Eduardo Cavazos | 1 Apr 2009 04:02
Picon

srfi-27 on chicken 4

Hello,

No pressure or rush. :-) I'm just planning ahead. Do y'all intend to 
offer SRFI-27 for Chicken 4?

Ed
Kon Lovett | 1 Apr 2009 04:25

Re: srfi-27 on chicken 4


On Mar 31, 2009, at 7:02 PM, Eduardo Cavazos wrote:

> Hello,
>
> No pressure or rush. :-) I'm just planning ahead. Do y'all intend to  
> offer SRFI-27 for Chicken 4?

Short answer - Yes.

Long answer - Need to port 'lexmod' 1st and since 'lexmod' is a module  
system need to figure out how to co-exist w/ Chicken 4 module system.  
So a more difficult port than just slapping 'module srfi-29' around  
the routines.

>
>
> Ed
>
>
> _______________________________________________
> Chicken-users mailing list
> Chicken-users <at> nongnu.org
> http://lists.nongnu.org/mailman/listinfo/chicken-users

Best Wishes,
Kon
Eduardo Cavazos | 2 Apr 2009 16:48
Picon

Abstracting

Hello,

Abstracting is a Scheme framework with an emphasis on OpenGL programming.

Features:

     * Bindings to GL and GLUT

     * Runs on Ypsilon, Larceny, Chicken, and Gambit

     * Collection of examples

Github page:

     http://github.com/dharmatech/abstracting

Git repository:

     git://github.com/dharmatech/abstracting.git

To begin, follow the directions in the README.

Feedback welcome!

I'm looking to add more examples. If you write a cool demo, you are
welcome to submit it! :-)

News:

2009-04-02
(Continue reading)

Matt Jones | 5 Apr 2009 16:03
Picon

tcp-connect open ports problem

Hi folks,

I'm having an issue with the tcp-connect function.  When it fails to 
connect, it seems not to clean up properly, so that I end up with lots 
of open ports.

Try this code:

   (require-extension tcp)

   (define (error-info e)
     ((condition-property-accessor 'exn 'message) e))

   (define (test-tcp-connect)
     (condition-case
       (tcp-connect "localhost" 8000)
       (e (exn i/o net) (print (error-info e)))))

assuming there is nothing on port 8000.  Then,

   (let loop ((i 0)) (when (< i 2000) (test-tcp-connect) (loop (+ i 1))))

I get,

   can not create socket - Connection refused
   can not create socket - Connection refused
   ...
   can not create socket - Too many open files
   can not create socket - Too many open files
   ...
(Continue reading)

Kon Lovett | 5 Apr 2009 17:28

Re: tcp-connect open ports problem


On Apr 5, 2009, at 7:03 AM, Matt Jones wrote:

> Hi folks,
>
> I'm having an issue with the tcp-connect function.  When it fails to  
> connect, it seems not to clean up properly, so that I end up with  
> lots of open ports.
>
> Try this code:
>
>  (require-extension tcp)
>
>  (define (error-info e)
>    ((condition-property-accessor 'exn 'message) e))
>
>  (define (test-tcp-connect)
>    (condition-case
>      (tcp-connect "localhost" 8000)
>      (e (exn i/o net) (print (error-info e)))))
>
> assuming there is nothing on port 8000.  Then,
>
>  (let loop ((i 0)) (when (< i 2000) (test-tcp-connect) (loop (+ i  
> 1))))
>
> I get,
>
>  can not create socket - Connection refused
>  can not create socket - Connection refused
(Continue reading)

felix winkelmann | 5 Apr 2009 23:16
Picon

Re: -prologue

On Thu, Mar 26, 2009 at 5:08 AM, Eduardo Cavazos <wayo.cavazos <at> gmail.com> wrote:
> Hello,
>
> I just noticed that something like this:
>
>  csc somefile.scm -dynamic -prologue SOME-DIRECTORY
>
> doesn't produce an error. Does it make sense to pass a directory to
> '-prologue'?
>
> Just wondering if maybe 'csc' should print an error in the above case.
>

It should indeed. Thanks for pointing this out.

cheers,
felix
Matt Jones | 5 Apr 2009 23:29
Picon

Re: tcp-connect open ports problem

Kon Lovett wrote:
 >
 > Versions before 3.0 are unsupported AFAIK. Packaged sources are
 > available at the Chicken website:
 > http://call-with-current-continuation.org as are instructions for
 > accessing the source repository.

Thanks for the tip.  I have upgraded to version 3.4.  The problem 
persists, though.  I looked at the source and the relevant part is the 
same.  The function 'fail appears to close the port -- (##net#close s) 
-- but somehow these ports end up open and accumulate.

Best
Matt
felix winkelmann | 5 Apr 2009 23:30
Picon

Re: Re: Trac moving to new server

On Tue, Mar 31, 2009 at 9:24 PM, Arto Bendiken <arto.bendiken <at> gmail.com> wrote:
>
> Also, I've asked Mario to change the Galinha repository rsync so that
> http://trac.callcc.org/timeline will start showing the latest updates
> again.
>
> Unless anyone is experiencing some new unexpected problems, this
> concludes the server move.
>

Hi!

Thanks, Arto. I must admit that I'm not using the trac anymore - it is
often not reachable (usually when I needed it most). Ivan is going
to set up a new one on a dedicated server. Will we still need the
current trac instance? What do others think?

cheers,
felix
felix winkelmann | 5 Apr 2009 23:32
Picon

Re: srfi-27 on chicken 4

On Wed, Apr 1, 2009 at 4:25 AM, Kon Lovett <klovett <at> pacbell.net> wrote:
>
> On Mar 31, 2009, at 7:02 PM, Eduardo Cavazos wrote:
>
>> Hello,
>>
>> No pressure or rush. :-) I'm just planning ahead. Do y'all intend to offer
>> SRFI-27 for Chicken 4?
>
> Short answer - Yes.
>
> Long answer - Need to port 'lexmod' 1st and since 'lexmod' is a module
> system need to figure out how to co-exist w/ Chicken 4 module system. So a
> more difficult port than just slapping 'module srfi-29' around the routines.
>

Hi!

I recommend not to worry too much about lexmod and use the real module
system.

cheers,
felix
felix winkelmann | 6 Apr 2009 00:12
Picon

Re: tcp-connect open ports problem

On Sun, Apr 5, 2009 at 11:29 PM, Matt Jones <mj <at> momoweb.co.uk> wrote:
> Kon Lovett wrote:
>>
>> Versions before 3.0 are unsupported AFAIK. Packaged sources are
>> available at the Chicken website:
>> http://call-with-current-continuation.org as are instructions for
>> accessing the source repository.
>
>
> Thanks for the tip.  I have upgraded to version 3.4.  The problem persists,
> though.  I looked at the source and the relevant part is the same.  The
> function 'fail appears to close the port -- (##net#close s) -- but somehow
> these ports end up open and accumulate.
>

Is it possible that the condition-case handling somehow skips the
closing of the port? (A lame question, I know - but I don't
know what to do besides trying the code myself, for which I currently haven't
got the time)

cheers,
felix

Gmane