August 2006 - slime-devel - mailman3.common-lisp.net

[slime-devel] swank-sbcl.lisp
by Donald Lindsay 15 Aug '06

15 Aug '06

Hi, ------- This is SBCL 0.9.5, an implementation of ANSI Common Lisp. More information about SBCL is available at <http://www.sbcl.org/>. (...) ; compiling file "/home/forwardarc/slime/swank-sbcl.lisp" (written 14 AUG 2006 10:49:18 PM): compilation aborted because of fatal error: READ failure in COMPILE-FILE: READER-ERROR at 16174 (line 444, column 70) on #<SB-SYS:FD- STREAM for "file /home/forwardarc/slime/swank-sbcl.lisp" {96F3869}>: Symbol "FIND-DEFINITION-SOURCES-BY-NAME" not found in the SB- INTROSPECT package. ----- When compiling a pull from cvs today, I run into this error on compilation. I have tried deleting old fasl files in my .slime directory. The specific symptom I see is a failure of slime to activate the debugger correctly. Aquamacs 0.9.9d tells me: ---------- debugger invoked on a SIMPLE-ERROR in thread #<THREAD "initial thread" {90032D1}>: SWANK-BACKEND:CALL-WITH-DEBUGGING-ENVIRONMENT not implementated Type HELP for debugger help, or (SB-EXT:QUIT) to exit from SBCL. restarts (invokable by number or by possibly-abbreviated name): 0: [ABORT-REQUEST] Abort handling SLIME request. 1: [ABORT ] Exit debugger, returning to top level. (SB-INT:EVAL-IN-LEXENV *DEFAULT-PATHNAMES-DEFAULTS* #<NULL-LEXENV>) 0] ----- Thanks, Don

2 1

[slime-devel] keep getting these messages with latest CVS of SBCL and Slime
by Lars Rune Nøstdal 14 Aug '06

14 Aug '06

In *Messages*-buffer: cond: Wrong type argument: integerp, 0.01 error in process filter: if: tag = slime-result-14-62004 eval-tags = nil sexp = (swank:completions "for" (quote "COMMON-LISP-USER")) error in process filter: tag = slime-result-14-62004 eval-tags = nil sexp = (swank:completions "for" (quote "COMMON-LISP-USER")) This seems to happen whenever Slime wants to show something in the status-buffer (minibuffer?), like argument-lists. I tried downgrading to Slime tagged as FAIRLY-STABLE, but this didn't help. I'm using the package `emacs-snapshot-gtk' (Emacs 22.0.50.1) under Ubuntu Dapper. Anyone else having troubles or know how to solve this?

2 3

[slime-devel] swank-openmcl.lisp broken?
by Ben Hyde 13 Aug '06

13 Aug '06

After many months I did a cvs up on my slime directory, and then as I was recompiling things I discovered swank-openmcl can't compile. The initial problem is that many of the defimplementation foo forms in there do return-from foo. I presume that somebody changed defimplementation do use an anonymous lambda rather than named function. After cobbling together a fix to that by injecting block foo forms here and there in swank-openmcl I then ran into some other problem I unwound this stack and went back to sbcl, where my real currently work is. - ben

