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

[slime-devel] Standard output no longer appears in REPL
by Steven E. Harris 25 Jun '11

25 Jun '11

It's been a couple of months since I last used SLIME. Today I noticed that when I evaluate expressions in the REPL that write to *standard-output*, the output appears in the *inferior-lisp* buffer, but not in the REPL as it used to. I'm using GNU Emacs on Windows atop the CLISP Cygwin build. SLIME is up to date with CVS. Is there some new configuration required to restore the old behavior with output streams feeding into the REPL output? -- Steven E. Harris

5 12

[slime-devel] Proposed patch for asdf-loading under ALLEGRO
by Robert Goldman 05 Jan '10

05 Jan '10

ASDF operations using SLIME and SWANK won't work under Allegro's Modern (case-sensitive) lisp, because the symbol names are not in all caps as in ANSI lisp. Here's a small patch that makes it work under ACL mlisp as well as under normal ANSI lisps: Index: swank-asdf.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/contrib/swank-asdf.lisp,v retrieving revision 1.7 diff -b -u -F '^(def' -r1.7 swank-asdf.lisp --- swank-asdf.lisp 23 Oct 2008 21:28:03 -0000 1.7 +++ swank-asdf.lisp 25 May 2009 16:40:10 -0000 @@ -26,7 +26,9 @@ (defun operate-on-system (system-name op Example: \(operate-on-system \"SWANK\" \"COMPILE-OP\" :force t)" (with-compilation-hooks () - (let ((operation (find-symbol operation-name :asdf))) + (let ((operation (or (find-symbol operation-name :asdf) + #+allegro + (find-symbol (string-downcase operation-name) :asdf)))) (when (null operation) (error "Couldn't find ASDF operation ~S" operation-name)) (apply #'asdf:operate operation system-name keyword-args))))

2 1

Re: [slime-devel] random returning the same value
by Tobias C. Rittweiler 30 Nov '09

30 Nov '09

Patrick May <pjm(a)spe.com> writes: > Hi, > > I'm using Slime in Aquamacs with CCL. When I evaluate (random > 100) repeatedly in CCL started from a shell, I get a randomly > generated number between 0 and 99 inclusive. When I do the same with > C-x C-e within Slime, I get the same number every time. (random 100 > (make- > random-state)) gives the same behavior, but (random 100 (make-random- > state t)) works properly. > > Is Slime caching something or have I hosed up my environment? Sorry for the late response. CCL's swank backend uses threads by default, and threads inherit values from the global environment. *RANDOM-STATE* probably becomes thread-local automatically in CCL, resulting in the behaviour you described. It should work when you type in the form at the Slime REPL which evaluates all forms in the same thread. -T.

4 4

[slime-devel] bug: slime-buffer-package not set on connection
by Jeffrey Chu 27 Sep '09

27 Sep '09

Hi, It looks like in slime-buffer-package is not being set after connection and (swank:create-repl nil) is called / returned. This causes things that call (slime-current-package), like slime-c-p-c, to be a little sad when the REPL's default package isn't "COMMON-LISP-USER". >From what I can tell, a good place to apply it is at the end of the slime-repl-connected-hook-function: diff a/contrib/slime-repl.el b/contrib/slime-repl.el --- a/contrib/slime-repl.el +++ b/contrib/slime-repl.el @@ -1519,9 +1519,11 @@ expansion will be added to the REPL's history.)" (let ((slime-current-thread t)) (slime-eval '(swank:create-repl nil))) (setf (slime-lisp-package) package) - (setf (slime-lisp-package-prompt-string) prompt)) - (slime-hide-inferior-lisp-buffer) - (slime-init-output-buffer (slime-connection))) + (setf (slime-lisp-package-prompt-string) prompt) + (slime-hide-inferior-lisp-buffer) + (slime-init-output-buffer (slime-connection)) + (with-current-buffer (slime-output-buffer) + (setf slime-buffer-package package)))) Please let me know if you need any additional information, - Jeff

2 1

[slime-devel] [BUG] Switching default connection does not work correctly
by Tobias C. Rittweiler 27 Sep '09

27 Sep '09

Steps to reproduce: M-x slime M-x slime "Create an additional *inferior-lisp*? (y or n)" -> y ,quit in the new REPL Currently there's no default connection, so the next Slime command with an RPC call asks for switching to a new default. At the other REPL, M-. on DEFUN. "No default connection selected. Switch to <foo>?" Now, when you answer that question with no, it asks the question again, and again, and after a few times (always negating the question), it performs the M-. in the non-selected connection even though it shouldn't. -T.

