- Gambit-list - mailman.iro.umontreal.ca

open-tcp-server MacOSX
by François Magnan 15 Mai '07

15 Mai '07

Hi, I found a strange behavior with open-tcp-server on MacOSX of GambitC4b20 and GambitC4b22. I was able to isolate the problem up to this point... Eval the following code (define (rpc-server-start) (let ((rpc-command-queue (open-tcp-server 5000))) (thread-start! (make-thread (lambda () (let loop () (let ((incoming-connection (read rpc-command-queue))) (debug "Got connection\n") (serve-rpc-command incoming-connection) ) (loop))))) rpc-command-queue)) (define (serve-rpc-command incoming-connection) (let* ((command (read-line incoming-connection))) (display (string-append command "\n") incoming-connection) (display "ok!\n\n" incoming-connection) (force-output incoming-connection) (close-port incoming-connection))) (define (main) (rpc-server-start) (loop-indefinitely) ) (define (loop-indefinitely) (thread-sleep! 1) (loop-indefinitely)) (main) Then in a Terminal type telnet 127.0.0.1 5000 type something and then return. You should see: Trying 127.0.0.1... Connected to localhost. Escape character is '^]'. sdj sdj ok! Now comment out the call to "loop-indefinitely" in the (main) function. If you try the telnet again it will not work... This is not the case in the MingW (WinXP) versions. It seems I cannot start a tcp-server in a thread and still get the read-eval-print loop. Anybody can help? Thank you, Francois Magnan

3 3

SGDL, where art thou?
by Bradley Lucier 15 Mai '07

15 Mai '07

I followed Dominique's e-mail about the next MSLUG meeting to the MSLUG home page and clicked on some "Companies known to use Lisp/ Scheme". It seems that the SGDL and Logik3d pages are no more, although there are http:/www.sgdl.com/ (permanently under construction) and http://logik3d.free.fr/ I found the SGDL paper from the Scheme 2000 workshop (was it really so long ago?), and I still find it interesting. And the current version of Gambit, with inline fixnum arithmetic jumping to the bignum routines when necessary, would be a much better platform for such a system. It appears that Dominique wrote a lot of code for the SGDL system, at least from some messages that went back and forth on the PLT mail list. Is there anyone still trying to do stuff like this in Scheme and/or Lisp (and don't tell me AutoCAD!). Brad

2 1

Null pointer in scheme
by David St-Hilaire 12 Mai '07

12 Mai '07

Hi all, I would like to know how it is possible to write the equivalent to the C/C++ NULL pointer when interfacing some C/C++ librairies... I ask this question because I have the following problem in my interface: ... (c-define-type IEventReceiver* (pointer "IEventReceiver")) ... (define createDevice (c-lambda (E_DRIVER_TYPE dimension2d<s32> unsigned-int32 bool bool bool IEventReceiver*) IrrlichtDevice* "___result_voidstar = createDevice (___CAST(E_DRIVER_TYPE,___arg1), ___arg2, ___arg3, ___arg4, ___arg5, ___arg6, ___arg7);")) ... (set! dev (createDevice EDT_OPENGL (vector 800 600) 16 #f use-shadows? #f 0)) which when executed gives me the following problem: [dave@localhost demo]$ ./demo *** ERROR IN demo-main -- (Argument 7) Can't convert to C pointer (createDevice 5 '#(800 600) 16 #f #f #f 0) Can anyone tell me would I should do...? I know that the same line in c++ works fine: createDevice( video::EDT_OPENGL, dimension2d<s32>(640, 480), 16, false, false, false, 0); Thanks for your help!! David St-Hilaire

2 1

Adding engineering notation support?
by Shawn Hyam 12 Mai '07

12 Mai '07

I'm very interesting in adding support for engineering notation (M, k, m, u, etc.) to numbers in Gambit. I'm curious if Gambit's readtables would provide me with this kind of power, but I haven't been able to glean too much information about them so far. Any pointers would be greatly appreciated. thanks Shawn

2 1

gambit doesn't disable reading of gambcext while running tests
by Marijn Schouten (hkBst) 11 Mai '07

