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

[slime-devel] Daily ChangeLog diff
by Helmut Eller 13 Sep '11

13 Sep '11

Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.491 slime/contrib/ChangeLog:1.492 --- slime/contrib/ChangeLog:1.491 Wed Aug 31 23:33:23 2011 +++ slime/contrib/ChangeLog Mon Sep 12 09:23:59 2011 @@ -1,3 +1,10 @@ +2011-09-12 Christophe Rhodes <csr21(a)cantab.net> + + * slime-media.el (slime-dispatch-media-event): allow swank to + popup a buffer with a given name, mode and contents. Intended to + support display of information formatted by the inferior process + as a return value: in particular, R help_files_with_topic objects. + 2011-09-01 Anton Kovalenko <anton(a)sw4me.com> * slime-c-p-c.el (slime-complete-symbol*-fancy-bit): "imitate" a

1 0

[slime-devel] Bothersome *slime-description* buffer
by Sebastian Tennant 09 Sep '11

09 Sep '11

Hi list, Recently the *slime-description* buffer started popping up every time a condition is signalled in SLIME's REPL (although I can see nothing about this in the ChangeLog file). --- *slime-repl sbcl* --- ; SLIME 2011-07-03 CL-USER> foo ; Evaluation aborted on #<UNBOUND-VARIABLE FOO {CE10179}>. CL-USER> ------------------------- --- *slime-description* --- The variable FOO is unbound. --------------------------- Try as I might, I cannot get used to this behaviour as it completely messes with my buffer/window configuration. Furthermore, I've had the function: (defun sldb-print-condition () (interactive) (unless sldb-condition (error "No condition known (wrong buffer?)")) (slime-write-string (format "%s\n%s\n" (first sldb-condition) (second sldb-condition)))) defined in my ~/.emacs since it was silently commented out in slime.el roughly a couple of years ago. It was activated by the following additional line in my ~/.emacs: (add-hook 'sldb-hook 'sldb-print-condition) and it performed the task of automatically printing the SLDB condition in SLIME's REPL after SLDB exits. There's no error, but this no longer works, i.e., the condition is no longer printed in the REPL. How can I prevent the *slime-description* buffer forn poppping up all the time and restore the functionality that I used to gain from adding 'sldb-print-condition to sldb-hook? Sebastian -- Emacs' AlsaPlayer - Music Without Jolts Lightweight, full-featured and mindful of your idyllic happiness. http://home.gna.org/eap

2 1

[slime-devel] find-unbalanced-parentheses functionality?
by UU UUUUU 06 Sep '11

06 Sep '11

ACL emacs has a function "fi:find-unbalanced-parentheses" that highlights the parenthesis in a region that is unbalanced. How do I do this in slime? Also, is there a way to indent a region or s-expression correctly (accounting for things like loop keywords etc.)?

2 1

[slime-devel] Restarting saved Lisp image.
by Andrew Myers 06 Sep '11

06 Sep '11

I'm wondering if there's a way to restart a Lisp image that already has Swank running. When using Franz's Emacs interaction package I can call dumplisp to save an image and then reload the image and continue running later. If I do this in slime the image hangs, it *appears* to me to be that the Swank server is running in the image and I would simply need a way to reconnect to it. I'm restarting the lisp image using this elisp function. It works with images built using the command line lisp. (defun run-image (image-name) (interactive "fImage name: ") (when (slime-connected-p) (slime-repl-quit)) (slime (concat inferior-lisp-program " -I " (expand-file-name image-name)))) How would I do this? slime-connect doesn't seem to do the trick but maybe I'm using it wrong. Thanks, Andrew

1 0

[slime-devel] Daily ChangeLog diff
by Helmut Eller 01 Sep '11

01 Sep '11

Index: slime/ChangeLog diff -u slime/ChangeLog:1.2212 slime/ChangeLog:1.2214 --- slime/ChangeLog:1.2212 Fri Aug 26 04:10:47 2011 +++ slime/ChangeLog Wed Aug 31 23:29:30 2011 @@ -1,3 +1,14 @@ +2011-08-31 Anton Kovalenko <anton(a)sw4me.com> + + * swank-sbcl.lisp (preferred-communication-style): check for + :sb-thread before :win32, so :spawn is preferred for threaded + Windows builds. + +2011-08-31 Nikodemus Siivola <nikodemus(a)random-state.net> + + * swank-sbcl.lisp (make-socket-io-stream): Backwards compatibility with + SBCL < 1.0.42.43 -- and bugfix. + 2011-08-26 Mark Evenson <evenson(a)panix.com> * swank-abcl.lisp (class-slots): Use exported symbol allowing Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.490 slime/contrib/ChangeLog:1.491 --- slime/contrib/ChangeLog:1.490 Tue Aug 30 08:34:10 2011 +++ slime/contrib/ChangeLog Wed Aug 31 23:33:23 2011 @@ -1,3 +1,11 @@ +2011-09-01 Anton Kovalenko <anton(a)sw4me.com> + + * slime-c-p-c.el (slime-complete-symbol*-fancy-bit): "imitate" a + close-paren, or a space, with exec-kbd-macro instead of inserting + them. Makes slime-complete-symbol*-fancy compatible with paredit + and probably other smart or "electric" stuff that could be bound + to these keys. + 2011-08-30 Nikodemus Siivola <nikodemus(a)random-state.net> * slime-repl.el (slime-repl-suppress-prompt): New variable. Bind to T to

