April 2007 - slime-devel - mailman3.common-lisp.net

[slime-devel] [PATCH] improved display of symbols of a package in the Inspector
by Tobias C. Rittweiler 17 Apr '07

17 Apr '07

To make oneself familiar with someone else's code, it's often desirable to see what symbols are defined in a package. I've added a special inspector page for that purpose that will pop up if you inspect the package, and then click on the links to the external/internal symbols. The flags are the same as those that are presented in Fuzzy Completion. Here's an snippet from all present symbols of #<PACKAGE "SWANK">: All present symbols of package "SWANK" [type: %PACKAGE-SYMBOLS-CONTAINER] -------------------- A symbol is considered present in a package if it's "accessible in that package directly, rather than being inherited from another package." (CLHS glossary entry for `present') Symbols: Flags: --------------------------------------------------- ------- %CONTAINER ------- %MAKE-FUZZY-MATCHING -f----- %PACKAGE-SYMBOLS-CONTAINER ---c--- %PACKAGE-SYMBOLS-CONTAINER- ------- %PACKAGE-SYMBOLS-CONTAINER-DESCRIPTION -f----- %PACKAGE-SYMBOLS-CONTAINER-P -f----- %PACKAGE-SYMBOLS-CONTAINER-SYMBOLS -f----- %PACKAGE-SYMBOLS-CONTAINER-TITLE -f----- &UNKNOWN-JUNK ------- *ACTIVE-THREADS* b------ *AFTER-INIT-HOOK* b------ ... Hope you'll find this useful, -T.

1 2

[slime-devel] slime-display-compilation-output and slime-eval-with-transcript
by Yoni Rabkin Katzenell 17 Apr '07

17 Apr '07

Hello, C-c C-k takes the value of slime-display-compilation-output into account. I wanted the same to be true of C-M-x so I changed slime-eval-with-transcript: (defun slime-eval-with-transcript ... ... (slime-eval-async form (lambda (value) (with-current-buffer (slime-output-buffer) => (when slime-display-compilation-output (slime-show-last-output)) (cond (fn (funcall fn value)) (t (message "%s" value)))))))) Is there a simpler way to get the same result? -- "Cut your own wood and it will warm you twice"

2 1

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 17 Apr '07

17 Apr '07

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1106 slime/ChangeLog:1.1111 --- slime/ChangeLog:1.1106 Thu Apr 12 15:00:09 2007 +++ slime/ChangeLog Mon Apr 16 10:48:14 2007 @@ -1,3 +1,65 @@ +2007-04-16 Takehiko Abe <keke(a)gol.com> + + * swank-openmcl.lisp (accept-connection, find-external-format): + utf-8 support. + +2007-04-16 Marco Baringer <mb(a)bese.it> + + * slime.el (slime-with-xref-buffer): Added missing , + +2007-04-16 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime.el: Pressing `q' in *compiler notes* after a `C-c C-k' or + `C-c M-k' would not probably restore the original window + configuration. Fix that. + + (slime-get-temp-buffer-create): New &key arg WINDOW-CONFIGURATION. + (slime-with-xref-buffer): Likewise. + + (slime-compilation-finished): New &optional arg WINDOW-CONFIG. + (slime-maybe-show-xrefs-for-notes): Likewise. + (slime-show-xrefs) Likewise. + (slime-maybe-list-compiler-notes): Likewise. + (slime-list-compiler-notes): Likewise. + + (slime-compilation-finished-continuation): Renamed to + SLIME-MAKE-COMPILATION-FINISHED-CONTINUATION. + + (slime-make-compilation-finished-continuation): Now takes two + args, the current buffer and optionally the current window config + to be restored. + + (slime-compile-file): Save current window configuration before + popping up the REPL for compilation output, pass it down. + (slime-easy-menu): Add entry for SLIME-UNTRACE-ALL. + +2007-04-16 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank.lisp (fuzzy-find-matching-packages): Fix a small typo that + prevented interpreting NIL as the argument TIME-LIMIT-IN-MEC to + mean an infinite time limit. This bug propagated up to explicit + calls to FUZZY-COMPLETIONS, like + (swank:fuzzy-completions "mvb" "COMMON-LISP") => (NIL, T) + + (format-fuzzy-completions): Renamed to FORMAT-FUZZY-COMPLETION-SET + + (format-fuzzy-completion-set): Accomodated to recent changes of + the return value of FUZZY-COMPLETIONS; changed the docstring to + make it explicit that this function is supposed to take the return + value of FUZZY-COMPLETION-SET. + + * slime.el (slime-compilation-finished): Don't use + MULTIPLE-VALUE-BIND for list destructuring, only because multiple + values happen to be implemented via lists in elisp! + (slime-fuzzy-completions-mode): Added an detailed explanation to + the docstring of how Fuzzy Completion works and how it'ss supposed + to be worked with. + (slime-fuzzy-explanation): Shortened to reference to + SLIME-FUZZY-COMPLETIONS-MODE for help on fuzzy completion. + (slime-fuzzy-choices-buffer): Set BUFFER-QUIT-FUNCTION to + SLIME-FUZZY-ABORT to make it correctly exit fuzzy completion when + pressing `Esc Esc Esc' (`M-Esc Esc'). + 2007-04-12 Nikodemus Siivola <nikodemus(a)random-state.net> * swank-sbcl.lisp (emacs-buffer-source-location): Add &allow-other-keys

1 0

[slime-devel] [PATCH] fix restoration of window config when quitting from *compiler notes*
by Tobias C. Rittweiler 16 Apr '07

16 Apr '07

2007-04-12 Tobias C. Rittweiler <INSERT-EMAIL-HERE> * slime.el: Pressing `q' in *compiler notes* after a `C-c C-k' or `C-c M-k' would not probably restore the original window configuration. Fix that. (slime-get-temp-buffer-create): New &key arg WINDOW-CONFIGURATION. (slime-with-xref-buffer): Likewise. (slime-compilation-finished): New &optional arg WINDOW-CONFIG. (slime-maybe-show-xrefs-for-notes): Likewise. (slime-show-xrefs) Likewise. (slime-maybe-list-compiler-notes): Likewise. (slime-list-compiler-notes): Likewise. (slime-compilation-finished-continuation): Renamed to SLIME-MAKE-COMPILATION-FINISHED-CONTINUATION. (slime-make-compilation-finished-continuation): Now takes two args, the current buffer and optionally the current window config to be restored. (slime-compile-file): Save current window configuration before popping up the REPL for compilation output, pass it down.

3 4

[slime-devel] [PATCH] add easymenu entry for slime-untrace-all
by Tobias C. Rittweiler 16 Apr '07

16 Apr '07

2007-04-11 Tobias C. Rittweiler <INSERT-EMAIL-HERE> * slime.el (slime-easy-menu): Add entry for SLIME-UNTRACE-ALL.

2 1

[slime-devel] Connecting swank from lisp
by Volkan YAZICI 14 Apr '07

14 Apr '07

Hi, [I tried to search the archives but couldn't find anything similar to my needs. Therefore execuse me if this subject is discussed before. And I'll be appreciated if you can point me to the related thread.] I've been trying to establish a start/stop init script to fire up a swank server. IRC folks suggested to use interrupts while shutting the swank server down, but I'm looking for if there is any other portable way of it. What I was wondering is, would it be possible to connect to a swank server from another lisp image and execute arbitrary commands (e.g., QUIT, SWANK:QUIT-LISP) on it? Regards. P.S. Yep, I checked the scripts at boinkor.net, but I want to use GNU screen instead. (For functionality purposes.)

