headers
Mark Evenson | 1 Feb 2009 10:17
Picon
Favicon

Re: [armedbear-cvs] r11605 - in trunk/abcl: . src/org/armedbear/lisp test/lisp/abcl

Erik Huelsmann wrote:
>>   public static String getVersion()
>>   {
>> -    return "0.13.0-dev";
>> +    return "0.12.37";
>>   }
>>  }
> 
> 0.12.37? What's wrong with 0.13.0-dev? :-)

Reverted my inadvertent publishing of my "internal" versioning for ABCL.

Sorry about that.

--

-- 
"A screaming comes across the sky.  It has happened before, but there is
nothing to compare to it now."

------------------------------------------------------------------------------
This SF.net email is sponsored by:
SourcForge Community
SourceForge wants to tell your story.
http://p.sf.net/sfu/sf-spreadtheword
Permalink | Reply | 0 comments |
headers
Chun Tian (binghe | 4 Feb 2009 15:01
Picon

New Java-based ResearchCyc released

Hi, ABCL

This post is not about ABCL, but related to Common Lisp and Java.

ResearchCyc [1], the full-powered Cyc version, the world's biggest  
upper ontology platform, has changed to Java-based releases from 2009.  
If you have a ResearchCyc license which is either commercial or free,  
you should try the new version which runs on any Java-based platform  
now!

[1] http://researchcyc.cyc.com/releases/

As I know, the Cyc platform is fully written in Common Lisp. The Lisp  
version should run on LispWorks and Allegro CL. Years ago, the Cycorp  
developed a Lisp->C compiler, which converted Cyc source code into C  
code and compile it. The C version runs on at least Windows and Linux  
platform, then only Linux is supported during 2007-2008. Now seems  
they go further, a stronger Lisp->Java compiler has been written to  
convert Cyc source code from Common Lisp to Java, and make Cyc runs on  
JVM now.

Unfortunately, I cannot get touch with this Lisp->Java compiler. ABCL  
is just what we have in hand, hopes it will be stronger in the future:)

--
Chun Tian (binghe)
NetEase.com, Inc.
P. R. China

------------------------------------------------------------------------------
(Continue reading)

Permalink | Reply | 1 comment |
headers
logicmoo | 4 Feb 2009 18:17
Picon

Re: New Java-based ResearchCyc released


----- Original Message ----- 
From: "Chun Tian (binghe)" <binghe.lisp <at> gmail.com>
To: <armedbear-j-devel <at> lists.sourceforge.net>
Sent: Wednesday, February 04, 2009 6:01 AM
Subject: [j-devel] New Java-based ResearchCyc released

> Hi, ABCL
>
> This post is not about ABCL, but related to Common Lisp and Java.
>
> ResearchCyc [1], the full-powered Cyc version, the world's biggest
> upper ontology platform, has changed to Java-based releases from 2009.
> If you have a ResearchCyc license which is either commercial or free,
> you should try the new version which runs on any Java-based platform
> now!
>
> [1] http://researchcyc.cyc.com/releases/
>
> As I know, the Cyc platform is fully written in Common Lisp. The Lisp
> version should run on LispWorks and Allegro CL. Years ago, the Cycorp
> developed a Lisp->C compiler, which converted Cyc source code into C
> code and compile it. The C version runs on at least Windows and Linux
> platform, then only Linux is supported during 2007-2008. Now seems
> they go further, a stronger Lisp->Java compiler has been written to
> convert Cyc source code from Common Lisp to Java, and make Cyc runs on
> JVM now.
>
> Unfortunately, I cannot get touch with this Lisp->Java compiler. ABCL
> is just what we have in hand, hopes it will be stronger in the future:)
(Continue reading)

Permalink | Reply | 0 comments |
headers
Ville Voutilainen | 9 Feb 2009 18:19
Picon

Current maxima result..

Maxima once again builds, but the tests fail to run. This is with our SVN head.

