gmane.lisp.scheme.bigloo

Bigloo object system : default values ?

Saul Malesac — 2008-11-27T13:21:51

Hi, Still more newbie questions : - How do default values work with classes ? $bigloo 1:=> (module toto (export (class test (i::elong (default #e0))))) #unspecified 1:=> (make-test) *** ERROR:eval: Wrong number of arguments: 1 expected, 0 provided -- ... make-test 0.... make-test 1:=> (define z (make-test (unspecified))) z 1:=> (test-i z) #unspecified So, I could not understand from the documentation when or how these default values are used. - my second question is about constructors and inheritance : constructors of subclasses seem to have as many parameters as the sum of all the inherited fields. This may become cumbersome (and hard to follow) with huge class hierarchies. How do you cope with this ? Sincerely yours, Saul

compilation of Bigloo program under Windows XP

2008-11-25T17:12:20

Hi, After installation of Yannis Brès Windows port of Bigloo 2.8c ( http://www.yaya.fr/Bigloo/ ), interpreter is OK, but any attempt to compile anything results in : *** ERROR:run-process: Specified file not found -- cl 0.win32-cc 1.cc 2.cc-compiler 3.backend-link 4.backend-walk 5.compiler 6.engine 7.main The free Microsoft Visual C++ Toolkit 2005 is installed. Any hint will be appreciated. Thanks.

Sorry

Saul Malesac — 2008-11-25T16:39:03

Hi, Please excuse me for the last two mails I sent to this mail address : they were intended to be sent to the mailing list, but I have fumbled with my contacts. Sincerely yours, Saul

Yet another question

Saul Malesac — 2008-11-25T16:37:26

Hello all, I have a (may be stupid) question about Scheme programming or, let's say, programming pattern : still struggling with the PDF format, I'm currently wondering about how to implement the "right" "PDF structure". I mean, PDF objects are somewhat like XML nodes, in the sense that most of them have attributes that only make sense within the context of a PDF structure : parents objects, a cross-reference table shared between all the PDF objects, inherited attributes, etc. My first call would be to include, in a PDF object, something like a "pointer" to the common PDF structure, but I'm looking for a way to do this in Scheme without having multiple copies (one "in" each object) wandering around, especially because the shared data may be large (it's more or less the whole PDF file). I do not want to have a shared global variable, neither. Of course, there is still the option to have two args for each function on the PDF objects (one arg. for the object, one arg. for the shared PDF structure), which would, by the way, lead me to see whether that could be written in a monadic-style way in Scheme (I hope I'm clear here), but I would prefer not doing this. Do you guys know if there is a "standard" coding pattern in Scheme for that kind of situation ? (I've read Oleg's paper on pointers in SXML) Sincerely yours, Saul

Yet another question

Saul Malesac — 2008-11-25T16:36:00

issue with zlib

Saul Malesac — 2008-11-17T20:50:40

Good evening all, I hope this is not too off-topic, but I have a bug I cannot manage to understand using a binding of mine with the zlib library. In short, I'm trying to write a quick-and-dirty binding to the zlib in order to be able to inflate (decompress) deflated streams with bigloo. My failure to do so may come from my misunderstanding of (1) how to use the zlib, (2) how to use the C interface, (3) lack of scheme knowledge. Because cigloo was confused with the original zlib.h file, I cut-and-pasted what seemed to be the bare minimum needed for me (that is, the inflate methods) ; my patched zlib.h is joined to this message. I then used cigloo to get the zlib.sch file, which needed some redefinition (the correct type signature for the inflate functions that are macros in the zlib.h file could not be inferred). Then I wrote my simple zlib.scm test routine. Everything seems to work fine except that the inflated stream gets corrupted at some point : this depends on the size of the local buffer. I started with 8192, and I could decode short streams without issue ; but, when trying a larger file, I got Z_DATA_ERROR (-3) at some point. The inflated was garbled before the error, but not at the beginning of the stream. I've reduced the size of the local buffer (*BUF_SIZE* parameter) so that the behavior occurs with short files, so I suspect some pointer stuff or something that could get garbage-collected ? May be there is also something which does not work well in my implementation about the way open-input-procedure works. Please note that I am sure that my stream is correct : I have used perl scripts to deflate and inflate them, so I'm sure that my test files are right. If someone knows the zlib enough to help find what is the issue, that would be nice. Otherwise, do not bother yourself with that inelegant code. In short, here is what the code does : the function port->inflated-port is supposed to receive a port as its argument, and it builds an inner function (inflated-port) which returns string chunks of inflated data from that port. That function is used by Bigloo's open-input-procedure to create a port. The function inflated-port works as follows : to zlib-inflate data, you need to create a z_stream C structure, which contains a input buffer and an output buffer. You feed the input buffer (which I do from the original port), you call the "inflate" C function, which updates the structure. You then check how many bytes were inflated. When one of the two buffers is filled up, you need to re-initialize it. Note : the compilation of the C file contains two warnings, which concern the type of the zalloc and zfree functions. I could not manage to get the pointer types right. This could be the source of the bug, but I would find it strange as these are pointers to C functions that do not interact with Bigloo. Thank you for reading me, and especially this ugly code, Saul

