On a related note, I was wondering how one typically handles the case
of a function that returns multiple values, but also needs to signal
failure. I can think of a few ways:
1. Returning #f as one of the values, and testing for that;
2. out-of-band -- an exception or continuation to error handli
meant to address.
On 5/13/05, Zbigniew <[EMAIL PROTECTED]> wrote:
> It makes sense to throw some kind of exception, but this seems
> exceedingly rare from what I've seen. Returning #f is so much more
> common.
___
Chicken-users
I'm returning multiple values with something like C_values(4, 0, C_k,
C_fix(1), C_fix(2)). Is C_values the right function to call? It
works fine, but I wonder if this is for internal use only.
On 5/9/05, felix winkelmann <[EMAIL PROTECTED]> wrote:
> Sorry, this is currently not possible. Using
I was writing CPS code directly with foreign-declare (by studying
compiled output) before discovering the much nicer foreign-primitive.
It was a good way to learn the internals... ;)
On 5/14/05, felix winkelmann <[EMAIL PROTECTED]> wrote:
> On 5/14/05, Zbigniew <[EMAIL PROTECTED]>
nal" (i.e. not part of the documented interface). So I've just
studied chicken and the eggs and used my best judgement.
On 5/17/05, felix winkelmann <[EMAIL PROTECTED]> wrote:
> On 5/15/05, Zbigniew <[EMAIL PROTECTED]> wrote:
> > I was writing CPS code directly with
Your example works for me on 1.935.
#;18> (use regex)
; loading library regex ...
#;19> (string-match-positions ".*needle.*" "hay needle stack" 2)
((2 16))
On 5/15/05, Christian Jaeger <[EMAIL PROTECTED]> wrote:
> Hello
>
> Is this a bug?:
>
> (string-match-positions ".*needle.*" "hay needle
Correct, I am using libc regex (on OS X).
On 5/18/05, felix winkelmann <[EMAIL PROTECTED]> wrote:
> I assume Christian uses pcre and you not right?
___
Chicken-users mailing list
Chicken-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/chicken
I was wondering if csi keeps some kind of hidden reference to objects
you create. Heap memory never seems to get freed, as shown below.
Running this code as an executable, or compiling it into a shared
library and running that in csi, does free the memory.
#;1> (memory-statistics)
#(50 44084
Hi, I am starting to document an egg I'm working on. I notice many of
the html docs on the web page look similar and was curious if there's
a script or template used to generate them. Or are they being written
by hand?
___
Chicken-users mailing list
C
Oh, I thought that was just command history so I never tried it!
That's really neat.
On 5/23/05, felix winkelmann <[EMAIL PROTECTED]> wrote:
> On 5/21/05, Zbigniew <[EMAIL PROTECTED]> wrote:
> > I was wondering if csi keeps some kind of hidden reference to objects
&g
Hello,
I've ported the final reference implementation of SRFI 43 to Chicken,
along with several additions/fixes described in the documentation. I
hope the result is good enough to be put on the eggs page.
http://jiyu.gnook.org/~zbigniew/vector-lib/vector-lib.scm [scheme source]
sion has proved
very useful to me.
Source: http://jiyu.gnook.org/~zbigniew/hostinfo/hostinfo.scm
HTML: http://jiyu.gnook.org/~zbigniew/hostinfo/hostinfo.html
Zb
___
Chicken-users mailing list
Chicken-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/chicken-users
This should fix a small bug in libc regex where a case-insensitivity
request i.e. (regexp "." #t) would be ignored.
--- chicken-1.935/regex.scm.orig2005-05-30 15:39:20.0 -0500
+++ chicken-1.935/regex.scm 2005-05-30 15:39:38.0 -0500
@@ -268,7 +268,7 @@
into an egg? Is
there another, better implementation somewhere?
http://gnook.org/~zbigniew/args/args.scm
http://gnook.org/~zbigniew/args/args-examples.scm
Zb
___
Chicken-users mailing list
Chicken-users@nongnu.org
http://lists.nongnu.org/mailman
I like your recipe. Given a robust enough parser and perhaps some
very minimal markup to help it along, it would cover a lot of cases.
