May 2009 - slime-devel - mailman3.common-lisp.net

[slime-devel] slime-autodoc stopping because of invalid characters.
by Mark Cox 28 May '09

28 May '09

Hi all, Occasionally I manage to input a strange character into the slime-repl and try and send it to the lisp. Fortunately, there is a condition in SLIME-NET-SEND in slime.el which detects for strange characters and refuses to send them. However, after signalling the "Coding system ..." error in that function, the slime-autodoc/eldoc support for the REPL stops working. The attached patch fixes that issue. Is it correct to send CONTINUATION the value (LIST :ABORT)? I have seen uses of SLIME-REX which look for it. Thanks Mark P.S. I'm a newbie to emacs lisp and the inner workings of slime.

2 1

[slime-devel] Daily ChangeLog diff
by Helmut Eller 25 May '09

25 May '09

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1770 slime/ChangeLog:1.1772 --- slime/ChangeLog:1.1770 Sat May 23 12:48:16 2009 +++ slime/ChangeLog Sun May 24 08:58:50 2009 @@ -1,3 +1,17 @@ +2009-05-24 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime.el ([test] fancy-symbol-names): Add cases involving #| + ... |# style comments. Currently failing. + + Reported by Madhu. + +2009-05-24 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime.el (slime-goto-location-position): Only go to match-beginning if + search succeeded. + + Patch by Madhu. + 2009-05-23 Helmut Eller <heller(a)common-lisp.net> * swank-openmcl.lisp (break-in-sldb): Honor *break-on-signals*.

1 0

[slime-devel] parser state bugs
by Madhu 24 May '09

24 May '09

Consider the following content in a slime buffer with the cursor at the point indicated by the caret |# (defun foo () (getf ^ (slime-symbol-at-point) returns nil. It should return "getf". [I thought I had isolated myself from this department by setting `slime-highlight-suppressed-forms' to nil. Could all buffer fontification code please be moved to an optional contrib? including the complexities involved in department] -- Madhu

2 5

[slime-devel] Daily ChangeLog diff
by Helmut Eller 24 May '09

24 May '09

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1769 slime/ChangeLog:1.1770 --- slime/ChangeLog:1.1769 Thu May 21 05:41:50 2009 +++ slime/ChangeLog Sat May 23 12:48:16 2009 @@ -1,3 +1,11 @@ +2009-05-23 Helmut Eller <heller(a)common-lisp.net> + + * swank-openmcl.lisp (break-in-sldb): Honor *break-on-signals*. + That means that we can't use SIGNAL here and we have to invoke + SLDB directly. + (condition-for-break): New helper. + Reported by Bill St. Clair. + 2009-05-19 Tobias C. Rittweiler <tcr(a)freebits.de> * slime.el (slime-symbol-at-point): Sometimes we can be too good,

1 0

[slime-devel] SLIME has stopped fontifying #+foo #-foo based on *FEATURES*
by graham＠gorgeous.org 23 May '09

23 May '09

I just looked at some SBCL code of mine that has different implementations of a function according to whether we're on Windows or not: #-win32 (defun cumnorm (x) "The cumulative normal distribution function - zero mean and unit variance" (declare (type double-float x)) (/ (#+sbcl sb-alien:alien-funcall #+cmu alien:alien-funcall (#+sbcl sb-alien:extern-alien #+cmu alien:extern-alien "erfc" (function double-float (double-float))) (- (/ (coerce x 'double-float) (sqrt 2)))) 2)) #+win32 (defun cumnorm (x) "The cumulative normal distribution function - zero mean and unit variance" (declare (type double-float x)) (/ (error-function-complement (- (/ (coerce x 'double-float) (sqrt 2)))) 2.0)) erfc is not in the C library on Windows. I'm sure that on Windows, SBCL used to highlight the #-win32 form in font-lock-comment-face (red in my Emacs) because WIN32 is in *FEATURES*, but for some reason this is no longer working. I recently CVS updated SLIME from about November (?) 2008 to 2009-05-19, so perhaps something in there broke it. I haven't yet got around to doing the CVS binary search thing to work out if and when it did. Any suggestions? Graham

2 2

[slime-devel] *break-on-signals* in ccl doesn't break if handler present
by Bill St. Clair 23 May '09

23 May '09

I'm running the latest CVS version of slime, "2009-05-19" in a fairly recent CCL (OpenMCL), "Version 1.3-r11936 (DarwinX8664)". I would expect the following code to bring up an SLDB buffer with a break loop in the repl: (setq *break-on-signals* t) (handler-case (error "foo") (error () 1)) Instead, it returns 1. This is not surprising given this in swank-openmcl.lisp: (defun break-in-sldb (x y &rest args) (let ((*sldb-stack-top-hint* (or *sldb-stack-top-hint* (ccl::%get-frame-ptr)))) (apply #'cerror y (if args "Break: ~a" x) args))) I'm debugging some server code that really needs to catch all errors. This makes it rather difficult. The only workaround I've discovered is to set swank-backend::*break-in-sldb* to NIL and use CCL's builtin debugger. Not my favorite debugging environment. Clues, anyone? I could probably fix this myself, and if nobody else has done it, I may be forced to, but I'd rather make progress on my application code. -Bill

