List for users and developers of ECL Common Lisp

Pascal J. Bourguignon | 5 Mar 2012 21:27

decoding error on fas file compiled from utf-8 files.

Hello,

This evening I pulled from git the latest release of ecl, compiled it,
and now I'm trying to boot it, unfortunately it fails after compiling my
files (ql:quickload :com.informatimago.common-lisp.cesarum) with:

…
[package com.informatimago.common-lisp.cesarum.file]
[package com.informatimago.common-lisp.cesarum.peek-stream].
[package com.informatimago.common-lisp.cesarum.cache].
..................................................
[package com.informatimago.common-lisp.cesarum.float-binio].
An error occurred during initialization:
decoding error on stream #<sequence-input stream 000000000348b3c0> (:external-format (:utf-8 :lf)):
the octet sequence (233 112 117) cannot be decoded..
[pjb <at> kuiper :0 lisp]$ echo $?
1

It looks like the error is produced after compiling (since quicklisp
prints the name of the last package defined).

All my files are stored in UTF-8, and none of them contain that octet
sequence (some of them contain the characters "épu" to which this octet
sequence decodes in iso-8859-1.

I tried with and without 
(setf ext:*default-external-format* '(:utf-8 :lf))
in ~/.eclrc.lisp.

I tried deleting the .fas and recompiling everything.  I always get

(Continue reading)

headers

Dima Pasechnik | 6 Mar 2012 04:36

Fwd: Sage 4.8 build failed: Error: Failed to build Maxima as an ECL library.

Hi,
this is about a problem with building Maxima as an ECL library using
ECL 11.1.1 on a rather old Redhat system with gcc 4.1.2.
Is ECL tested on that kind of systems?
Does this issue look familiar?
Thanks,
Dima

The complete sage-devel discussion is here:
https://groups.google.com/group/sage-devel/browse_thread/thread/33580987797084dd

---------- Forwarded message ----------
From: Dima Pasechnik <dimpase <at> gmail.com>
Date: 6 March 2012 11:31
Subject: Re: Sage 4.8 build failed: Error: Failed to build Maxima as
an ECL library.
To: sage-devel <sage-devel <at> googlegroups.com>
Cc: Dima Pasechnik <dimpase <at> gmail.com>

Well, I don't see what goes wrong there (except that the compiler you
use is so outdated that surely Sage is not tested with it --- version
alone does not say this, as the same version on MacOSX is a different
beast). You mention "people talking about this error on Lion".
Any links to any posts/forums on this?

I guess it is an ECL bug (or feature :)). I'll cc'in this to the ECL
list, just in case they are aware of these issues.

On Mar 5, 1:26 pm, "syd.lavas... <at> gmail.com" <syd.lavas... <at> gmail.com>
wrote:

(Continue reading)

headers

Raymond Toy | 6 Mar 2012 18:15

Re: Ecl version of with-float-traps-masked and vector-sap?

On Wed, Feb 8, 2012 at 6:01 AM, Juan Jose Garcia-Ripoll

I'm looking at trying to get matlisp to run with ecl, and now that
someone has contributed an ffi using cffi, I thought getting matlisp
working with ecl would be nice.

That would indeed be very nice.

FYI: matlisp will now compile and run with ecl. I only did a couple of simple tests with 32-bit ecl 11.1.x, but they worked. I still need to incorporate some changes like with-float-traps-masked, but it looks like the basic functionality is there. More testing is needed though.

Ray

------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d

_______________________________________________
Ecls-list mailing list
Ecls-list@...
https://lists.sourceforge.net/lists/listinfo/ecls-list

headers

Juan Jose Garcia-Ripoll | 6 Mar 2012 22:48

Re: Ecl version of with-float-traps-masked and vector-sap?

On Tue, Mar 6, 2012 at 12:15 PM, Raymond Toy <toy.raymond <at> gmail.com> wrote:

On Wed, Feb 8, 2012 at 6:01 AM, Juan Jose Garcia-Ripoll

I'm looking at trying to get matlisp to run with ecl, and now that

someone has contributed an ffi using cffi, I thought getting matlisp
working with ecl would be nice.

That would indeed be very nice.

FYI: matlisp will now compile and run with ecl. I only did a couple of simple tests with 32-bit ecl 11.1.x, but they worked. I still need to incorporate some changes like with-float-traps-masked, but it looks like the basic functionality is there. More testing is needed though.

Great, thanks for reporting!