The only problem is, it does not allow you to add a processor
function for each option.
In practice, though, I wonder how useful the processors are---the
exampl
lack of data vs code distinction)
I would be interested in an outline of your idea even if you can't
find your code.
On 6/6/05, Michele Simionato <[EMAIL PROTECTED]> wrote:
> On 6/6/05, Zbigniew <[EMAIL PROTECTED]> wrote:
> > I like your recipe. Given a robust enough
ttp://gnook.org/~zbigniew/args/args.scm
http://gnook.org/~zbigniew/args/args-examples.scm
Here is an illustration [the output is spaced correctly under a
fixed-width font]:
(use args)
(define opt-help (args:make-option (h help) #:none "Display this text" (usage)))
(define opts
(list (a
On 6/7/05, Zbigniew <[EMAIL PROTECTED]> wrote:
> (set!-values (options operands)
> (args:parse '("help") opts opt-help))
Minor correction, that should have been
(set!-values (options operands)
(args:parse '("--help") opts opt-help))
As written, &qu
Reed,
Although it's tangential to your question, if you wind up with a macro
shadowing a procedure, you can remove the binding with
(undefine-macro! foo). I have used this a few times during long
interpreter sessions.
On 6/6/05, Reed Sheridan <[EMAIL PROTECTED]> wrote:
> A self-explanatory snipp
I made the current version into an egg:
http://jiyu.gnook.org/~zbigniew/args/args.egg
You may load the code (args.scm and args-support.scm, also available
at that URL) into the interpreter without compiling or installing the
extension. However, you must explicitly load args-support.scm (the
http://jiyu.gnook.org/~zbigniew/patches/chicken-1.935-hpux-sl-so.patch
This patch enables compilation on HP-UX (tested on 11i). It removes
the -R flag to the linker (which HP does not support) and changes the
shared library extension to ".sl" to match what is generated by
libtool.
hostinfo egg is also available (host, service, and protocol lookup).
I don't believe I officially announced that before.
On 6/16/05, felix winkelmann <[EMAIL PROTECTED]> wrote:
> New eggs:
>
> args (thanks to Zbigniew)
> web-scheme (thanks to Carlos)
>
> Several f
close-input-pipe discards the return value of the process you invoked,
so you can't figure out if it failed. It only signals an error (via
exception) if the system call failed.
In Perl, close(pipe) will fail on a non-zero exit code (though the
code is returned out of band, in $?).
An easy fix is
On 6/20/05, Jörg F. Wittenberger <[EMAIL PROTECTED]> wrote:
> Am Montag, den 20.06.2005, 11:51 -0500 schrieb Zbigniew:
> > An easy fix is to have close-input-pipe return the value from pclose()
> > (stored in the variable "r"), since it currently returns nothing at
would have to be done out of band (should it
be deemed desirable.)
On 6/20/05, Zbigniew <[EMAIL PROTECTED]> wrote:
> Or possibly a really great idea that would even catch
> non-zero return codes in with-input-from-pipe etc.
___
Chicken-u
Works for me. I was about to patch my tree like that.
On 6/20/05, felix winkelmann <[EMAIL PROTECTED]> wrote:
> To be honest, I can't think of anything better than returning the status code
> in the moment.
___
Chicken-users mailing list
Chicken-users
Use return type c-string* and chicken will free() the memory for you.
Also, your function does not perform a callback to Scheme, so you can
use foreign-lambda instead of foreign-callback-lambda [which is itself
deprecated in favor of foreign-safe-lambda].
On 6/24/05, Andrey Fomichev <[EMAIL PROTE
Let's say I am writing an extension and I have some common code I want
available to all the .scm files in the extension, but not visible to
the user. Since these functions are internal, I similarly may not
want them visible to -every- .scm file in the extension. And it is
possible that other exte
://jiyu.gnook.org/~zbigniew/sxml-tools/sxml-tools.egg
http://jiyu.gnook.org/~zbigniew/sxml-tools/sxml-tools.html [HTML doc]
Zb
___
Chicken-users mailing list
Chicken-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/chicken-users
> Thank you for your irrefragable answer
No, thank -you- for the vocabulary lesson. :)
___
Chicken-users mailing list
Chicken-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/chicken-users
A SXML to HTML documentation tool for eggs.