2 2

[slime-devel] openmcl utf-8 support
by Takehiko Abe 14 Apr '07

14 Apr '07

OpenMCL 1.1 now supports unicode. The following adds utf-8 support to slime. ;;; swank-openmcl.lisp (defimplementation accept-connection (socket &key external-format buffering timeout) (declare (ignore buffering timeout #-openmcl-unicode-strings external-format)) #+openmcl-unicode-strings (when external-format (let ((keys (ccl::socket-keys socket))) (setf (getf keys :external-format) external-format (slot-value socket 'ccl::keys) keys))) (ccl:accept-connection socket :wait t)) #+openmcl-unicode-strings (defvar *external-format-to-coding-system* '((:iso-8859-1 "latin-1" "latin-1-unix" "iso-latin-1-unix" "iso-8859-1" "iso-8859-1-unix") (:utf-8 "utf-8" "utf-8-unix"))) #+openmcl-unicode-strings (defimplementation find-external-format (coding-system) (car (rassoc-if (lambda (x) (member coding-system x :test #'equal)) *external-format-to-coding-system*)))

1 0

[slime-devel] incompatible protocol?
by Joe Corneli 13 Apr '07

13 Apr '07

Hi -- I am getting weird results with my remote slime connection. I am running and SBCL 0.9.16 and swank from SLIME 2.0 on a remote linux box, and connecting with SLIME 2.0 on GNU Emacs 22.0.91.3 from 2006-12-16. When I `slime-connect' I get this prompt: Protocol version mismatch. Continue anyway? (yes or no) If I type yes, then I can do things: CL-USER> (asdf:operate 'asdf:load-op :elephant) ; loading system definition from ; /usr/local/lib/sbcl/site-systems/elephant.asd into #<PACKAGE "ASDF0"> [...] But I can't seem to get values back! -- CL-USER> (+ 1 1) CL-USER> On the other hand, I can print: CL-USER> (format t "~a" (+ 1 1)) 2 Is there a way to get the normal expected behavior back?

2 1

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 13 Apr '07

13 Apr '07

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1105 slime/ChangeLog:1.1106 --- slime/ChangeLog:1.1105 Sun Apr 8 18:56:32 2007 +++ slime/ChangeLog Thu Apr 12 15:00:09 2007 @@ -1,3 +1,9 @@ +2007-04-12 Nikodemus Siivola <nikodemus(a)random-state.net> + + * swank-sbcl.lisp (emacs-buffer-source-location): Add &allow-other-keys + to the descructuring of the source location plist in order to accept + :emacs-directory. + 2007-04-09 Marco Baringer <mb(a)bese.it> * swank.lisp (inspector-content-for-emacs): Look for refresh

1 0

[slime-devel] [PATCH] Fuzzy Completion cleanup
by Tobias C. Rittweiler 11 Apr '07