Juanjo

--
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com

------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d

_______________________________________________
Ecls-list mailing list
Ecls-list@...
https://lists.sourceforge.net/lists/listinfo/ecls-list

headers

Dima Pasechnik | 15 Mar 2012 11:46

a bug (or feature?) in (directory) implementation

This was posted on sage-devel:
https://groups.google.com/d/msg/sage-devel/M1gJh3lwhN0/G04AygcbreUJ

Below I cut the relevant part, which appears to demonstrate a problem
in calls like
(directory '/A/B/C/*')
 which boils down to the fact that is the top part (say '/A/B/') of
the path contains
an unreadable to the user directory, then the (directory '/A/B/C/*')
returns NIL,
even though it's perfectly readable to the user (tested on 11.1.1 and
on 12.**, too)

-----------------------------------------------------------------------
The problem is that my home directory is:

/files3/home/sahosse/

but I only have execution permission to the directory "home":

[sahosse <at> msx0 ~]$ ls -ld /files3/home
drwx--x--x 192 root msstaff 4096 Feb 15 13:23 /files3/home
[sahosse <at> msx0 ~]$ ls -l /files3/home
ls: /files3/home: Permission denied

Off course I have access to my home directory but (directory) function
of ecl always return NIL as long as "home" appears in the path of the
requested directory, which means it fails on my entire home directory:

[sahosse <at> msx0 ~]$ ~/sage/local/bin/ecl
ECL (Embeddable Common-Lisp) 11.1.1 (git:UNKNOWN)
Copyright (C) 1984 Taiichi Yuasa and Masami Hagiya
Copyright (C) 1993 Giuseppe Attardi
Copyright (C) 2000 Juan J. Garcia-Ripoll
ECL is free software, and you are welcome to redistribute it
under certain conditions; see file 'Copyright' for details.
Type :h for Help.
Top level.
> (directory "/*")

(#P"/.autorelabel" #P"/.autofsck" #P"/.rnd")
> (directory "/files3/*")

NIL
> (directory "/files3/home/sahosse/*")

NIL

--------------
Best,
Dima

------------------------------------------------------------------------------
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure

headers

Juan Jose Garcia-Ripoll | 18 Mar 2012 13:13

Re: Bug: Incorrect closure over call-next-method in 12.2.1

Thanks for reporting the bug, Pascal. It will be fixed in git/CVS today.

Juanjo

On Sat, Feb 25, 2012 at 9:02 PM, Pascal Costanza <pc-99OXBJU6cIpeoWH0uzbU5w@public.gmane.org> wrote:

Hi,

Consider the following test case:

(defclass c () ())
(defclass d (c) ())

(defgeneric test (obj))

(defmethod test ((obj c)) 42)

(defmethod test ((obj d))
(flet ((do-it ()
(let ((next-value (call-next-method)))
next-value)))
(lambda () (do-it))))

> (test (make-instance 'd))

#<bytecompiled-closure #<bytecompiled-function 000000010db4d280>>
> (funcall *)

Condition of type: SIMPLE-ERROR
No next method.

Available restarts:

1. (RESTART-TOPLEVEL) Go back to Top-Level REPL.

Similar code worked before in 11.1.1. (Instead of the error, the last funcall should return 42.)

Best,
Pascal

--
Pascal Costanza

------------------------------------------------------------------------------
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/
_______________________________________________
Ecls-list mailing list
Ecls-list-5NWGOfrQmnetEtDZOKyKiw@public.gmane.orgrge.net
https://lists.sourceforge.net/lists/listinfo/ecls-list

--
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com

------------------------------------------------------------------------------
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure

_______________________________________________
Ecls-list mailing list
Ecls-list@...
https://lists.sourceforge.net/lists/listinfo/ecls-list

headers

Juan Jose Garcia-Ripoll | 18 Mar 2012 13:15

Re: Calling cl_boot() from Qt plugin (Win XP, Vista only)

On Sun, Feb 26, 2012 at 10:06 PM, Polos Ruetz <polos.ruetz <at> gmail.com> wrote:

2012/2/26, Polos Ruetz <polos.ruetz-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>:
> But it crashes on Windows XP, Vista if called from a plugin.
> Simply calling GC_init() crashes too.

Solved!

Ingredients for the recipe:
- QTimer::singleShot()
- cl_boot() inside a Qt slot
- shoot to boot!

Hi Polos, did you get any insight on why this precise order? Is it due to something ECL does wrong?

Juanjo

--
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com

------------------------------------------------------------------------------
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure

_______________________________________________
Ecls-list mailing list
Ecls-list@...
https://lists.sourceforge.net/lists/listinfo/ecls-list

headers

Juan Jose Garcia-Ripoll | 18 Mar 2012 16:09

Re: Two little unicode fixes.

On Mon, Feb 20, 2012 at 12:52 AM, Andy Hefner <ahefner-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:

I've tracked down the cause of some encoding-related problems
encountered while using string literals containing special characters
(squared, cubed, and degree symbols) in source files encoded as
latin-1. I specify :external-format :latin-1 to compile-file, but the
resulting objects/fasls error at load-time during their
initialization, the error indicating an attempt to decode latin-1
characters as UTF-8. A peek at the .data file confirmed the characters
were not UTF-8 encoded in the compiler output.

Thanks for tracking this down to the root. I believe this is related to a different bug report in the mailing list later this month. Unfortunately I had not completed the fixes to the locks/condition-variable codes until today. I am uploading this evening a slightly different patch that covers this bug and the other one you found.

Juanjo

--
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com

------------------------------------------------------------------------------
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure

_______________________________________________
Ecls-list mailing list
Ecls-list@...
https://lists.sourceforge.net/lists/listinfo/ecls-list

headers

Juan Jose Garcia-Ripoll | 18 Mar 2012 16:16

Re: decoding error on fas file compiled from utf-8 files.