http://jiyu.gnook.org/~zbigniew/eggs/eggdoc.egg
http://jiyu.gnook.org/~zbigniew/eggs/eggdoc.html
All of my extensions have been converted to use eggdoc, and the source
code for each is contained within this egg. If Felix could also
upload the new HTML
The SXML transformations (to XML, SXML, and HTML) from the SSAX
project at Sourceforge.
http://jiyu.gnook.org/~zbigniew/eggs/sxml-transforms.egg
http://jiyu.gnook.org/~zbigniew/eggs/sxml-transforms.html
___
Chicken-users mailing list
Chicken-users
A simple extension that provides several XHTML and HTML doctypes as strings.
http://jiyu.gnook.org/~zbigniew/eggs/doctype.egg
http://jiyu.gnook.org/~zbigniew/eggs/doctype.html
___
Chicken-users mailing list
Chicken-users@nongnu.org
http
Toby, you might want to look at the hostinfo egg for examples of
passing vector data back to scheme from C. In general there are a few
ways to accomplish what you want. However, note that srfi-4 values
cannot be used as return types from foreign-lambda.
It is definitely possible to build an obje
Thought I got rid of all those invisible embedded control chars.
Let's try again.
Toby,
Here are two examples. In one I create the string on the heap and
return it as a scheme-object from the C code. In the other I write
directly into a scheme string, using a locative as the pointer.
Neither
Toby,
Here are two examples. In one I create the string on the heap and
return it as a scheme-object from the C code. In the other I write
directly into a scheme string, using a locative as the pointer.
Neither probably works with UTF-8. You didn't specify how the to-len
was computed, so I ass
I too am quite interested in a decent module system, especially after
kludging a few things in the sxml-tools egg to get around the lack of
one. I have a couple general questions/observations though:
- Startup time of either simple-macros or syntax-case is very long --
1.7s as opposed to 0.1s wit
Works for me. In fact (expt 100 1) gives the wrong magnitude,
about 10^305, without the patch. Interestingly it is even faster than
"calc", the "C-style arbitrary precision calculator (version
2.11.10.1)":
time csi -batch -eval '(use numbers) (display (expt 100 1))
(newline)' > /dev/null
Also I've found if you override a built-in function such as + or
length, previous behaviour will be wiped out. So after:
(define-method (+ ( x) ( y)) (string-append x y))
you lose the ability to do (+ 2 3) unless you define the appropriate
method on integers, and then + will only take 2 arguments
[re: Aquaterm adapter] Nice-- I have made heavy use of gnuplot's
AquaTerm driver and the possibility of writing to AquaTerm directly
from Scheme is enticing. Thanks.
On 9/12/05, Thomas Christian Chust <[EMAIL PROTECTED]> wrote:
> Hello,
>
> I have updated my sqlite3 egg with a minor bugfix to
The list got all three of your messages.
On 9/25/05, Brandon J. Van Every <[EMAIL PROTECTED]> wrote:
> [apologies if this is a repost. I've tried to send this 2 times now and
> am still not seeing it. Black hole?]
___
Chicken-users mailing list
Chick
Note that Campbell's proposal only covers the functionality of DSSSL
#!optional, not #!key [which Shawn liked I presume because of the
out-of-order named parameters].
One might also consider using get-keyword on the rest-list (see
section 5.7.6, Keywords). I used this method in args.egg for named
The web interface to gmail will show (in your inbox) your replies to a
thread already in your inbox. Thus it was surprising that no messages
would appear at all, since the first response to a thread you initiate
will pull that thread into your inbox. Now it makes sense, if the web
client isn't be
You could use fx+ if you're guaranteed to be operating on fixnums.
(You probably knew this already though)
On 9/29/05, Daishi Kato <[EMAIL PROTECTED]> wrote:
> Original Subject: wishlist: numbers egg (random) support
>
> At Wed, 28 Sep 2005 20:09:03 +0200,
> felix winkelmann wrote:
> >
> > That s
These changes---is someone looking at adding Cocoa support, or
otherwise interfacing with OS X stuff?
