On Fri, Jun 22, 2012 at 12:21 PM, Matthew Mondor
mm_li...@pulsar-zone.netwrote:
I think that the ANSI symbol to C function indexes are very nice. I
did not understand the sorting order (perhaps there simply is none
though).
It just copy-pastes the order in the ANSI specification. That is
On Fri, Jun 22, 2012 at 1:06 PM, Stanislav Frolov frolosof...@gmail.comwrote:
Thank you, Matt! I just miss finalization section in the ECL manual.
I am sorry: the references had disappeared almost entirely due to an
upgrade of the manual to support Docbook 5.0. If you go back to
On Sun, Jun 24, 2012 at 1:31 PM, Alessandro Serra gas.h...@gmail.comwrote:
Hi,
there is a typing error in legacy.h: FINXUMP.
#define FIXNUMP(o) ECL_FINXUMP(o)
Thanks, I will correct it.
A question. There is any function to create a string object from a
wchar_t array?
May
On Mon, Jun 25, 2012 at 4:19 AM, lookwong look.wangl...@gmail.com wrote:
http://sourceforge.net/tracker/?func=detailaid=3536628group_id=30035atid=398053
With the latest slime and ecl something is wrong.
The Slime problems disappear with the recent patches -- basically some
debug
On Tue, Jun 26, 2012 at 10:47 PM, Eric Marsden eric.mars...@free.fr wrote:
Hi,
ECL (Embeddable Common-Lisp) 12.5.1
(git:6e1c455e5207b3592125c52a7ec39fa353bc8a9d)
(lambda (p1)
(declare (type (short-float 0.0 43225.234) p1))
(functionp p1))
#bytecompiled-function 03ad1370
On Tue, Jun 26, 2012 at 10:53 PM, Eric Marsden eric.mars...@free.fr wrote:
ECL (Embeddable Common-Lisp) 12.5.1
(git:6e1c455e5207b3592125c52a7ec39fa353bc8a9d)
(lambda (p3)
(declare (optimize (speed 1) (safety 0) (debug 2) (space 0))
(type (eql 4432671931722342132) p3))
On Sat, Jun 30, 2012 at 5:40 AM, Matthew Mondor mm_li...@pulsar-zone.netwrote:
Another interesting thing to consider is Common Lisp's UNWIND-PROTECT
capability, which can be used with a clean WITH macro to avoid
forgetting to explicitely free/destroy an FFI-pointed object and/or OS
resource,
Yo didn't get my email announcing the change, did you? I am getting
messages from mm_lists... with bounces. Could you please confirm if you get
this one... Or anybody in the list for that matter.
On Sun, Jul 1, 2012 at 4:34 PM, Matthew Mondor mm_li...@pulsar-zone.netwrote:
On Thu, 28 Jun 2012
I have managed to code a couple of macros that do the equivalent of
RESTART-CASE in C. I believe the same thing can be done for HANDLER-CASE.
Could you guys have a look at the manual and tell me whether it looks fine
for regular use?
http://ecls.sourceforge.net/new-manual/ (under I.6 Conditions)
On Mon, Jul 2, 2012 at 3:29 PM, Juan Jose Garcia-Ripoll
juanjose.garciarip...@gmail.com wrote:
Could you guys please emit your opinion about this request? How do other
implementations behave in this respect? Do they all offer an ABORT restart
on every thread?
No need for polling. In the end
On Tue, Jul 3, 2012 at 7:34 PM, Matthew Mondor mm_li...@pulsar-zone.netwrote:
I also noticed the similar handler-case related macros but I've not
seen them in the conditions dictionary.
I just finished the equivalent of HANDLER-CASE a few hours ago. I will try
to change the documentation
You cannot build with gengc and multithreaded ECL. Did you deactivate
threads?