regular-grammar help

w r — 2008-11-16T19:01:37

I'm trying to learn the use of regular-grammar, and getting unexpected results: (module test (main main)) (define (read-decl inp) (let ((grammar (regular-grammar ((decl-open "")) ((+ blank) (ignore)) ;; xml declaration: ((: decl-open (* all) decl-close) (the-string)) (else #f)))) (read/rp grammar inp))) (define (main args) (let ((p (open-input-string "non-matching string"))) (print (input-port-position p)) ; => 0 (print (read-decl p)) ; => #f ;; expecting 0 here: (print (input-port-position p)) ; => 1 )) I'd like to make read-decl so that if it fails, the port remains at the first char. What am I missing? Wayne

zlib bindings and other questions

Saul Malesac — 2008-11-13T09:21:15

Hi, I have been discovering Scheme and Bigloo for few weeks now, and I wanted to share my first feelings, if you guys do not mind. To begin, I should explain that my goal in exlporing Scheme is to broaden my experience with programming languages : I have some knowledge of other functional languages, like Haskell, ML, and mainstream-bloated ones, like Java, C++ and C, though I mainly wrote Haskell and Ruby code for the past few years. When coding in Scheme, I find that I tend to use functional coding habits. For instance, I used a lot "named let" for loops where "do" clauses would have been cleaner. Hence, my first question for you all is where may I find some examples or use-cases for Scheme-specific features, in particular, macros and continuations ? I mean, I currently feel that I'm coding in Scheme just like I would have written in Haskell or ML, which is not my objective. I've read some tutorials on continuations, for instance, but I still do not understand what they are really useful for. As I said, this may be caused by my functional habits, which make me write CPS code naturally when the need arises... OK, all this is a little theoretical. More practicaly, here are some questions : 1) Is there a zlib library binding for Bigloo somewhere (see below why I need this) ? 2) What code documenting tool may be used with Bigloo ? Compared to other languages, I find strange that Scheme does not have that kind of javadoc-like software. 3) Is there some (unit) testing framework for Bigloo ? All the ones I have found, I did not manage to compile them, and testing is mandatory for me. 4) My personnal project is to write a PDF-parsing library (hence my need for zlib support), at least, to see how far I may go on this. This also raises the following question : the PDF format defines what it calls "streams", which is merely binary-encoded data of fixed length. I'm not sure it's the right way to do this, but I think that it would be natural to represent such stream by a Scheme port. Currently, the only way I see to implement them is to write something like this, assuming that "port" is an open port to the PDF file : ; size = length of stream data (define (make-pdf-stream size port) (open-input-procedure (lambda () (read-chars! size port)))) I'm just wondering whether there is another way, may be more elegant, to do this, especially to prevent intensive memory allocation if the stream is huge (the above method reads all the stream at once, it seems) ? Could one use a generator here, to read the stream piece by piece ? Thank you for reading me, Saul