1 0

[slime-devel] error running slime + openmcl
by Otto Diesenbacher 13 Aug '06

13 Aug '06

Hi slimers, :) after doing a cvs update I have problems running slime with openmcl. (having deleted ~/.slime and also tried a complete new cvs checkout) the error happens after trying to start slime: Welcome to OpenMCL Version 1.0-p060223 (DarwinPPC32)! ? ;Loading #P"/Users/okflo/lisp/slime/swank-loader.lisp"... ;Compiling "/Users/okflo/lisp/slime/swank-backend.lisp"... ;Loading #P"/Users/okflo/.slime/fasl/2006-08-11/openmcl-1.0-darwin-powerpc/swank-backend.dfsl"... ;Compiling "/Users/okflo/lisp/slime/nregex.lisp"... ;Loading #P"/Users/okflo/.slime/fasl/2006-08-11/openmcl-1.0-darwin-powerpc/nregex.dfsl"... ;Compiling "/Users/okflo/lisp/slime/metering.lisp"... ;Loading #P"/Users/okflo/.slime/fasl/2006-08-11/openmcl-1.0-darwin-powerpc/metering.dfsl"... ;Compiling "/Users/okflo/lisp/slime/swank-openmcl.lisp"... ;Compiler warnings for "/Users/okflo/lisp/slime/swank-openmcl.lisp" : ; Unused lexical variable STREAM, in an anonymous lambda form. > Error in process listener(1): While compiling an anonymous function : > Can't RETURN-FROM block : FRAME-VAR-VALUE. > Type :POP to abort. Type :? for other options. 1 > Unfortunately I am not an as expierenced lisp-user to suggest a fix by myself. I hope you can use this report and thanks for than incredible piece of software! br okflo -- Otto Diesenbacher mailto:diesenbacher@gmail.com Salzburg, Austria http://okflo.homelinux.net/

1 0

[slime-devel] auto-abbreviated-package-name
by Lars Brinkhoff 12 Aug '06

12 Aug '06

Sometimes I would like to include a version number in a package name, for example: "FOO.BAR-2.1". This isn't handled well by auto-abbreviated-package-name when *auto-abbreviate-dotted-packages* is true. Would it be a good idea to add some heuristic to handle this better? E.g., reject a short string consisting of only digits: (when *auto-abbreviate-dotted-packages* (let ((last-dot (position #\. (package-name package) :from-end t))) (when last-dot (let ((string (subseq (package-name package) (1+ last-dot)))) (when (or (> (length string) 2) (some (complement #'digit-char-p) string)) string))))) or exclude the last few charcters in the search for a period: (when *auto-abbreviate-dotted-packages* (let* ((name (package-name package)) (length (length name)) (last-dot (position #\. name :from-end t :end (- length 2)))) (when last-dot (subseq (package-name package) (1+ last-dot)))))

1 1

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 12 Aug '06

12 Aug '06

Index: slime/ChangeLog diff -u slime/ChangeLog:1.926 slime/ChangeLog:1.927 --- slime/ChangeLog:1.926 Thu Aug 10 15:00:50 2006 +++ slime/ChangeLog Fri Aug 11 12:29:11 2006 @@ -1,3 +1,15 @@ +2006-08-11 Helmut Eller <heller(a)common-lisp.net> + + * swank.lisp (close-connection, swank-error): Include backtraces + in our own errors. + (simple-serve-requests): Don't try to enter the + debugger if the connection is closed. + + * slime.el (disconnect): Test disconnecting. + + * swank-cmucl.lisp (startup-idle-and-top-level-loops): Initialize + MP only once. + 2006-08-10 Helmut Eller <heller(a)common-lisp.net> * swank-allegro.lisp (fspec-definition-locations): Improve

1 0

[slime-devel] Problem with SLIME & CLISP
by Lei Tang 12 Aug '06

12 Aug '06

I have just configured slime with xemacs associated with Clisp. It works fine when I switch to "slime mode" if I type M-x slime. However, when I exit xemacs, there's always coming up with a process "lisp.exe" which exhaust all the cpu resources and never ends. I have to kill it manually. My OS is windows xp. I installed Xemacs 21.4.19 and CLISP 2.39; the slime package is the latest cvs version. The following is the init file I used: ;============= SLIME Setup ==================================== (add-to-list 'load-path "D:/XEmacs/slime") ; your SLIME directory (setq inferior-lisp-program "D:/Clisp/clisp-2.39/clisp.exe -I") ; (require 'slime) (slime-setup) Any suggestions? Thanks a lot for your help! -Lei