On Fri, Jul 13, 2012 at 9:10 AM, Eric Marsden eric.mars...@free.fr wrote:
Hi,
ECL from git does not build on Linux/AMD64 with gengc (it builds fine
without). The symptom is a segfault in ecl_min
Program received
On Fri, Jul 13, 2012 at 10:26 AM, Eric Marsden eric.mars...@free.fr wrote:
jg You cannot build with gengc and multithreaded ECL. Did you deactivate
jg threads?
My bad; I did not realize these were incompatible (perhaps could be
added to configure --help output?).
In principle it
On Mon, Jul 16, 2012 at 1:02 PM, Pascal J. Bourguignon
p...@informatimago.com wrote:
clisp has a true interpreter, and it will macroexpand several times the
same macro.
I think it was either Bruno Haible or Sam Steingold who explained to me
once that the trick in CLISP is not to interpret,
On Mon, Jul 16, 2012 at 11:07 PM, e...@ecky.fr wrote:
@Juan : Well now that is an interesting remark, as what I read (unless I'm
mixing something up) I thought that ECL uses several steps to produce
executable code (see http://ecls.sourceforge.net/ecldev/devel_5.html) and
naively I figured
On Tue, Jul 17, 2012 at 1:25 PM, Mark Cox markco...@gmail.com wrote:
I am having a problem where the ecl process enters the sleeping state upon
exiting. It does not do it every time.
There was a hang in the garbage collector because ECL closed down all
signals before exiting. I have removed
On Sun, Jul 22, 2012 at 4:12 PM, H Durer h.due...@gmail.com wrote:
So, should we remove the stuff added by ql:add-to-init-file from our
~/.eclrc and just add a (require :ecl-quicklisp)?
I actually never used ql:add-to-init-file. What is it?
Juanjo
--
Instituto de Física Fundamental, CSIC
On Sun, Jul 22, 2012 at 6:32 PM, H Durer h.due...@gmail.com wrote:
| ;;; The following lines added by ql:add-to-init-file:
| #-quicklisp
| (let ((quicklisp-init (merge-pathnames quicklisp/setup.lisp
| (user-homedir-pathname
| (when (probe-file
On Sun, Jul 22, 2012 at 7:45 PM, H Durer h.due...@gmail.com wrote:
After I compiled my own version of ecl, I sudo make install'ed it to
/usr/local/...
.. the patch took some time to get into the repo. My apologies for posting
so fast.
Juanjo
--
Instituto de Física Fundamental, CSIC
c/
On Sun, Jul 22, 2012 at 8:13 PM, Juan Jose Garcia-Ripoll
juanjose.garciarip...@gmail.com wrote:
In short: it is not compulsory, it allows me to plug in customizations
such as enforcing monolithic libraries, it also allows to plug in
additional steps that quicklisp need not consider
On Sun, Jul 22, 2012 at 8:48 PM, Pascal J. Bourguignon
p...@informatimago.com wrote:
People will blog about how they used (require :ecl-quicklisp) in ecl,
newbies
will complain that it doesn't work (on their implementation), then may
blog about the social problems of #lisp or cll…
By that
On Tue, Jul 17, 2012 at 1:25 PM, Mark Cox markco...@gmail.com wrote:
I am having a problem where the ecl process enters the sleeping state upon
exiting. It does not do it every time.
As explained on a separate email this is due to a race condition in the
garbage collector. Unfortunately I do
Would it be possible to distinguish between test failure and build failure?
That would set up priorities straight. Many of the libraries seem to have
failed tests also on other ports.
Juanjo
--
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
On Tue, Jul 24, 2012 at 8:52 PM, Steve Graham jsgraha...@yahoo.com wrote:
Where can one get an executable for Windows?
I am working on that.
Juanjo
--
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com
I just uploaded a Windows build, but I only have a free antivirus. It would
be very appreciated if someone who runs Windows regularly could check it.
