Is there a portable way to quit a CL process?
abort only terminates the current load,
and sbcl keeps quit in a special package.
--
Dan
www.prairienet.org/~dsb/
Dan Bensen <··········@cyberspace.net> writes:
> Is there a portable way to quit a CL process?
> abort only terminates the current load,
> and sbcl keeps quit in a special package.
Here's a start:
(defun portable-quit ()
#+allegro(excl:exit)
#+sbcl(sb-ext:quit)
#+clisp(ext:quit)
#-(or allegro sbcl clisp)(error "don't know how to quit."))
Or use SLIME and then ,sayoonara will quit for you. (For that matter,
you could probably look in the slime sources to fill out the
PORTABLE-QUIT function above. However there it's probably spread over
several files.)
-Peter
--
Peter Seibel : ·····@gigamonkeys.com
Gigamonkeys Consulting : http://www.gigamonkeys.com/
Peter Seibel wrote:
> Dan Bensen <··········@cyberspace.net> writes:
>> Is there a portable way to quit a CL process?
> Here's a start:
> (defun portable-quit ()
> #+allegro(excl:exit)
> #+sbcl(sb-ext:quit)
> #+clisp(ext:quit)
> #-(or allegro sbcl clisp)(error "don't know how to quit."))
Thanks :)
--
Dan
www.prairienet.org/~dsb/
Dan Bensen <··········@cyberspace.net> writes:
> Peter Seibel wrote:
>> Dan Bensen <··········@cyberspace.net> writes:
>>> Is there a portable way to quit a CL process?
>> Here's a start:
>> (defun portable-quit ()
>> #+allegro(excl:exit)
>> #+sbcl(sb-ext:quit)
>> #+clisp(ext:quit)
#+movitz(contrib:pull-the-plug)
#+universe(sys:apocalypse)
>> #-(or allegro sbcl clisp movitz universe)(error "don't know how to quit."))
>
> Thanks :)
Quitting is not always meaningful, that's why it's not standard.
--
__Pascal Bourguignon__
http://www.informatimago.com
http://pjb.ogamita.org
On Apr 3, 2:47 am, Pascal Bourguignon <····@informatimago.com> wrote:
> Dan Bensen <··········@cyberspace.net> writes:
> > Peter Seibel wrote:
> >> Dan Bensen <··········@cyberspace.net> writes:
> >>> Is there a portable way to quit a CL process?
> >> Here's a start:
> >> (defun portable-quit ()
> >> #+allegro(excl:exit)
> >> #+sbcl(sb-ext:quit)
> >> #+clisp(ext:quit)
>
> #+movitz(contrib:pull-the-plug)
> #+universe(sys:apocalypse)
>
> >> #-(or allegro sbcl clisp movitz universe)(error "don't know how to quit."))
>
> > Thanks :)
>
> Quitting is not always meaningful, that's why it's not standard.
Sorry, but can you explain a bit more on what you mean? Is it because
a lisp system could be running on the metal?
·······@gmail.com" <······@gmail.com> writes:
> On Apr 3, 2:47 am, Pascal Bourguignon <····@informatimago.com> wrote:
>> Quitting is not always meaningful, that's why it's not standard.
>
> Sorry, but can you explain a bit more on what you mean? Is it because
> a lisp system could be running on the metal?
Indeed. Quitting is a valid notion when you are in an independent
process with an underlying system to return to. But for software
running on the metal, it is not (you might "reboot" or "halt"). But
also, there are a lot of processes that should never quit anyways.
There's a current of thought that actually we don't need any "quit" or
"clean way to exit", that merely making the systems and software crash
resilient is enough and actually simplier than implementing quit and
clean up functions. Said otherwise, the only "quit" you need is a
plug to pull.
--
__Pascal Bourguignon__
http://www.informatimago.com
http://pjb.ogamita.org