Gambit-list Juin 2005

gambit-list@iro.umontreal.ca

9 participants
19 discussions

A module system - update
by Christian 14 Jui '05

14 Jui '05

Hello I've packaged a new version which corrects a dumb bug which I introduced while creating the first release (the "import" function gave an error), improved the note about fully qualified names in the README file, and added 7 more modules: cj-math, cj-u8vector-util, cj-realrandom, cj-realrandom-formatted, cj-webserver-saveload-serialize, cj-webserver-saveload-hashtable, and cj-webserver (the latter being a modified version of Marc's web-server example, which doesn't send the continuations to the user but stores them locally in the filesystem -- do "mkdir tmp" before running it from the repl with "(main 1234)"; by replacing cj-webserver-saveload-serialize with cj-webserver-saveload-hashtable in cj-webserver-module.scm you can switch to a version which keeps them in memory). Christian. (PS. I've also corrected an error in my email setup; mail being sent to me yesterday should still reach me if relays waited long enough to retry now; in case you've got an error, please try again and accept my apologies.)

1 0

Binary serializer/deserializer
by Marc Feeley 13 Jui '05

13 Jui '05

I have just completed a first implementation of the binary serializer/ deserializer. The interface is (object->u8vector obj) (u8vector->object u8vect) The space used is typically 30%-50% of the space used by the textual serializer (if you assume 1 character = 1 octet). Here are the results of running the calculator web-continuation example with the new procedures. I have run the example two ways: with everything compiled, and with web-server.scm interpreted. This is on a 400 MHz PowerBook and without -D___SINGLE_HOST. compiled interpreted size of continuation after serialization and base64 encoding (octets) 840 38900 (u8vector->object cont) secs 0.000 0.021 (object->u8vector k) secs 0.003 0.042 base64 encoding/decoding secs 0.000 0.028 As you can see, the compiled version is really very fast, and you can barely see a delay in the browser. The interpreted version is slower but still acceptable, less than 1/10 of a second CPU time per interaction and about 2 seconds per interaction real-time when using Safari (I assume Safari spends most of this time processing the long URLs, which must be displayed in the address box... must be a dumb algorithm). The binary serializer/deserializer and the updated web-server example will be part of the next beta release. Marc

1 0

gcc-4.0 compiler from Apple's Xcode 2.1 seems to work fine
by Bradley Lucier 13 Jui '05

13 Jui '05

Previously I sent somewhat cryptic instructions on how to build a later version of Apple's gcc-4.0.0 that would build Gambit 4.0beta13. Well, you don't have to do that any more; I suggest you sign up as an apple developer and download and install the free Xcode 2.1, which installs [descartes:~/programs/gambc40b13] lucier% gcc -v Using built-in specs. Target: powerpc-apple-darwin8 Configured with: /private/var/tmp/gcc/gcc-5026.obj~19/src/configure -- disable-checking --prefix=/usr --mandir=/share/man --enable- languages=c,objc,c++,obj-c++ --program-transform-name=/^[cg][^+.-]*$/ s/$/-4.0/ --with-gxx-include-dir=/include/gcc/darwin/4.0/c++ -- build=powerpc-apple-darwin8 --host=powerpc-apple-darwin8 -- target=powerpc-apple-darwin8 Thread model: posix gcc version 4.0.0 (Apple Computer, Inc. build 5026) This version of Apple's gcc seems to build gambc40b13 just fine; I configured a 64-bit version with env CC='gcc -mcpu=970 -m64 -force_cpusubtype_ALL' ./configure -- enable-single-host and a 2.7 GHz G5 smokes a 3.6 GHz P4, both running 64-bit code. Too bad IBM didn't seem to be able to supply Jobs with a low-power G5 :-P. Brad

1 0

byte operations on ports
by Christian 13 Jui '05

13 Jui '05

Hello I can't find out how to read bytes from a device file: > (define urandom-port (open-input-file "/dev/urandom")) > urandom-port #<input-port #2 "/dev/urandom"> > (read-byte urandom-port) *** ERROR IN ##read-byte -- Invalid argument (read-char '#<input-port #2 "/dev/urandom">) 1> btw note that the error message mentions read-char instead of read-byte > (read-char urandom-port) *** ERROR IN (console)(a)5.1 -- Invalid argument (read-char '#<input-port #2 "/dev/urandom">) 1> > (input-port-byte-position urandom-port) *** ERROR IN (console)(a)6.1 -- Unknown error (input-port-byte-position '#<input-port #2 "/dev/urandom">) 1> whereas with a normal file: > (define urandom-port (open-input-file "chj.scm")) > urandom-port #<input-port #3 "/home/chris/schemedevelopment/gambit/chj.scm"> > (read-byte urandom-port) 10 > (read-char urandom-port) #\( > (input-port-byte-position urandom-port) 0 > (read-byte urandom-port) 100 > (input-port-byte-position urandom-port) 0 > (read-byte urandom-port) 101 > (input-port-byte-position urandom-port) 0 (but isn't it strange that input-port-byte-position doesn't advance?) Thanks for your help Christian.