On Tue, Jul 24, 2012 at 10:07 PM, Juan Jose Garcia-Ripoll
juanjose.garciarip...@gmail.com wrote:
On Tue, Jul 24, 2012 at 9:31 PM, Steve Graham
On Sun, Jul 29, 2012 at 8:02 PM, Pascal Costanza p...@p-cos.net wrote:
I just tried to configure ECL on Mac OS X 10.8, and got the error message
C compiler cannot create executable.
Is this a known issue?
I don't know. Is 10.8 out already? I presume it also requires a different
set of
This is a probably simpler solution:
http://stackoverflow.com/questions/11654702/osx-10-8-xcode-4-4-make-and-gcc-gone-from-environment
On Sun, Jul 29, 2012 at 9:52 PM, Sylvain Ageneau agen...@gmail.com wrote:
Hi,
In my case I had to reinstall Xcode + command line tools after upgrading
to
On Thu, Aug 2, 2012 at 2:17 PM, Stanislav Frolov frolosof...@gmail.comwrote:
It there a problem with GMT detection?
Could it be rather a problem with summer time? ECL relies on the C library
for the timezone and the summer time (see src/c/mislib.lsp) and it seems to
work for Spain
On Thu, Aug 2, 2012 at 4:33 PM, Jean-Pierre Flori jpfl...@gmail.com wrote:
I tried to directly submit a bug on sf.net but can not devise how to
do that (I created an account and went to the bugs page of the ECL
project but there was no obvious button to click on).
If you go to
On Fri, Aug 3, 2012 at 7:18 AM, Stanislav Frolov frolosof...@gmail.comwrote:
Looks like localtime returns 3 hours instead of 4.
And datetimes for current time_t:
[...]Difference between gmtime and localtime is 4 hours. Bug? Feature?
Obviously a bug in the C library or a misconfiguration from
Bug filed https://sourceforge.net/p/ecls/bugs/197/
Juanjo
--
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com
--
Live Security Virtual Conference
On Fri, Aug 3, 2012 at 9:42 AM, Stanislav Frolov frolosof...@gmail.comwrote:
Right. But In 1970 Moscow time was UTC+3 and now it's UTC+4. So localtime
returns correct time for 1970 and for 2012. In mislib we calculate timezone
for 1970 that wrong, isn't it?
I see... The timezone computation
On Fri, Aug 3, 2012 at 10:22 AM, Jean-Pierre Flori jpfl...@gmail.comwrote:
As I still seems unable to directly submit the patch on sf.net (no
button yet although I'm logged in, just View stats and some Groups
and Searches and Help buttons on the left panel... maybe some openid
It feels like a deja-vu
https://sourceforge.net/mailarchive/forum.php?thread_name=CANejTzqFLY2uR5YTWJHsxQkkJ2h3oAW4jW2WNVAHefMEU3FpPg%40mail.gmail.comforum_name=ecls-list
On Tue, Jul 31, 2012 at 5:03 PM, Antonio Bonifati
antonio.bonif...@gmail.com wrote:
I need to test your patience again,
Just wanted to warn you that I will be away without computer or internet
access (well, perhaps just email) for a week or so. Feel free to submit bug
reports, though, for they will be read when I come back :-)
Best
Juanjo
--
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006
On Tue, Aug 7, 2012 at 7:22 PM, Jean-Pierre Flori jpfl...@gmail.com wrote:
Still working on building Sage on Cygwin, we realized that the
produced ECL shared library built on Cygwin is called ecl.dll.
While this should not be problematic, we also have another file called
ecl.dll in Sage which
On Mon, Aug 13, 2012 at 6:24 PM, bob bobat...@gmail.com wrote:
/usr/include/octave-3.2.4/octave/lo-mappers.h:33:26: error: expected ‘)’
before ‘’ token
All have the same error occurring in different header files. I have tried
to compile pure C++ code containing oct.h with the same compiler
On Sun, Aug 12, 2012 at 6:22 PM, bob bobat...@gmail.com wrote:
Its solved now.. just put a .cc file as the intermediate :c-file.