3 2

[slime-devel] Semantic of slime-repl-quit command
by David Vazquez 22 May '09

22 May '09

Hello, Currently I am using slime to hack a remote lisp process using the swank server with the parameter :dont-close to T. However, when I am typing commands at repl, if I use `, quit' the lisp process ends. I know it is a documented thing, but I think `slime-repl-quit' should not request to quit the inferior process. It should be done by swank if DONT-CLOSE is NIL. Otherwise we should close emacs buffers by hand if we want to keep alive the swank connection, I think. I would like if a patch for this would be wise. Thanks in advance. David.

3 4

[slime-devel] no ripple effect in GNU Emacs 22.3.1+SLIME (CVS download of May 20, 2009, at 5:42 am)+CLISP 2.45 on Windows XP Professional, Service Pack 3
by Benjamin L. Russell 22 May '09

22 May '09

When I type M-x slime RET in GNU Emacs 22.3.1 with SLIME (CVS download of May 20, 2009, at 5:42 am) with CLISP 2.45 on Windows XP Professional, Service Pack 3, although the REPL is invoked correctly, there is no ripple effect upon REPL startup. However, when I use lispbox-0.7-clisp-2.37 on the same OS, the ripple effect appears correctly upon REPL startup. How can I invoke the ripple effect upon SLIME REPL startup in regular GNU Emacs (as opposed to Lispbox)? In my .emacs file, I have the following relevant code: >;; Setup for SLIME >(setq inferior-lisp-program "C:/bin/clisp-2.45/full/lisp.exe >-B C:/bin/clisp-2.45/full >-M C:/bin/clisp-2.45/full/lispinit.mem >-ansi -q") >(add-to-list 'load-path "C:/bin/emacs/site-lisp/slime/") >(require 'slime) >(slime-setup '(slime-fancy slime-asdf)) Any help would be greatly appreciated. -- Benjamin L. Russell -- Benjamin L. Russell / DekuDekuplex at Yahoo dot com http://dekudekuplex.wordpress.com/ Translator/Interpreter / Mobile: +011 81 80-3603-6725 "Furuike ya, kawazu tobikomu mizu no oto." -- Matsuo Basho^

2 2

[slime-devel] Daily ChangeLog diff
by Helmut Eller 22 May '09

22 May '09

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1768 slime/ChangeLog:1.1769 --- slime/ChangeLog:1.1768 Tue May 19 16:42:22 2009 +++ slime/ChangeLog Thu May 21 05:41:50 2009 @@ -1,5 +1,14 @@ 2009-05-19 Tobias C. Rittweiler <tcr(a)freebits.de> + * slime.el (slime-symbol-at-point): Sometimes we can be too good, + e.g. in "|# (defun foo () (getf" the above would return nil + because the vertical bar is not terminated. The user probably + wants "getf" nontheless. + + Reported by Madhu. + +2009-05-19 Tobias C. Rittweiler <tcr(a)freebits.de> + * slime.el (sldb-restartable-frame-line-face): Set a default value. (sldb-frame-restartable-p): New. (sldb-compute-frame-face): Use it.

1 0

[slime-devel] CMUCL: how to start slime in cl-user
by N. Raghavendra 21 May '09

21 May '09

Slime with CMUCL starts the Slime REPL in the `swank' package, and the inferior lisp in the `cl-user' package. Is there a way to get the Slime REPL also to start CMUCL in the `cl-user' package? I've tried without success (defun nr-cmucl-init-command (port-filename coding-system) "Return a string to initialize CMUCL. This is a modified version of `slime-init-command' from `slime.el'." (let ((loader (if (file-name-absolute-p slime-backend) slime-backend (concat slime-path slime-backend))) (encoding (slime-coding-system-cl-name coding-system))) (format "%S\n\n" `(progn (load ,(expand-file-name loader) :verbose t) (funcall (read-from-string "swank-loader:init")) (funcall (read-from-string "swank:start-server") ,port-filename :coding-system ,encoding) (in-package "CL-USER"))))) (setq slime-lisp-implementations '((cmucl ("/opt/bin/lisp") :init nr-cmucl-init-command))) Thanks, Raghavendra. -- N. Raghavendra <raghu(a)mri.ernet.in> | http://www.retrotexts.net/ Harish-Chandra Research Institute | http://www.mri.ernet.in/ See message headers for contact and OpenPGP information.

2 2