Interfacing with C functions that accept const char*

Deepankar Sharma — 2008-11-13T00:28:28

I am working on the glfw bindings and writing some examples. While working on an example that loads textures I ran into the following problem - The function I want to interface with is: GLFWAPI int GLFWAPIENTRY glfwLoadTexture2D( const char *name, int flags ); // defined in glfw.h When I wrap it using cigloo the wrapper that gets generated is: (glfwLoadTexture2D::int (string int) "glfwLoadTexture2D") When i try to use this function in my scheme code the resulting C file produced by bigloo contains this line - extern int glfwLoadTexture2D(char *, int); gcc refuses to compile the c file generated by bigloo because the two declarations (one in glfw.h and in the c file generated by bigloo) have different signatures. How do I get bigloo to generate extern int glfwLoadTexture2D(const char *, int); Thanks, Deep

Foreign functions do not link

Eduardo Costa — 2008-11-12T18:35:43

Could you people tell what I am doing wrong? The idea is to write a function in C that calls a function in Bigloo. I tried to compile the example from Foreign, but it does not compile either (although for a different reason). Here is the function in Bigloo: (module bigside (export (fib::int ::int)) (extern (testc::int () "testc")) (main main)) (define (main argv) (testc)) (define (fib x) (* x x )) And that is the function in C: extern int fib(int); int chama(int x) { return(fib(x)); } int testc() { printf("Hello."); } And here is how I tried to compile them: D:\Tcl\docs>bigloo -c ctes.c libdir:(. d:\bigloo\lib\bigloo\3.1b) - /d/bigloo/lib/bigloo/3.1b gcc -D_MINGW_VER -D_MINGW_VER -c -I. -I. -Id:\bigloo\lib\bigloo\3.1b ctes.c 0 D:\Tcl\docs>bigloo -c stes.scm libdir:(. d:\bigloo\lib\bigloo\3.1b) - /d/bigloo/lib/bigloo/3.1b gcc -D_MINGW_VER -D_MINGW_VER -c -I. -I. -Id:\bigloo\lib\bigloo\3.1b stes.c 0 D:\Tcl\docs>bigloo ctes.o stes.o -o tes.exe *** WARNING:bigloo:link No Bigloo module found for -- ctes.o ctes.o:ctes.c:(.text+0xd): undefined reference to `fib' collect2: ld returned 1 exit status 1 D:\Tcl\docs>bigloo stes.o ctes.o -o tes.exe *** WARNING:bigloo:link No Bigloo module found for -- ctes.o ctes.o:ctes.c:(.text+0xd): undefined reference to `fib' collect2: ld returned 1 exit status 1

Question about types

Deepankar Sharma — 2008-11-04T12:29:32

Hello, I am trying to write some bindings to glfw and opengl since bigloo-lib doesnt work on newer bigloo versions. I currently have an issue with types as defined in a module. I have reduced the problem down to this test case: (module glfw-wrapper (extern (include "GL/gl.h")) (extern (include "GL/glfw.h")) (extern (type uchar* (pointer uchar) "unsigned char *") (type s-glfw__s1076 (struct (Width::int "Width") (Height::int "Height") (Format::int "Format") (BytesPerPixel::int "BytesPerPixel") (Data::uchar* "Data")) "GLFWimage"))) The error I get is : ---------------------------------------------------------------------------------------------------------------- make[2]: Entering directory `/home/deep/programs/scheme/one_gl_function' File "glfw-wrapper.scm", line 13, character 350: # (Data::uchar* "Data")) # ^ # *** ERROR:top-level:use-type! # Can't find type -- uchar* ------------------------------------------------------------------------------------------------------------------ Any reason why the uchar* type declared as the first statement in the second extern block doesnt make uchar* available to the declaration of s-glfw__s1076? Oddly enough - Before moving these declarations to a seperate glfw-wrapper.scm file I was using them in a test file with a main function in it. And the code was compiling without any errrors and glfw was working. Any pointers will be appreciated. I will gladly release these bindings for everyone to use. Thanks, Deepankar

