January 2008 - slime-devel - mailman3.common-lisp.net

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 11 Jan '08

11 Jan '08

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1265 slime/ChangeLog:1.1266 --- slime/ChangeLog:1.1265 Wed Jan 9 12:08:59 2008 +++ slime/ChangeLog Thu Jan 10 10:30:24 2008 @@ -1,3 +1,9 @@ +2008-01-10 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime.el (slime-delete-and-extract-region): New + function. Portable version of `delete-and-extract-region' which + returned NIL instead of "", as experienced by Matthias Koeppe. + 2008-01-09 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> * slime.el (slime-repl-mode-map): Bind C-c C-t to Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.71 slime/contrib/ChangeLog:1.77 --- slime/contrib/ChangeLog:1.71 Wed Jan 9 19:43:14 2008 +++ slime/contrib/ChangeLog Thu Jan 10 15:23:41 2008 @@ -1,5 +1,75 @@ 2008-01-10 Tobias C. Rittweiler <tcr(a)freebits.de> + * slime-parse.el (slime-make-form-spec-from-string): Correctly + handle quoted things and other non-proper "(...)" forms. + + * swank-arglist.lisp (read-form-spec): Added assertion against + receiving junk form specs from Emacs. + +2008-01-10 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime-editing-commands.el (slime-close-all-parens-in-sexp): Use + new portability function `slime-delete-and-extract-region'. + +2008-01-10 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank-parse.lisp (slime-incomplete-form-at-point): Hopefully + better fix than before. + +2008-01-10 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> + + Add keyboard commands (starting with C-c C-v) and a top-level menu + for presentation-related commands. Add a command (C-c C-v M-o) to + forget all objects associated with presentations, without + clearing the REPL buffer. + + * slime-presentations.el + (slime-presentation-around-or-before-point-or-error): New + function. + (slime-inspect-presentation): New function, factored out from + slime-inspect-presentation-at-mouse. + (slime-inspect-presentation-at-mouse): Use it here. + (slime-inspect-presentation-at-point): New command. + (slime-copy-presentation-to-repl): New function, factored out + from slime-copy-presentation-at-mouse. + (slime-copy-presentation-at-mouse-to-repl): Renamed from + slime-copy-presentation-at-mouse; use the new function + slime-copy-presentation-to-repl. + (slime-copy-presentation-at-point-to-repl): New command. + (slime-copy-presentation-to-kill-ring): New function, factored + out from slime-copy-presentation-at-mouse-to-kill-ring. + (slime-copy-presentation-at-point-to-kill-ring): New command. + (slime-describe-presentation): New function, factored out from + slime-describe-presentation-at-mouse. + (slime-describe-presentation-at-mouse): Use it here. + (slime-describe-presentation-at-point): New command. + (slime-pretty-print-presentation): New function, factored out + from slime-pretty-print-presentation-at-mouse. + (slime-pretty-print-presentation-at-mouse): Use it here. + (slime-pretty-print-presentation-at-point): New command. + (slime-mark-presentation): New command. + (slime-previous-presentation, slime-next-presentation): New + commands. + (slime-presentation-command-map, slime-presentation-bindings): + New variables. + (slime-presentation-init-keymaps): New function. + (slime-presentation-around-or-before-point-p): New function. + (slime-presentation-easy-menu): New variable. + (slime-presentation-add-easy-menu): New function. + (slime-clear-presentations): Make interactive, remove + presentation markup from all presentations in the REPL buffer. + (slime-presentations-init): Call slime-presentation-init-keymaps + and slime-presentation-add-easy-menu. + +2008-01-10 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank-parse.lisp (slime-incomplete-form-at-point): Take the + arglist index the user's point is located at correctly into + account. Previously `C-c C-s' on `(defun |foo' would have inserted + `args body...)', now it inserts `name args body...)' + +2008-01-10 Tobias C. Rittweiler <tcr(a)freebits.de> + * swank-arglists.lisp (read-form-spec): Changed "cons" clause to "list" clause in etypecase. Fix for error on arglist display on `(declare (ftype (|)))', | being point. @@ -13,7 +83,7 @@ * swank-fuzzy.el: Updated some comments. (fuzzy-generate-matchings): Sort package matchings before traversal, such that they're traversed in the order of their - score. (Important with time limit exhausts during traversal.) + score. (Important when time limit exhausts during traversal.) 2008-01-09 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de>