11 Apr '07

Hi, the attached patch cleans up parts of the relevant code for fuzzy completion in swank.lisp. There are two user visible changes: i) The flags in the *Fuzzy Completions* buffer (indicating if the respective symbol is boundp, fboundp, represeting a class, macro &c) are now extended by the "p" flag indicating the there exists a package accessible via that symbol. ii) the symbols in the *Fuzzy Completions* are now sorted alphabetically before sorted by their score. From my ChangeLog entry: Affects especially the list of all possible completions when the user hits fuzzy-completion on an empty string within Emacs; also makes the potential limitness of the listed completions clearer to the end user of SLIME. Stay tuned, -T. Index: ChangeLog =================================================================== RCS file: /project/slime/cvsroot/slime/ChangeLog,v retrieving revision 1.1076 diff -u -r1.1076 ChangeLog --- ChangeLog 26 Feb 2007 09:56:06 -0000 1.1076 +++ ChangeLog 26 Feb 2007 19:58:56 -0000 @@ -1,3 +1,53 @@ +2007-02-26 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank.lisp: Cleanup of parts of the fuzzy completion code. + Additionally a couple of enhancements. As follows: + + (fuzzy-completions, fuzzy-completion-selected): Minor + stylistic and clarifying modifications of the docstrings. + + (fuzzy-find-matching-symbols): Huge code reorganization. + Organizing relevant code into local function TIME-EXHAUSTED-P, + renaming local function SYMBOL-MATCH to PERFORM-FUZZY-MATCH, + making previously required argument EXTERNAL to new &key + argument :EXTERNAL-ONLY, clarifying docstring. + + (fuzzy-find-matching-packages): Making its return value + conformant to that of FUZZY-FIND-MATCHING-SYMBOLS, i.e. + instead of returning, among others, a package's name as + string, it now returns a symbol representing the package. + Accomodates the docstring accordingly. + + (fuzzy-completion-set): Minor typographical fix in docstring. + Changing local function CONVERT to use MAP-INTO instead of + doing it essentially manually. Accomodating to changes of + FUZZY-FIND-MATCHING-SYMBOLS, resp. -PACKAGES. + + (fuzzy-completion-set): Additional new feature: + The returned completions are sorted alphabetically by the + matched completion string before sorted by its score. + Affects especially the list of all possible completions when + the user hits fuzzy-completion on an empty string within Emacs; + also makes the potential limitness of the listed completions + clearer to the end user of SLIME. + + (classify-symbol): New function. Returns a list with keywords + that classifies a given symbol. (E.g. :BOUNDP, :MACRO &c) + Supersedes parts of CONVERT-FUZZY-COMPLETION-RESULT, + implementing them in a more straightforward and proper way; + removes prior KLUDGE in that part of the original function. + + (convert-fuzzy-completion-result): The above changes made + it possible to simplify this function drastically. Now uses + the newly introduced function CLASSIFY-SYMBOL. + + * slime.el: Minor stylistic changes. Additionally: + (slime-fuzzy-insert-completion-choice): + (slime-fuzzy-fill-completions-buffer) : Adding use of the + :PACKAGE classification flag returned by SWANK:FUZZY-COMPLETIONS. + This flag is called "p". + + 2007-02-26 Nikodemus Siivola <nikodemus(a)random-state.net> * swank.lisp (inspect-for-emacs): Add support for inspecting Index: slime.el =================================================================== RCS file: /project/slime/cvsroot/slime/slime.el,v retrieving revision 1.766 diff -u -r1.766 slime.el --- slime.el 25 Feb 2007 15:59:34 -0000 1.766 +++ slime.el 26 Feb 2007 19:58:57 -0000 @@ -6412,7 +6412,7 @@ (slime-fuzzy-done)) (goto-char end) (cond ((= (length completion-set) 1) - (insert-and-inherit (caar completion-set)) + (insert-and-inherit (caar completion-set)) ; insert completed string (delete-region beg end) (goto-char (+ beg (length (caar completion-set)))) (slime-minibuffer-respecting-message "Sole completion") @@ -6430,7 +6430,7 @@ "Click <mouse-2> on a completion to select it. In this buffer, type n and p to navigate between completions. Type RET to select the completion near point. Type q to abort. -Flags: boundp fboundp generic-function class macro special-operator +Flags: boundp fboundp generic-function class macro special-operator package \n" "The explanation that gets inserted at the beginning of the *Fuzzy Completions* buffer.") @@ -6440,11 +6440,11 @@ completion choice into the current buffer, and mark it with the proper text properties." (let ((start (point)) - (symbol (first completion)) + (symbol-name (first completion)) (score (second completion)) (chunks (third completion)) (flags (fourth completion))) - (insert symbol) + (insert symbol-name) (let ((end (point))) (dolist (chunk chunks) (put-text-property (+ start (first chunk)) @@ -6454,13 +6454,14 @@ (put-text-property start (point) 'mouse-face 'highlight) (dotimes (i (- max-length (- end start))) (insert " ")) - (insert (format " %s%s%s%s%s%s %8.2f" + (insert (format " %s%s%s%s%s%s%s %8.2f" (if (member :boundp flags) "b" "-") (if (member :fboundp flags) "f" "-") (if (member :generic-function flags) "g" "-") (if (member :class flags) "c" "-") (if (member :macro flags) "m" "-") (if (member :special-operator flags) "s" "-") + (if (member :package flags) "p" "-") score)) (insert "\n") (put-text-property start (point) 'completion completion)))) @@ -6522,9 +6523,12 @@ (setf max-length (max max-length (length (first completion))))) (insert "Completion:") (dotimes (i (- max-length 10)) (insert " ")) - (insert "Flags: Score:\n") + ;; Flags: Score: + ;; ... ------- -------- + ;; bfgcmsp + (insert "Flags: Score:\n") (dotimes (i max-length) (insert "-")) - (insert " ------ --------\n") + (insert " ------- --------\n") (setq slime-fuzzy-first (point)) (dolist (completion completions) (slime-fuzzy-insert-completion-choice completion max-length)) Index: swank.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank.lisp,v retrieving revision 1.460 diff -u -r1.460 swank.lisp --- swank.lisp 26 Feb 2007 09:56:06 -0000 1.460 +++ swank.lisp 26 Feb 2007 19:58:57 -0000 @@ -3455,6 +3455,9 @@ ;;;; Fuzzy completion +;;; For nomenclature of the fuzzy completion section, please read +;;; through the following docstring. + (defslimefun fuzzy-completions (string default-package-name &key limit time-limit-in-msec) "Return an (optionally limited to LIMIT best results) list of fuzzy completions for a symbol designator STRING. The list will @@ -3463,13 +3466,17 @@ The result is a list of completion objects, where a completion object is: (COMPLETED-STRING SCORE (&rest CHUNKS) FLAGS) -where a CHUNK is a description of a matched string of characters: - (OFFSET STRING) -and FLAGS is a list of keywords describing properties of the symbol. -For example, the top result for completing \"mvb\" in a package -that uses COMMON-LISP would be something like: - (\"multiple-value-bind\" 42.391666 ((0 \"mul\") (9 \"v\") (15 \"b\")) +where a CHUNK is a description of a matched substring: + (OFFSET SUBSTRING) +and FLAGS is a list of keywords describing properties of the +symbol (see CLASSIFY-SYMBOL). + +E.g., completing \"mvb\" in a package that uses COMMON-LISP would +return something like: + + ((\"multiple-value-bind\" 26.588236 ((0 \"m\") (9 \"v\") (15 \"b\")) (:FBOUNDP :MACRO)) + ...) If STRING is package qualified the result list will also be qualified. If string is non-qualified the result strings are @@ -3489,57 +3496,50 @@ :limit limit :time-limit-in-msec time-limit-in-msec) 'list)) -(defun convert-fuzzy-completion-result (result converter + +(defun convert-fuzzy-completion-result (fuzzy-matching converter internal-p package-name) "Converts a result from the fuzzy completion core into something that emacs is expecting. Converts symbols to strings, fixes case issues, and adds information describing if the symbol is :bound, :fbound, a :class, a :macro, a :generic-function, a :special-operator, or a :package." - (destructuring-bind (symbol-or-name score chunks) result + (destructuring-bind (symbol score chunks) fuzzy-matching (multiple-value-bind (name added-length) (format-completion-result - (if converter - (funcall converter - (if (symbolp symbol-or-name) - (symbol-name symbol-or-name) - symbol-or-name)) - symbol-or-name) - internal-p package-name) - (list name score - (mapcar - #'(lambda (chunk) - ;; fix up chunk positions to account for possible - ;; added package identifier - (list (+ added-length (first chunk)) - (second chunk))) - chunks) - (loop for flag in '(:boundp :fboundp :generic-function - :class :macro :special-operator - :package) - if (if (symbolp symbol-or-name) - (case flag - (:boundp (boundp symbol-or-name)) - (:fboundp (fboundp symbol-or-name)) - (:class (find-class symbol-or-name nil)) - (:macro (macro-function symbol-or-name)) - (:special-operator - (special-operator-p symbol-or-name)) - (:generic-function - (typep (ignore-errors (fdefinition symbol-or-name)) - 'generic-function))) - (case flag - (:package (stringp symbol-or-name) - ;; KLUDGE: depends on internal - ;; knowledge that packages are - ;; brought up from the bowels of - ;; the completion algorithm as - ;; strings! - ))) - collect flag))))) + (funcall (or converter #'identity) (symbol-name symbol)) + internal-p package-name) + (list name + score + (mapcar #'(lambda (chunk) + ;; fix up chunk positions to account for possible + ;; added package identifier + (list (+ added-length (first chunk)) + (second chunk))) + chunks) + (classify-symbol symbol))))) + +(defun classify-symbol (symbol) + "Returns a list of classifiers that classify SYMBOL according +to its underneath objects (e.g. :BOUNDP if SYMBOL constitutes a +special variable.) The list may contain the following classification +keywords: :BOUNDP, :FBOUNDP, :GENERIC-FUNCTION, :CLASS, :MACRO, +:SPECIAL-OPERATOR, and/or :PACKAGE" + (check-type symbol symbol) + (let (result) + (when (boundp symbol) (push :boundp result)) + (when (fboundp symbol) (push :fboundp result)) + (when (find-class symbol nil) (push :class result)) + (when (macro-function symbol) (push :macro result)) + (when (special-operator-p symbol) (push :special-operator result)) + (when (find-package symbol) (push :package result)) + (when (typep (ignore-errors (fdefinition symbol)) + 'generic-function) + (push :generic-function result)) + result)) (defun fuzzy-completion-set (string default-package-name &key limit time-limit-in-msec) - "Prepares list of completion obajects, sorted by SCORE, of fuzzy + "Prepares list of completion objects, sorted by SCORE, of fuzzy completions of STRING in DEFAULT-PACKAGE-NAME. If LIMIT is set, only the top LIMIT results will be returned." (declare (type (or null (integer 0 #.(1- most-positive-fixnum))) limit time-limit-in-msec)) @@ -3547,23 +3547,28 @@ (parse-completion-arguments string default-package-name) (flet ((convert (vector &optional converter) (when vector - (loop for idx below (length vector) - for el = (aref vector idx) - do (setf (aref vector idx) (convert-fuzzy-completion-result - el converter internal-p package-name)))))) + (map-into vector + #'(lambda (fuzzy-matching) + (convert-fuzzy-completion-result fuzzy-matching converter + internal-p package-name)) + vector)))) (let* ((symbols (and package - (fuzzy-find-matching-symbols name - package - (and (not internal-p) - package-name) + (fuzzy-find-matching-symbols name package :time-limit-in-msec time-limit-in-msec - :return-converted-p nil))) - (packs (and (not package-name) + :external-only (and (not internal-p) + package-name)))) + (packages (and (not package-name) (fuzzy-find-matching-packages name))) (results)) - (convert symbols (completion-output-symbol-converter string)) - (convert packs) - (setf results (sort (concatenate 'vector symbols packs) #'> :key #'second)) + (convert symbols (completion-output-symbol-converter string)) + (convert packages #'(lambda (package-name) + (let ((converter (completion-output-package-converter string))) + ;; Present packages with a trailing colon for maximum convenience! + (concatenate 'string (funcall converter package-name) ":")))) + ;; Sort alphabetically before sorting by score. (Especially useful when + ;; STRING is empty, and SYMBOLS is a list of all possible completions.) + (setf results (sort (concatenate 'vector symbols packages) #'string-lessp :key #'first)) + (setf results (stable-sort results #'> :key #'second)) (when (and limit (> limit 0) (< limit (length results))) @@ -3572,59 +3577,62 @@ (setf results (make-array limit :displaced-to results)))) results)))) -(defun fuzzy-find-matching-symbols (string package external &key time-limit-in-msec return-converted-p) - "Return a list of symbols in PACKAGE matching STRING using the -fuzzy completion algorithm. If EXTERNAL is true, only external -symbols are returned." +(defun fuzzy-find-matching-symbols (string package &key external-only time-limit-in-msec) + "Returns a vector of fuzzy matchings (that is a list of the symbol in +PACKAGE that's matching STRING, its score, and a list of its completion +chunks), using the fuzzy completion algorithm. If EXTERNAL-ONLY is true, +only external symbols are considered." (let ((completions (make-array 256 :adjustable t :fill-pointer 0)) + (converter (completion-output-symbol-converter string)) (time-limit (if time-limit-in-msec (ceiling (/ time-limit-in-msec 1000)) 0)) (utime-at-start (get-universal-time)) - (count 0) - (converter (completion-output-symbol-converter string))) - (declare (type (integer 0 #.(1- most-positive-fixnum)) count time-limit) - (type function converter)) - (flet ((symbol-match (symbol converted) - (and (or (not external) - (symbol-external-p symbol package)) - (compute-highest-scoring-completion - string converted)))) - (block loop - (do-symbols* (symbol package) - (incf count) - (when (and (not (zerop time-limit)) - (zerop (mod count 256)) ; ease up on calling get-universal-time like crazy - (>= (- (get-universal-time) utime-at-start) time-limit)) - (return-from loop)) - (let* ((converted (funcall converter (symbol-name symbol))) - (result (if return-converted-p converted symbol))) - (if (string= "" string) - (when (or (and external (symbol-external-p symbol package)) - (not external)) - (vector-push-extend (list result 0.0 (list (list 0 ""))) completions)) - (multiple-value-bind (match-result score) (symbol-match symbol converted) - (when match-result - (vector-push-extend (list result score match-result) completions))))))) - completions))) + (count 0)) + (declare (type (integer 0 #.(1- most-positive-fixnum)) count time-limit)) + (declare (type function converter)) + (flet ((time-exhausted-p () + (and (not (zerop time-limit)) + (zerop (mod count 256)) ; ease up on calling get-universal-time like crazy + (incf count) + (>= (- (get-universal-time) utime-at-start) time-limit))) + (perform-fuzzy-match (string symbol-name) + (let ((converted-symbol-name (funcall converter symbol-name))) + (compute-highest-scoring-completion string converted-symbol-name)))) + (prog1 completions + (block loop + (do-symbols* (symbol package) + (when (time-exhausted-p) (return-from loop)) + (when (or (not external-only) (symbol-external-p symbol package)) + (if (string= "" string) + (vector-push-extend (list symbol 0.0 (list (list 0 ""))) completions) + (multiple-value-bind (match-result score) + (perform-fuzzy-match string (symbol-name symbol)) + (when match-result + (vector-push-extend (list symbol score match-result) completions))))))))))) (defun fuzzy-find-matching-packages (name) - "Return a list of package names matching NAME using the fuzzy -completion algorithm." + "Returns a vector of relevant fuzzy matchings (that is a list +consisting of a symbol representing the package that matches NAME, +its score, and its completions chunks.)" (let ((converter (completion-output-package-converter name)) (completions (make-array 32 :adjustable t :fill-pointer 0))) (declare ;;(optimize (speed 3)) (type function converter)) (loop for package in (list-all-packages) - for package-name = (concatenate 'string - (funcall converter - (package-name package)) - ":") - for (result score) = (multiple-value-list + for package-name = (package-name package) + for converted-name = (funcall converter package-name) + for package-symbol = (or (find-symbol package-name) + (make-symbol package-name)) ; INTERN'd be + for (result score) = (multiple-value-list ; too invasive. (compute-highest-scoring-completion - name package-name)) + name converted-name)) + ;; We return a symbol that represents the package, a) to make + ;; the type of the returned value consistent with the one of + ;; FUZZY-FIND-MATCHING-SYMBOLS, and b) to be able to call + ;; CLASSIFY-SYMBOL upon it later on. when result do - (vector-push-extend (list package-name score result) completions)) + (vector-push-extend (list package-symbol score result) completions)) completions)) (defslimefun fuzzy-completion-selected (original-string completion) @@ -3657,7 +3665,7 @@ "Finds the highest scoring way to complete the abbreviation SHORT onto the string FULL, using CHAR= as a equality function for letters. Returns two values: The first being the completion -chunks of the high scorer, and the second being the score." +chunks of the highest scorer, and the second being the score." (let* ((scored-results (mapcar #'(lambda (result) (cons (score-completion result short full) result))

