gzipped) file at

<http://people.csail.mit.edu/riastradh/tmp/s48-64-initial.image.gz>

for `build/initial.image' instead of bootstrapping it yourself (step 3
in my previous email).

I'm not aware of any bugs in the new patches, but of course I haven't
tested everything; it would be great if anyone out there with 64-bit
environments could try running lots of code in this new 64-bit-native
Scheme48 variant.

Now, while this seems to work (although I haven't tested it
extensively), it's miserably unsatisfactory.  All that I've done is to
change a number of hard-coded numerical constants, some of which are
painfully obscure (e.g., shifting a stob header right by eleven
instead of ten to get the length it encodes), and some C types.  The
(Continue reading)

Jonathan A Rees | 8 Sep 2006 18:26
Favicon

conditions have no debug menu

The purpose of the ,condition command is to allow inspection of the  
constituents of a condition. Apparently this facility was a victim of  
the condition system upgrade.
Jonathan

scheme48
Welcome to Scheme 48 1.3 (made by jar on Sun Jul 31 17:42:31 EDT 2005)
Copyright (c) 1993-2005 by Richard Kelsey and Jonathan Rees.
Please report bugs to scheme-48-bugs <at> s48.org.
Get more information at http://www.s48.org/.
Type ,? (comma question-mark) for help.
 > ,open signals
 > (error "Interesting object" (cons 1 2))

Error: Interesting object
        (1 . 2)
        (&error)
1> ,condition
#{Condition (#{Condition-type &message} (message . "Interesting  
object")) (#{Condition-type &error}) (#{Condition-type &irritants}  
(values (1 . 2))) (#{Condition-type &simple-condition} (type . error)  
(stuff "Interesting object" (1 . 2)))}
1>

Michael Sperber | 8 Sep 2006 19:36
Picon

Re: conditions have no debug menu


Jonathan A Rees <jar <at> mumble.net> writes:

> The purpose of the ,condition command is to allow inspection of the
> constituents of a condition. Apparently this facility was a victim of
> the condition system upgrade. [...]

>> ,open signals
>> (error "Interesting object" (cons 1 2))
>
> Error: Interesting object
>        (1 . 2)
>        (&error)
> 1> ,condition
> #{Condition (#{Condition-type &message} (message . "Interesting
> object")) (#{Condition-type &error}) (#{Condition-type &irritants}
> (values (1 . 2))) (#{Condition-type &simple-condition} (type . error)
> (stuff "Interesting object" (1 . 2)))}
> 1>

Did you type ",inspect"?

--

-- 
Cheers =8-} Mike
Friede, Völkerverständigung und überhaupt blabla

Jonathan A Rees | 8 Sep 2006 20:47
Favicon

Re: conditions have no debug menu

Foo.

Jonathan A Rees | 12 Sep 2006 23:06
Favicon

Attempt to call a template?

Under what circumstances does this happen?

Error: attempt to call a non-procedure
        ('#{Template 8357} ...)

Michael Sperber | 13 Sep 2006 06:53
Picon

Re: Attempt to call a template?


Jonathan A Rees <jar <at> mumble.net> writes:

> Under what circumstances does this happen?
>
> Error: attempt to call a non-procedure
>        ('#{Template 8357} ...)

Some kind of internal error, I'd guess: Some procedure application
received a template instead of a procedure.  User programs normally
shouldn't be seeing templates.

Do you have ,preview or ,debug output?

--

-- 
Cheers =8-} Mike
Friede, Völkerverständigung und überhaupt blabla

Mitchell Wand | 15 Sep 2006 08:40
Favicon

Report from R6RS Editors Committee

I am extremely pleased to announce that a draft version of R6RS is now
available at www.r6rs.org .  A copy will also be posted on
schemers.org .

The charter provides for a six-month public comment period.  Therefore
the editors, in consultation with the steering committee, have
provided a mechanism for comment and discussion.  Details are also at
www.r6rs.org .  

The comment period is now open and will continue until March 15, 2007.

The steering committee thanks the editors for their intensive work on
the draft R6RS, and looks forward to the public comment period.

Enjoy!

For the Steering Committee,
--Mitch Wand

Taylor R Campbell | 15 Sep 2006 09:26
Favicon

floating-point printer does not support non-decimal radices


 (NUMBER->STRING 1.0 16) formerly yielded "1.", but now signals an
 error.

Michael Sperber | 15 Sep 2006 09:30
Picon

Re: floating-point printer does not support non-decimal radices


Taylor R Campbell <campbell <at> mumble.net> writes:

>  (NUMBER->STRING 1.0 16) formerly yielded "1.", but now signals an
>  error.

I don't know what exactly you mean by "floating-point printer" and by
"formerly."  The printer for inexact reals represented as fractions
yields the result you describe, but I don't believe the floating-point
printer ever did.  (And it's not required by R5RS, AFAIR.)

--

-- 
Cheers =8-} Mike
Friede, Völkerverständigung und überhaupt blabla

Mark Smith | 19 Sep 2006 18:04
Picon

Problem with - and / ?

Hi,

This has to be a bug, in the other five to so Scheme's I've used - and
/ take any number of arguments. In Scheme48 this isn't true?

> (+ 1 2 3)
6
> (* 1 2 3)
6
> (/ 1 2 3)

Warning: invalid arguments
         (/ 1 2 3)
         (procedure wants: (:number &opt :number))
         (arguments are: (:exact-integer :exact-integer :exact-integer))
         (&warning)

Error: wrong number of arguments
       ('#{Procedure 90 /} 1 2 3)
1> ^D
> (- 1 2 3)

Warning: invalid arguments
         (- 1 2 3)
         (procedure wants: (:number &opt :number))
         (arguments are: (:exact-integer :exact-integer :exact-integer))
         (&warning)

Error: wrong number of arguments
       ('#{Procedure 84 -} 1 2 3)
(Continue reading)


Gmane