usocket-devel December 2009

usocket-devel@common-lisp.net

5 participants
6 discussions

[usocket-devel] wait-for-input support on SBCL/Windows
by Elliott Slaughter 13 Jan '10

13 Jan '10

Hi, I noticed while reading sbcl.lisp that wait-for-input is not yet implemented on Windows. I'm wondering whether anyone is working on this and/or what it would take to get this working. Thanks. -- Elliott Slaughter "Don't worry about what anybody else is going to do. The best way to predict the future is to invent it." - Alan Kay

2 1

Re: [usocket-devel] ECL and MSVC
by Chun Tian (binghe) 02 Jan '10

02 Jan '10

Hi, there I just built out a ECL binary on MSVC, now I'm starting looking into this issue. ECL team maintains a really good code base on all supported platforms ... --binghe > Ive actually done that and hunchentoot displays nothing on the simple test > website. Oh well... Let me know how your test of the 'patch' goes... > > -----Original Message----- > From: Chun Tian (binghe) [mailto:binghe.lisp@gmail.com] > Sent: Tuesday, December 29, 2009 2:33 AM > To: Seth Burleigh > Subject: Re: [usocket-devel] ECL and MSVC > > As I know these test code never been used for long time, we're working on > bring them back ... > > I think you should use a real networking package which built upon USOCKET, > i.e. hunchentoot > > --binghe > > 在 2009-12-29，01:43， Seth Burleigh 写道： > >> Perhaps im doing these tests wrong since I really have never used sockets. >> All I did was load the usocket-test.asd and go (usocket:run-usocket-tests) >> If im reading the code right, it attempt to make a connection on 127.0.0.0 >> at port 15213 for the first test. Surely there has to be something > listening >> for it in order for it to receive it correctly? Using commview and the >> loopback sniffer, no activity is reported at 127.0.0.0. >> >> >> -----Original Message----- >> From: Chun Tian (binghe) [mailto:binghe.lisp@gmail.com] >> Sent: Monday, December 28, 2009 10:19 AM >> To: Seth Burleigh >> Subject: Re: [usocket-devel] ECL and MSVC >> >> Hi, Seth >> >> Thanks for these tests. I suggest you using WireShark (or something else >> eq. to "tcpdump") to try to capture if there's really IP packets sent out > by >> USOCKET functions. If USOCKET really send them, things should be easier >> then. >> >> --binghe >> >> 在 2009-12-28，23:26， Seth Burleigh 写道： >> >>> Hmm.. I tried testing it on clisp and ecl. On ecl it failed 13/15 times. >> On >>> clisp , currently, it has failed the first 3 tests with timeout errors. > I >>> even had clisp running with admin rights. Maybe it’s the firewall or my >>> win7 operating system. I"ll have to look into this deeper. >>> >>> -----Original Message----- >>> From: Chun Tian (binghe) [mailto:binghe.lisp@gmail.com] >>> Sent: Monday, December 28, 2009 8:48 AM >>> To: Seth Burleigh >>> Cc: usocket-devel(a)common-lisp.net >>> Subject: Re: [usocket-devel] ECL and MSVC >>> >>> Hi, Seth >>> >>> Sorry for the delay response. I still doesn't have a chance to try build > a >>> MSVC-based ECL by myself and test USOKET on it, but I think you're right. >>> However, after searching in MSDN, I found a better presentation: >>> >>> #include <time.h> >>> >>> That is, <time.h> instead of "time.h", better followed C programming >> style. >>> >>> Please give me two days try it by myself, and I'll merge it soon. >>> >>> Regards, >>> >>> Chun Tian (binghe) >>> >>> 在 2009-12-28，06:20， Seth Burleigh 写道： >>> >>>> So I searched for it, it looks like you just >>>> #+(and :ecl :msvc) #include “time.h” >>>> For a msvc build, I think, instead of just including “sys/time.h” >>>> >>>> Though im not sure, it builds, but I haven’t tried any tests. >>>> _______________________________________________ >>>> usocket-devel mailing list >>>> usocket-devel(a)common-lisp.net >>>> http://common-lisp.net/cgi-bin/mailman/listinfo/usocket-devel >>> >>> >> >> > >

1 1

[usocket-devel] ECL and MSVC
by Seth Burleigh 28 Dec '09

28 Dec '09

So I searched for it, it looks like you just #+(and :ecl :msvc) #include "time.h" For a msvc build, I think, instead of just including "sys/time.h" Though im not sure, it builds, but I haven't tried any tests.

2 1

[usocket-devel] ECL and MSVC
by Seth Burleigh 27 Dec '09

27 Dec '09

When using ecl that has been built with msvc, the code in sbcl.lisp #-:wsock (ffi:clines "#include <errno.h>" "#include <sys/socket.h>") #+:wsock (ffi:clines "#ifndef FD_SETSIZE" "#define FD_SETSIZE 1024" "#endif" "#include <winsock2.h>") (ffi:clines "#include <sys/time.h>" "#include <ecl/ecl-inl.h>" ) Doesn't build because sys/time is not available in msvc. What header should I include instead?

1 0

[usocket-devel] wait-for-input's ready-only argument
by Mihai Bazon 18 Dec '09

18 Dec '09

Hi folks, What's with the ready-only arg to wait-for-input? It's not documented anywhere, but I noticed that if I don't pass T for it my code behaves strangely. Cheers, -Mihai

2 2

[usocket-devel] [discuss] usocket direction
by Chun Tian (binghe) 08 Dec '09

08 Dec '09

Hi, I'm "disappeared" for several months from public CL community, because I'm hired by Gensym for improving their big lisp products. Now I'm partly back and still want to do contributions to usocket & other projects. For usocket project, it's long since the last release. Before next release, I personal am thinking of following plans: 1) Finish the experimental-udp branch, by cleaning some code first (reliable UDP), then finish port on CLISP & ABCL, finally merged back to trunk. 2) Add support for MCL (Macintosh Common Lisp), because the new Rosetta based MCL is alive again to Mac Lispers. 3) Add some unit test code, I'll start the first part above from not later then next month (Jan 2010), help from ABCL & CLISP users are welcome. More ideas or suggestions are also welcome, especially notes from Erik & Hans. Regards, Chun Tian (binghe)

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

usocket-devel December 2009