slime-devel

slime-devel@common-lisp.net

4231 discussions

[slime-devel] return-from-frame for CMUCL
by Jan Rychter 04 Sep '04

04 Sep '04

The enclosed tiny patch enables support for return-from-frame ("R" in sldb or DEBUG-RETURN in CMUCL debugger) in CMUCL. You need the latest (09/2004) CMUCL snapshot and optimization set to (debug 3) to use it. --J. ChangeLog entry: 2004-09-05 Jan Rychter <jan(a)rychter.com> * swank-cmucl.lisp (return-from-frame): Add it. ------------------------------------------------------------------------- Index: swank-cmucl.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank-cmucl.lisp,v retrieving revision 1.113 diff -u -3 -p -u -r1.113 swank-cmucl.lisp --- swank-cmucl.lisp 28 Aug 2004 02:27:08 -0000 1.113 +++ swank-cmucl.lisp 2 Sep 2004 16:35:30 -0000 @@ -1511,6 +1511,14 @@ A utility for debugging DEBUG-FUNCTION-A (defimplementation frame-catch-tags (index) (mapcar #'car (di:frame-catches (nth-frame index)))) +;; This requires a custom patch to CMUCL at the moment, hopefully this +;; will be incorporated into CMUCL soon. +(defimplementation return-from-frame (index form) + (let* ((frame (nth-frame index)) + (probe (di::find-debug-tag-for-frame frame))) + (cond (probe (throw (car probe) (eval-in-frame form index))) + (t (format nil "Cannot return from frame: ~S" frame))))) + (defun set-step-breakpoints (frame) (when (di:debug-block-elsewhere-p (di:code-location-debug-block (di:frame-code-location frame)))

1 0

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 02 Sep '04

02 Sep '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.511 slime/ChangeLog:1.513 --- slime/ChangeLog:1.511 Tue Aug 31 00:34:24 2004 +++ slime/ChangeLog Thu Sep 2 08:00:38 2004 @@ -1,3 +1,32 @@ +2004-09-02 Wolfgang Mederle <wolfgang.mederle(a)stud.uni-muenchen.de> + + * swank-loader.lisp (*lisp-name*): Replace / with - in CMUCL + version strings. + +2004-09-01 John Paul Wallington <jpw(a)gnu.org> + + * slime.el (slime-oneliner): Don't use free variable. + (slime-recenter-window, slime-set-connection-info) + (slime-pprint-event, slime-compiler-notes-quit) + (slime-apropos-summary): Likewise. + (slime-connect): Tidy up handshake `message' call. + +2004-09-01 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> + + * slime.el (slime-repl-push-directory): Fix interactive spec. + (sldb-reference-properties): Take a the reference object as + argument instead of its parts. Fix callers accordingly. + (slime-fuzzy-choices-buffer): Remove assignment to unused variable + slime-fuzzy-target-mtime. + (slime-ed): Replace call to new-frame with make-frame. + (sldb-find-buffer): Cleanup. + (sldb-highlight-sexp): Fix regexp. It's now almost a full line. + + * swank.lisp (completion-set, tokenize-symbol-designator) + (tokenize-completion, fuzzy-completion-set) + (briefly-describe-symbol-for-emacs): Remove simple-base-string + declarations. + 2004-08-30 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> * PROBLEMS: We require SBCL 0.8.13. 0.8.12 is no longer

1 0

[slime-devel] Bug in directory creation
by Wolfgang Mederle 02 Sep '04

02 Sep '04

Hi, I encountered a bug in a fresh CVS slime with the latest cmucl binary for Darwin: CL-USER> (lisp-implementation-version) "19a+ Darwin/Exp 2004-07-25-090" ^ That's the problem. I'd like to suggest this change in swank-loader.lisp: ----- 8< ------------ --- swank-loader.lisp.~1.31.~ Fri Aug 6 02:43:35 2004 +++ swank-loader.lisp Thu Sep 2 02:32:22 2004 @@ -41,7 +41,7 @@ ))) (defparameter *lisp-name* - #+cmu (format nil "cmu-~A" (lisp-implementation-version)) + #+cmu (format nil "cmu-~A" (substitute #\- #\/ (lisp-implementation-version))) #+sbcl (format nil "sbcl-~A" (lisp-implementation-version)) #+openmcl "openmcl" #+lispworks (format nil "lispworks-~A" (lisp-implementation-version)) ------------- >8 ------------ (I hope, the above format is correct, it's the first time I try to send a patch.) Regards, Wolfgang -- Wolfgang Mederle wolfgang.mederle(a)stud.uni-muenchen.de * GPG key id E8EF7E7F <URL:http://www.mederle.de/> ICQ# 1435333 * AIM therealmadearl