4 8

[slime-devel] Some *PRINT-PRETTY* bindings in swank.lisp
by Madhu 11 Apr '07

11 Apr '07

Wanted to note two issues: 1. SWANK:INSPECT-OBJECT binds *PRINT-PRETTY* to NIL - This means that *PRINT-LINES* will not be obeyed. Which makes it impossible for the user to control the printing of huge strings in the inspector. Imagine slurping a huge file to a string, end emacs and lisp grinding this back and forth. The user ought to be have some mechanism to control the printing of strings in the inspector window, in particular to limit the output. It is preferable to allow this via *PRINT-LINES*. 2. SWANK::SAFE-CONDITION-MESSAGE binds *PRINT-PRETTY* to T - Typically SAFE-CONDITION-MESSAGE gets called in a dynamic scope where printer settings are bound to SWANK::*SLDB-PRINTER-BINDINGS*. which already sets *PRETTY-PRINT* to T (since an earlier patch from me. So at present explicitly binding this is redundant) If *PRINT-PRETTY* is non-NIL, *PRINT-LINES* are obeyed. And *SLDB-PRINTER-BINDINGS* sets *PRINT-LINES* to a non-NIL value by default. Which means that error messages reported when the condition is printed are truncated. The user ought to have a mechanism to control the printing of CONDITION reported strings in the debugger window, in particular to get the full output. It is preferable to allow this via *SLDB-PRINTER-BINDINGS*. The suggestion here is to remove the explicit binding, so the user's request to set *PRINT-PRETTY* to NIL is not overriden. Appreciate any comments, and do drop a note if I'm missing something. -- Madhu

1 0

[slime-devel] recording and compiling changed definitions
by Marco Baringer 10 Apr '07

10 Apr '07

attached is a patch which allows slime to track, list and recompile top level forms which have been edited since the last compilation. I've added a brief description to the manual, here's the executive summary: When the, customizable, variable slime-record-changed-definitions is non-nil we install before/after edit hooks which keep a list of all the toplevel forms that have been modified and their locations. The command slime-compile-changed-definitions is run, not bound to any key, we just recompile all the changed forms. The command slime-list-changed-definitions pops up a simplistic buffer with a list of all the forms which would be cempiled by a call to slime-compile-changed-definitions and the order in which they'd be compiled. comments welcome. i'll apply this shortly unless it breaks slimes someone's slimewhere slime-record-changed-definitions is NIL (iow i'm using this email to make sure the patch doesn't breax anything, i don't expect it to be 100% bug free or perfect from a UI stand point). -- -Marco Ring the bells that still can ring. Forget your perfect offering. There is a crack in everything. That's how the light gets in. -Leonard Cohen