Hmmm, this should probably be done by ECL itself.
Juanjo
--
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
On Wed, Aug 8, 2012 at 7:38 PM, Jerry James loganje...@gmail.com wrote:
On Wed, Aug 8, 2012 at 11:10 AM, Jerry James loganje...@gmail.com wrote:
So SI:WAIT-FOR-ALL-PROCESSES needs to accept a :process keyword
argument when ECL_THREADS is defined.
How does the attached patch look?
Fine!
On Sun, Aug 5, 2012 at 3:55 PM, Dima Pasechnik dimpase+...@gmail.comwrote:
This was mentioned last Dec on sage-devel, that on Cygwin ECL call
(directory )
sometimes can return NIL. This breaks ASDF scripts, etc.
(this is a symptom of a bigger problem, certainly).
I am unable to reproduce
On Fri, Aug 17, 2012 at 12:46 PM, Jean-Pierre Flori jpfl...@gmail.comwrote:
On Cygwin at least, when the linker is given the -lecl flag, it will
look for files in PATH in a certain order.
First libecl.dll.a, then ..., then cygecl.dll, then ... then ecl.dll
IIRC [...]
So if we only have
On Thu, Aug 9, 2012 at 1:31 PM, Simon Ortiz o.si...@gmail.com wrote:
Now, I would like to achieve the same using CFFI instead of FFI. Using
CFFI I tried several approaches, but I wasn't successful. Basically, I
compiled the Lisp code into a static library, but the produced library
seems not
I am a bit lost with the mingw64 toolchain, with so may different tarballs
and options. I must admit I have never managed to get it up and running.
What do I have to do to use it?
On Sun, Aug 19, 2012 at 11:56 AM, Gabriel Dos Reis
g...@integrable-solutions.net wrote:
* fresh check out of CVS
On Sun, Aug 19, 2012 at 11:56 AM, Gabriel Dos Reis
g...@integrable-solutions.net wrote:
After modifying c/symbols_list.h and c/symbols_list2.h to include
a definition for :WIN64,
I uploaded a fix for this
the build failed again because
the function ecl_make_ulong_long was mistyped as
On Sun, Aug 19, 2012 at 9:14 PM, Gabriel Dos Reis
g...@integrable-solutions.net wrote:
No, it wasn't. Browse to the source code for number.d
http://ecls.cvs.sourceforge.net/viewvc/ecls/ecl/src/c/number.d?revision=1.102view=markup
Have a look at lines 308, 326, and 394.
Ok, it seems my
On Sun, Aug 19, 2012 at 11:56 AM, Gabriel Dos Reis
g...@integrable-solutions.net wrote:
Unfortunately, it failed to execute properly, apparently for the
same reasons as 2 years ago.
There were not so many things to fix after all:
* the typo you found
* missing :win64 keyword
* typo in
On Sun, Aug 19, 2012 at 10:46 PM, Juan Jose Garcia-Ripoll
juanjose.garciarip...@gmail.com wrote:
With all these, ECL seems to build fine on mingw32.
I should have chosen my words with greater care :-) fine is not
appropriate, for there are some regressions compared to other ports when
running
On Sun, Jul 29, 2012 at 11:31 PM, Сергій Яковин syako...@gmail.com wrote:
I have just tried to build 64 bit version of ECL on Windows 7 x64 using
MSVC 10, but got a crash:
;;; Loading lsp/format.lsp
;;; About to load cmp/load.lsp
Condition of type: SEGMENTATION-VIOLATION
Detected access to
On Tue, Aug 14, 2012 at 10:19 AM, Yuito Murase themamedaif...@gmail.comwrote:
I tried to load ieee-floats using quicklisp, but failed with massage
below. (sorry for some words being Japanese)
According to error message, external symbol, _ecl_make_ulong_Long
seems to be undefined.
Any
On Mon, Aug 20, 2012 at 4:35 PM, Jean-Pierre Flori jpfl...@gmail.comwrote:
I found out that some header files in the src/h directory are
executable after cloning ECL git repo.