newbie on bigloo

2008-11-04T10:36:05

Hi, I'm toying with scheme and Bigloo, and I have (for the moment, more will surely come...) few questions on Bigloo's behavior. I'm using $ bigloo --------------- Bigloo (3.1b) on Windows platform. 1) My first question seems to have some link with some issues (?) concerning hashtables that I found in the mailing list's archives. I was trying to create an hashtable whose keys are pairs, and I could not manage to make them work. Here is a simple session which explains my problem : 1:=> (define h (make-hashtable)) h 1:=> (hashtable-put! h (cons 1 2) 3) 3 1:=> (hashtable-key-list h) ; my key is in the table, OK ((1 . 2)) 1:=> (hashtable-get h (cons 1 2)) ; really ? #f 1:=> (equal? (cons 1 2) (cons 1 2)) #t So, the key (1 . 2) is indeed a key in my hashtable, but "hashtable-get" seems unable to find the corresponding value. According to the documentation, equal? is used by default for hashtables, so where am I wrong ? Moreover, if I try this : 1:=> (hashtable-get h (car (hashtable-key-list h))) 3 I get the value. To me, it seems that eq? (or eqv?) is used by default ? My other questions are about the module system and the interpreter. 2) In the interpreter, (load "toto.scm") loads my file but none of the declarations are visible, even exported ones. $ cat foo.sch (define-struct point x y) $ bigloo ------------------------------------------------------ Bigloo (3.1b) `a practical Scheme compiler' Mon Sep 15 11:42:55 CEST 2008 Inria -- Sophia Antipolis email: bigloo< at >sophia.inria.fr url: http://www.inria.fr/mimosa/fp/Bigloo ------------------------------------------------------ 1:=> (module foo (include "foo.sch")) *** ERROR:eval: Unbound variable (from module `foo') -- define-struct Is this a documented limitation of the interpreter ? Sincerely yours, Saul

Scheme workshop survey deadline changed to november 2

Marc Feeley — 2008-11-01T00:49:19

The Scheme workshop has been held every year since 2000 (http://www.schemeworkshop.org/ ). It is one of the main venues where the Scheme community can come together to discuss the Scheme programming language. Traditionally the workshop is co-located with the ICFP conference (the only exception is in 2003 when it was held in Boston). The Scheme workshop steering committee (http://www.schemeworkshop.org/sc.html ) which oversees the organization of the workshop is searching for ways to increase the level of participation from the Scheme community. For this reason we are conducting a survey to determine at what location and time the next workshop should be held. One important issue is whether we should continue co-location with the ICFP conference. Please take a few minutes of your time to complete the online survey at the following URL (before the deadline of midnight sunday november 2): http://dynamo.iro.umontreal.ca/sw/survey Your feedback is very much appreciated! Marc Feeley for the Scheme workshop steering committee P.S. My apologies if you get this message multiple times.

A bunch of questions

deepankar — 2008-10-30T01:18:06

I have recently begun to use Bigloo and I have a some questions / comments about it. 1) Could the mailing list be moved / mirrored on groups.google.com / some other searchable interface ? This would allow people to easily search for content. Also the sign up page for the mailing list is currently in french and is hence slightly intimidating for non french speakers. 2) Is there a way to build bigloo on the Windows platform using the Microsoft VC++ compilers ? I know of http://www.yaya.fr/Bigloo/ but sadly that page is not current. 3) Are there any opengl bindings available (i tried bigloo-lib but that doesnt seem to compile with 3.1b or 3.2a under linux) 4) Is there anyway to cross compile windows executables from a linux installation of bigloo ? 5) And lastly - I see that bigloo has multimedia bindings like gstreamer. Are there any open source programs / examples available which use this part of bigloo. btw - Great job on bee. Under linux it works very well and gives the feeling of a really integrated environment.