On 11/2/05, felix winkelmann <[EMAIL PROTECTED]> wrote:
> - csc:
> * new option `-objc' compiles generated C files in Objective-C mode
> * accepts .m files
> * new option `-framework' is pas
I was testing out the RPC egg and noticed it didn't work between a Mac
and a PC. This stems from endian issues in the s11n egg. Now, I
don't know if serialization was ever intended to work across
architectures, but I went ahead and made a simple patch [attached] to
the s11n egg. With this, the R
I think you're describing a more complicated version of WRITE and READ.
On 11/12/05, Sean Doran <[EMAIL PROTECTED]> wrote:
> One approach would be to translate the binary output into a tagged/
> marked-up format like *gasp* XML...
>
> A sender would do something like write a tagged/marked-up (numb
Pupeno, although I didn't try your code, the tagged pointer
functionality seems to work for me (for artificially-generated
pointers). E.g.
Version 2, Build 207 - macosx-unix-gnu-ppc - [ dload ptables ]
...
#;9> (address->pointer #x80bf740)
#
#;10> (tag-pointer # 'x-display)
#
Could you try this
Please try the following patch:
--- runtime.c.orig 2005-11-16 19:53:45.0 -0600
+++ runtime.c 2005-11-16 19:53:49.0 -0600
@@ -7308,7 +7308,7 @@
void C_make_tagged_pointer(C_word c, C_word closure, C_word k, C_word tag)
{
- C_word ab[ 2 ], *a = ab,
+ C_word ab[ 3 ], *a =
This might answer your question:
http://www.mail-archive.com/users@ipv6.org/msg02107.html
On 11/18/05, Daishi Kato <[EMAIL PROTECTED]> wrote:
> Hi,
>
> I'm trying to compile hostinfo egg on windows.
> What I have done so far:
> - add #include and #if _WIN32
> - add ws2_32.lib for link
>
> howeve
cense)
> or just to use inet_ntoa instead and does not support IPv6?
>
> It is currently not easy to send you a patch,
> because the egg is single scm file, which does not
> include hostinfo.setup. Would you update it so that
> the egg will be hostinfo.egg, or I could do it.
&g
I would like to pass along an observation on returning multiple
values. I wrote a C function that needed to return two values to
scheme, an int and a pointer. Ultimately, I allocated a scheme object
for the pointer prior to the call, passed it as an argument, then
manipulated the contents in C, "
:
> On 11/22/05, Zbigniew <[EMAIL PROTECTED]> wrote:
> >
> > Unfortunately, I found that handling multiple values is very slow. In
> > fact, if you cannot use the method I mentioned above (the fastest), it
> > is much faster to construct and deconstruct a list or
Try the readline egg at
http://www.call-with-current-continuation.org/eggs/readline.html.
On 11/22/05, Matthew David Parker <[EMAIL PROTECTED]> wrote:
> PS. On a side question, is there some thing you can do in the csi chicken
> interpreter to make it so you can press the "up" arrow and cycle thr
I have a C function which calls another C function that creates and
returns a scheme-object. Because it returns (rather than continuing)
I cannot allocate this on the stack, rather I use malloc. The
original caller itself returns this object, along with a c-pointer
that can later be freed. As th
Here's another interesting performance note, and I assure you
everything is compiled this time.
I have a C function which returns a void * by reference (i.e. it takes
a void ** as argument). I was using let-location, but I found that if
you pass a # object as a scheme-pointer, it works just as
we
It looks like everybody is working on the same thing. I have been
porting and improving the gauche-objc bridge. It takes a highly
dynamic (read: slow but very interactive) approach, handling
everything at runtime. Method calls work fine; but I am currently
working on solving a memory allocation
That's odd, the chicken.info file has the correct signature of
(set-read-syntax! CHAR PROC).
On 11/24/05, felix winkelmann <[EMAIL PROTECTED]> wrote:
> On 11/24/05, Pupeno <[EMAIL PROTECTED]> wrote:
> > Note:
> > http://www.call-with-current-continuation.org/manual/Reader-extensions.html#index-set
OS X on PPC. Compilation flags are -O3 -fomit-frame-pointer
-fno-strict-aliasing -mcpu=7450 -DC_ENABLE_PTABLES.