2 2

Re: [slime-devel] Corman 2.51 (again)
by James Kirkland 11 Aug '06

11 Aug '06

From: Helmut Eller <heller(a)common-lisp.net> >> Am I the only one still using Corman? With the stable slime-2.0 (not >> CVS) Corman loads without error though there are a couple of >> warnings, so something's definitely changed. As I posted earlier, it >> breaks somewhere in swank.lisp but I'm too much of a n00b to tell >> where it's breaking. I could use some pointers. >Some of those arglist related tests fail for CormanLisp. I added a >workaround in the CVS version: it prints a message that the test >fails, but it continues instead of raising an error. Helmut, Thanks for the quick reply and code updates. However, I'm still getting an error, but this time there is no backtrace. See below: ?(load "c:/xemacs/slime/swank-loader.lisp") ;;; Warning: Unused variable G13342 in function COMPILE-FILES-IF-NEEDED-SERIALLY , File c:\xemacs\slime\swank-loader.lisp, line 136 ;; Loading c:\xemacs\slime\swank-backend.lisp ;; Loading c:\xemacs\slime\nregex.lisp ;; Loading c:\xemacs\slime\swank-corman.lisp ;; Loading c:\xemacs\slime\swank-gray.lisp ;; Loading c:\xemacs\slime\swank.lisp ;;; An error occurred in function STD-COMPUTE-EFFECTIVE-METHOD-FUNCTION: ;;; Error: No primary methods for the generic function #<Standard-Generic-Function DEFAULT-READTABLE-ALIST #x1076C2E0>. ;;; Entering Corman Lisp debug loop. ;;; Use :C followed by an option to exit. Type :HELP for help. ;;; Restart options: ;;; 1 Abort to top level. :c 1 ;;; Returning to top level loop. ?

2 5

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 11 Aug '06

11 Aug '06

Index: slime/ChangeLog diff -u slime/ChangeLog:1.924 slime/ChangeLog:1.926 --- slime/ChangeLog:1.924 Wed Aug 9 13:03:18 2006 +++ slime/ChangeLog Thu Aug 10 15:00:50 2006 @@ -1,3 +1,15 @@ +2006-08-10 Helmut Eller <heller(a)common-lisp.net> + + * swank-allegro.lisp (fspec-definition-locations): Improve + handling of (:internal ... n) like fspecs. + + * slime.el (slime-restart-inferior-lisp-aux): Remove the + interactive spec. + + * swank-backend.lisp (definterface): Drop that incredibly + unportable CLOS stuff. Use plists and plain functions instead. + Update backends accordingly. + 2006-08-09 Helmut Eller <heller(a)common-lisp.net> * slime.el (slime-find-filename-translators): CL:MACHINE-INSTANCE

1 0

[slime-devel] Questions about the Inspector documentation
by Jonathon McKitrick 10 Aug '06

10 Aug '06

I'm trying to figure out something in section as 3.2.8. The keys for the inspector do not seem to agree with what I can get to work. For example, it seems 'n' and 'SPC' are 'inverse' of 'l', but none of them see to do anything. Also, I don't see docs for 'p' but it does something similar to 'd'. Are there up-to-date docs that might explain these features? Jonathon McKitrick -- My other computer is your Windows box.

1 0

Re: [slime-devel] Corman 2.51 (again)
by James Kirkland 10 Aug '06

10 Aug '06

>A couple of people on the Corman forums have asked Roger about SLIME support, so >hopefully before 2.6 is released he will have a look at it and perhaps even >incorporate Espen's patches. But he's a busy guy, so don't hold your breath. Well, Roger's response was "we haven't had a chance to look at SLIME support yet" and since his target release date is "any day now" it doesn't look good. On the one hand, I believe part of the incompatibility issue is due to some non-standard stuff in CormanLisp, and so you might be tempted to think that the best way to fix this is to fix CormanLisp. On the other hand, SLIME is written specifically to support different Lisp implementations, not the other way around. If CormanLisp doesn't work with SLIME anymore, and it used to, then SLIME has broken something in it's support of CormanLisp. I would much rather see effort put into getting SLIME to work with 2.6 when it comes out later this month than to have valuable time wasted on 2.51 support. But chances are that whatever is causing SLIME to not work anymore with 2.51 will probably still be the case with 2.6. If it is fixed to work with 2.51 now, there will probably be much less to fix (if anything) with 2.6 later.