1 0

[slime-devel] source locations for clisp
by Andy Chambers 10 Jan '08

10 Jan '08

Hi, How does one define where slime should look for function definitions. The cliki page for slime shows how to do this for sbcl/cmucl but I can't find anywhere showing how to do it with clisp. It seems like maybe it somehow knows automatically but something is not working correctly. For example. I have cells loaded, and I'm working in the cells package, but when I put the cursor on defmodel and hit M-., I get the following error. PARSE-NAMESTRING: argument #1=(SYSTEM::DEFUN/DEFMACRO #P"C:\\Program Files\\clisp-2.42\\systems\\cells\\defmodel.fas" 20 123) should be a pathname designator #2=(OR STRING FILE-STREAM PATHNAME) [Condition of type SIMPLE-TYPE-ERROR] Cheers, Andy

2 1

[slime-devel] Behaviour of slime-inspect in slime-cvs
by Baishampayan Ghose 10 Jan '08

10 Jan '08

Hello, I have been using slime-2.0 all this while and I tried out slime-cvs recently. I discovered that C-c I (slime-inspect) doesn't inspect the object under the point any more as it used to do up to slime-2.0 I found the feature to be extremely useful as I could move up and down anywhere in the REPL buffer and inspect any object. Now I am limited to inspecting only * though **** and variables. In my humble opinion the old behaviour of slime-inspect should be in slime-cvs too; or there should at least be _some_ way to trivially enable the old behaviour. Regards, BG -- Baishampayan Ghose (concatenate 'string (mapcar #'code-char (reverse '(109 111 99 46 108 105 97 109 103 64 101 115 111 104 103 46 98))))

2 1

[slime-devel] M-x slime-version
by Terrence Brannon 10 Jan '08

10 Jan '08

Shouldnt there be a command to display the slime-version?

2 1

[slime-devel] slime-complete-form broken
by Matthias Koeppe 10 Jan '08

10 Jan '08

Hi Tobias, I noticed that slime-complete-form is broken in SLIME CVS. Since you did the most recent changes to arglist-related parsing, I hope you can fix it. Create this Lisp buffer: --------------------------------------------------------- (defun (defun foo (a) a) --------------------------------------------------------- Then move point after the first "defun" and press C-c C-s. SLIME now inserts "varlist body...)" and then signals "Wrong type argument: stringp, nil". It should have inserted "name varlist body...)" instead (it should also not signal an error). Apparently the "defun foo" form is taken as an actual argument to the first "defun" form and consumes the formal argument "name". This is wrong. For slime-complete-form, the parser should ignore everything that is beyond the token around point, as in earlier versions of the code (I remark it is NOT sufficient to just stop before the next top-level expression). The wrong-type-arg error is occuring in code related to the automatic closing of parens. Thanks, Matthias -- Matthias Köppe -- http://www.math.uni-magdeburg.de/~mkoeppe

2 8

[slime-devel] executing a lisp file from emacs
by Dave Pawson 10 Jan '08

10 Jan '08

New slime user. Running sbcl, emacs 21, FC8 I have a lisp file x.lisp $sbcl --load dovetailCalc.lisp and the program runs fine. I *think* C-c C-k should execute the program. >From the screen information it is simply compiled. http://common-lisp.net/project/slime/doc/html/Lisp-Evaluation.html#Lisp-Eva… Tried C-M-x from within the lisp buffer and I get an error report The function SELFTEST is undefined. [Condition of type UNDEFINED-FUNCTION] Seems it fails to find the top level definition? I guess I'm doing something silly, but I can't figure out what. Any help appreciated. regards -- Dave Pawson XSLT XSL-FO FAQ. http://www.dpawson.co.uk