10 19

[slime-devel] (re)connecting to swank server
by Andras Simon 09 Apr '07

09 Apr '07

Connecting to a swank server opened with (swank:create-server :port 4005 :dont-close t) I get the following: Debugger entered--Lisp error: (wrong-type-argument processp nil) set-process-filter(nil (lambda (process string) (slime-write-string string))) (lambda (&rest args) (set-process-filter (slime-inferior-process) (lambda ... ...)))() run-hooks(slime-connected-hook) (let ((args ...)) (when-let (name ...) (unless ... ...)) (slime-hide-inferior-lisp-buffer) (slime-init-output-buffer connection) (run-hooks (quote slime-connected-hook)) (when-let (fun ...) (funcall fun))) (let ((slime-dispatching-connection connection)) (destructuring-bind (&key pid style lisp-implementation machine features package version &allow-other-keys) info (or ... ... ... ...) (setf ... pid ... style ... features) (destructuring-bind ... package ...) (destructuring-bind ... lisp-implementation ...) (destructuring-bind ... machine ...)) (setq slime-state-name "") (let (...) (when-let ... ...) (slime-hide-inferior-lisp-buffer) (slime-init-output-buffer connection) (run-hooks ...) (when-let ... ...)) (message "Connected. %s" (slime-random-words-of-encouragement))) slime-set-connection-info(#<process SLIME Lisp> (:pid 6231 :style :spawn :lisp-implementation (:type "SBCL" :name "sbcl" :version "1.0.1.24") :machine (:instance "omnibus" :type "X86-64" :version "Dual Core AMD Opteron(tm) Processor 265") :features (:asdf :sb-thread :ansi-cl :common-lisp :sbcl :unix :sb-doc :sb-test :sb-ldb :sb-package-locks :sb-unicode :sb-eval :sb-source-locations :ieee-floating-point :x86-64 :elf :linux :gencgc :stack-grows-downward-not-upward :c-stack-is-control-stack :linkage-table :stack-allocatable-closures :alien-callbacks :os-provides-dlopen :os-provides-dladdr :os-provides-putwc) :package (:name "COMMON-LISP-USER" :prompt "CL-USER") :version "2007-04-09")) (lambda (G81915 info) (slime-set-connection-info (symbol-value G81915) info))(--proc-- (:pid 6231 :style :spawn :lisp-implementation (:type "SBCL" :name "sbcl" :version "1.0.1.24") :machine (:instance "omnibus" :type "X86-64" :version "Dual Core AMD Opteron(tm) Processor 265") :features (:asdf :sb-thread :ansi-cl :common-lisp :sbcl :unix :sb-doc :sb-test :sb-ldb :sb-package-locks :sb-unicode :sb-eval :sb-source-locations :ieee-floating-point :x86-64 :elf :linux :gencgc :stack-grows-downward-not-upward :c-stack-is-control-stack :linkage-table :stack-allocatable-closures :alien-callbacks :os-provides-dlopen :os-provides-dladdr :os-provides-putwc) :package (:name "COMMON-LISP-USER" :prompt "CL-USER") :version "2007-04-09")) apply((lambda (G81915 info) (slime-set-connection-info (symbol-value G81915) info)) --proc-- (:pid 6231 :style :spawn :lisp-implementation (:type "SBCL" :name "sbcl" :version "1.0.1.24") :machine (:instance "omnibus" :type "X86-64" :version "Dual Core AMD Opteron(tm) Processor 265") :features (:asdf :sb-thread :ansi-cl :common-lisp :sbcl :unix :sb-doc :sb-test :sb-ldb :sb-package-locks :sb-unicode :sb-eval :sb-source-locations :ieee-floating-point :x86-64 :elf :linux :gencgc :stack-grows-downward-not-upward :c-stack-is-control-stack :linkage-table :stack-allocatable-closures :alien-callbacks :os-provides-dlopen :os-provides-dladdr :os-provides-putwc) :package (:name "COMMON-LISP-USER" :prompt "CL-USER") :version "2007-04-09")) (lambda (&rest --cl-rest--) (apply (lambda ... ...) (quote --proc--) --cl-rest--))((:pid 6231 :style :spawn :lisp-implementation (:type "SBCL" :name "sbcl" :version "1.0.1.24") :machine (:instance "omnibus" :type "X86-64" :version "Dual Core AMD Opteron(tm) Processor 265") :features (:asdf :sb-thread :ansi-cl :common-lisp :sbcl :unix :sb-doc :sb-test :sb-ldb :sb-package-locks :sb-unicode :sb-eval :sb-source-locations :ieee-floating-point :x86-64 :elf :linux :gencgc :stack-grows-downward-not-upward :c-stack-is-control-stack :linkage-table :stack-allocatable-closures :alien-callbacks :os-provides-dlopen :os-provides-dladdr :os-provides-putwc) :package (:name "COMMON-LISP-USER" :prompt "CL-USER") :version "2007-04-09")) funcall((lambda (&rest --cl-rest--) (apply (lambda ... ...) (quote --proc--) --cl-rest--)) (:pid 6231 :style :spawn :lisp-implementation (:type "SBCL" :name "sbcl" :version "1.0.1.24") :machine (:instance "omnibus" :type "X86-64" :version "Dual Core AMD Opteron(tm) Processor 265") :features (:asdf :sb-thread :ansi-cl :common-lisp :sbcl :unix :sb-doc :sb-test :sb-ldb :sb-package-locks :sb-unicode :sb-eval :sb-source-locations :ieee-floating-point :x86-64 :elf :linux :gencgc :stack-grows-downward-not-upward :c-stack-is-control-stack :linkage-table :stack-allocatable-closures :alien-callbacks :os-provides-dlopen :os-provides-dladdr :os-provides-putwc) :package (:name "COMMON-LISP-USER" :prompt "CL-USER") :version "2007-04-09")) (if (symbol-value G81920) (funcall (symbol-value G81920) result)) (let* ((--cl-rest-- rand-81918) (result ...)) (if (symbol-value G81920) (funcall ... result))) (cond ((eql op-81917 ...) (let* ... ...)) ((eql op-81917 ...) (let* ... ... ...)) (t (error "Elisp destructure-case failed: %S" tmp-81919))) (let* ((tmp-81919 G81916) (op-81917 ...) (rand-81918 ...)) (cond (... ...) (... ...) (t ...))) (lambda (G81920 G81916) (let* (... ... ...) (cond ... ... ...)))(--cont-- (:ok (:pid 6231 :style :spawn :lisp-implementation (:type "SBCL" :name "sbcl" :version "1.0.1.24") :machine (:instance "omnibus" :type "X86-64" :version "Dual Core AMD Opteron(tm) Processor 265") :features (:asdf :sb-thread :ansi-cl :common-lisp :sbcl :unix :sb-doc :sb-test :sb-ldb :sb-package-locks :sb-unicode :sb-eval :sb-source-locations :ieee-floating-point :x86-64 :elf :linux :gencgc :stack-grows-downward-not-upward :c-stack-is-control-stack :linkage-table :stack-allocatable-closures :alien-callbacks :os-provides-dlopen :os-provides-dladdr :os-provides-putwc) :package (:name "COMMON-LISP-USER" :prompt "CL-USER") :version "2007-04-09"))) apply((lambda (G81920 G81916) (let* (... ... ...) (cond ... ... ...))) --cont-- (:ok (:pid 6231 :style :spawn :lisp-implementation (:type "SBCL" :name "sbcl" :version "1.0.1.24") :machine (:instance "omnibus" :type "X86-64" :version "Dual Core AMD Opteron(tm) Processor 265") :features (:asdf :sb-thread :ansi-cl :common-lisp :sbcl :unix :sb-doc :sb-test :sb-ldb :sb-package-locks :sb-unicode :sb-eval :sb-source-locations :ieee-floating-point :x86-64 :elf :linux :gencgc :stack-grows-downward-not-upward :c-stack-is-control-stack :linkage-table :stack-allocatable-closures :alien-callbacks :os-provides-dlopen :os-provides-dladdr :os-provides-putwc) :package (:name "COMMON-LISP-USER" :prompt "CL-USER") :version "2007-04-09"))) (lambda (&rest --cl-rest--) (apply (lambda ... ...) (quote --cont--) --cl-rest--))((:ok (:pid 6231 :style :spawn :lisp-implementation (:type "SBCL" :name "sbcl" :version "1.0.1.24") :machine (:instance "omnibus" :type "X86-64" :version "Dual Core AMD Opteron(tm) Processor 265") :features (:asdf :sb-thread :ansi-cl :common-lisp :sbcl :unix :sb-doc :sb-test :sb-ldb :sb-package-locks :sb-unicode :sb-eval :sb-source-locations :ieee-floating-point :x86-64 :elf :linux :gencgc :stack-grows-downward-not-upward :c-stack-is-control-stack :linkage-table :stack-allocatable-closures :alien-callbacks :os-provides-dlopen :os-provides-dladdr :os-provides-putwc) :package (:name "COMMON-LISP-USER" :prompt "CL-USER") :version "2007-04-09"))) slime-dispatch-event((:return (:ok (:pid 6231 :style :spawn :lisp-implementation ... :machine ... :features ... :package ... :version "2007-04-09")) 1) #<process SLIME Lisp>) slime-process-available-input(#<process SLIME Lisp>) slime-net-filter(#<process SLIME Lisp> "in-addr\" . 2) (\"with-vector-sap\" . 1) (\"with-sockaddr-for\" . 1) (\"with-alien-termios\" . 2) (\"with-alien-passwd\" . 2) (\"define-designator\" . 2) (\"with-alien-utimbuf\" . 2) (\"with-alien-stat\" . 2) (\"define-entry-point\" . 2) (\"with-dirent\" . 2) (\"with-alien-timeval\" . 2) (\"with-buffer-syntax\" . 1) (\"with-describe-settings\" . 1) (\"with-connection\" . 1) (\"defslimefun\" . 2) (\"do-symbols*\" . 1) (\"without-interrupts\" . 0) (\"with-reader-error-handler\" . 1) (\"with-temp-package\" . 1) (\"with-bindings\" . 1) (\"with-io-redirection\" . 1)))000294(:return (:ok (:pid 6231 :style :spawn :lisp-implementation (:type \"SBCL\" :name \"sbcl\" :version \"1.0.1.24\") :machine (:instance \"omnibus\" :type \"X86-64\" :version \"Dual Core AMD Opteron(tm) Processor 265\") :features (:asdf :sb-thread :ansi-cl :common-lisp :sbcl :unix :sb-doc :sb-test :sb-ldb :sb-package-locks :sb-unicode :sb-eval :sb-source-locations :ieee-floating-point :x86-64 :elf :linux :gencgc :stack-grows-downward-not-upward :c-stack-is-control-stack :linkage-table :stack-allocatable-closures :alien-callbacks :os-provides-dlopen :os-provides-dladdr :os-provides-putwc) :package (:name \"COMMON-LISP-USER\" :prompt \"CL-USER\") :version \"2007-04-09\")) 1)") But I also get a working *slime-repl*. However, after slime-disconnect, slime-connect gives error in process filter: Wrong type argument: processp, nil and I can only reconnect from a fresh emacs. After an unsuccessful connect, *slime-events* looks just like it does after a connection is established. Is slime-disconnect the correct way of disconnecting? GNU Emacs 22.0.50.1, slime is cvs head, the lisps I tried this with are sbcl 1.0.1.24 and LWL Personal 5.0.1. Andras