1 0

[slime-devel] Profiling worked once, then quit
by Jonathon McKitrick 10 Aug '06

10 Aug '06

Hi all, I tried slime-profile-package on one of my packages, and it worked fine. But every attempt after that, with this package and others failed. It says 'the variable NIL does not designate any package'. Looking at the backtrace, the variable 'package' is bound correctly at one point, and then the very next frame up, it is NIL. I'm running 0.9.14 SBCL with Aquamacs 0.9.9d/SLIME on a MacBook (Intel). Any help would be appreciated. Jonathon McKitrick -- My other computer is your Windows box.

1 0

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 10 Aug '06

10 Aug '06

Index: slime/ChangeLog diff -u slime/ChangeLog:1.923 slime/ChangeLog:1.924 --- slime/ChangeLog:1.923 Fri Jul 28 11:05:52 2006 +++ slime/ChangeLog Wed Aug 9 13:03:18 2006 @@ -1,3 +1,21 @@ +2006-08-09 Helmut Eller <heller(a)common-lisp.net> + + * slime.el (slime-find-filename-translators): CL:MACHINE-INSTANCE + can return nil. Silently accept that case for now. + + * swank.lisp (test-print-arglist): Print a message instead of + signalling an error. This should avoid startup problems, in + particular with CormanLisp. + (setup-stream-indirection): Disable it for now. We should fix it, + if there is a need for this functionality or just remove it. + + * swank-backend.lisp (definterface): Bring the old implementation + based on NO-APPLICABLE-METHOD back. It avoids lots of redefintion + warnings (but it creates more "noise" in backtraces). + + * swank-*.lisp (inspect-for-emacs): Don't use defimplementation + for real generics. + 2006-07-28 Helmut Eller <heller(a)common-lisp.net> * slime.el (slime-thread-quit): Call swank:quit-thread-browser.