2 2

[slime-devel] slime and utf-8
by Угодай n/a 10 Jan '08

10 Jan '08

hi, all. I have such the linux system: ~% uname -a Linux rys 2.6.23-ARCH #1 SMP PREEMPT Fri Dec 21 20:26:31 CET 2007 x86_64 AMD Athlon(tm) 64 X2 Dual Core Processor 4400+ AuthenticAMD GNU/Linux ~% locale LANG=ru_RU.UTF-8 LC_CTYPE="ru_RU.UTF-8" LC_NUMERIC="ru_RU.UTF-8" LC_TIME="ru_RU.UTF-8" LC_COLLATE=C LC_MONETARY="ru_RU.UTF-8" LC_MESSAGES="ru_RU.UTF-8" LC_PAPER="ru_RU.UTF-8" LC_NAME="ru_RU.UTF-8" LC_ADDRESS="ru_RU.UTF-8" LC_TELEPHONE="ru_RU.UTF-8" LC_MEASUREMENT="ru_RU.UTF-8" LC_IDENTIFICATION="ru_RU.UTF-8" LC_ALL= ~% pacman -Qs emacs local/emacs-cvs 20071029-1 Gnu Emacs from CVS local/slime-cvs 20070115-3 The Superior Lisp Interaction Mode for Emacs ~% pacman -Qs sbcl local/sbcl 1.0.12-1 Steel Bank Common Lisp My emacs work with MULE and I wrote in ~/.emacs: ;;lisp ;;; The SBCL binary and command-line arguments (setq inferior-lisp-program "/usr/bin/sbcl --noinform") ;;slime (setq slime-net-coding-system 'utf-8-unix) (add-to-list 'load-path "/usr/share/emacs/site-lisp/slime") (require 'slime) (slime-setup) (slime-setup :autodoc t) but if I try to run the such code (pushnew #p"/home/ugoday/.sbcl/systems/uffi/" asdf:*central-registry*b. :test #'equal) (asdf:operate 'asdf:load-op :uffi) (pushnew #p"/home/ugoday/.sbcl/systems/curl/" asdf:*central-registry* :test #'equal) (asdf:operate 'asdf:load-op :curl) (curl:with-connection-returning-string (:cookies t) (curl:set-option :url "http://ya.ru") (curl:perform)) slime tells me c-string decoding error (:external-format :UTF-8): the octet sequence 2 cannot be decoded. [Condition of type SB-INT:C-STRING-DECODING-ERROR] what can I do to fix this?

2 1

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 10 Jan '08

10 Jan '08

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1264 slime/ChangeLog:1.1265 --- slime/ChangeLog:1.1264 Thu Jan 3 22:34:34 2008 +++ slime/ChangeLog Wed Jan 9 12:08:59 2008 @@ -1,3 +1,11 @@ +2008-01-09 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> + + * slime.el (slime-repl-mode-map): Bind C-c C-t to + slime-toggle-trace-fdefinition (as in Lisp buffers) instead of + slime-repl-clear-buffer. This binding is useful for untracing + functions directly from the trace output. Move + slime-repl-clear-buffer to the keybinding C-c M-o. + 2008-01-04 Juho Snellman <jsnell(a)iki.fi> * swank-sbcl.lisp (source-file-source-location): Use the Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.69 slime/contrib/ChangeLog:1.71 --- slime/contrib/ChangeLog:1.69 Sun Dec 30 06:32:06 2007 +++ slime/contrib/ChangeLog Wed Jan 9 19:43:14 2008 @@ -1,3 +1,27 @@ +2008-01-10 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank-arglists.lisp (read-form-spec): Changed "cons" clause to + "list" clause in etypecase. Fix for error on arglist display on + `(declare (ftype (|)))', | being point. + +2008-01-10 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime-fuzzy.el (slime-fuzzy-completion-time-limit-in-msec): + Update docstring: Its value isn't rounded to nearest second, but + is really interpreted as msecs. + + * swank-fuzzy.el: Updated some comments. + (fuzzy-generate-matchings): Sort package matchings before + traversal, such that they're traversed in the order of their + score. (Important with time limit exhausts during traversal.) + +2008-01-09 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> + + Restore support for Scheme programs that was removed from core + SLIME on 2007-09-19, as a "slime-scheme" contrib. + + * slime-scheme.el: New file. + 2007-12-30 Tobias C. Rittweiler <tcr(a)freebits.de> * swank-arglists.lisp: Fix for `(cerror "FOO" 'type-error ...)'

