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

[slime-devel] Difficulty connecting to external process
by Dan Knapp 20 Apr '05

20 Apr '05

Much like in the recent thread on ssh, I'm trying to use Slime by connecting to a running Lisp. In theory, this ought to be simpler, because I'm not using ssh for it. I am not having much luck. I'm using the latest CVS version of Slime (just ran an update and repeated my test); OpenMCL 1.14.3; and the Emacs version: GNU Emacs 21.3.50.1 (powerpc-apple-darwin7.3.0) of 2004-05-08 on iBook.local In the host process, I load Slime through asdf, then I start it by invoking (setq swank:*use-dedicated-output-stream* nil) (swank:create-server :port 9090 :dont-close t :style nil) The *use-dedicated-output-stream* is there because an old thread on this list advised it. I have also tried without the :style parameter, not that that should have any relevance. Then in Emacs, I do M-x slime-connect RET 127.0.0.1 RET 9090 RET. Emacs says this: Connecting to Swank on port 9090.. net-read error: (end-of-file) error in process-filter: byte-code: PANIC! error in process-filter: PANIC! Lisp says this (I have it set up to dump into syslog): ;; Connection to Emacs lost. ;; [Short read: length=3158064 pos=58] If I try to connect again, I get the same error messages on both ends, so at least I know the loop isn't falling through... Both processes happen to be on the same computer, but are using different copies of the same version of Slime (one of them is inside a .app bundle...), in case that's relevant. Please advise? -- Dan Knapp

2 2

[slime-devel] Daily ChangeLog diff
by lgorrie＠common-lisp.net 19 Apr '05

19 Apr '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.666 slime/ChangeLog:1.667 --- slime/ChangeLog:1.666 Mon Apr 18 21:39:12 2005 +++ slime/ChangeLog Tue Apr 19 22:24:49 2005 @@ -1,3 +1,15 @@ +2005-04-19 heller <heller@xaital> + + * /project/slime/cvsroot/slime/PROBLEMS: Warn about old kernels. +2005-04-19 Helmut Eller <heller(a)common-lisp.net> + + * PROBLEMS: Warn about old kernels. + + * swank-backend.lisp: Fix some typos. + + * swank-sbcl.lisp (preferred-communication-style): Don't test for + sb-futex, it has lost its meaning in 0.8.21. + 2005-04-18 Helmut Eller <heller(a)common-lisp.net> * slime.el (inferior-lisp-program): Defvar it here, in case it is

1 0

[slime-devel] Re: Slime and :sb-futex on recent CVS SBCLs
by Christophe Rhodes 19 Apr '05

19 Apr '05

Andreas Fuchs <asf(a)boinkor.net> writes: > I'm not entirely sure which of SBCL or SLIME needs to be fixed. Either > SBCL should start including :sb-futex on its *feature* list again (or > some other feature saying that threads are non-broken on this lisp > build), or SLIME should stop checking for :sb-futex (which will break > on SBCLs older than 0.8.21). If it's possible given the time constraints and cvs tagging activity, I'd be happy if the preferred-communication-style checked only :sb-thread, and if the slime release notes for 1.2 required an sbcl of 0.8.21 or later. Would that be OK with you guys? Sorry for the eleventh-hour (or even later) change request... Cheers, Christophe

4 5

[slime-devel] Slime and :sb-futex on recent CVS SBCLs
by Andreas Fuchs 18 Apr '05

18 Apr '05

Hi, current SLIME CVS's swank-sbcl.lisp, function preferred-communication-style contains an assumption that is no longer true about recent SBCLs. Since 0.8.21, usefully threaded SBCL builds don't include :sb-futex in their feature list any more. Still, Swank checks for the presence of :sb-futex on the *features* list before adding the :spawn communication-style. I'm not entirely sure which of SBCL or SLIME needs to be fixed. Either SBCL should start including :sb-futex on its *feature* list again (or some other feature saying that threads are non-broken on this lisp build), or SLIME should stop checking for :sb-futex (which will break on SBCLs older than 0.8.21). Here's the (proof of concept) patch to make slime choose the right communication-style for 0.8.21 or later (but not earlier. I know no practical way to find out if we're running in SBCL version >=X): --- swank-sbcl.lisp 9 Apr 2005 07:07:00 -0000 1.129 +++ swank-sbcl.lisp 18 Apr 2005 22:27:45 -0000 @@ -42,8 +42,7 @@ ;;; TCP Server (defimplementation preferred-communication-style () - (if (and (sb-int:featurep :sb-thread) - (sb-int:featurep :sb-futex)) + (if (and (sb-int:featurep :sb-thread)) :spawn :fd-handler)) Cheers, -- Andreas Fuchs, <asf(a)boinkor.net>, asf(a)jabber.at, antifuchs