1 1

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 09 Apr '07

09 Apr '07

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1091 slime/ChangeLog:1.1105 --- slime/ChangeLog:1.1091 Fri Apr 6 12:06:44 2007 +++ slime/ChangeLog Sun Apr 8 18:56:32 2007 @@ -1,3 +1,232 @@ +2007-04-09 Marco Baringer <mb(a)bese.it> + + * swank.lisp (inspector-content-for-emacs): Look for refresh + keyword argument in :action links. + (inspect-whole-thing-action, inspect-show-more-action): Update for + new :action argument handling. + (inspect-for-emacs stream, inspect-for-emacs stream-error): Pass + :refresh nil on :action links. + (action-part-for-emacs): Set both lambda and refresh in the + *inspectee-actions* array. + (inspector-call-nth-action): *inspectee-actions* now holds both + the function and the boolean specifying whether to refresh or not. + + * swank-backend.lisp (inspect-for-emacs): Docstring update. + + * slime.el (slime-inspector-operate-on-point): Allow the action + calls to return nil. + +2007-04-08 Marco Baringer <mb(a)bese.it> + + * .cvsignore: Added *.lx64fsl (openmcl on linux fasls). + +2007-04-08 Marco Baringer <mb(a)bese.it> + + * swank.lisp (inspect-for-emacs): Added 'jump to source' action + for symbols in the new package-symbol browser. + +2007-04-08 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank.lisp: Implemented a new special inspector page for + displaying internal (external, &c) symbols that display + classification flags additionally to each symbol, similiar to the + content of a *Fuzzy Completion* buffer. Furthermore, added the + possibility to display all symbols that are /present/ in a + package. Combined with cleanup of the code parts in question. + + (symbol-status): New function. Returns the status of a symbol in a + given package (:internal, :external &c.) + + (symbol-external-p): Adapted to use new function SYMBOL-STATUS. + + (symbol-classification->string): New function. Converts a list of + classification flags into a concise string representation. + + (%package-symbols-container): New struct. We need a unique type to + dispatch in INSPECT-FOR-EMACS for the new inspector page, use this + as a wrapper structure. + + (inspect-for-emacs package): Reorganized to not cause too much eye + cancer; now with a saner maximum column width. Changed to make use + of new SYMBOL-STATUS, for code reuse. Also changed to make use of + new %PACKAGE-SYMBOLS-CONTAINER to let a new page pop up in Emacs + if the user wants to access the list of symbols of the package. + Added such a possibility to access all `present' symbols. + + (inspect-for-emacs %package-symbols-container): New method. + Displays all symbols wrapped up in the container structure + combined with their classification flags as determined by + CLASSIFY-SYMBOL. + +2007-04-08 Luís Oliveira <loliveira(a)common-lisp.net> + + * swank-backend.lisp (compute-sane-restarts): New interface. + * swank-clisp.lisp: Fix tabs and trailing whitespace. + (compute-sane-restarts): Implement new interface. + +2007-04-08 Takehiko Abe <keke(a)gol.com> + + * swank-openmcl.lisp (xref-locations): + +2007-04-08 Marco Baringer <mb(a)bese.it> + + * swank.lisp (fuzzy-completion-set): Use two check-type forms + instead of a place like (values limit time-limit-in-msec). While + sbcl seems to accept this form openmcl doesn't and it's not clear + from the spec that this is allowed. + +2007-04-07 Harald Hanche-Olsen <hanche(a)math.ntnu.no> + + * slime.el (sldb-mode-map): Added key definition for follow-link. + +2007-04-06 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank.lisp: Making fuzzy completion regard the time limit + correctly. Also make it properly use microseconds as time + granularity and inform the Emacs side if the time limit has + exhausted. Additionally, over all minor and cosmetic changes: + + (fuzzy-completions, fuzzy-completion-set): Returns now + additionally a flag indicating whether the time limit has + exhausted under the hood. Accomodated docstring accordingly. + + (fuzzy-create-completion-set): Changed to correctly catch and + propagate the remaining time limit to the actual match functions, + and return once time limit has exhausted. Some aesthetical code + reorganization. + + (get-real-time-in-msecs): New function. + + (fuzzy-find-matching-symbols, fuzzy-find-matching-packages): + Correctly regard the time limit. Use new function + GET-REAL-TIME-IN-MSECS for that purpose. Return the remaining + time limit as second value. + + * slime.el (slime-fuzzy-complete-symbol): Accomodated to deal with + the additionally returned flag of SWANK:FUZZY-COMPLETIONS. Pass + the flag by. + (slime-fuzzy-choices-buffer): Pass interruption flag by. + (slime-fuzzy-fill-completions-buffer): If time limit has exhausted + during completion retrieval, show an informational indication as + last entry in *Fuzzy Completion*. + (slime-fuzzy-last): New variable. To hold the last real completion + choice previous to the (possible) Time Limit Exhausted information. + (slime-fuzzy-next): Accomodated to not go beneath SLIME-FUZZY-LAST. + +2007-04-06 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank.lisp (tokenize-symbol, tokenize-symbol-thoroughly): + Previously these functions said a string representing a symbol is + internal exactly if it contained "::" as substring. Now they say + additionally so for symbols without any package identifier, as + they are internal to am implicit current package. (Otherwise + will break fuzzy completion.) + + (tokenize-symbol): Added docstring. + + * swank.lisp (format-completion-result): Fixed formation + for the case that PACKAGE-NAME is NIL but INTERNAL-P is T. + +2007-04-06 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank.lisp: Making fuzzy completion semantically right from a + user perspective. As an example on SBCL, "sb:with- C-c M-i" will + display all exported "with"-style macros in all sb-* packages from + now on. :) + + (parse-completion-arguments): Replacing with a semantically-sound + implementation, as the previous one was a bit confused. Clarifying + docstring. Adding commentary table of various constellations of + returned values for thorough explanation. + + (carefully-find-package): Removed. Obsolete by above change. + + (defstruct fuzzy-matching): Introduced to make internally-used + datastructure explicit. Distinguishing ``completion chunks'' + between those pertaining to the symbol itself and those to the + package identifier. + + (convert-fuzzy-completion-result): Renamed to + FUZZY-CONVERT-MATCHING-FOR-EMACS. + + (fuzzy-convert-matching-for-emacs): Accomodating for the new + datastructure. Only the chunks pertaining to the symbol itself are + fixed up positionally, the package-chunks are untouched. + Necessary for letting package identifiers be highlighted within + *Fuzzy Completions* in cases like "sb:with- C-c M-i." + + (fuzzy-completion-set): Taking out most code to become new + function FUZZY-CREATE-COMPLETION-SET. + + (fuzzy-create-completion-set): Doing all the hard work. Crux of + this changeset. so to speak. Largly rewritten to accomodate all + different cases of PARSE-COMPLETION-ARGUMENT. + + (fuzzy-find-matching-symbols, fuzzy-find-matching-packages): + Accomodating to new datatstructure FUZZY-MATCHING. Adapting + docstring accordingly. + + * swank-backend.lisp: Export WITH-STRUCT. + + * swank.lisp (eval-for-emacs, fuzzy-completions): + Various trivia like fixing spelling and indentation. + +2007-04-06 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime.el (slime-fuzzy-highlight-current-completion): Fix + off-by-one error that causes the currently selected + completion in the *Fuzzy Completion* buffer be highlighted + one char too far. + +2007-04-06 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank.lisp: Cleanup of parts of the fuzzy completion code. + Additionally a couple of enhancements. As follows: + + (fuzzy-completions, fuzzy-completion-selected): Minor + stylistic and clarifying modifications of the docstrings. + + (fuzzy-find-matching-symbols): Huge code reorganization. + Organizing relevant code into local function TIME-EXHAUSTED-P, + renaming local function SYMBOL-MATCH to PERFORM-FUZZY-MATCH, + making previously required argument EXTERNAL to new &key + argument :EXTERNAL-ONLY, clarifying docstring. + + (fuzzy-find-matching-packages): Making its return value + conformant to that of FUZZY-FIND-MATCHING-SYMBOLS, i.e. + instead of returning, among others, a package's name as + string, it now returns a symbol representing the package. + Accomodates the docstring accordingly. + + (fuzzy-completion-set): Minor typographical fix in docstring. + Changing local function CONVERT to use MAP-INTO instead of + doing it essentially manually. Accomodating to changes of + FUZZY-FIND-MATCHING-SYMBOLS, resp. -PACKAGES. + + (fuzzy-completion-set): Additional new feature: + The returned completions are sorted alphabetically by the + matched completion string before sorted by its score. + Affects especially the list of all possible completions when + the user hits fuzzy-completion on an empty string within Emacs; + also makes the potential limitness of the listed completions + clearer to the end user of SLIME. + + (classify-symbol): New function. Returns a list with keywords + that classifies a given symbol. (E.g. :BOUNDP, :MACRO &c) + Supersedes parts of CONVERT-FUZZY-COMPLETION-RESULT, + implementing them in a more straightforward and proper way; + removes prior KLUDGE in that part of the original function. + + (convert-fuzzy-completion-result): The above changes made + it possible to simplify this function drastically. Now uses + the newly introduced function CLASSIFY-SYMBOL. + + * slime.el: Minor stylistic changes. Additionally: + (slime-fuzzy-insert-completion-choice): + (slime-fuzzy-fill-completions-buffer) : Adding use of the + :PACKAGE classification flag returned by SWANK:FUZZY-COMPLETIONS. + This flag is called "p". + 2007-04-06 Neil Van Dyke <neil(a)neilvandyke.org> * slime.el (sldb-insert-frame): Added mouse-face to frame label