I will try on my x86 box later.
On 11/24/05, felix winkelmann <[EMAIL PROTECTED]> wrote:
> On 11/24/05, Zbigniew <[EMAIL PROTECTED]> wrote:
> > Here's anoth
Both signatures are correct in the info file.
On 11/24/05, Pupeno <[EMAIL PROTECTED]> wrote:
> On Thursday 24 November 2005 12:29, you wrote:
> > That's odd, the chicken.info file has the correct signature of
> > (set-read-syntax! CHAR PROC).
> But we are talking about set-dispatch-read-syntax!
>
sable ideas into your code. Of course, if the past is
any guide, you should have a fully-functional implementation inside of
3 days anyway. :D
On 11/25/05, felix winkelmann <[EMAIL PROTECTED]> wrote:
> On 11/24/05, Zbigniew <[EMAIL PROTECTED]> wrote:
> >
> > If you want I
Oh, to clarify---unsigned char is always a character. So uchar return
/ arg types do not accept or convert to predicate values, i.e. an
unsigned char always returns C_make_character(*(unsigned char *)obj);
On 11/26/05, Zbigniew <[EMAIL PROTECTED]> wrote:
> Yes, they use th
Yes, they use the natural representation, so if you know from the
methodReturnType call that a return type is _C_DBL, you can cast the
return value directly to a double. E.g. C_flonum(&a, *(double
*)rbuf); And _C_CLASS is a Class (struct objc_class *) and _C_ID is
an id (struct objc_object *).
I got libffi to work, and ran some tests on the objc egg. In my
opinion, if lazy-ffi is the way you choose to go, it will need some
modifications to arg and return value conversion, because the current
state acts unexpectedly as compared to the native FFI. I appended
some test results below (I se
I have updated the testeez egg to the latest upstream version (0.3)
and made a couple small additions. A new eggdoc is included.
Get it at: http://jiyu.gnook.org/~zbigniew/eggs/testeez.egg
___
Chicken-users mailing list
Chicken-users@nongnu.org
http
the
first Cocoa application for Chicken. :) That code is included below,
and also in the egg (cd tests/; make).
Get zobjc at: http://jiyu.gnook.org/~zbigniew/eggs/zobjc.egg
Oh, and it's a cold, snowy 19F here... or -7.222 C.
;;; --
;;; test f->
I renamed zobjc to objc, and added an eggdoc.
http://jiyu.gnook.org/~zbigniew/eggs/objc.egg
http://jiyu.gnook.org/~zbigniew/eggs/objc.html
Zb
___
Chicken-users mailing list
Chicken-users@nongnu.org
http://lists.nongnu.org/mailman/listinfo/chicken
Thanks Raffael. The unit test was for internal use only at the moment
and I forgot to fix it. I wrote it when this egg was a port of the
Gauche-ObjC bridge. Later I rewrote from scratch but kept the unit
test for verification. That particular test is expected to "fail"
right now, so no worries.
I've written up a document which walks you through creating the
Currency Converter application in Apple's Objective C tutorial, using
the objc egg.
It uses XCode to create the application skeleton, but would be easily
tweaked for Project Builder. A completed application is included,
which has its
I have updated the objc egg to support NSRect, NSPoint, NSSize, and
NSRange, fixed a few bugs and done a bit of renaming. I have also
added a new application, Color View, which demonstrates creating a
custom NSView (NSRect support was required for this).
I'll update the SVN repository as soon as
I have updated the objc egg in the SVN repository.
On 12/23/05, Zbigniew <[EMAIL PROTECTED]> wrote:
> I have updated the objc egg to support NSRect, NSPoint, NSSize, and
> NSRange, fixed a few bugs and done a bit of renaming. I have also
> added a new application, Color View, whi
Previously (and as per the manual) I could define a macro alias with
(define-macro new-name old-name)
This doesn't work any more---now, old-name is evaluated as a variable.
Has this been changed?