On Mon, Mar 5, 2012 at 9:27 PM, Pascal J. Bourguignon <pjb <at> informatimago.com> wrote:

This evening I pulled from git the latest release of ecl, compiled it,
and now I'm trying to boot it, unfortunately it fails after compiling my
files (ql:quickload :com.informatimago.common-lisp.cesarum) with:[...]
[package com.informatimago.common-lisp.cesarum.float-binio].
An error occurred during initialization:
decoding error on stream #<sequence-input stream 000000000348b3c0> (:external-format (:utf-8 :lf)): the octet sequence (233 112 117) cannot be decoded..

This error will be solved by a patch I will upload later on. The problem was already pointed out in the mailing list: the compiler was producing source code in latin-1 encoding while the binary loader expects constants in utf-8 encoding.

Juanjo

--
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com

------------------------------------------------------------------------------
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure

_______________________________________________
Ecls-list mailing list
Ecls-list@...
https://lists.sourceforge.net/lists/listinfo/ecls-list

headers

Juan Jose Garcia-Ripoll | 18 Mar 2012 17:05

Re: a bug (or feature?) in (directory) implementation

On Thu, Mar 15, 2012 at 11:46 AM, Dima Pasechnik <dimpase+ecl-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:

Below I cut the relevant part, which appears to demonstrate a problem in calls like (directory '/A/B/C/*') which boils down to the fact that is the top part (say '/A/B/') of the path contains an unreadable to the user directory, then the (directory '/A/B/C/*') returns NIL,
even though it's perfectly readable to the user (tested on 11.1.1 and on 12.**, too)

Fixed in local (see below). Will upload a patch later on.

$ ecl -norc

ECL (Embeddable Common-Lisp) 12.2.1 (git:f074dd441f7500f9fd96607443e27d6b6afe04c9)

ECL is free software, and you are welcome to redistribute it

under certain conditions; see file 'Copyright' for details.

Type :h for Help.

Top level in: #<process TOP-LEVEL>.

> (directory "~/foo/faa/*.*")

(#P"/Users/jjgarcia/foo/faa/lsp.txt")

$ ls ~/foo

ls: foo: Permission denied

--
Instituto de Física Fundamental, CSIC
c/ Serrano, 113b, Madrid 28006 (Spain)
http://juanjose.garciaripoll.googlepages.com

------------------------------------------------------------------------------
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure

_______________________________________________
Ecls-list mailing list
Ecls-list@...
https://lists.sourceforge.net/lists/listinfo/ecls-list

27	May 2013
50	April 2013
54	March 2013
18	February 2013
89	January 2013
135	December 2012
119	November 2012
61	October 2012
57	September 2012
97	August 2012
100	July 2012
69	June 2012
62	May 2012
40	April 2012
39	March 2012
136	February 2012
64	January 2012
90	December 2011
72	November 2011
13	October 2011
23	September 2011
92	August 2011
90	July 2011
38	June 2011
33	May 2011
32	April 2011
50	March 2011
164	February 2011
246	January 2011
55	December 2010
52	November 2010
84	October 2010
55	September 2010
93	August 2010
130	July 2010
144	June 2010
133	May 2010
120	April 2010
350	March 2010
324	February 2010
90	January 2010
75	December 2009
202	November 2009
143	October 2009
91	September 2009
203	August 2009
175	July 2009
133	June 2009
66	May 2009
46	April 2009
82	March 2009
98	February 2009
56	January 2009
94	December 2008
37	November 2008
130	October 2008
182	September 2008
258	August 2008
56	July 2008
82	June 2008
296	May 2008
178	April 2008
81	March 2008
74	February 2008
113	January 2008
72	December 2007
68	November 2007
53	October 2007
57	September 2007
26	August 2007
17	July 2007
42	June 2007
69	May 2007
50	April 2007
43	March 2007
86	February 2007
97	January 2007
60	December 2006
46	November 2006
133	October 2006
79	September 2006
72	August 2006
22	July 2006
66	June 2006
154	May 2006
46	April 2006
80	March 2006
45	February 2006
68	January 2006
40	December 2005
142	November 2005
136	October 2005
28	September 2005
131	August 2005
39	July 2005
108	June 2005
67	May 2005
46	April 2005
54	March 2005
115	February 2005
100	January 2005
35	December 2004
60	November 2004
34	October 2004
3	September 2004
28	August 2004
12	July 2004
24	June 2004
35	May 2004
15	April 2004
48	March 2004
15	February 2004
22	January 2004
54	December 2003
43	November 2003
39	October 2003
14	September 2003
11	August 2003
43	July 2003
44	June 2003
23	May 2003
19	April 2003
27	March 2003
1	February 2003

Mon	Tue	Wed	Thu	Fri	Sat	Sun

			1	2	3	4
5	6	7	8	9	10	11
12	13	14	15	16	17	18
19	20	21	22	23	24	25
26	27	28	29	30	31