1 0

[slime-devel] Corman 2.51 (again)
by John Stoneham 09 Aug '06

09 Aug '06

I've got a fresh install of Corman Lisp 2.51 on Windows XP SP2, and a fresh install of SLIME CVS. I followed the instruction on http://www.grumblesmurf.org/lisp/corman-patches to create a patched image, but it seems something may have changed in the SLIME CVS (again) because trying to load swank-loader.lisp breaks. Back in October 2005, there was a change in swank.lisp which required an updated patch to Corman Lisp (in misc-ansi.lisp I believe, posted on grumblesmurf.org) to get it to work again. Has something else now broken it once more? Here's the error I get (which probably doesn't help much): -------------------------- ;; Corman Lisp 2.51 Copyright (c) 2005 Corman Technologies. All rights reserved. ;; Unlicensed version: For evaluation and personal use only. ;; Some limitations apply. Type :quit to exit. ?(load "load-bits.lisp") 6 ?(load "c:/xemacs/slime/swank-loader.lisp") ;;; Warning: Unused variable G11641 in function COMPILE-FILES- IF-NEEDED-SERIALLY, File c:\xemacs\slime\swank-loader.lisp, line 136 ;;; Warning: No architecture feature found in (POWERPC PPC X86 X86-64 AMD64 I686 I586 I486 PC386 IAPX386 SPARC64 SPARC HPPA64 HPPA). ;; Loading c:\xemacs\slime\swank-backend.lisp ;; Loading c:\xemacs\slime\nregex.lisp ;; Loading c:\xemacs\slime\swank-corman.lisp ;; Loading c:\xemacs\slime\swank-gray.lisp ;; Loading c:\xemacs\slime\swank.lisp Warning: T is used as a key in (T (PRINC ARG)), assuming you mean (T) ;;; Warning: Unused variable ARGUMENT-FORMS in function COMPUTE- ENRICHED-DECODED-ARGLIST, File c:\xemacs\slime\swank.lisp, line 2014 ;;; An error occurred in function #< COMPILED-FUNCTION: #xDF16F0 >: ;;; Error: #<Error #xDF7B10> ;;; Entering Corman Lisp debug loop. ;;; Use :C followed by an option to exit. Type :HELP for help. ;;; Restart options: ;;; 1 Abort to top level. :c 1 ;;; Warning: Unused variable G14600 in function INSPECT-FOR-EMACS, File c:\xemacs\slime\swank.lisp, line 3924 ;;; Warning: Unused variable G14597 in function INSPECT-FOR-EMACS, File c:\xemacs\slime\swank.lisp, line 3924 ;;; Warning: Unused variable G14596 in function INSPECT-FOR-EMACS, File c:\xemacs\slime\swank.lisp, line 3924 ;;; Warning: the following function(s) are called from forms ;;; in file "#P"c:\xemacs\slime\swank.lisp"" but have not yet been defined: ;;; SERVE ;;; WITHOUT-DEBUGGER-HOOK ;;; MAKE-BROADCAST-STREAM ;;; PRINT-ARG ;;; COMPUTE-APPLICABLE-METHODS ;;; STREAM-ERROR-STREAM ;;; Warning: These Swank interfaces are unimplemented: (ACTIVATE-STEPPING ADD-FD-HANDLER ADD-SIGIO-HANDLER CALLS-WHO DISASSEMBLE-FRAME INTERRUPT-THREAD LIST-CALLEES LIST-CALLERS PROFILE PROFILE-PACKAGE PROFILE-REPORT PROFILE-RESET PROFILED-FUNCTIONS REMOVE-FD-HANDLERS REMOVE-SIGIO-HANDLERS RESTART-FRAME RETURN-FROM-FRAME SLDB-BREAK-AT-START SLDB-BREAK-ON-RETURN TOGGLE-TRACE UNPROFILE WHO-BINDS WHO-CALLS WHO-MACROEXPANDS WHO-REFERENCES WHO-SETS WHO-SPECIALIZES) 21 ?

3 2

[slime-devel] minor patch to
by Gary King 09 Aug '06

09 Aug '06

The patch below alters the lisp-version-string to accommodate 64 bit Allegro and 64 bit OpenMCL. Any comments... Index: swank-loader.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank-loader.lisp,v retrieving revision 1.59 diff -u -w -r1.59 swank-loader.lisp --- swank-loader.lisp 13 Apr 2006 10:41:59 -0000 1.59 +++ swank-loader.lisp 7 Aug 2006 14:30:26 -0000 @@ -66,12 +66,15 @@ #+scl (lisp-implementation-version) #+sbcl (lisp-implementation-version) #+ecl (lisp-implementation-version) - #+openmcl (format nil "~d.~d" + #+openmcl (format nil "~d.~d~@[-~d~]" ccl::*openmcl-major-version* - ccl::*openmcl-minor-version*) + ccl::*openmcl-minor-version* + #+ppc64-target 64 + #-ppc64-target nil) #+lispworks (lisp-implementation-version) #+allegro (concatenate 'string (if (eq 'h 'H) "A" "M") ; ANSI vs MoDeRn - excl::*common-lisp-version-number*) + excl::*common-lisp-version-number* + #+64bit "-64") #+clisp (let ((s (lisp-implementation-version))) (subseq s 0 (position #\space s))) #+armedbear (lisp-implementation-version) thanks, -- Gary Warren King metabang.com [http://www.metabang.com/] (413) 885 9127 * (206) 338-4052 [Fax] gwking on #lisp (occasionally)