Of course, one can still get the effect of an alias with
(define-macro (new-name . args) `(old-name
Hi all. I updated the sxml-transforms egg (SVN repo) with the
following two changes:
1. Do not prefix inline elements with a newline--fixes a problem with
extraneous whitespace. E.g. (p "in" (em "cred") "ible") no longer
renders as "in credible".
2. Add (& ENTITY-NAME ...) rule. Allows quoting
wrote:
> On 12/28/05, Zbigniew <[EMAIL PROTECTED]> wrote:
> > Previously (and as per the manual) I could define a macro alias with
> > (define-macro new-name old-name)
> >
> > This doesn't work any more---now, old-name is evaluated as a variable.
> >
t;Use the". This is the same behaviour as
seen in HTML.
So, if you regenerate an eggdoc after updating sxml-transforms, watch
for words thatrun together.
On 12/28/05, Zbigniew <[EMAIL PROTECTED]> wrote:
> Hi all. I updated the sxml-transforms egg (SVN repo) with the
> following two chan
Additionally, it would become infeasible to splice a pair into or out
of the list [an O(1) operation], given only a pointer into the middle
of the list, because you cannot update the counts of earlier list
elements. This would be "possible" with a doubly-linked list---but
insert and remove would n
I'm in the middle of adding a bunch of stuff to the objc egg, and have
encountered a problem. For the first time, I happened to write some
code which consed together a few thousand Objective C objects
(#) into a list. Suddenly, I took a gigantic speed hit
using cons, as compared to code which sim
It looks like Thomas is right about the callback, although I can't get
your example to crash. I'm running Mac OS X PPC, which sometimes
allows memory accesses that would cause segfaults elsewhere, so I have
another possibility if you still get crashes. Before testing for the
scheme string length
6, felix winkelmann <[EMAIL PROTECTED]> wrote:
> On 1/18/06, Zbigniew <[EMAIL PROTECTED]> wrote:
> The current solution is to increase the size of the pending-finalizers
> buffer, done with the "-:fXXX" runtime option. You can also run your
> code with "-:d"
I've updated the objc egg to v0.4 in the SVN repository. An eggdoc is
forthcoming, hopefully very soon. In the meantime, here's a summary
of changes:
- Class proxies. The objc:class wrapper for each class is now unique
(rather than simply containing a pointer) and contains metadata useful
to th
On 1/21/06, Kon Lovett <[EMAIL PROTECTED]> wrote:
> objc:box/unbox instead of objc:wrap/unwrap
That's not a bad idea.
> obj-c:* instead of objc:*)?
The "objc" prefix is, near as I can tell, universal in Lisp--OpenMCL,
Gauche--and other implementations---Python, Ruby, even Emacs'
bug] running 1 finalizers (0 live, 7 allocated) ...
bye
169284
#;14>
On 1/20/06, felix winkelmann <[EMAIL PROTECTED]> wrote:
> On 1/18/06, Zbigniew <[EMAIL PROTECTED]> wrote:
> Ok. Finalizers work like this:
>
> There is a list of "live" finalizers, that is fin
Your other message came through. I was able to compile syntax case on
10.4.4 without a problem (and I can't figure out what exit code 10
is). Perhaps run the chicken command (not chicken-setup, but the
failing invocation of chicken) by hand, removing -quiet and adding
-verbose instead.
On 1/22/0
Sorry if I'm being dense---why does it need to release any finalizers
at all, if the pending buffer is totally empty? I thought the size of
the pending buffer was the issue here.
On 1/23/06, felix winkelmann <[EMAIL PROTECTED]> wrote:
> The runtime system doesn't know at that time how many of the
zers (5623 live, 5624 allocated) ...
594284
#;7> (define L '())
#;8> (gc #t)
[debug] running 5617 finalizers (6 live, 5624 allocated) ...
[debug] resizing heap dynamically from 1e8688 tof4344 bytes...
272588
On 1/23/06, felix winkelmann <[EMAIL PROTECTED]> wrote:
> On 1/
I've updated the eggdoc now in SVN and at http://3e8.org/zb/eggs/objc.html.
I'm curious if updating the egg repository automatically results in
stuff being updated on the web page, or if someone (like Felix) needs
to push an update by hand.