Trace:
java.lang.NoSuchFieldError: G44646
	at org.armedbear.lisp.abcl51039.execute(Unknown Source)
	at org.armedbear.lisp.LispThread.execute(LispThread.java:618)
	at org.armedbear.lisp.quadpack_11._execute(quadpack.lisp:77)
	at org.armedbear.lisp.ClosureTemplateFunction.execute(ClosureTemplateFunction.java:77)
	at org.armedbear.lisp.CompiledClosure.execute(CompiledClosure.java:63)
	at org.armedbear.lisp.LispThread.execute(LispThread.java:618)
	at org.armedbear.lisp.dqk21_2._execute(dqk21.lisp:10)
	at org.armedbear.lisp.ClosureTemplateFunction.execute(ClosureTemplateFunction.java:128)
	at org.armedbear.lisp.CompiledClosure.execute(CompiledClosure.java:114)
	at org.armedbear.lisp.Symbol.execute(Symbol.java:827)
	at org.armedbear.lisp.LispThread.execute(LispThread.java:783)
	at org.armedbear.lisp.dqagse_2._execute(dqagse.lisp:10)
	at org.armedbear.lisp.ClosureTemplateFunction._execute(ClosureTemplateFunction.java:182)
	at org.armedbear.lisp.ClosureTemplateFunction.execute(ClosureTemplateFunction.java:84)
	at org.armedbear.lisp.CompiledClosure.execute(CompiledClosure.java:70)
	at org.armedbear.lisp.Lisp.funcall(Lisp.java:154)
	at org.armedbear.lisp.Lisp.multipleValueCall1(Lisp.java:714)
	at org.armedbear.lisp.dqagse_1.execute(dqagse.lisp:10)
	at org.armedbear.lisp.Symbol.execute(Symbol.java:863)
	at org.armedbear.lisp.LispThread.execute(LispThread.java:839)
	at org.armedbear.lisp.dqags_2._execute(dqags.lisp:10)
	at org.armedbear.lisp.ClosureTemplateFunction._execute(ClosureTemplateFunction.java:182)
	at org.armedbear.lisp.ClosureTemplateFunction.execute(ClosureTemplateFunction.java:84)
	at org.armedbear.lisp.CompiledClosure.execute(CompiledClosure.java:70)
	at org.armedbear.lisp.Lisp.funcall(Lisp.java:154)
	at org.armedbear.lisp.Lisp.multipleValueCall1(Lisp.java:714)
(Continue reading)

Permalink | Reply | 3 comments |
headers
Erik Huelsmann | 9 Feb 2009 20:19
Picon
Gravatar

Re: [armedbear-ticket] [armedbear] #46: Compilation failure from several numeric arguments to MIN

Hi Mark,

While this was true last weekend, I don't have a problem compiling the
come below today:

(defun fail () (min 1 2 3))
(compile 'fail)
--> NIL

(fail)
--> 1

Are you sure you're using the most recent there is?

Bye,

Erik.

On Mon, Feb 9, 2009 at 8:29 AM, armedbear
<armedbear-devel <at> common-lisp.net> wrote:
> #46: Compilation failure from several numeric arguments to MIN
> ----------------------+-----------------------------------------------------
>  Reporter:  mevenson  |       Owner:  ehuselmann
>     Type:  defect    |      Status:  new
>  Priority:  major     |   Milestone:  unscheduled
> Component:  compiler  |     Version:  1.0
>  Keywords:            |
> ----------------------+-----------------------------------------------------
>  After the recent (JAN/FEB) changes to the compiler to support numeric
>  types more efficiently, the following code fails to compile:
(Continue reading)

Permalink | Reply | 9 comments |
headers
Ville Voutilainen | 9 Feb 2009 20:39
Picon

Re: [armedbear-ticket] [armedbear] #46: Compilation failure from several numeric arguments to MIN

On Mon, Feb 9, 2009 at 9:19 PM, Erik Huelsmann <ehuels <at> gmail.com> wrote:
> Hi Mark,
> While this was true last weekend, I don't have a problem compiling the
> come below today:
> (defun fail () (min 1 2 3))
> (compile 'fail)
> --> NIL
> (fail)
> --> 1

This works for me, our svn trunk on linux. I guess it's been fixed
after the ticket's been
created, but not quickly enough. :D

------------------------------------------------------------------------------
Create and Deploy Rich Internet Apps outside the browser with Adobe(R)AIR(TM)
software. With Adobe AIR, Ajax developers can use existing skills and code to
build responsive, highly engaging applications that combine the power of local
resources and data with the reach of the web. Download the Adobe AIR SDK and
Ajax docs to start building applications today-http://p.sf.net/sfu/adobe-com
Permalink | Reply | 8 comments |
headers
Erik Huelsmann | 9 Feb 2009 23:06
Picon
Gravatar

Re: Current maxima result..

The issue shown below has been present in the code base for a long
time (definitely not of the past half year). We just never came this
far into the Maxima tests. I guess that's a good thing.

BTW: The issue has been fixed in svn trunk now.

Let's find more Maxima issues ! :-)

Bye,

Erik.