2 4

[slime-devel] Removing some allegro warnings
by Gary King 07 Aug '06

07 Aug '06

The patch below wraps #'excl:without-redefinition-warnings around a number of forms that cause allegro to complain... Any comments? Index: swank-allegro.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank-allegro.lisp,v retrieving revision 1.89 diff -u -w -r1.89 swank-allegro.lisp --- swank-allegro.lisp 28 Jul 2006 15:04:53 -0000 1.89 +++ swank-allegro.lisp 7 Aug 2006 14:32:44 -0000 @@ -28,8 +28,9 @@ ;;;; TCP Server +(excl:without-redefinition-warnings (defimplementation preferred-communication-style () - :spawn) + :spawn)) (defimplementation create-socket (host port) (socket:make-socket :connect :passive :local-port port @@ -64,40 +65,48 @@ (setf (stream-external-format stream) (find-external-format external-format))) +(excl:without-redefinition-warnings (defimplementation format-sldb-condition (c) - (princ-to-string c)) + (princ-to-string c))) +(excl:without-redefinition-warnings (defimplementation condition-references (c) (declare (ignore c)) - '()) + '())) +(excl:without-redefinition-warnings (defimplementation call-with-syntax-hooks (fn) - (funcall fn)) + (funcall fn))) ;;;; Unix signals +(excl:without-redefinition-warnings (defimplementation call-without-interrupts (fn) - (excl:without-interrupts (funcall fn))) + (excl:without-interrupts (funcall fn)))) (defimplementation getpid () (excl::getpid)) +(excl:without-redefinition-warnings (defimplementation lisp-implementation-type-name () - "allegro") + "allegro")) +(excl:without-redefinition-warnings (defimplementation set-default-directory (directory) (let* ((dir (namestring (truename (merge-pathnames directory))))) (setf *default-pathname-defaults* (pathname (excl:chdir dir))) - dir)) + dir))) +(excl:without-redefinition-warnings (defimplementation default-directory () - (namestring (excl:current-directory))) + (namestring (excl:current-directory)))) ;;;; Misc +(excl:without-redefinition-warnings (defimplementation arglist (symbol) (handler-case (excl:arglist symbol) - (simple-error () :not-available))) + (simple-error () :not-available)))) (defimplementation macroexpand-all (form) (excl::walk form)) @@ -129,8 +138,9 @@ (:class (describe (find-class symbol))))) +(excl:without-redefinition-warnings (defimplementation make-stream-interactive (stream) - (setf (interactive-stream-p stream) t)) + (setf (interactive-stream-p stream) t))) ;;;; Debugger @@ -578,14 +588,15 @@ (when doc `("Documentation:" (:newline) ,doc)))))) -(defmethod inspect-for-emacs ((o t) (inspector acl-inspector)) - inspector - (values "A value." (allegro-inspect o))) - +#+(or) (defmethod inspect-for-emacs ((o function) (inspector acl-inspector)) inspector (values "A function." (allegro-inspect o))) +(defmethod inspect-for-emacs ((o t) (inspector acl-inspector)) + inspector + (values "A value." (allegro-inspect o))) + (defmethod inspect-for-emacs ((o standard-object) (inspector acl- inspector)) inspector (values (format nil "~A is a standard-object." o) (allegro- inspect o))) @@ -613,8 +624,9 @@ ;;;; Multithreading +(excl:without-redefinition-warnings (defimplementation initialize-multiprocessing () - (mp:start-scheduler)) + (mp:start-scheduler))) (defimplementation spawn (fn &key name) (mp:process-run-function name fn)) @@ -632,21 +644,26 @@ (find id mp:*all-processes* :key (lambda (p) (getf (mp:process-property-list p) 'id)))) +(excl:without-redefinition-warnings (defimplementation thread-name (thread) - (mp:process-name thread)) + (mp:process-name thread))) +(excl:without-redefinition-warnings (defimplementation thread-status (thread) (format nil "~A ~D" (mp:process-whostate thread) - (mp:process-priority thread))) + (mp:process-priority thread)))) +(excl:without-redefinition-warnings (defimplementation make-lock (&key name) - (mp:make-process-lock :name name)) + (mp:make-process-lock :name name))) +(excl:without-redefinition-warnings (defimplementation call-with-lock-held (lock function) - (mp:with-process-lock (lock) (funcall function))) + (mp:with-process-lock (lock) (funcall function)))) +(excl:without-redefinition-warnings (defimplementation current-thread () - mp:*current-process*) + mp:*current-process*)) (defimplementation all-threads () (copy-list mp:*all-processes*)) @@ -654,8 +671,9 @@ (defimplementation interrupt-thread (thread fn) (mp:process-interrupt thread fn)) +(excl:without-redefinition-warnings (defimplementation kill-thread (thread) - (mp:process-kill thread)) + (mp:process-kill thread))) (defvar *mailbox-lock* (mp:make-process-lock :name "mailbox lock")) @@ -751,8 +769,10 @@ ;;;; Weak hashtables +(excl:without-redefinition-warnings (defimplementation make-weak-key-hash-table (&rest args) - (apply #'make-hash-table :weak-keys t args)) + (apply #'make-hash-table :weak-keys t args))) +(excl:without-redefinition-warnings (defimplementation make-weak-value-hash-table (&rest args) - (apply #'make-hash-table :values :weak args)) + (apply #'make-hash-table :values :weak args))) thanks, -- Gary Warren King metabang.com [http://www.metabang.com/] (413) 885 9127 * (206) 338-4052 [Fax] gwking on #lisp (occasionally)