1 0

[slime-devel] Daily ChangeLog diff
by Helmut Eller 31 Aug '11

31 Aug '11

Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.489 slime/contrib/ChangeLog:1.490 --- slime/contrib/ChangeLog:1.489 Fri Aug 5 01:05:47 2011 +++ slime/contrib/ChangeLog Tue Aug 30 08:34:10 2011 @@ -1,3 +1,11 @@ +2011-08-30 Nikodemus Siivola <nikodemus(a)random-state.net> + + * slime-repl.el (slime-repl-suppress-prompt): New variable. Bind to T to + temporarily suppress the prompt. + (slime-eval-last-expression-in-repl): New command. Sends form from + another buffer to REPL for evaluation. + (slime-mode-map): Bind slime-eval-last-expression-in-repl to C-c C-j. + 2011-08-05 Nikodemus Siivola <nikodemus(a)random-state.net> * swank-indentation.lisp (macro-indentation): Tweak so that

1 0

[slime-devel] Daily ChangeLog diff
by Helmut Eller 27 Aug '11

27 Aug '11

Index: slime/ChangeLog diff -u slime/ChangeLog:1.2211 slime/ChangeLog:1.2212 --- slime/ChangeLog:1.2211 Thu Aug 18 02:41:15 2011 +++ slime/ChangeLog Fri Aug 26 04:10:47 2011 @@ -1,3 +1,8 @@ +2011-08-26 Mark Evenson <evenson(a)panix.com> + + * swank-abcl.lisp (class-slots): Use exported symbol allowing + classes with non-standard metaobjects to be inspected. + 2011-08-18 Helmut Eller <heller(a)common-lisp.net> Fix line numbers for compiler notes if is narrowing was in effect.

1 0

[slime-devel] invalid protocol message / illegal sharp macro character
by Tamas Papp 25 Aug '11

25 Aug '11

When compiling a code snippet with C-c C-c, I get the following error in SLIME: Invalid protocol message: SB-INT:SIMPLE-READER-ERROR at 191 (line 3, column 41) on #<SB-IMPL::STRING-INPUT-STREAM {10049826B1}>: illegal sharp macro character: #\ "(:emacs-rex (swank:compile-string-for-emacs \"(defun show-pdf (pathname) (external-program:start \\\"xpdf\\\" (list (namestring pathname) \\\"-z page\\\"))) \" #(\"external.lisp\" 0 13 (buffer-name #2 buffer #<buffer external.lisp> font-lock-face buffer-menu-buffer mouse-face highlight help-echo \"mouse-2: select this buffer\")) (quote ((:position 1951) (:line 43 1))) \"/home/tamas/src/lisp/cl-pgf/external.lisp\" (quote nil)) \":cl-pgf\" t 88) " I pasted *slime-events* at http://paste.lisp.org/display/124254. C-c C-k compiles everything cleanly, there are no # characters in the entire file. I am using the latest SLIME from the repo, GNU Emacs 24.0.50.1 "(x86_64-pc-linux-gnu, GTK+ Version 2.24.5) of 2011-08-21 on keller, modified by Debian", and SBCL 1.0.50.0.debian. The error shows up with other version of SLIME, but not with Emacs23, and it is not specific to the particular code snippet. Thanks, Tamas

2 3

[slime-devel] (PREFERRED-COMMUNICATION-STYLE) on SBCL/Windows
by Anton Kovalenko 22 Aug '11

22 Aug '11

Hello SLIME developers, Please consider swapping COND clauses for :win32 and :sb-thread for SBCL backend's PREFERRED-COMMUNICATION-STYLE, in order to select :spawn communication style if :sb-thread is available. Like this: (defimplementation preferred-communication-style () (cond ((member :sb-thread *features*) :spawn) ;; fixme: when SBCL/win32 gains better select() support, remove ;; this. ((member :win32 *features*) nil) (t :fd-handler))) For now, it makes no difference for mainline SBCL users, but this change will be required when the thread support for Windows will be merged, anyway. (And, as you can see, the fact that no one but me is responsible for unofficial Windows fork's behavior won't protect you from questions and (maybe) bug reports for problems with SLIME + unofficial SBCL). What we have now is the code with a /meaning/ like "even if these guys will add threading to the windows port, it will surely be broken, so let's avoid using it". If the order of clauses were intentional, it would be a funny example of rude and offensive Common Lisp. -- Regards, Anton Kovalenko +7(916)345-34-02 | Elektrostal' MO, Russia

2 1

[slime-devel] Daily ChangeLog diff
by Helmut Eller 18 Aug '11

18 Aug '11

Index: slime/ChangeLog diff -u slime/ChangeLog:1.2209 slime/ChangeLog:1.2211 --- slime/ChangeLog:1.2209 Tue Aug 9 03:27:25 2011 +++ slime/ChangeLog Thu Aug 18 02:41:15 2011 @@ -1,3 +1,16 @@ +2011-08-18 Helmut Eller <heller(a)common-lisp.net> + + Fix line numbers for compiler notes if is narrowing was in effect. + + * slime.el (slime-canonicalized-location): Widen before calling + line-number-at-pos. + +2011-08-17 Stas Boukarev <stassats(a)gmail.com> + + * swank-allegro.lisp (with-redirected-y-or-n-p): Fix modern-mode + issues. + Patch by Andrew Myers. + 2011-08-09 Nikodemus Siivola <nikodemus(a)random-state.net> * swank-sbcl.lisp (make-socket-io-stream): Use :SERVE-EVENTS T when

1 0