2 2

[slime-devel] Setting FASL directory
by David Neu 24 Sep '09

24 Sep '09

Hi, I'd like to be able to set the location of the *.fasl files that are ending up in ~/.slime/fasl/..., preferably from my .emacs. I tried setting slime-compile-file-options in my .emacs file as follows (eval-after-load "slime" '(progn (setq slime-compile-file-options (append '(:fasl-directory "/tmp/") slime-compile-file-options)) ...)) but that didn't seem to do it. Then I tried to put the following in my ~/.swank.lisp file (setf swank:*fasl-pathname-function* (lambda (input-file options) (declare (ignore options)) (let* ((str "/tmp/") (dir (filename-to-pathname str))) (assert (char= (aref str (1- (length str))) #\/)) (compile-file-pathname input-file :output-file dir)))) but that didn't seem to do it. Many thanks for any suggestions! Cheers, David

4 8

[slime-devel] TYPE ERROR: The value 9 is not of type LIST
by Sebastian Tennant 02 Sep '09

02 Sep '09

Hi all, I'm running a threaded SBCL version 1.0.30 with SLIME version 2009-08-19 in GNU Emacs verison 22.2.1. Non-ASCII characters break the connection to the REPL: CL-USER> "ğ" "ğ" The REPL is hung at this point. When I hit RET, the *inferior-lisp* buffer is flooded with text and the minibuffer reports: Lisp connection closed unexpectedly: connection broken by remote peer Relevant content of *inferior-lisp* buffer attached. Any suggestions as to how to solve/debug this problem much appreciated. Regards, Sebastian -- Emacs' AlsaPlayer - Music Without Jolts Lightweight, full-featured and mindful of your idyllic happiness. http://home.gna.org/eap

4 13

[slime-devel] [patch] add a new protocol message for outputting images
by Stanislaw Halik 29 Aug '09