This must have been cygwin's git rsync, when importing the changes I made
for windows. I will correct that
Juanjo
On Mon, Aug 20, 2012 at 4:52 PM, Tara Lorenz t...@terralonza.de wrote:
while trying to get cl-num-utils running on ECL i encountered a problem
with let-plus in combination with define-modify-macro.
Informal description: When define-modify-macro defines a three-argument
modifier, ECL
On Mon, Aug 20, 2012 at 4:52 PM, Tara Lorenz t...@terralonza.de wrote:
while trying to get cl-num-utils running on ECL i encountered a problem
with let-plus in combination with define-modify-macro.
My previous diagnostic was wrong. There was an ancient piece of code using
NREVERSE on a
Simon, I am not able of reproducing your problems. See the files here
https://docs.google.com/folder/d/0B2brKGbNKZYDY0hiMXUyUy1aMUk/edit
and how I build and run them
bash-3.2$ cd build/ship-test/
bash-3.2$ ls
cffi-test ffi-test
bash-3.2$ (cd cffi-test; ls -l; make ./test_exe)
total 11904
On Fri, Aug 17, 2012 at 2:21 PM, bob bobat...@gmail.com wrote:
I am trying to build ecl (both stable from git) on MinGW (latest stable)
with gcc 4.7.0. But I am presented with different errors during the make
stage. Using the stable release (12.7.1) I hit upon make error while
building
On Sat, Aug 25, 2012 at 8:14 PM, bob bobat...@gmail.com wrote:
I git pull'ed the latest ECL but these configure flags didn't work:
--with-sse=yes --with-cxx=yes. It had problems compiling libffi
I upgraded mingw using mingw-get update and mingw-get upgrade and ECL does
compile with libffi and
On Sun, Aug 26, 2012 at 2:57 PM, Simon Ortiz o.si...@gmail.com wrote:
Condition of type: SIMPLE-ERROR
FIND-FOREIGN-SYMBOL: Could not load foreign symbol sin from module
:DEFAULT (Error: NIL)
Sorry, there must have been a glitch in my system -- I was building your
example with an ECL that
On Sat, Aug 25, 2012 at 2:08 AM, Mark Cox markco...@gmail.com wrote:
G'day,
The hyperspec [1] states that DELETE-FILE should signal an error of type
FILE-ERROR if it cannot succeed in deleting a file. ECL currently signals a
SIMPLE-ERROR.
[...] there are other functions in that file that
On Sun, Sep 9, 2012 at 7:11 PM, Pascal Costanza p...@p-cos.net wrote:
It seems this bug report has not been acknowledged. Did I report it
correctly to this mailing list, or is there a better way to report it?
No, the bug has been received, but I have been simply abroad travelling,
without
On Sun, Sep 2, 2012 at 4:38 PM, Pascal Costanza p...@p-cos.net wrote:
The fact that ext-generic-function-method-class is compiled seems to be
important here, and it seems that the dispatch is fixed to the methods
known at the time it is compiled.
I still do not understand what is happening,
Sorry, forget the noise: I took your ext-generic-function... name for a
typo of ext:generic-function..., but obviously it is not
Juanjo
--
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com
On Sun, Sep 2, 2012 at 4:38 PM, Pascal Costanza p...@p-cos.net wrote:
The fact that ext-generic-function-method-class is compiled seems to be
important here, and it seems that the dispatch is fixed to the methods
known at the time it is compiled.
Not exactly, but compilation was a problem.
On Sun, Sep 16, 2012 at 10:18 PM, Juan Jose Garcia-Ripoll
juanjose.garciarip...@gmail.com wrote:
Not exactly, but compilation was a problem. ECL hardcodes the name of
certain functions and they can be directly called from C code, instead of
dispatching via the symbol. Unfortunately, some
This is was latest commit
commit ce653d889d818ba8b6709812a83d3246762030a6
Do not hard-code the name of generic function accessors in compiled code
seems it did not get all the way to SF. I have re-uploaded it. Thanks for
the warning.