On 1/21/06, Zbigniew <[EMAIL PROTECTED]&g
Just noticed that find-files traverses symbolic links to directories
(since directory? returns true for a directory symlink). I'm not sure
this behaviour is desirable---at least there should be the option to
skip symlinks or treat them strictly as files.
_
You're right about limit, of course. I looked at using the
predicate---which isn't called for directories---and forgot limit,
which is.
On 2/1/06, felix winkelmann <[EMAIL PROTECTED]> wrote:
> On 2/1/06, Zbigniew <[EMAIL PROTECTED]> wrote:
> > Just noticed tha
I uploaded an article about reading and modifying the iTunes Music
Library property list using the objc egg. It's saturated with
commentary and builds a simple library from the ground up, so it might
also be useful outside the domain of plists and iTunes, and to
beginning Schemers as well.
http:/
Hi Arto. That is very strange. string-upcase is included in SRFI-13,
which I included in cocoa.scm (hopefully correctly) using
(require-for-syntax 'srfi-13). This works for me on 2.207 and there
have been at least a couple people successfully building it on 2.216.
I just built it again as I wro
I believe there is a serious regression in this patch. The URL is
being canonicalized in its entirety at the beginning of
http:decode-url, before it is taken apart. This is not correct,
because escaped characters should only be unescaped -after- breaking
the URL down into pathname and key/value p
The patch works for me--in fact it is critical for decoding complex
referer URLs, such as from Google Images--but I want to be sure I did
not miss something (like a problem with http-server).
The "url" egg also should probably be changed to remove the + to space
translation, as this translation sh
Very nice! Thanks very much to both of you.
On 2/18/06, Thomas Chust <[EMAIL PROTECTED]> wrote:
> On Sat, 18 Feb 2006, [EMAIL PROTECTED] wrote:
>
> > [...] SQLite has been fixed so that IF EXISTS and IF NOT EXISTS always
> > generate a non-NULL statement if there are no errors. [...]
>
> Wow, tha
I have written a CGI script using Chicken, along with a backend script
which parses Apache logs and deposits search data in a SQLite
database. This was partially a feasibility test, as most people are
using Spiffy for page generation, and only plain CGI is allowed at my
webhost.
If anyone is inte
Arto: please download another copy of the objc egg. This problem was
fixed a week ago. Thanks!
On 2/19/06, Arto Bendiken <[EMAIL PROTECTED]> wrote:
> Hi,
>
> I ran into a small glitch trying to use the objc egg with the brand
> new Chicken 2.3 (though this also applies to 2.216):
>
> #;1> (use o
I don't recall seeing a response here and someone else had the same
problem. The "solution" is to set
DYLD_LIBRARY_PATH=~/scheme/chicken/lib
[or wherever your chicken libs are]. It seems that when the chicken
libraries are not installed in a standard search path such as
/usr/local/lib, dlopen w
Ben,
I haven't yet attacked the problem of distributing binaries to people
without a copy of Chicken installed, and I've never been able to get
-static working on OS X. Hopefully we can get something working
starting with Felix's instructions. It is possible that copies of
Chicken's dynamic lib
For what it's worth, I put a snapshot of darcs HEAD at
http://3e8.org/pub/chicken-darcs-20060410.tar.gz [3.2M]
On 4/10/06, John Cowan <[EMAIL PROTECTED]> wrote:
> felix winkelmann scripsit:
>
> > Adding an extra slot to symbols would be pretty handy, and would make
> > it easy to implement diffe
-Excellent- work! As usual.
On 4/21/06, felix winkelmann <[EMAIL PROTECTED]> wrote:
> Hi!
>
> I have put together a walkthrough for creating a self-contained
> cocoa application that does not need specially compiled eggs.
> The idea is to put all required libraries and eggs into the application b
I thought the non-FFI limit was 126? Not that it applies to Reed's question.
On 4/18/06, felix winkelmann <[EMAIL PROTECTED]> wrote:
> Currently the limit is 120 vs 1000 with libffi installed (the latter
> is a completely arbitrary limit).
___
Chicken
1 - 100 of 278 matches
Mail list logo