29 Aug '09

Here's a new protocol message for outputting images. Unlike the previous approach, it doesn't rely upon EVAL-IN-EMACS. The Bad Thing is that it puts REPL-specific code inside DISPATCH-EVENT which should be contrib-agnostic. There seems to be no way to add protocol messages in contribs, maybe there's some better way not involving protocol messages. I added the `fake-p' stuff because if it isn't there, the image is displayed right before the last REPL result (maybe dependent upon presentations?). PS in the last commit I got credited as "StanisBaw". The "ł" character apparently got turned into this "B". It can be transliterated to ASCII as "l" if the changelog has problems with Unicode. diff --git a/contrib/slime-repl.el b/contrib/slime-repl.el index 9e05531..3bcb894 100644 --- a/contrib/slime-repl.el +++ b/contrib/slime-repl.el @@ -719,23 +719,27 @@ balanced." (goto-char (point-max)) (recenter -1)))) -(defun slime-repl-send-input (&optional newline) +(defun slime-repl-send-input (&optional newline fake-p) "Goto to the end of the input and send the current input. -If NEWLINE is true then add a newline at the end of the input." +If NEWLINE is true then add a newline at the end of the input. +When FAKE-P is true it doesn't get added to history. Useful for +adding images to the REPL." (unless (slime-repl-in-input-area-p) (error "No input at point.")) (goto-char (point-max)) (let ((end (point))) ; end of input, without the newline - (slime-repl-add-to-input-history - (buffer-substring slime-repl-input-start-mark end)) + (unless fake-p + (slime-repl-add-to-input-history + (buffer-substring slime-repl-input-start-mark end))) (when newline (insert "\n") (slime-repl-show-maximum-output)) - (let ((inhibit-modification-hooks t)) - (add-text-properties slime-repl-input-start-mark - (point) - `(slime-repl-old-input - ,(incf slime-repl-old-input-counter)))) + (unless fake-p + (let ((inhibit-modification-hooks t)) + (add-text-properties slime-repl-input-start-mark + (point) + `(slime-repl-old-input + ,(incf slime-repl-old-input-counter))))) (let ((overlay (make-overlay slime-repl-input-start-mark end))) ;; These properties are on an overlay so that they won't be taken ;; by kill/yank. @@ -745,7 +749,8 @@ If NEWLINE is true then add a newline at the end of the input." (goto-char (point-max)) (slime-mark-input-start) (slime-mark-output-start) - (slime-repl-send-string input))) + (unless fake-p + (slime-repl-send-string input)))) (defun slime-repl-grab-old-input (replace) "Resend the old REPL input at point. diff --git a/slime.el b/slime.el index c392996..e60a022 100644 --- a/slime.el +++ b/slime.el @@ -2392,6 +2392,12 @@ Debugged requests are ignored." (let ((slime-dispatching-connection (or process (slime-connection)))) (or (run-hook-with-args-until-success 'slime-event-hooks event) (destructure-case event + ((:write-image pathname) + (if (file-exists-p pathname) + (with-current-buffer (slime-repl-buffer) + (insert-image (create-image pathname)) + (slime-repl-send-input t t)) + (error "No such file: '%s'" pathname))) ((:emacs-rex form package thread continuation) (when (and (slime-use-sigint-for-interrupt) (slime-busy-p)) (slime-display-oneliner "; pipelined request... %S" form)) diff --git a/swank.lisp b/swank.lisp index 9b1f0f4..a259797 100644 --- a/swank.lisp +++ b/swank.lisp @@ -1174,7 +1174,16 @@ The processing is done in the extent of the toplevel restart." ((:reader-error packet condition) (encode-message `(:reader-error ,packet ,(safe-condition-message condition)) - (current-socket-io))))) + (current-socket-io))) + ((:write-image pathname) + (handler-case (namestring (truename pathname)) + (error (e) + (declare (ignore e)) + nil) + (:no-error (string) + (when string + (encode-message `(:write-image ,string) + (current-socket-io)))))))) (defvar *event-queue* '()) (defvar *events-enqueued* 0) @@ -3845,4 +3854,4 @@ Collisions are caused because package information is ignored." (defun init () (run-hook *after-init-hook*)) -;;; swank.lisp ends here \ No newline at end of file +;;; swank.lisp ends here

3 2

[slime-devel] Daily ChangeLog diff
by Helmut Eller 29 Aug '09

29 Aug '09

Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.238 slime/contrib/ChangeLog:1.239 --- slime/contrib/ChangeLog:1.238 Thu Aug 27 16:15:43 2009 +++ slime/contrib/ChangeLog Fri Aug 28 19:50:48 2009 @@ -1,3 +1,7 @@ +2009-08-28 Stas Boukarev <stassats(a)gmail.com> + + * slime-c-p-c.el (slime-contextual-completions): Remove debugging code. + 2009-08-27 Stas Boukarev <stassats(a)gmail.com> * slime-c-p-c.el (slime-contextual-completions): Detect characters properly.

1 0

[slime-devel] Two minor bugs
by James Wright 28 Aug '09

28 Aug '09

Hi everyone, I've noticed a couple of bugs in the most recent version of slime: 1. If I type #\Sp<Tab> in the REPL, I get the message "string=: Symbol's function definition is void: dbgmsg". The following change in `slime-contextual-completions' seems to fix it: diff -r1.11 slime-c-p-c.el 131c131 < (string= (dbgmsg (subseq token 0 2)) "#\\")) --- > (string= (subseq token 0 2) "#\\")) 2. The documentation for `slime-maybe-show-compilation-log' says that it displays the compilation log if compilation fails or if "NOTES is non-NIL". That doesn't match its behaviour; in the current CVS version, it shows the compilation log only if compilation has failed. Is this intended? If not, the following simple change fixes it: diff -r1.1217 slime.el 2957c2957 < (unless successp --- > (when (or (null successp) notes) Thanks, James

2 1