1 0

[slime-devel] Daily ChangeLog diff
by lgorrie＠common-lisp.net 18 Apr '05

18 Apr '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.664 slime/ChangeLog:1.666 --- slime/ChangeLog:1.664 Thu Apr 14 17:41:17 2005 +++ slime/ChangeLog Mon Apr 18 21:39:12 2005 @@ -1,3 +1,20 @@ +2005-04-18 Helmut Eller <heller(a)common-lisp.net> + + * slime.el (inferior-lisp-program): Defvar it here, in case it is + not defined in loaddefs and inf-lisp is not loaded. (That's the + case in XEmacs.) + + * mkdist.sh: update version number. + + * doc/slime.texi: Update version numbers for SBCL and ACL. + +2005-04-17 Peter Seibel <peter(a)gigamonkeys.com> + + * swank-loader.lisp (*implementation-features*): Added features + for GCL and ECL ... + (lisp-version-string): ... and code to compute version + string. (Supplied by someone who's email I've misplaced.) + 2005-04-14 Helmut Eller <heller(a)common-lisp.net> * slime.el (slime-selector): Discard input after sleeping.

1 0

[slime-devel] custom indentation of macros
by Chris Capel 18 Apr '05

18 Apr '05

I think there should be a way to specify, in a source file for a CL macro, how SLIME should indent the macro. There is already a robust and capable, if under-documented, way to specify the indentation for arbitrary forms in cl-indent.el, and so it would be nice to wrap this up in a nice way and give it some SLIME support. My idea is to have SWANK look for a property on the CL symbols, custom-indent-spec, that, if present, has a list (or perhaps a string that can be read as a list in Emacs) that can be assigned to the common-lisp-indent-function property on the Emacs symbol to modify the indentation. It'd be nice to have some documentation for the format of this list, too. Anyway, here's the sort of thing I think would work. (setf (get 'defclass-ext 'custom-indent-spec) '((6 4 defclass-slots-options-indent) (defun defclass-slots-options-indent (path state indent-point sexp-column normal-indent) (if (> (length path) 1) (1+ sexp-column) 2)))) The above form is Common Lisp and contained in the source file, either after the macro definition or somewhere else. The quoted list (caddr of the setf form) is written in Emacs lisp. (Symbol case issues? Should everything in the Emacs-destined form be downcased by SLIME or something?) The first form in that list is the value the common-lisp-indent-function property on the Emacs symbol gets set to. The subsequent forms are function definitions that can be used in the first. My example would allow the following indentation: (defclass-ext some-class () (slot-a :a :ia slot-b :a (:if 2))) There should be some attention to name collision across packages--Emacs lisp symbols don't have packages, so two different CL authors with a macro with the same name and custom indentation could potentially collide in unpleasant ways. I think what has to happen is that SLIME adds and removes the common-lisp-indent-function property on Emacs symbols every time a package is switched, based on what is set up for the inferior lisp's symbols in the new package. Is this last bit feasible? Perhaps SLIME could check to make sure the Emacs symbols have been set up with the correct plists before every indentation, resetting them all if the package has changed since then. Other ideas? Chris Capel

1 0

[slime-devel] Release?
by Helmut Eller 17 Apr '05

17 Apr '05

Is there anything left that we should fix before the release? If not, can we release the baby in the next few days? I used 1.2 as version number in the documentation, because there was already a 1.1 CVS tag (from November, before the Unicode changes). I hope that's ok with everybody. Helmut.

2 1

[slime-devel] Double newlines sent to clisp on user input
by Alex Farran 17 Apr '05

17 Apr '05

I'm running through Practical Common Lisp using SLIME + clisp on slackware, and I've run into a problem in chapter 3 with the prompt-for-cd function: (defun prompt-for-cd () (make-cd (prompt-read "Title") (prompt-read "Artist") (or (parse-integer (prompt-read "Rating") :junk-allowed t)) (y-or-n-p "Ripped [y/n]: "))) If I run clisp in a terminal it works as expected: [6]> (prompt-for-cd) Title: La Peste Artist: Alabama 3 Rating: 5 Ripped [y/n]: (y/n) y (:TITLE "La Peste" :ARTIST "Alabama 3" :RATING 5 :RIPPED T) But in SLIME this happens: CL-USER> (prompt-for-cd) Title: La Peste Artist: Rating: 5 Ripped [y/n]: (y/n) Please answer with y or n : y (:TITLE "La Peste" :ARTIST "" :RATING 5 :RIPPED T) It seems that for every one newline I enter two are sent. Is there something I can configure to fix it? I've tried running SLIME with my existing emacs and clisp and also with the versions compiled as part of the lisp-in-a-box package. I get the same problem either way. Alex