2 1

[slime-devel] minor slime.el nits
by John Paul Wallington 01 Sep '04

01 Sep '04

Here's a few minor nits: 2004-09-01 John Paul Wallington <jpw(a)gnu.org> * slime.el (slime-oneliner): Don't use free variable. (slime-recenter-window): Likewise. (slime-connect): Tidy up handshake `message' call. Index: slime.el =================================================================== RCS file: /project/slime/cvsroot/slime/slime.el,v retrieving revision 1.395 diff -u -r1.395 slime.el --- slime.el 30 Aug 2004 22:23:53 -0000 1.395 +++ slime.el 1 Sep 2004 08:46:40 -0000 @@ -965,9 +965,9 @@ (defun slime-oneliner (string) "Return STRING truncated to fit in a single echo-area line." - (substring msg 0 (min (length msg) - (or (position ?\n msg) most-positive-fixnum) - (1- (frame-width))))) + (substring string 0 (min (length string) + (or (position ?\n string) most-positive-fixnum) + (1- (frame-width))))) ;; Interface (defun slime-set-truncate-lines () @@ -1178,7 +1178,7 @@ (message "Connecting to Swank on port %S.." port) (let* ((process (slime-net-connect host port)) (slime-dispatching-connection process)) - (message "Initial handshake..." port) + (message "Initial handshake...") (slime-setup-connection process) (slime-hide-inferior-lisp-buffer) (message "Connected. %s" (slime-random-words-of-encouragement)))) @@ -5947,7 +5947,7 @@ (start (save-excursion (loop repeat line do (forward-line -1)) (point)))) - (set-window-start w start))) + (set-window-start window start))) (defun sldb-highlight-sexp (&optional start end) "Highlight the first sexp after point."

2 1

[slime-devel] simple-base-string declarations
by Christophe Rhodes 31 Aug '04

31 Aug '04

Hi, In a number of places in swank.lisp various strings are declared as SIMPLE-BASE-STRING. Is there a reason for this declaration -- in particular, is it expected that these strings will be specialized to hold only BASE-CHAR? Or are they in fact general strings? (at least the version in BRIEFLY-DESCRIBE-SYMBOL-FOR-EMACS leads me to suspect that they are fairly general strings). If I'm right, could the declaration be changed to reflect this, maybe to SIMPLE-STRING (or even STRING, if they're not in performance-critical areas)? It is likely that sbcl, in the near future, will have multiple string representations, where SIMPLE-BASE-STRING means (SIMPLE-ARRAY BASE-CHAR (*)), which will be distinct from (SIMPLE-ARRAY CHARACTER (*)) [ they're currently the same in all known implementations, I believe]. Cheers, Christophe -- http://www-jcsu.jesus.cam.ac.uk/~csr21/ +44 1223 510 299/+44 7729 383 757 (set-pprint-dispatch 'number (lambda (s o) (declare (special b)) (format s b))) (defvar b "~&Just another Lisp hacker~%") (pprint #36rJesusCollegeCambridge)

2 2

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 31 Aug '04

31 Aug '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.510 slime/ChangeLog:1.511 --- slime/ChangeLog:1.510 Sat Aug 28 04:27:08 2004 +++ slime/ChangeLog Tue Aug 31 00:34:24 2004 @@ -1,3 +1,19 @@ +2004-08-30 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> + + * PROBLEMS: We require SBCL 0.8.13. 0.8.12 is no longer + supported. + + * swank-allegro.lisp (find-fspec-location): Catch errors in + excl:source-file. + + * swank.lisp (send-to-socket-io): Add some ignore declarations. + + * slime.el (sldb-fetch-all-frames, sldb-end-of-backtrace) + (sldb-beginning-of-backtrace): New commands. + (slime-search-suppressed-forms): Change the start regexp so that + reader conditionals in single line comments, like "; #+foo", are + ignored. + 2004-08-27 Peter Seibel <peter(a)javamonkey.com> * swank-backend.lisp (swank-compile-string): Add directory @@ -20,8 +36,8 @@ 2004-08-18 Matthew Danish <mrd+nospam(a)cmu.edu> - * slime/swank-allegro.lisp (swank-compile-string): Use #\; instead - of #\: to separate the position from the buffer-name. This avoids + * swank-allegro.lisp (swank-compile-string): Use #\; instead of + #\: to separate the position from the buffer-name. This avoids troubles on Windows. 2004-08-16 Luke Gorrie <luke(a)bluetail.com>

