On Fri, 12 Oct 2007 00:05:46 -0400, "Andrei Stebakov" <[EMAIL PROTECTED]> wrote:
> I run across this error time after time but it's really hard to > reproduce but I can replicate exactly the same backtrace of the > problem if instead of some valid http domain I choose any fake one. > If I run: > (drakma:http-request > "http://some-fake-domain.com"; > :method :get > :parameters `(("param1" . "text"))) > > The backtrace will be: > > The value 0 > is not of type > (OR (SINGLE-FLOAT (0.0)) (DOUBLE-FLOAT (0.0d0)) (INTEGER 1)). > [Condition of type TYPE-ERROR] > > Restarts: > 0: [ABORT] Return to SLIME's top level. > 1: [ABORT] Exit debugger, returning to top level. > > Backtrace: > 0: (RANDOM 0) > 1: (USOCKET::GET-RANDOM-HOST-BY-NAME "some-fake-domain.com") > 2: (USOCKET:SOCKET-CONNECT "some-fake-domain.com" 80) > 3: (DRAKMA:HTTP-REQUEST "http://some-fake-domain.com";) > 4: (SB-INT:SIMPLE-EVAL-IN-LEXENV > (DRAKMA:HTTP-REQUEST "http://some-fake-domain.com"; :METHOD :GET > :PARAMETERS '(("param1" . "text"))) > #<NULL-LEXENV>) > 5: ((LAMBDA ())) > > In my case it looks like sometimes even for the real domain it calls > (RANDOM 0) which gives this log. Even if the domain is wrong do you > think it should throw an exception? I use SBCL 1.0.5. with the > latest drakma and usocket (usocket-0.3.2). I'm afraid this looks like a usocket problem and you should ask on their mailing list. Andreas Fuchs and others convinced me to switch Drakma from trivial-sockets to usocket, and one of their arguments was that trivial-sockets is unmaintained while usocket has an active developer community. So, you can now check if they were right... :) Cheers, Edi. _______________________________________________ drakma-devel mailing list drakma-devel@common-lisp.net http://common-lisp.net/cgi-bin/mailman/listinfo/drakma-devel