1 0

[slime-devel] swank format-iso8601-time bug in non-integer timezones
by Bob Halley 07 Aug '06

07 Aug '06

(resending, since I sent this from the wrong Sender id the first time...) In format-iso8601-time in swank.lisp, the following code is used to compute the timezone offset to print: (multiple-value-bind (h m) (truncate (abs zone) 1.0) ;; Tricky. Sign of time zone is reversed in ISO 8601 ;; relative to Common Lisp convention! (format nil "~:[+~;-~]~2,'0D:~2,'0D" (> zone 0) h (round m))) (round m) does the wrong thing in timezones which are not integers. E.g. at the moment I am in Adelaide, Australia, and the timezone is +09:30, which corresponds to a Common Lisp timezone value of -19/2. As written the code prints my timezone as +09:00. Rather than (round m), we want (round (* m 60)). Regards, /Bob P.S. Slime rocks, thanks! :)

1 0

[slime-devel] patch for editing method definitions from inspector
by Earl J. Wagner 01 Aug '06

01 Aug '06

This fixes a problem in the inspector where I'd hit M-. on the method name, then have to choose the method even though the method specializers were right there. Also, the previous patch is for keyword symbol completion. I'm not sure if my first (incorrect) patch was posted to the list or blocked. It it arrives, ignore it. -Earl Index: slime.el =================================================================== RCS file: /project/slime/cvsroot/slime/slime.el,v retrieving revision 1.635 diff -u -r1.635 slime.el --- slime.el 24 Jul 2006 14:01:03 -0000 1.635 +++ slime.el 1 Aug 2006 22:22:03 -0000 @@ -6319,12 +6319,15 @@ (:type list)) dspec location) + (defun slime-edit-definition (name &optional where) "Lookup the definition of the name at point. If there's no name at point, or a prefix argument is given, then the function name is prompted." (interactive (list (slime-read-symbol-name "Name: "))) - (let ((definitions (slime-eval `(swank:find-definitions-for- emacs ,name)))) + (let* ((method-context (slime-extract-method-context name)) + (full-name (if method-context (format "%s" method-context) name)) + (definitions (slime-eval `(swank:find-definitions-for- emacs ,full-name)))) (cond ((null definitions) (if slime-edit-definition-fallback-function @@ -6801,15 +6804,7 @@ ;;a setf-definition, but which? (backward-up-list 1) (slime-parse-context `(setf ,name))) - ((slime-in-expression-p '(defmethod *)) - (unless (looking-at "\\s ") - (forward-sexp 1)) ; skip over the methodname - (let (qualifiers arglist) - (loop for e = (read (current-buffer)) - until (listp e) do (push e qualifiers) - finally (setq arglist e)) - `(:defmethod ,name ,@qualifiers - ,(slime-arglist-specializers arglist)))) + ((slime-in-expression-p '(defmethod *)) (slime-extract- method-context name)) ((and (symbolp name) (slime-in-expression-p `(,name))) ;; looks like a regular call @@ -6829,6 +6824,22 @@ (t name)))) +(defun slime-extract-method-context (name) + (save-excursion + (let ((method-p (when (slime-in-expression-p '(method *)) 'method)) + (defmethod-p (when (slime-in-expression-p '(defmethod *)) 'defmethod))) + (when (or method-p defmethod-p) + (unless (looking-at "\\s ") + (forward-sexp 1)) ; skip over the methodname + (let (qualifiers arglist) + (loop for e = (read (current-buffer)) + until (listp e) do (push e qualifiers) + finally (setq arglist e)) + (if method-p + `(method ,name ,@qualifiers ,arglist) + `(:defmethod ,name ,@qualifiers + ,(slime-arglist-specializers arglist)))))))) + (defun slime-in-expression-p (pattern) "A helper function to determine the current context. The pattern can have the form:

