July 2005 - slime-devel - mailman3.common-lisp.net

[slime-devel] Bug reading strings in REPL
by Edi Weitz 26 Jul '05

26 Jul '05

Unfortunately I have no time to investigate this but here is how to reproduce it (seen with LispWorks as well as with CMUCL, SLIME from CVS): At the REPL prompt type " (quote) and then RETURN. At that point I get an END-OF-FILE error. Cheers, Edi.

2 1

[slime-devel] Tiny patch for the SLIME inspector
by Matthias Koeppe 26 Jul '05

26 Jul '05

Hi, I am sending a tiny patch for the SLIME inspector. Because Emacs merges adjacent text properties of the same value, adjacent "action" buttons are glued together, so they are all highlighted when mouse is over one of them. (This can be seen, for instance, when one inspects an internal symbol.) The patch below fixes this bug by separating all "action" buttons by whitespace without the `highlight' text property. Cheers, Matthias 2005-07-17 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> * swank.lisp (inspect-for-emacs): Don't make whitespace surrounding :action buttons part of the highlighted region. --- swank.lisp.~1.309.~ 2005-06-26 20:38:06.000000000 +0200 +++ swank.lisp 2005-07-17 16:58:41.000000000 +0200 @@ -3342,9 +3343,11 @@ " to the package: " (:value ,package ,(package-name package)) ,@(if (eq :internal status) - `((:action " [export it]" + `(" " + (:action "[export it]" ,(lambda () (export symbol package))))) - (:action " [unintern it]" + " " + (:action "[unintern it]" ,(lambda () (unintern symbol package))) (:newline)) '("It is a non-interned symbol." (:newline))) @@ -3356,7 +3359,8 @@ (if (find-class symbol nil) `("It names the class " (:value ,(find-class symbol) ,(string symbol)) - (:action " [remove]" + " " + (:action "[remove]" ,(lambda () (setf (find-class symbol) nil))) (:newline))) ;; @@ -3477,7 +3481,8 @@ `((:value ,method ,(inspector-princ ;; drop the name of the GF (cdr (method-for-inspect-value method)))) - (:action " [remove method]" + " " + (:action "[remove method]" ,(let ((m method)) ; LOOP reassigns method (lambda () (remove-method gf m)))) -- Matthias Koeppe -- http://www.math.uni-magdeburg.de/~mkoeppe

2 1

[slime-devel] specifying charset of a file to compile
by frederic.gobry＠epfl.ch 26 Jul '05

26 Jul '05

Hi, I'm working on a simple web app whose source contains utf-8 characters, and I try to configure my environment so that everything plays nicely. What works: - when using emacs in utf-8 environment, I can properly use the REPL and compile functions with C-c C-c. They play nicely together. What does not work: - when I use C-c C-k, the functions seem to be compiled as if the source code was in latin-1 I think. How can I configure slime to behave similarly whether it compiles a file or a function? The only slime config I have is: (setq slime-net-coding-system 'utf-8-unix) I'm using: - sbcl 0.9.2 - slime CVS Thanks for your help, Frédéric

2 5

[slime-devel] Re: [Fix] Slime-1.2.1-Clisp-2.34-Win32 (XP)
by Hoehle, Joerg-Cyril 26 Jul '05

26 Jul '05

Hi, Luke Gorrie wonders: >"VAUCHER Laurent" <VAUCHER(a)fermat.fr> writes: >> 92c92 >> < (funcall (find-symbol "gETcURRENTpROCESSiD" :win32))) >> --- >> > (funcall (find-symbol "GetCurrentProcessId" :win32))) > >is this different between cygwin and win32 versions? should we check >for each in swank-clisp? No. Inverted case derives from the documentation on so-called MODERN packages via DEFPACKAGE or MAKE-PACKAGE in CLISP. Of course, there's no guarantee (i.e. documentation) that the package named "WIN32" uses modern mode. http://clisp.cons.org/impnotes/pack-dict.html#defpack Remember that this excursion to GetCurrentProcessId is only needed for clisp <2.34. Since CVS May 2005, you can use sys::process-id in all implementations of CLISP. BTW, I promised you a patch around the getpid issue, but that has to wait until my laptop comes out of repair. What about (and (find-package "WIN32") (read-from-string "WIN32:GetCurrentProcessId")) to not suffer from sLurdYCaPS? Regards, Jorg Hohle.