11 Mai '07

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 It seems that gambit doesn't disable reading of gambcext while running tests. My gambcext does (load "~~/syntax-case"). Without gambcext tests run fine, but when syntax-case is loaded, the following happens: make[1]: Entering directory `/var/tmp/portage/dev-scheme/gambit-4.0_beta22/work/gambc-4.0b22/tests' - ------------ TEST 1 (debugging support) LD_LIBRARY_PATH=../lib: ../gsi/gsi -f debug.scm > test1.out diff test1.ok test1.out && rm -f test1.out - ------------ TEST 2 (error handling) LD_LIBRARY_PATH=../lib: ../gsi/gsi -:h4000 -f error.scm < error.scm > test2.out *** ERROR IN ##gen-glo-ref -- Heap overflow make[1]: *** [test2] Error 70 make[1]: Leaving directory `/var/tmp/portage/dev-scheme/gambit-4.0_beta22/work/gambc-4.0b22/tests' make: *** [check] Error 2 Marijn -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFGRF9pp/VmCx0OL2wRAv11AKC3D4hyl+JNirZayXzP1dcHTephZACeLTkC 5on7DPabNJINKonMLE5C4Mo= =IvJN -----END PGP SIGNATURE-----

2 1

Bug
by Bradley Lucier 10 Mai '07

10 Mai '07

Marc: Untar the following and load the two files: euler-92% gsc Gambit Version 4.0 beta 22 > (load "all.scm") "/export/users/lucier/lang/scheme/Arrays/srfi3/BV-fast/all.scm" > (load "generate-circle.scm") *** ERROR IN ##load -- (Argument 1) VECTOR expected (vector-length '#<readtable #2>) 1> This is on x86_64-unknown-linux-gnu (RHEL 4.0). Brad

1 0

Re: [gambit-list] Compile scheme to executable code...
by David St-Hilaire 07 Mai '07

07 Mai '07

James Long wrote: > You shouldn't need those extra libraries. Make sure you have compiled > Gambit as c++ code (some option in configure). Thank you again, I didn't know that gambit could be compile with g++...! But, as I try to compile it with g++, I get an error at compilation. I'm now compiling the version 4 beta 22. I tried to compile gambit version 4 beta 20 and it worked fine... Here is what the error I got: $ ./configure --prefix=/opt/g4 --enable-cplusplus $ make making all in include make[1]: Entering directory `/home/dave/apps-src/gambc-4.0b22/include' make[1]: Leaving directory `/home/dave/apps-src/gambc-4.0b22/include' making all in lib make[1]: Entering directory `/home/dave/apps-src/gambc-4.0b22/lib' g++ -I../include -I. -Wall -W -Wno-unused -O1 -fno-math-errno -fschedule-insns2 -fno-trapping-math -fno-strict-aliasing -fwrapv -fexpensive-optimizations -fforce-addr -fpeephole2 -falign-jumps -falign-functions -fno-function-cse -ftree-copyrename -ftree-fre -ftree-dce -fregmove -fgcse-las -freorder-functions -fcaller-saves -fno-if-conversion2 -foptimize-sibling-calls -fcse-skip-blocks -funit-at-a-time -finline-functions -fomit-frame-pointer -fPIC -fno-common -mieee-fp -DHAVE_CONFIG_H -D___PRIMAL -D___LIBRARY -D___GAMBCDIR=\"/opt/g4/4.0b22\" -c _io.c g++ -I../include -I. -Wall -W -Wno-unused -O1 -fno-math-errno -fschedule-insns2 -fno-trapping-math -fno-strict-aliasing -fwrapv -fexpensive-optimizations -fforce-addr -fpeephole2 -falign-jumps -falign-functions -fno-function-cse -ftree-copyrename -ftree-fre -ftree-dce -fregmove -fgcse-las -freorder-functions -fcaller-saves -fno-if-conversion2 -foptimize-sibling-calls -fcse-skip-blocks -funit-at-a-time -finline-functions -fomit-frame-pointer -fPIC -fno-common -mieee-fp -DHAVE_CONFIG_H -D___PRIMAL -D___LIBRARY -D___GAMBCDIR=\"/opt/g4/4.0b22\" -c _num.c g++ -I../include -I. -Wall -W -Wno-unused -O1 -fno-math-errno -fschedule-insns2 -fno-trapping-math -fno-strict-aliasing -fwrapv -fexpensive-optimizations -fforce-addr -fpeephole2 -falign-jumps -falign-functions -fno-function-cse -ftree-copyrename -ftree-fre -ftree-dce -fregmove -fgcse-las -freorder-functions -fcaller-saves -fno-if-conversion2 -foptimize-sibling-calls -fcse-skip-blocks -funit-at-a-time -finline-functions -fomit-frame-pointer -fPIC -fno-common -mieee-fp -DHAVE_CONFIG_H -D___PRIMAL -D___LIBRARY -D___GAMBCDIR=\"/opt/g4/4.0b22\" -c _std.c os.h:537: error: previous declaration of ‘char** environ’ with ‘C++’ linkage /usr/include/unistd.h:496: error: conflicts with new declaration with ‘C’ linkage make[1]: *** [_std.o] Error 1 make[1]: Leaving directory `/home/dave/apps-src/gambc-4.0b22/lib' make: *** [all-recursive] Error 1 I would prefer to install gambit 4 beta 22 instead of beta 20 (to be able to use snowfort), am I the only one which this problem occured? Thank you very again for your help! David St-Hilaire