1 0

[slime-devel] [Patch] Rebind C-c C-t in REPL to slime-toggle-trace-fdefinition
by Matthias Koeppe 09 Jan '08

09 Jan '08

Unless there are objections, I will make the change below in SLIME CVS. 2007-12-12 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> * slime.el (slime-repl-mode-map): Bind C-c C-t to slime-toggle-trace-fdefinition (as in Lisp buffers) instead of slime-repl-clear-buffer. This binding is useful for untracing functions directly from the trace output. --- slime.el.~1.882.~ 2007-12-11 17:20:50.000000000 +0100 +++ slime.el 2007-12-12 13:57:41.509816000 +0100 @@ -2873,7 +2873,7 @@ ("\C-c\C-w" slime-who-map) ("\C-\M-x" 'slime-eval-defun) ("\C-c\C-o" 'slime-repl-clear-output) - ("\C-c\C-t" 'slime-repl-clear-buffer) + ("\C-c\C-t" 'slime-toggle-trace-fdefinition) ("\C-c\C-u" 'slime-repl-kill-input) ("\C-c\C-n" 'slime-repl-next-prompt) ("\C-c\C-p" 'slime-repl-previous-prompt) Matthias -- Matthias Köppe -- http://www.math.uni-magdeburg.de/~mkoeppe

3 4

[slime-devel] [RfC] Changing behaviour of `C-w' in Isearch
by Tobias C. Rittweiler 08 Jan '08

08 Jan '08

Hi, I often find it quite annoying that `C-w' during Isearch just inserts the word at point; the reason is that variable identifiers in Lisp contain non-word constituents very often. E.g. If point is at *special-variables*, you'd have to press `C-w' three times to eventually get it inserted as the search string. The following will make `C-w' respect symbol constituents as well as word constituents when isearching in Lisp buffers, such that one `C-w' will suffice for gripping special variables. (defun isearch-yank-symbolic-word-or-char () (interactive) (isearch-yank-internal (lambda () (let ((distance (skip-syntax-forward "w_"))) (when (zerop distance) (forward-char 1)) (point))))) (add-hook 'lisp-mode-hook (lambda () (make-local-variable 'isearch-mode-map) (define-key isearch-mode-map "\C-w" 'isearch-yank-symbolic-word-or-char))) I consider this generally useful, and wanted to ask if I should commit something like that to Slime? -T.

2 1

[slime-devel] [BUG] Inspecting LPNs with SLIME
by Michael Weber 05 Jan '08

05 Jan '08

Inspecting logical pathnames with SLIME's inspector appears to fail. Test case: (setf (logical-pathname-translations "FOO") '(("FOO:**;*.*.*" "/foo/**/*.*"))) C-c I #p"FOO:bar;" RET #<SB-KERNEL:LOGICAL-HOST "FOO"> fell through ETYPECASE expression. Wanted one of (CONS STRING NULL). [Condition of type SB-KERNEL:CASE-FAILURE] Backtrace: 0: (SWANK::INSPECTOR-CONTENT-FOR-EMACS ("Namestring" #1=": " (:VALUE "FOO:BAR;") #2=(:NEWLINE) "Physical pathname: " #1# (:VALUE #P"/foo/bar/") #2# "Host: " #<SB-KERNEL:LOGICAL-HOST "FOO"> ...)) 1: (SWANK::INSPECT-OBJECT #P"FOO:BAR;" #<SWANK::FANCY-INSPECTOR {15EB7981}>) [...] Cheers, Michael

1 0

[slime-devel] Re: Bug#457648: fails to compile when using with slime
by Luca Capello 05 Jan '08

05 Jan '08

reassign 457648 cl-swank found 457648 1:20070927-3 found 457648 1:20071202-1 tags 457648 + upstream thanks Hello! Cc:ing the SLIME-devel mailing list because it's an upstream bug. On Mon, 24 Dec 2007 08:50:10 +0100, anoop aryal wrote: > when using cl-arnesi with slime, > > (require :arnesi) > > causes: > > ;; loading #P"/var/cache/common-lisp-controller/1000/sbcl/swank/fasl/sbcl-1.0.12.debian-linux-x86/swank-arglists.fasl" > WARNING: change in instance length of class SB-PRETTY:PRETTY-STREAM: > current length: 27 > compile time length: 25 > ;; > ;; Error while loading: /var/cache/common-lisp-controller/1000/sbcl/swank/fasl/sbcl-1.0.12.debian-linux-x86/swank-arglists.fasl > ;; Condition: The class SB-PRETTY:PRETTY-STREAM was not changed, and there's no guarantee that > ;; the loaded code (which expected another layout) will work. > ;; Aborting. > ;; ; > ; compilation unit aborted > ; caught 1 fatal ERROR condition This is reproducible on my sid-amd64 with a simple (require 'swank) in SLIME and SBCL_1:1.0.12.0-1 on emacs-snapshot_1:20080101-1 (GNU Emacs 23.0.50.1). For this reason it's not an arnesi bug, but a SLIME one, reassigning. The bug doesn't happen with SLIME_1:20070628-1, but only starting with version 1:20070927-3, which unconditionally loads all the fancy stuff: the WARNING happears with contrib/swank-presentation-streams.lisp. Thx, bye, Gismo / Luca