Juanjo
--
Instituto de Física Fundamental, CSIC
c/
You have to load the C compiler first with (require :cmp) in order to be
able to produce DLLs or binary files. Did you do that?
Juanjo
--
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com
On Fri, Sep 21, 2012 at 12:14 AM, Matthew Mondor
mm_li...@pulsar-zone.netwrote:
So would a C profiler currently be my best resort?
Right now I would say it is. Do not be scared: function names are easily
identified and you will get insight on functions that not even the Lisp
layer can profile,
Could you people have a look at the bug report and my answer? I believe I
am right, though SBCL is reporting a different order and this (once more)
confuses users
*bugs:206 string-lessp does not behave as string ignoring case*
*Status:* open *Created:* Thu Sep 20, 2012 09:43 PM UTC by Rafael
On Wed, Aug 29, 2012 at 1:20 PM, Mark Cox markco...@gmail.com wrote:
The issue relevant to ECL is that commit 469c8a5 introduces a problem that
results in one of the following errors:
I do not see the problem with this patch:
commit 469c8a520f8e3a053775597f4d82a32e656161bf
Author: Juan Jose
On Mon, Sep 24, 2012 at 9:21 PM, Juan Jose Garcia-Ripoll
juanjose.garciarip...@gmail.com wrote:
I do not see the problem with this patch:
I now see what you mean. There are a couple of mistakes in the routines
that store and retrieve the errors. I have made some simplifications
On Fri, Sep 21, 2012 at 12:11 AM, Matthew Mondor
mm_li...@pulsar-zone.netwrote:
I noticed that functions such as WRITE-SEQUENCE will signal a condition
of type SIMPLE-ERROR if EPIPE is returned when writing. This is
admitedly in non-standard territory, as only TYPE-ERROR is specified
for
On Wed, Sep 26, 2012 at 2:34 PM, Stanislav Frolov frolosof...@gmail.comwrote:
#includeecl/ecl.h
#includepthread.h
Could you try exchanging these two, or removing pthread.h completely? ECL
is including gc.h which overrides several thread functions, including
creation and destruction -- not
On Thu, Sep 27, 2012 at 1:13 PM, Stanislav Frolov frolosof...@gmail.comwrote:
There is a bug in error processing, I think. But root of my problem is in
FElibc_error:pthread_getspecific() failed.. I don't understand ECL's
threading but looks like you call pthread_getspecific (ecl_process_env)
On Fri, Sep 21, 2012 at 12:11 AM, Matthew Mondor
mm_li...@pulsar-zone.netwrote:
I noticed that functions such as WRITE-SEQUENCE will signal a condition
of type SIMPLE-ERROR if EPIPE is returned when writing. This is
admitedly in non-standard territory, as only TYPE-ERROR is specified
for
On Mon, Oct 1, 2012 at 11:23 PM, Juan Jose Garcia-Ripoll
juanjose.garciarip...@gmail.com wrote:
I am going to upload some patches that fix this. ECL will signal
FILE-ERROR and STREAM-ERROR conditions when stream operations fail due to
the C library.
It would be nice if you could provide
On Sun, Oct 21, 2012 at 11:17 AM, Simon Ortiz o.si...@gmail.com wrote:
I was wondering of you figured out the cause of the problem. Did you try
using an ECL without DFFI?
Sorry, no answer yet. I mark your email -- it dropped from the todo list by
mistake.
Juanjo
--
Instituto de Física
On Tue, Oct 23, 2012 at 5:54 PM, Matthew Mondor mm_li...@pulsar-zone.netwrote:
So I today had some time to do a small test and the following function
indeed seems to work better in general.