2 1

Compile scheme to executable code...
by David St-Hilaire 07 Mai '07

07 Mai '07

Hi everyone! I'm rather new to the scheme world so that, so far, I've only needed to use a scheme interpreter to do the work I needed to get done. But now, I would be interested to build executable code from my scheme code. I've looked the example provided in the gambit documentation, but unfortunately, it does not work on my system (fedora core 6 laptop). The exemple has 3 source files: m1.c, m2.scm and m3.scm. The instructions tells us to compile first the m2 and m3 scheme source files either by using $ *gsc -c m2.scm /# create m2.c (note: .scm is optional)/* $ *gsc -c m3.scm /# create m3.c (note: .scm is optional)/* $ *gsc m2.c m3.c /# create the incremental link file m3_.c /* or simply using: $ *gsc m2 m3 * The result of compiling the source files like this should create 3 new files: m2.c, m3.c and m3_.c, but on my system, only the m2.c and m3.c files gets created... Then when I try to compile my 3 'c' files with gcc: $ gcc -I/home/dave/diro/ift3060/g4/current/include -L/home/dave/diro/ift3060/g4/current/lib m1.c m2.c m3.c -lgambc -lm -ldl -lutil /usr/lib/gcc/i386-redhat-linux/4.1.1/../../../crt1.o: In function `_start': (.text+0x18): undefined reference to `main' /tmp/cc5ZPJZX.o: In function `___H__20_m2': m2.c:(.text+0x59): undefined reference to `___G_pow2' m2.c:(.text+0x64): undefined reference to `___G_twice' /tmp/cc5ZPJZX.o: In function `___init_proc': m2.c:(.text+0x4d6): undefined reference to `___G__20_m2' m2.c:(.text+0x4db): undefined reference to `___G__20_m2' m2.c:(.text+0x4e0): undefined reference to `___G__20_m2' m2.c:(.text+0x4ed): undefined reference to `___G__20_m2_23_0' m2.c:(.text+0x4f2): undefined reference to `___G__20_m2_23_0' m2.c:(.text+0x4f7): undefined reference to `___G__20_m2_23_0' /tmp/ccKAvQIK.o: In function `___H__20_m3': m3.c:(.text+0xa5): undefined reference to `___G_twice' m3.c:(.text+0xb2): undefined reference to `___G_pow2' /tmp/ccKAvQIK.o: In function `___init_proc': m3.c:(.text+0x779): undefined reference to `___G__20_m3' m3.c:(.text+0x77e): undefined reference to `___G__20_m3' m3.c:(.text+0x783): undefined reference to `___G__20_m3' collect2: ld returned 1 exit status What's going on? How can I compile this simple example? Thank you very much for your help!! __ David St-Hilaire

2 1

c style #ifdef
by paul bite 04 Mai '07

04 Mai '07

hi there is a way to do #ifdef like c. I really need something that works like: #ifdef DEBUG ... #endif

2 1

how to make "trace" print everything instead of "..."?
by dillo gimp 01 Mai '07

01 Mai '07

hi 1. how to make "trace" print everything instead of "..."? 2. Is there a way to search gambit-c archive? The site has only gzipped files : https://webmail.iro.umontreal.ca/pipermail/gambit-list/ don't know if this is a good idea or not, maybe moving to google's group is easier to search?

2 1