On Mon, Feb 9, 2009 at 6:19 PM, Ville Voutilainen
<ville.voutilainen <at> gmail.com> wrote:
> Maxima once again builds, but the tests fail to run. This is with our SVN head.
>
> Trace:
> java.lang.NoSuchFieldError: G44646
>        at org.armedbear.lisp.abcl51039.execute(Unknown Source)
>        at org.armedbear.lisp.LispThread.execute(LispThread.java:618)
>        at org.armedbear.lisp.quadpack_11._execute(quadpack.lisp:77)
>        at org.armedbear.lisp.ClosureTemplateFunction.execute(ClosureTemplateFunction.java:77)
>        at org.armedbear.lisp.CompiledClosure.execute(CompiledClosure.java:63)
>        at org.armedbear.lisp.LispThread.execute(LispThread.java:618)
>        at org.armedbear.lisp.dqk21_2._execute(dqk21.lisp:10)
>        at org.armedbear.lisp.ClosureTemplateFunction.execute(ClosureTemplateFunction.java:128)
>        at org.armedbear.lisp.CompiledClosure.execute(CompiledClosure.java:114)
>        at org.armedbear.lisp.Symbol.execute(Symbol.java:827)
>        at org.armedbear.lisp.LispThread.execute(LispThread.java:783)
>        at org.armedbear.lisp.dqagse_2._execute(dqagse.lisp:10)
(Continue reading)

Permalink | Reply | 0 comments |
headers
Robert Dodier | 9 Feb 2009 22:33
Picon
Favicon

Re: Current maxima result..

--- On Mon, 2/9/09, Ville Voutilainen <ville.voutilainen <at> gmail.com> wrote:

> Maxima once again builds, but the tests fail to run. This is
> with our SVN head.
> 
> Trace:
> java.lang.NoSuchFieldError: G44646

Thanks for trying ABCL + Maxima. I think I've seen this error
before (don't remember if I reported it to j-devel).
IIRC this error is triggered by attempting to compile some
code which contains gensyms declared special. 
(The code in question is an automatically generated function.)

Now declaring gensyms as special might be a silly thing to do,
and indeed it's probably possible to avoid them.
All the same, the Lisp compiler should be able to handle them.

Thanks for your work on ABCL.

Robert Dodier

------------------------------------------------------------------------------
Create and Deploy Rich Internet Apps outside the browser with Adobe(R)AIR(TM)
software. With Adobe AIR, Ajax developers can use existing skills and code to
build responsive, highly engaging applications that combine the power of local
resources and data with the reach of the web. Download the Adobe AIR SDK and
Ajax docs to start building applications today-http://p.sf.net/sfu/adobe-com
Permalink | Reply | 1 comment |
headers
Ville Voutilainen | 9 Feb 2009 22:39
Picon

Re: Current maxima result..

On Mon, Feb 9, 2009 at 11:33 PM, Robert Dodier <robert_dodier <at> yahoo.com> wrote:
> Thanks for trying ABCL + Maxima. I think I've seen this error
> before (don't remember if I reported it to j-devel).

Erik is investigating this currently, so we're hoping to have a solution for it
soonish. Our long-term target is to run maxima flawlessly, some day. :)

------------------------------------------------------------------------------
Create and Deploy Rich Internet Apps outside the browser with Adobe(R)AIR(TM)
software. With Adobe AIR, Ajax developers can use existing skills and code to
build responsive, highly engaging applications that combine the power of local
resources and data with the reach of the web. Download the Adobe AIR SDK and
Ajax docs to start building applications today-http://p.sf.net/sfu/adobe-com
Permalink | Reply | 0 comments |
headers
Robert Dodier | 10 Feb 2009 06:55
Picon
Favicon

Re: more special variable strangeness

Encouraged by recent developments I tried this example again:

> (setf (symbol-function 'bar)
>   (let ((x (gensym)))
>     (coerce `(lambda (,x) (declare (special ,x)) (* 2 ,x))
> 'function)))
> 
> (let ((f (compile nil #'bar ))) (funcall f 1d0))

When I tried this a few weeks ago I got this error:

>  => java.lang.NoSuchFieldError: G75

However today after updating from svn I get:

java.lang.ClassFormatError: Illegal class name "Lorg/armedbear/lisp/Symbol;" in class file <Unknown>
        at java.lang.ClassLoader.defineClass1(Native Method)
        ....

If I replace (gensym) with 'xx, then it runs successfully
and returns 2.0d0.

Maybe someone can look into this. Thanks for your help!

Robert Dodier

------------------------------------------------------------------------------
Create and Deploy Rich Internet Apps outside the browser with Adobe(R)AIR(TM)
software. With Adobe AIR, Ajax developers can use existing skills and code to
build responsive, highly engaging applications that combine the power of local
(Continue reading)

Permalink | Reply | 1 comment |

Gmane