Call for Submissions, International Lisp Conference 2009

2008-10-29T16:54:01

CALL FOR SUBMISSIONS INTERNATIONAL LISP CONFERENCE 2009 Lisp: The Next 50 Years http://www.international-lisp-conference.org Massachusetts Institute of Technology Cambridge, Massachusetts, USA March 22-25, 2009 Sponsored by the Association of Lisp Users General Information: The Association of Lisp Users is pleased to announce the 2009 International Lisp Conference will be held in Cambridge, Massachusetts, at the Massachusetts Institute of Technology, Sunday through Wednesday, March 22-25, 2009. The emphasis will be on present and future applications of technologies that have been or might soon be associated with the Lisp programming language and/or related languages and software. We encourage submissions in diverse areas, including but not limited to: language design and implementation, memory management, software engineering, mathematical and scientific computing, artificial intelligence, database processing and data mining, business intelligence, performance analysis, parallel processing, quantum computing, bioinformatics, telecommunications and networking, the semantic web, music, domain-specific languages, and entertainment technologies. ILC09 is not limited to topics discussed in previous symposia. Authors concerned about the appropriateness of a topic may communicate by electronic mail with the program chair prior to submission. Explaining a known idea in a new way may make as strong a contribution as inventing a new one. We encourage the submission of "pearls": elegant essays that illustrate an idea, for example by developing a short program. (A pearl should be concise, instructive, self-contained, and entertaining. Your pearl is likely to be rejected if your readers get bored, if the material gets too complicated, if too much specialized knowledge is needed, or if the writing is inelegant. The key to writing a good pearl is polishing.) There is no formal separation of categories and no need to explicitly label pearls as such: all papers, whether pearl or otherwise, will be judged on a combination of correctness, significance, novelty, clarity, and elegance. Each paper should explain its contributions in both general and technical terms, identifying what has been accomplished, explaining why it is significant, and comparing it with previous work. Authors should strive to make their papers understandable to a broad audience. Alongside our usual four-day program of tutorials, prominent invited speakers, and excellent technical sessions, this year we will also consider demonstration sessions. The official language of the conference is English. Further details are available at the conference web site: http://www.international-lisp-conference.org Technical Program: Original submissions in all areas related to the conference themes are invited for the following categories. Papers: Technical papers of up to 15 pages that describe original results ("research papers") or explain known ideas in new ways ("pearls"). Demonstrations: Abstracts of up to 4 pages for demonstrations of tools, libraries, and applications. Tutorials: Abstracts of up to 4 pages for in-depth presentations about topics of special interest for at least 90 minutes and up to 180 minutes. Panel discussions: Abstracts of up to 4 pages for discussions about current themes. Panel discussion proposals must mention panel members who are willing to partake in a discussion. Lightning talks: Abstracts of up to one page for talks to last for no more than 5 minutes. Important Dates: Please send contributions before the submission deadline to the program committee (ilc09-program-committee at alu.org). Deadline for submissions: December 9, 2009 Notification of acceptance or rejection: January 9, 2009 Deadline for final paper submissions: February 9, 2009 Organizing Committee: Conference Chair: Daniel Weinreb (ITA Software) General correspondence: ilc09-organizing-committee at alu.org Program Chair: Guy L. Steele Jr. (Sun Microsystems Laboratories) Contact: ilc09-program-committee at alu.org Local chair: Howard Shrobe (MIT) Technical Program Committee: to be announced soon

[3.1b Doc] make-client-socket

igli — 2008-10-21T15:56:18

Hi, Minor doc error (was told to report as and when, and not bundle them all into one) encountered while upgrading to 3.1b: Example gives: (make-client-socket "www.inria.fr" 80 :buffer #f) When the actual API is: make-client-socket::socket host port #!key (inbuf #t) (outbuf #t) (timeout 0) (ie no buffer keyword arg.) Thanks for bigloo; it makes roadsend rock ;-) Regards, igli. PS: Thank you to chandler for being so helpful in #scheme. A good community makes a *big* difference.