1 0

[slime-devel] Error handling missed in favor of next swank request
by Samium Gromoff 08 Apr '07

08 Apr '07

What i see here looks like a swank request which results in a lisp error signalled, does not cause entrance into the debugger in the context of that request, but rather causes a next request to be served, for some obscure reason. the .emacs snippet i invoke looks like that: (defun foo () (interactive) (slime-start :program "sbcl" :init-function (lambda () (cd "/some/where/") (slime-load-file "package.lisp") (slime-repl-set-package ":lispdb")))) I.e, we're talking about two swank requests here -- load-file, and then, once that completes, repl-set-package. when i run it, the following might or might not happen, due to circumstances i still did not have figured out: (note that the error appearing in frame 11 is expected -- it is not that error which is the problem -- the problem is that swank silently skips processing that request, and goes on with the next one (which causes another error)) The assertion (PACKAGEP SWANK::P) failed. [Condition of type SIMPLE-ERROR] Restarts: 0: [CONTINUE] Retry assertion. 1: [ABORT] Return to sldb level 1. 2: [ABORT] Return to SLIME's top level. 3: [ABORT] Exit debugger, returning to top level. Backtrace: 0: (SB-KERNEL:ASSERT-ERROR (PACKAGEP SWANK::P) NIL NIL) 1: (SWANK:SET-PACKAGE ":lispdb") 2: (SB-INT:SIMPLE-EVAL-IN-LEXENV (SWANK:SET-PACKAGE ":lispdb") #<NULL-LEXENV>) 3: ((LAMBDA ())) 4: ((LAMBDA (SWANK-BACKEND::HOOK SWANK-BACKEND::FUN)) #<FUNCTION SWANK:SWANK-DEBUGGER-HOOK> #<CLOSURE (LAMBDA #) {A6AD52D}>) 5: (SWANK::SLDB-LOOP 1) 6: (SWANK::CALL-WITH-BINDINGS ((*PRINT-PRETTY* . T) (*PRINT-LEVEL* . 4) (*PRINT-LENGTH* . 10) (*PRINT-CIRCLE* . T) (*PRINT-READABLY*) (*PRINT-PPRINT-DISPATCH* . #<SB-PRETTY:PPRINT-DISPATCH-TABLE {A7B9999}>) (*PRINT-GENSYM* . T) (*PRINT-BASE* . 10) (*PRINT-RADIX*) (*PRINT-ARRAY* . T) ...) #<FUNCTION (LAMBDA #) {ADFC5A5}>) 7: ((LAMBDA (SWANK-BACKEND::DEBUGGER-LOOP-FN)) #<FUNCTION (LAMBDA #) {ADFC555}>) 8: (SWANK::DEBUG-IN-EMACS #<SIMPLE-ERROR {B2EBAC9}>) 9: (INVOKE-DEBUGGER #<SIMPLE-ERROR {B2EBAC9}>) 10: (INVOKE-DEBUGGER #<SIMPLE-ERROR {B2EBAC9}>) 11: (ERROR "ASDF_ROOT is not set: where are the ASDF modules to be found?") 12: (SB-INT:SIMPLE-EVAL-IN-LEXENV (ERROR "ASDF_ROOT is not set: where are the ASDF modules to be found?") #<NULL-LEXENV>) 13: (SB-INT:SIMPLE-EVAL-IN-LEXENV (PROGN NIL (ERROR "ASDF_ROOT is not set: where are the ASDF modules to be found?")) #<NULL-LEXENV>) 14: (SB-INT:SIMPLE-EVAL-IN-LEXENV (UNLESS *ASDFROOT* (ERROR "ASDF_ROOT is not set: where are the ASDF modules to be found?")) #<NULL-LEXENV>) 15: (SB-FASL::LOAD-AS-SOURCE #<SB-SYS:FD-STREAM for "file /home/deepfire/source/lispdb/package.lisp" {B2E98A1}> NIL NIL) 16: (SB-FASL::INTERNAL-LOAD #P"/home/deepfire/source/lispdb/package.lisp" #P"/home/deepfire/source/lispdb/package.lisp" :ERROR NIL NIL :SOURCE :DEFAULT) 17: (SB-FASL::INTERNAL-LOAD #P"/home/deepfire/source/lispdb/package.lisp" #P"/home/deepfire/source/lispdb/package.lisp" :ERROR NIL NIL NIL :DEFAULT) 18: (LOAD "/home/deepfire/source/lispdb/package.lisp") 19: (SWANK:LOAD-FILE "/home/deepfire/source/lispdb/package.lisp") SLIME is about two weeks old, or so. regards, Samium Gromoff