(defimplementation arglist (name)
(let ((arglist (ext:get-annotation name :lambda-list nil)))
On Tue, Oct 30, 2012 at 2:41 AM, Matthew Mondor mm_li...@pulsar-zone.netwrote:
With in swank-ecl.lisp:
(defimplementation arglist (name)
(let ((arglist (ext:get-annotation name :lambda-list nil)))
(if (null arglist)
:not-available
arglist)))
And your latest changes,
On Wed, Oct 31, 2012 at 1:02 PM, Juan Jose Garcia-Ripoll
juanjose.garciarip...@gmail.com wrote:
On Wed, Oct 31, 2012 at 8:07 AM, Chris Bagley chris.bag...@gmail.comwrote:
Have I done this incorrectly or is there something else I have missed?
Any advice would be amazingly helpful!
Try
On Sat, Aug 18, 2012 at 7:09 AM, Simon Ortiz o.si...@gmail.com wrote:
I would like to reproduce the functionality of ffi-test in cffi-test.
That is, I would like to produce a static library that uses CFFI to
access foreign functions.
I'm sure I'm missing something in cffi-test. I've tried
I am fwd this to the mailing list because I do not want to repeat these
answers or restart this discussion for every user that might find such
problems. This is what the mailing list archives are fore
On Fri, Nov 2, 2012 at 3:46 PM, Peter Enerccio enerc...@gmail.com wrote:
Well, I can get my
On Sat, Nov 3, 2012 at 10:36 PM, Eric Marsden eric.mars...@free.fr wrote:
This is pretty, and hopefully useful to you for tracking performance
regressions.
Indeed, it is important to realize whether there were regressions and a
very important one was just spotted in clos/complex-method
Thanks for the test case. I will look into that along the week,
On Mon, Nov 5, 2012 at 9:24 AM, Peter Enerccio enerc...@gmail.com wrote:
Should be the latest:
ECL (Embeddable Common-Lisp) 12.7.1
(git:99b1e54eb07373bda77110761c5143d2ad614e6c)
Copyright (C) 1984 Taiichi Yuasa and Masami
On Tue, Nov 6, 2012 at 12:31 AM, Peter Enerccio enerc...@gmail.com wrote:
I am using this kind of macro:
#define DEFUN( name, argcount, fpointer) do { \
cl_def_c_function(c_string_to_object(name), \
(cl_objectfn_fixed)fpointer, \
argcount); \
} while (0)
On Wed, Oct 24, 2012 at 11:14 PM, Matthew Mondor
mm_li...@pulsar-zone.netwrote:
It seems that the function above is called twice; I suppose that one of
both #-ecl-bytecmp should be #+ecl-bytecmp?
Thanks for reporting. I just uploaded a fix to slime
Juanjo
--
Instituto de Física
On Wed, Oct 31, 2012 at 1:17 AM, Matthias Hölzl t...@xantira.com wrote:
when compiling the Snark theorem prover (
http://www.ai.sri.com/~stickel/snark.html, or with ASDF system
definition at https://github.com/hoelzl/Snark), ECL (version 12.7.1 on
OSX 10.8.2, installed using homebrew) fails
On Sun, Nov 11, 2012 at 10:26 PM, Peter Enerccio enerc...@gmail.com wrote:
I know I can redefine the builtin functions by simply redefining them, but
is that fool proof?
Yes it is possible, and the right approach would be more useful than just
for ECL. Creation of domain specific languages is
Dear Peter,
I would appreciate if, instead of directing the questions to my private
email, you would send them to the mailing list for future use of other
people. This is not a private consulting office and if I answer some
question to you, it is most likely that I will have to answer the same
On Wed, Nov 14, 2012 at 2:42 AM, Matthew Mondor mm_li...@pulsar-zone.netwrote:
I forgot to write about another alternative using bytecode though... if
all code compiles to bytecode only, and the bytecode interpreter
permits to impose all the limits you need
The problem is not the bytecodes,
On Wed, Nov 14, 2012 at 8:30 AM, Ralph Möritz ralph.moer...@outlook.comwrote:
I don't think 64k is such a severe limitation; for code using very
long string literals couldn't we use embedded resources instead?