1 0

A module system
by Christian 13 Jui '05

13 Jui '05

Hello Over the course of the last few months I've started using Gambit for "real" stuff. The first thing I've done has been a (simple?) module handling system. It has served me quite well and I'm glad to have it done. It's not "finished", it still contains bugs and misses a number of things that would be useful, but maybe you can check it out and see yourself if you find it interesting. I'd like to get feedback on how I should proceed in the light of the possible prospective R6RS module system. I'm willing to adapt it to the latter, or help merge the interesting parts of it. The tar also contains 39 actual modules adapted or written for Gambit (not counting an additional 14 unfinished ones), 8 by myself (small ones; the most useful ones: sorting functions (cj-sort), and a number of macros and utilities (cj-env and cj-listutil)). For more information, the list of features, and the code, see: http://scheme.mine.nu/gambit/chjmodule/ (BTW this has been my first real Scheme program; I would be happy getting tips for coding improvements.) Thanks for any feedback Christian. PS. during it's implementation, a big list of questions has accumulated. I'll ask them bit by bit when I find time in the next days or weeks.

1 0

stop external filters
by Thomas Hafner 11 Jui '05

11 Jui '05

Hello, I'm experimenting about using external commands, which act as filters (i.e. read from standard input and write to standard output): (define p (open-process (list path: "/bin/head"))) (define t (make-thread (lambda () (let loop () (define line (read-line p)) (if (eof-object? line) line (begin (display "[") (display line) (display "]\n") (thread-yield!) (loop))))))) (thread-start! t) (display "a\nb\nc\n" p) (display "d\ne\nf\ng\nh\ni\nj\nk\nl\n" p) (force-output p) (thread-join! t) Here's the corresponding output: [a] [b] [c] [d] [e] [f] [g] [h] [i] [j] No problem so far. This case is easy, because the external command /bin/head terminates automatically after having filtered the first 10 lines. But there are filters, which will never terminate automatically, e.g. /bin/cat. In that case I want the main thread stopping filtering. How can that be done? By adding (close-port p) somewhere? How should the example look like, if /bin/head is replaced by /bin/cat, and the sending thread decides to terminate after having sent 10 lines. The output should be the same as from the original example, of course. Regards Thomas

2 1

Threads IO question
by leo lencioni 05 Jui '05

05 Jui '05

I'am a scheme newbie and trying to learm scheme. I have the following function: (define (disp-msg msg) (display msg) (newline)) I'am using it inside inside a thread to print debug info. (thread-start! (make-thread (lambda () (disp-msg "Hello"))) Running it inside GSI gives no problem and everithing goes well. But when I compile it they show no output. I'am compiling with gsc with no options. Linux/debian. Gambit 4.0 b13. What I'am doing wrong? I'am missing compiler options. Thanks for your help

3 4

Gambit 4, SLIB
by Thomas Hafner 04 Jui '05

04 Jui '05

Hello, can SLIB be used with Gambit 4? The initialization file »gambit.ini« shipped with SLIB claims to be suitable for Gambit 3. Has anybody already ported it for Gambit 4? Regards Thomas

1 0

configure must use -fwrapv as a gcc option for building gambit
by Bradley Lucier 03 Jui '05

03 Jui '05

Here's the documentation for -fwrapv in gcc-4.0 in http://gcc.gnu.org/onlinedocs/gcc-4.0.0/gcc/Code-Gen-Options.html#Code- Gen-Options -fwrapv This option instructs the compiler to assume that signed arithmetic overflow of addition, subtraction and multiplication wraps around using twos-complement representation. This flag enables some optimizations and disables other. This option is enabled by default for the Java front-end, as required by the Java language specification. Since we assume this of all processors, we need to enable this flag. Brad

2 2

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Gambit-list Juin 2005