2 1

[slime-devel] Re: *sldb sbcl/0* infinite loop?
by Jeffrey Cunningham 08 Apr '07

08 Apr '07

Jeffrey Cunningham wrote: > Hi, > > I have been running into a problem off and on for several weeks that > seems to come from an interaction between slime and SBCL resulting in an > infinite loop requiring me to killall sbcl to stop it. Until now I > haven't been able to isolate it to a simple case. Here's one: > > * I start up emacs (21.4.1) > * I start up M-x slime (2007-03-29). > * In the slime-repl I type the bogus command 'lt' <enter>. > > The speaker starts beeping and slime and emacs are utterly hung. I have > to kill sbcl which releases emacs to keyboard commands. > > When I hit C-x b a window opens up at the bottom of emacs which shows > the following: > > iswitch > {imagelib.lisp,*scratch*,*Messages*,*Compile-Log*,*slime-events*,*sldb > sbcl/0*,*sldb sbcl/0*<2>,*sldb sbcl/0*<3>,... > where the number inside the braces keeps incrementing forever > <4>,...,<5>... etc. > > I will attach a screen capture in case it isn't clear. I can't navigate > to that window and copy it. > > The steps I've taken to isolate the problem are to update sbcl to todays > version (1.0.4) compiled from source (from 1.0.3 - both act the same > way), and to remove my .sbclrc file. > > Any ideas how to debug this? > > Thanks > > --Jeff Cunningham > > ------------------------------------------------------------------------ > I have a feeling the attachment may have blocked this from reaching the list, so I'm resending it without the image. I also managed to get into the *Messages* buffer and here's what I found there: error in process filter: sldb-mode: Symbol's function definition is void: delay-mode-hooks error in process filter: Symbol's function definition is void: delay-mode-hooks ; pipelined request... (swank:debugger-info-for-emacs 0 10) error in process filter: sldb-mode: Symbol's function definition is void: delay-mode-hooks error in process filter: Symbol's function definition is void: delay-mode-hooks ; pipelined request... (swank:debugger-info-for-emacs 0 10) error in process filter: sldb-mode: Symbol's function definition is void: delay-mode-hooks This repeats over and over - 51 times. --Jeff

2 1

[slime-devel] [PATCH] remove CLISP's infamous ABORT restart from SLDB's restart list
by Luis Oliveira 08 Apr '07

08 Apr '07

Hello, I usually type 'a' when I want to pick the "Return to SLIME's top level." restart. On the rare occasions I use CLISP, it annoys me a lot that 'a' picks CLISP's ABORT request which returns to the REPL and kills SWANK. The attached patch should fix that. -- Luís Oliveira http://student.dei.uc.pt/~lmoliv/

2 1

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 07 Apr '07

07 Apr '07

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1088 slime/ChangeLog:1.1091 --- slime/ChangeLog:1.1088 Thu Mar 29 12:12:38 2007 +++ slime/ChangeLog Fri Apr 6 12:06:44 2007 @@ -1,3 +1,20 @@ +2007-04-06 Neil Van Dyke <neil(a)neilvandyke.org> + + * slime.el (sldb-insert-frame): Added mouse-face to frame label + and expression in Backtrace. + (sldb-insert-frames): Added mouse-face to "--more--" label in + Backtrace. + +2007-04-06 Michael Weber <michaelw+slime(a)foldr.org> + + * slime.el (slime-call-defun): insert the closing parenthesis for + the form. + +2007-04-06 Marco Baringer <mb(a)bese.it> + + * swank-openmcl.lisp (package swank-mop): Added + slot-makunbound-using-class. + 2007-03-29 Nikodemus Siivola <nikodemus(a)random-state.net> * swank-sbcl.lisp (swank-compile-string): save the original

1 0

[slime-devel] small patch for more mouse-face highlighting in Backtrace
by Neil Van Dyke 06 Apr '07

06 Apr '07

Hi. This is one small bit of user interface feedback that I really like to have in the Backtrace. Thanks, Neil cvs diff -u ChangeLog slime.el Index: ChangeLog =================================================================== RCS file: /project/slime/cvsroot/slime/ChangeLog,v retrieving revision 1.1076 diff -u -r1.1076 ChangeLog --- ChangeLog 26 Feb 2007 09:56:06 -0000 1.1076 +++ ChangeLog 1 Mar 2007 09:17:08 -0000 @@ -1,3 +1,10 @@ +2007-03-01 Neil Van Dyke <neil(a)neilvandyke.org> + + * slime.el (sldb-insert-frame): Added mouse-face to frame label + and expression in Backtrace. + (sldb-insert-frames): Added mouse-face to "--more--" label in + Backtrace. + 2007-02-26 Nikodemus Siivola <nikodemus(a)random-state.net> * swank.lisp (inspect-for-emacs): Add support for inspecting Index: slime.el =================================================================== RCS file: /project/slime/cvsroot/slime/slime.el,v retrieving revision 1.766 diff -u -r1.766 slime.el --- slime.el 25 Feb 2007 15:59:34 -0000 1.766 +++ slime.el 1 Mar 2007 09:17:19 -0000 @@ -8303,7 +8303,8 @@ sldb-previous-frame-number ,num point-entered sldb-fetch-more-frames start-open t - face sldb-section-face) + face sldb-section-face + mouse-face highlight) " --more--") (insert "\n")))) @@ -8313,10 +8314,11 @@ (destructuring-bind (number string) frame (let ((props `(frame ,frame sldb-default-action sldb-toggle-details))) (slime-propertize-region props - (insert " " (in-sldb-face frame-label (format "%2d:" number)) " ") - (slime-insert-possibly-as-rectangle - (slime-add-face (or face 'sldb-frame-line-face) - string)) + (slime-propertize-region '(mouse-face highlight) + (insert " " (in-sldb-face frame-label (format "%2d:" number)) " ") + (slime-insert-possibly-as-rectangle + (slime-add-face (or face 'sldb-frame-line-face) + string))) (insert "\n"))))) (defun sldb-fetch-more-frames (&rest ignore)

2 1

[slime-devel] (no subject)
by Michael Weber 06 Apr '07

06 Apr '07

Hi, attached is a tiny patch which inserts balanced parentheses for slime- call-defun (C-c C-y), to make it play better with paredit. Cheers, Michael

2 1