2 2

[slime-devel] selective macroexpansion patch
by Chris Capel 16 Apr '05

16 Apr '05

I have a patch and new file that implements a few functions that allow the selective expansion of subforms. I have the beginnings of a new mode, but my very limited familiarity with Emacs/SLIME programming ensures that it's almost useless. I'm hoping the improvements will be obvious and trivial to make this more usable. I'm kind of puzzled about why my returned forms are printing out with every symbol qualified by its package. Also, the new function slime-macroexpand-expand-subform should probably make the expansion be undoable with one undo, instead of two. I don't have the keybindings set up for the new mode, and slime-mode could probably use a key and menu entry to open the buffer. swank-loader.lisp will need to be modified appropriately. I wasn't sure what to do here, as swank-source-path-parser.lisp was only loaded for SBCL and CMUCL. Is it really that unportable? I guess this is a good first stab. I might look at it some more, see if I can clean it up some, if no one else does. Chris Capel New file: swank-expander.lisp ;;;; Selective macro expander ;;; These are some macros that will macroexpand a specific subform of a form ;;; once (destructively), respecting macros locally established by macrolets and ;;; symbol-macrolets. ;;; There are currently some limitations. swank-source-path-parser.lisp doesn't ;;; currently return the path of the specific atom the cursor is on--only the ;;; list. So it's not possible to expand symbol-macros until this situation is ;;; improved. Also, forms that expand into macrolets that are not expanded ;;; before expanding the subform will not respect the macrolet. ;;; For example, with ;;; ;;; (defmacro foo (&body body) ;;; `(macrolet ((bar (x) `(y ,x))) ;;; ,@body))) ;;; ;;; expanding the subform in ;;; ;;; (foo (bar xyz)) ;;; ;;; will not be effective unless FOO is added to ;;; *macroexpand-macro-binding-form* (in-package :swank-backend) (defparameter *macroexpand-macro-binding-form* '((macrolet 2) (symbol-macrolet 2)) "This is a specification for forms that are expected to expand into forms that bind symbols to macros. Each element of the list contains, first, the name of the macro, and second, the number of elements (including the name) to keep before replacing the rest with the specific form to be expanded.") (defmacro expand (form &environment e) "Return a form that represents a macroexpand-1'd FORM when evaluated in the current environment." (list 'quote (macroexpand-1 form e))) (defun get-list-to-expand (list source-path) "Get a sublist based on the source-path argument, but wrap it in all forms that bind macros, using *macroexpand-macro-binding-form*" (flet ((recurse () (get-list-to-expand (nth (car source-path) list) (cdr source-path)))) (if (null source-path) (list 'expand list) (let ((sym (find (car list) *macroexpand-macro-binding-form* :key #'car))) (if sym (append (subseq list 0 (cadr sym)) (list (recurse))) (recurse)))))) (defun nsubst-list (source substitution source-path) "Destructively replace the form in SOURCE at the give source path with SUBSTITUTION." (if (null source-path) substitution (progn (setf (nth (car source-path) source) (nsubst-list (nth (car source-path) source) substitution (cdr source-path))) source))) (defun subst-all-tree (tree obj-a obj-b &key (test #'equal)) "Non-destructively replace all instances of OBJ-A in TREE with OBJ-B (compared using TEST)." (mapcar (lambda (i) (if (consp i) (subst-all-tree i obj-a obj-b :test test) (if (funcall test obj-a i) obj-b i))) tree)) (defun expand-part (list source-path) "Expand, once, the form located at SOURCE-PATH and replace it with its expansion." (nsubst-list list (eval (get-list-to-expand list source-path)) source-path)) (defun expand-part-from-position (form-string position) (with-input-from-string (s form-string) (multiple-value-bind (list hash) (read-and-record-source-map s) (expand-part list (source-position-source-path position list hash))))) Index: slime.el =================================================================== RCS file: /project/slime/cvsroot/slime/slime.el,v retrieving revision 1.482 diff -u -r1.482 slime.el --- slime.el 9 Apr 2005 07:05:54 -0000 1.482 +++ slime.el 16 Apr 2005 04:54:28 -0000 @@ -6137,6 +6137,56 @@ "Display the recursively macro expanded sexp at point." (interactive) (slime-eval-macroexpand 'swank:swank-macroexpand-all)) + + +;;;; Selective macroexpansion + +(defvar slime-macroexpand-mode-map) + +(setq slime-macroexpand-mode-map (make-sparse-keymap)) +(set-keymap-parent slime-macroexpand-mode-map lisp-mode-map) + +(slime-define-keys slime-macroexpand-mode-map + ("\C-m" 'slime-macroexpand-expand-subform)) + +(defun slime-macroexpand-selective () + "Start a selective macroexpansion buffer with whatever's at the current +point." + (interactive) + (let ((string (slime-sexp-at-point)) + (buffer (get-buffer-create "*slime-macroexpand*"))) + (with-current-buffer buffer + (slime-macroexpand-mode) + (erase-buffer) + (pp (first (read-from-string string)) + buffer)))) + +(defun slime-macroexpand-expand-subform () + "In macroexpand mode, expand the currently selected subform." + (interactive) + (let ((i (point)) + (text (buffer-string))) + ;;how to make this erase and replacement a single operation? + (erase-buffer) + (let ((new-form (slime-eval `(swank-backend::expand-part-from-position ,text ,i)))) + ;;prints package qualifier on everything. Not very nice. + (pp new-form (current-buffer)) + (goto-char i)))) + +(defun slime-macroexpand-mode () + "Major mode for expanding parts of a form interactively. +\\{slime-macroexpand-mode-map}" + (interactive) + ;;need to set package somewhere... + (kill-all-local-variables) + (setq major-mode 'slime-macroexpand-mode) + (use-local-map slime-macroexpand-mode-map) + (lisp-mode-variables t) + (set (make-local-variable 'lisp-indent-function) + 'common-lisp-indent-function) + (setq font-lock-defaults nil) + (setq mode-name "Macroexpand") + (run-hooks 'slime-macroexpand-hook)) ;;;; Subprocess control Index: swank-source-path-parser.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank-source-path-parser.lisp,v retrieving revision 1.13 diff -u -r1.13 swank-source-path-parser.lisp --- swank-source-path-parser.lisp 1 Apr 2005 13:59:48 -0000 1.13 +++ swank-source-path-parser.lisp 16 Apr 2005 04:54:29 -0000 @@ -111,3 +111,26 @@ finally (destructuring-bind ((start . end)) positions (return (values (1- start) end)))))) +(defun source-position-source-path (position form source-map + &optional parent-form) + "Return the source path of the form most tightly surrounding the given +position." + (print form) + (if (and (not (null form)) (listp form)) + (destructuring-bind ((start . end)) (gethash form source-map) + (if (or (< position start) (> position end)) + nil + (let ((sub-path + (find-if #'identity + (mapcar (lambda (sub-form) + (source-position-source-path + position sub-form source-map form)) + form)))) + (cond + ((and sub-path parent-form) + (cons (position form parent-form) sub-path)) + (sub-path sub-path) + (parent-form + (list (position form parent-form))) + (t nil))))) + nil)) Index: swank.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank.lisp,v retrieving revision 1.293 diff -u -r1.293 swank.lisp --- swank.lisp 9 Apr 2005 07:07:46 -0000 1.293 +++ swank.lisp 16 Apr 2005 04:54:38 -0000 @@ -34,6 +34,7 @@ #:*swank-pprint-bindings* #:*default-worker-thread-bindings* #:*macroexpand-printer-bindings* + #:*macroexpand-macro-binding-forms* ;; These are re-exported directly from the backend: #:buffer-first-change #:frame-source-location-for-emacs

1 0

[slime-devel] Daily ChangeLog diff
by lgorrie＠common-lisp.net 14 Apr '05

14 Apr '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.663 slime/ChangeLog:1.664 --- slime/ChangeLog:1.663 Sat Apr 9 09:08:21 2005 +++ slime/ChangeLog Thu Apr 14 17:41:17 2005 @@ -1,3 +1,7 @@ +2005-04-14 Helmut Eller <heller(a)common-lisp.net> + + * slime.el (slime-selector): Discard input after sleeping. + 2005-04-09 Helmut Eller <heller(a)common-lisp.net> * slime.el (sldb-get-buffer): Create a fresh buffer if there's no @@ -8,8 +12,8 @@ because they are potentially useful. * swank.lisp (with-io-redirection, with-connection) - (with-buffer-syntax): Implement macros with a `call-with' - functions to avoid some code bloat. + (with-buffer-syntax): Implement macros with `call-with' functions + to avoid some code bloat. (call-with-connection, maybe-call-with-io-redirection) (call-with-buffer-syntax): New functions. (interactive-eval): Use from-string instead of read-from-string to

1 0