1 0

[slime-devel] sbcl and compiler note annotation
by Zach Beane 26 Jul '05

26 Jul '05

Recently, slime and SBCL have stopped collaborating to correctly place compiler notes in source buffers. For example, I compile this function with C-c C-c: (defun funk (x) (+ 1 'x)) The generated notes are stuck on the arglist of the form. If I use cmucl, for example, all the notes are correctly placed. What should I do? Zach

2 3

[slime-devel] [Fix] Slime-1.2.1-Clisp-2.34-Win32 (XP)
by VAUCHER Laurent 26 Jul '05

26 Jul '05

This is a small fix I made to swank-clisp.lisp. 92c92 < (funcall (find-symbol "gETcURRENTpROCESSiD" :win32))) --- > (funcall (find-symbol "GetCurrentProcessId" :win32))) Don't ask me why the case of symbols in the Win32 package is reversed... I mainly work with the Cygwin version of Clisp, but the native Win32 is 30% faster on one of my project. Laurent.

2 1

[slime-devel] inferior-lisp-program cannot contain spaces
by Ury Marshak 25 Jul '05

25 Jul '05

Looking at CVS HEAD's slime-inferior-lisp functions it seems that it's impossible to use paths containing spaces, since they will be split-string'd in several parts. Trying to run on Win32, where most programs have space-containing paths ("c:/program files/....") confirms the problem. Are there any fixes/workarounds? Thanks

3 2

[slime-devel] Re: [Sbcl-devel] Would McCLIM benefit from this pretty printer patch for CMUCL/SBCL?
by Paolo Amoroso 25 Jul '05

25 Jul '05

Matthias Koeppe <mkoeppe(a)merkur.math.uni-magdeburg.de> writes: > I have worked on a patch for the pretty printer of CMUCL and SBCL that > helps support the new presentation-like features of SLIME in > conjunction with the pretty printer. I probably can't comment on your patch. But you might be interested in some experimental work Gilbert Baumann did on the CMUCL pretty printer and McCLIM. Here are some screen shots: http://bauhh.dyndns.org:8000/mcclim/screenshots/pprint-1.png http://bauhh.dyndns.org:8000/mcclim/screenshots/pprint-2.png I think the code is here: http://bauhh.dyndns.org:8000/mcclim/pprint/ Paolo -- Lisp Propulsion Laboratory log - http://www.paoloamoroso.it/log

1 0

[slime-devel] SB-EXT:PACKAGE-LOCKED-ERROR
by Tim Johnson 25 Jul '05

25 Jul '05

Hello: Using cvs ver last update july 21. Inferior lisp is SBCL 0.9.0 When loading lisp file and typing M-x slime, I get the following (abbreviated) error message debugger invoked on a SB-EXT:PACKAGE-LOCKED-ERROR in thread 2427: Lock on package SB-VM violated when interning THREAD-OS-THREAD-SLOT. See also: The SBCL Manual, Node "Package Locks" ..... (SB-EXT:PACKAGE-LOCK-VIOLATION #<PACKAGE "SB-VM">) I'm new to lisp and slime, and relatively callow with emacs. I'm currently looking at http://www.sbcl.org/manual/Operations-Violating-Package-Locks.html#Operatio… but given my unfamiliarity with the topic, I'm currently unsure of a fix. Any help, advice or pointers to docs is welcome. Thanks tim -- Tim Johnson <tim(a)johnsons-web.com> http://www.alaska-internet-solutions.com

2 2

[slime-devel] Daily ChangeLog diff
by lgorrie＠common-lisp.net 24 Jul '05

24 Jul '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.724 slime/ChangeLog:1.725 --- slime/ChangeLog:1.724 Fri Jul 22 12:53:11 2005 +++ slime/ChangeLog Sun Jul 24 17:30:46 2005 @@ -1,3 +1,10 @@ +2005-07-24 Tom Pierce <tlpierce(a)gmail.com> + + * swank.lisp (format-iso8601-time): New functions. Properly + formats a universal-time as an iso8601 string. + (inspect-for-emacs integer): Use the new + format-iso8601 function when printing an integer as a date. + 2005-07-22 Marco Baringer <mb(a)bese.it> * swank-openmcl.lisp (frame-catch-tags): Remove some debugging

1 0