1 0

[slime-devel] MOP access
by Rui Patrocínio 31 Aug '04

31 Aug '04

Hi, Is there any access to MOP-like functionality in swank? I've not found it so I create a simple interface, similar to the xref one. swank.lisp: ;;; MOP (defslimefun mop (type symbol-name) (let ((symbol (parse-symbol-or-lose symbol-name *buffer-package*))) (ecase type (:subclasses (class-direct-subclasses symbol)) (:superclasses (class-direct-superclasses symbol))))) Maybe 'mop' isn't the best name. swank-backend.lisp: ;;; MOP (definterface class-direct-subclasses (class-name) "Return the list of subclasses of class.") (definterface class-direct-superclasses (class-name) "Return the list of superclasses of class.") swank-cmucl.lisp: ;;; CLOS MOP (defimplementation class-direct-subclasses (class-name) (handler-bind ((simple-error (lambda (c) (declare (ignore c)) (return-from class-direct-subclasses nil)))) (let* ((class (find-class class-name)) (subclasses (pcl:class-direct-subclasses class))) (mapcar #'(lambda (x) (symbol-name (pcl:class-name x))) subclasses)))) (defimplementation class-direct-superclasses (class-name) (handler-bind ((simple-error (lambda (c) (declare (ignore c)) (return-from class-direct-superclasses nil)))) (let* ((class (find-class class-name)) (superclasses (pcl:class-direct-superclasses class))) (mapcar #'(lambda (x) (symbol-name (pcl:class-name x))) superclasses)))) The handler is to catch the errors of find-class with a class that doesn't exist. Maybe there's a better behaviour in that situation. I have a generic browser in emacs that can be used to see the stuff above, xrefs,etc. I'll release the code soon. Rui Patrocínio

2 1

[slime-devel] Bug report and bug reporting
by smishra＠speakeasy.net 30 Aug '04

30 Aug '04

Hi, I've encountered a bug when using slime with ACL 6.2 on windows. I get an error if in my slime repl I run (tpl:do-command "zoom"). Is this replicable? Any suggestions? Also, is there any mechanism to report slime bugs other than subscribing to this mailing list? I've never used gmane, is it less of a hassle for the end user? Perhaps a note to that effect would be good on the web page. But I feel there ought to be something on the web page or documentation that says something about bug reporting while slime is still imperfect. Thanks! Sunil

4 4

[slime-devel] how to see the full backtrace???
by Pascal J.Bourguignon 30 Aug '04

30 Aug '04

I've got a big backtrace. Of couse in these cases, the interesting part is deep down at the beginning of the recursion. So, bar from spending two hours scrolling down line by line, (I've already spent 3 minutes and I only got 838 frames, how can I get it to show the full backtrace at once? -- __Pascal Bourguignon__ http://www.informatimago.com/ Our enemies are innovative and resourceful, and so are we. They never stop thinking about new ways to harm our country and our people, and neither do we.

2 2

Re: [slime-devel] how to see the full backtrace???
by Bill_Clementson＠peoplesoft.com 29 Aug '04

29 Aug '04

Pascal J.Bourguignon <pjb(a)informatimago.com> writes: > I've got a big backtrace. Of couse in these cases, the interesting > part is deep down at the beginning of the recursion. > > So, bar from spending two hours scrolling down line by line, (I've > already spent 3 minutes and I only got 838 frames, how can I get it to > show the full backtrace at once? One option is to press "B" (sldb-break-with-default-debugger) in the sldb buffer - that will take you to the default debugger in the repl buffer and you will no longer be subject to the constraints of sldb. - Bill Clementson

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

slime-devel