Scheme workshop survey

Marc Feeley — 2008-10-17T18:04:54

The Scheme workshop has been held every year since 2000 (http://www.schemeworkshop.org/ ). It is one of the main venues where the Scheme community can come together to discuss the Scheme programming language. Traditionally the workshop is co-located with the ICFP conference (the only exception is in 2003 when it was held in Boston). The Scheme workshop steering committee (http://www.schemeworkshop.org/sc.html ) which oversees the organization of the workshop is searching for ways to increase the level of participation from the Scheme community. For this reason we are conducting a survey to determine at what location and time the next workshop should be held. One important issue is whether we should continue co-location with the ICFP conference. Please take a few minutes of your time to complete the online survey at the following URL (before the deadline of midnight friday October 31): http://dynamo.iro.umontreal.ca/sw/survey Your feedback is very much appreciated! Marc Feeley for the Scheme workshop steering committee P.S. My apologies if you get this message multiple times.

timing a function

Antonio, Fabio Di Narzo — 2008-10-14T19:20:13

Hi all. By reading at the manual, I see I should use the 'time' function, which mimics the posix 'time' command, and is exactly what I need. However, my bigloo setup doesn't seem to find it: 1:=> (module uno) 1:=> (time (lambda () (list))) *** ERROR:eval: Unbound variable (from module `uno') -- time 0.call-with-values Tried with: x86-Solaris bigloo-3.0c, x86-Linux bigloo-2.8c These are my very first attempts with bigloo, so it's very likely I'm missing something stupid. If true, pls apologize for the dumb question. Antonio.

Changes between 3.1b and 3.2a?

Sven Hartrumpf — 2008-10-14T18:41:27

Hi all. I have legacy code for a simple http server based on the example Manuel once packaged with bigloo. (BTW: when was it removed?) Switching to 3.2a, I see the following new crash: File "/home/sven/nli/bhttp.scm", character 4510: *** ERROR:read/rp: Grammar arity mismatch -- # 0.http-eval, bhttp.scm:97 1.make-http-server, bhttp.scm:58 2.bhttp-main, bhttp.scm:38 3.web-server-top-level, bhttp.scm:337 bhttp.scm at line 96 is: (define (http-eval s::socket num::int) (let* ((s-port (socket-input s)) (lines (let loop ((line (synchronous-read-characters s-port (lambda (c) (memv c '(#\newline #\return)))))) (if (<= (length line) 1) ; turned = into <= '() (cons (list->string line) ; drop last character (loop (synchronous-read-characters s-port (lambda (c) (memv c '(#\newline #\return))))))))) (line (fold-right (lambda (line2 result) (if (string=? result "") line2 (string-append line2 " " result))) "" lines))) (flush-output-port (current-output-port)) (string-case line ("VERSION" ... Looks like some function was changed w.r.t. arity? Any solutions? Sven P.S. Manuel, if needed, I can send you the whole file.

simple web server - memory

w r — 2008-10-10T17:36:27

I'd like to use bigloo 3.1b for a simple web server, but it seems to be growing by ~3K bytes with each client access. Am I doing this right? (module server (main main)) (define (http-read-headers p) (let loop ((line (read-line p)) (acc '())) (cond ((or (eof-object? line) (string-null? line)) (reverse acc)) (else (loop (read-line p) (cons line acc)))))) (define (make-unthreaded-server! port) (let ((s (make-server-socket port))) (let loop ((sock (socket-accept s))) (let ((inp (socket-input sock)) (outp (socket-output sock))) (http-read-headers inp) (fprintf outp "HTTP/1.1 200 OK\n\ntesting memory leaks") (flush-output-port outp) (socket-close sock) (loop (socket-accept s)))))) (define (main args) (make-unthreaded-server! 8080))

install handler for any error or exception

naruto canada — 2008-10-04T02:52:32

hi I'm looking for mechanism for installing handler for "any error or exception", I simply need to call (exit 1) on "all" errors or exception. For example, in mzscheme you can simply do: (uncaught-exception-handler (lambda (exn) (exit 1))) in petite: (error-handler (lambda (w m . a) (exit 1))) note: I've read the manual at: http://www-sop.inria.fr/mimosa/fp/Bigloo/doc/bigloo-17.html#Errors-Assertions-and-Traces It's not entirely clear what I should do to achieve this. Thanks