1 0

[slime-devel] How to 'start again'
by Dave Pawson 04 Jan '08

04 Jan '08

After running a program a few times (C-c C-k) the output is not longer showing the program output. Possibly due to errors in the program that I'm developing (compile errors show etc). Is there a way to 'clear' out the buffers and start again? I.e. to clean the lisp environment and restart? TIA -- Dave Pawson XSLT XSL-FO FAQ. http://www.dpawson.co.uk

4 6

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 04 Jan '08

04 Jan '08

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1263 slime/ChangeLog:1.1264 --- slime/ChangeLog:1.1263 Wed Jan 2 11:02:47 2008 +++ slime/ChangeLog Thu Jan 3 22:34:34 2008 @@ -1,3 +1,13 @@ +2008-01-04 Juho Snellman <jsnell(a)iki.fi> + + * swank-sbcl.lisp (source-file-source-location): Use the + debootstrap readtable when appropriate (fixes occasional reader + errors when using "v" on debugger frames that point to functions + defined in SBCL). Likewise for the debootstrapping packages. + (code-location-debug-source-name): Ensure that we always return a + physical namestring, Emacs won't like a pathname or a logical + namestring. + 2008-01-02 Luís Oliveira <loliveira(a)common-lisp.net> Use sane default values for slime-repl-set-package.

1 0

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 03 Jan '08

03 Jan '08

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1260 slime/ChangeLog:1.1263 --- slime/ChangeLog:1.1260 Sat Dec 22 08:24:49 2007 +++ slime/ChangeLog Wed Jan 2 11:02:47 2008 @@ -1,3 +1,29 @@ +2008-01-02 Luís Oliveira <loliveira(a)common-lisp.net> + + Use sane default values for slime-repl-set-package. + + Previously, when typing `,!p' at the REPL, the current package + would have been inserted as a default (although the whole intent + was to /change/ the current package in the first place), now + nothing is inserted anymore. + + * slime.el (slime-pretty-current-package): rename it to + slime-pretty-find-buffer-package and make it use + slime-find-buffer-package instead of slime-current-package. + (slime-repl-set-package, slime-set-package): use new function. + +2008-01-02 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime.el (slime-print-apropos): Simplified: Don't insert action + properties anymore for the symbol; they were ignored anyway, + because `apropos-follow' (bound to RET in the resulting + *SLIME Apropos* buffer) looks for buttons only. + +2008-01-02 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime.el (slime-apropos): Update docstring: Apropos doesn't + match on regular expressions anymore since 2007-11-24. + 2007-12-22 Douglas Crosher <dcrosher(a)common-lisp.net> * swank-scl.lisp (set-stream-timeout, make-socket-io-stream): update

3 3

[slime-devel] documentation
by Jean-Christophe Helary 03 Jan '08

03 Jan '08

I was looking for a comprehensive user manual for (Lisp beginners with) Slime (with Ready Lisp) and found that the HTML as well as the PDF available from the project page were quite outdated. I took the texi file from CVS but makeinfo says it can't build it since contributors.texi is missing. ... slime.texi:2633: @include `contributors.texi': No such file or directory. ... Is there a way to help here ? Jean-Christophe Helary ------------------------------------ http://mac4translators.blogspot.com/

1 0

[slime-devel] [PATCH] Don't use a default value in slime-repl-set-package
by Luis Oliveira 02 Jan '08

02 Jan '08

Hello, A recent change on 2007-11-29, "Prettify package names for slime-repl-set-package", broke the way this function works by making it use slime-current-package instead of slime-find-buffer-package as its default value. The attached patch should fix that. -- Luís Oliveira http://student.dei.uc.pt/~lmoliv/

2 1