1 0

[slime-devel] keyword symbol completion
by Earl J.Wagner 01 Aug '06

01 Aug '06

Ugh. Here's the fixed patch that works for symbols without specified packages as well. Index: swank.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank.lisp,v retrieving revision 1.387 diff -u -r1.387 swank.lisp --- swank.lisp 24 Jul 2006 14:01:15 -0000 1.387 +++ swank.lisp 1 Aug 2006 19:32:03 -0000 @@ -1270,7 +1270,9 @@ ;; FIXME: deal with #\| etc. hard to do portably. (defun tokenize-symbol (string) (let ((package (let ((pos (position #\: string))) - (if pos (subseq string 0 pos) nil))) + (cond + ((and pos (= pos 0)) "keyword") + (pos (subseq string 0 pos))))) (symbol (let ((pos (position #\: string :from-end t))) (if pos (subseq string (1+ pos)) string))) (internp (search "::" string))) @@ -3089,6 +3091,7 @@ (defun format-completion-result (string internal-p package-name) (let ((prefix (cond (internal-p (format nil "~A::" package-name)) + ((equal package-name "keyword") ":") (package-name (format nil "~A:" package-name)) (t "")))) (values (concatenate 'string prefix string)

1 0

[slime-devel] keyword symbol completion
by Earl J. Wagner 01 Aug '06

01 Aug '06

Hi all, Here is a patch to complete keyword symbols. -Earl Index: swank.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank.lisp,v retrieving revision 1.387 diff -u -r1.387 swank.lisp --- swank.lisp 24 Jul 2006 14:01:15 -0000 1.387 +++ swank.lisp 1 Aug 2006 19:32:03 -0000 @@ -1270,7 +1270,9 @@ ;; FIXME: deal with #\| etc. hard to do portably. (defun tokenize-symbol (string) (let ((package (let ((pos (position #\: string))) - (if pos (subseq string 0 pos) nil))) + (cond + ((= pos 0) "keyword") + (pos (subseq string 0 pos))))) (symbol (let ((pos (position #\: string :from-end t))) (if pos (subseq string (1+ pos)) string))) (internp (search "::" string))) @@ -3089,6 +3091,7 @@ (defun format-completion-result (string internal-p package-name) (let ((prefix (cond (internal-p (format nil "~A::" package-name)) + ((equal package-name "keyword") ":") (package-name (format nil "~A:" package-name)) (t "")))) (values (concatenate 'string prefix string)

1 0

[slime-devel] Problem using paredit mode with REPL
by Ralf Mattes 01 Aug '06

01 Aug '06

Hello list, i'm trying to use paredit-mode in Slime's repl but unfortunately enabling paredit mode results in a dysfunctional repl -- pressing 'enter'/'return' while the pointer is after the closing sexpression only inserts a newline but doesn't send the expression to the lisp process. BTW, i tried both supported types of slime-repl-return-behavior. Cheers Ralf Mattes

2 1