64k is pretty small for many libraries out there that build tons of data.
https://github.com/juanjosegarciaripoll/gtd-drive
It uses Safari as a front-end and works with toot and cl-who. The patches
for toot are available here
https://github.com/juanjosegarciaripoll/toot
Feedback welcome :-)
Juanjo
--
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid
I am attaching a patch for CFFI that allows ECL to build code that does not
use dlopen() but rather uses the function symbols as they are know to the C
compiler. This is best explained in the code comments, which I reproduce
below, and in the example.
The patched CFFI is available at my github
of the OpenGLES functions/extensions that are not in the pure
OpenGL spec but I think it shouldn't be too much trouble.
Thanks again!
On Nov 22, 2012, at 5:41 PM, Juan Jose Garcia-Ripoll
juanjose.garciarip...@gmail.com wrote:
On Thu, Nov 22, 2012 at 8:27 PM, Sylvain Ageneau agen
On Sun, Nov 18, 2012 at 11:38 PM, Pascal Costanza p...@p-cos.net wrote:
- It seems that slot-unbound is not correctly handled. Here is a test case:
Thanks for reporting this, Pascal. I just fixed it.
Is the other issue about slot optimization a problem? Note that I would
rather have _no_ code
On Tue, Nov 27, 2012 at 10:28 AM, Jean-Pierre Flori jpfl...@gmail.comwrote:
Oh, of course there are older releases!
I meant something between 12.2.1 and 12.7.1 if that exists, but I
think I just understood your numbering scheme...
For a couple of years I have been using date schemes. It
Hi Stas,
i am creating tickets for both bugs. One is half-solved, but I need this to
keep myself posted on what I do -- please apologize if I forget to report
the bug solution to the mailing list. It seems most efficient for me to
proceed this way.
Juanjo
On Mon, Nov 26, 2012 at 12:58 AM, Stas
.
But it is not something I want to impose on you all. If someone does not
want to deal with SF, he/she can report bugs here.
On Tue, Nov 27, 2012 at 5:15 PM, Stas Boukarev stass...@gmail.com wrote:
Juan Jose Garcia-Ripoll juanjose.garciarip...@gmail.com writes:
Hi Stas,
i am creating tickets for both bugs. One
On Tue, Nov 27, 2012 at 5:14 PM, Jean-Pierre Flori jpfl...@gmail.comwrote:
PS: just saw your email, I'll redirect this to the bug tracker.
Thanks for helping me track this issue. I think it can be considered
solved, but please report in either case, so that I can close the bug
report.
Juanjo
On Tue, Nov 27, 2012 at 4:01 PM, Anton Vodonosov avodono...@yandex.ruwrote:
In other words the version advertised on ECL web site as the release should
be the best known of existing versions.
Let me clarify the criterion I have used so far for making releases:
* Sufficient number of fixes.
*
On Wed, Nov 28, 2012 at 5:03 AM, Faré fah...@gmail.com wrote:
A shorter README explaining that asdf.lisp is from asdf (with URL)
and the other files from asdf-bundle (with URL), would also be nice.
http://common-lisp.net/project/asdf/
On Wed, Oct 3, 2012 at 12:54 AM, Matthew Mondor mm_li...@pulsar-zone.netwrote:
Unfortunately there's still no way to differenciate a timeout error
(EAGAIN) from a broken pipe one (EPIPE) when using STREAM-ERROR.
However, because of the design of the system (in HTTP we first read a
request
On Mon, Nov 26, 2012 at 11:06 PM, Anton Vodonosov avodono...@yandex.ruwrote:
I see that according reports situation is not always improving with
commits.
It turns out that the code I used for sorting reports was broken! Now I
fixed the code that separates dates from file names and the
401 - 500 of 633 matches
Mail list logo
