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

[slime-devel] slime-after-change-function performance issue
by Russell McManus 22 Sep '05

22 Sep '05

On XEmacs, this function appears to cause serious thrashing for me. When I take it out of after-change-functions, the problem goes away. -russ

3 2

[slime-devel] Daily ChangeLog diff
by lgorrie＠common-lisp.net 21 Sep '05

21 Sep '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.785 slime/ChangeLog:1.787 --- slime/ChangeLog:1.785 Mon Sep 19 10:20:48 2005 +++ slime/ChangeLog Wed Sep 21 22:55:50 2005 @@ -1,3 +1,49 @@ +2005-09-21 Helmut Eller <heller(a)common-lisp.net> + + * slime.el (slime-setup-command-hooks): Make + after-change-functions a buffer-local variable; it's by default + global in XEmacs. + + * swank.lisp (throw-to-toplevel): Invoke the `abort-restart' + request instead of throwing to the `slime-toplevel' catch tag. + (handle-request): Rename the restart from abort to abort-request. + (call-with-connection): Remove the slime-toplevel catch tag + because with-connection is used in far to many places which aren't + at "toplevel". + + * present.lisp (presentation-start, presentation-end): Use + finish-output instead of force-output. + + * swank-gray.lisp, swank-cmucl.lisp: Improve stream efficiency by + buffering more output. stream-force-output simply does nothing, if + the output buffer was flushed less than 200 millisecons before. + stream-finish-output can still be used to really flush the buffer. + (slime-output-stream): New slot last-flush-time. + (stream-finish-output): New function. Do what stream-force-output + did previously. + (stream-force-output): Buffer more output. + + * slime.el (slime-process-available-input): Oops, don't start a + timer for every event. + (slime-write-string): Renamed from slime-output-string. + (slime-dispatch-event): Rename :read-output to :write-string. + (slime-io-speed-test): New command. + (slime-open-stream-to-lisp): Fix parens. The coding system should + also be set if presentations are disabled. + + * swank.lisp (make-output-function): Rename :read-output to + :write-string. + (eval-for-emacs, interactive-eval, eval-region): Use finish-output + not force-output. + + * swank-sbcl.lisp, swank-openmcl.lisp, swank-allegro.lisp, + swank-lispworks: Import `stream-finish-output'. + + * swank-scheme48/io.scm (empty-swank-output-buffer): Rename + :read-output to :write-string. + + * swank-scheme48/load.scm (slime48-start): Fix '() vs. #f bug. + 2005-09-19 Luke Gorrie <luke(a)synap.se> * nregex.lisp: Released into the public domain by Lawrence E. Freil.

1 0

[slime-devel] Patch for stepper support in SBCL and (sort of) CLISP
by Svein Ove Aas 21 Sep '05

21 Sep '05

One patch, one new file. The stepper works in SBCL, and should be fairly self-explanatory. C-s to toggle stepping, c to step inwards and x to quit stepping but keep running. You need to compile with high debug quality. One gotcha: Slime will only display the source correctly if you compile the entire file with C-c C-k. The same works in CLISP, but only for uncompiled code, and you won't see your location. Fix the sldb view-source command, and it'll work. OpenMCL and CMUCL appear to have broken step implementations; if these work for anyone, please contact me so I can write backends. Finally, while working on this I had some ideas about improving other parts of Slime, and I've left commented-out code for this in at least one place. When the debugger pops up, we might want to (a) skip "internal" frames such as break when positioning the cursor, and (b) attempt to view the source for the resulting frame automatically. There is very little work left to do this, but is it a good idea? Also, the way the stepper currently works there are (at least in clisp) a couple of stepper-internal frames *between* those belonging to the user. Would filtering out those be okay?

1 1

[slime-devel] Daily ChangeLog diff
by lgorrie＠common-lisp.net 19 Sep '05

19 Sep '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.784 slime/ChangeLog:1.785 --- slime/ChangeLog:1.784 Sun Sep 18 23:16:51 2005 +++ slime/ChangeLog Mon Sep 19 10:20:48 2005 @@ -1,3 +1,7 @@ +2005-09-19 Luke Gorrie <luke(a)synap.se> + + * nregex.lisp: Released into the public domain by Lawrence E. Freil. + 2005-09-19 Helmut Eller <heller(a)common-lisp.net> * slime.el (slime48): New command.

1 0

[slime-devel] lag in posting to slime-devel
by Matthew D Swank 19 Sep '05

19 Sep '05

I often see a delay of up to several hours when I post to this list. I was wondering if my experience was atypical. Matt

2 1

[slime-devel] patch to load.scm
by Matthew D Swank 19 Sep '05

19 Sep '05

CAUTION: scheme ahead! From (define (slime48-start . port-file) ... Scheme list are never #f, so the test is always true: (eval (if port-file `(start-swank ,@port-file) '(slime48)) diff -ur swank-scheme48.orig/load.scm swank-scheme48/load.scm --- swank-scheme48.orig/load.scm 2005-09-19 11:55:46.000000000 -0500 +++ swank-scheme48/load.scm 2005-09-19 11:50:13.000000000 -0500 @@ -73,9 +73,9 @@ (in 'slime48 (lambda () (call-with-values (lambda () - (eval (if port-file - `(start-swank ,@port-file) - '(slime48)) + (eval (if (null? port-file) + '(slime48) + `(start-swank ,@port-file)) (interaction-environment))) (lambda (world server) (user (lambda ()

2 1

[slime-devel] trouble updating slime from cvs
by Alan Ruttenberg 19 Sep '05

19 Sep '05

cvs server: Updating slime/swank-scheme48 cvs server: failed to create lock directory for `/project/slime/ cvsroot/slime/swank-scheme48' (/project/slime/cvsroot/slime/swank- scheme48/#cvs.lock): Permission denied cvs server: failed to obtain dir lock in repository `/project/slime/ cvsroot/slime/swank-scheme48' cvs [server aborted]: read lock failed - giving up -Alan

2 2

[slime-devel] problem in slime-after-change-function w xemacs 21.4.17
by Russell McManus 19 Sep '05

19 Sep '05

cvs slime and gnus don't work well together. I wonder whether the loop inside slime-for-each-presentation-in-region doesn't handle narrowed buffers? I'm not sure how to fix this, and I'm hoping that the backtrace will get the discussion going. I get the following backtrace if I M-x slime, then run gnus. -russ Signaling: (args-out-of-range #<buffer " *nnmail incoming*"> 5995) next-single-property-change(5995 slime-repl-presentation #<buffer " *nnmail incoming*"> 6010) (setq point (next-single-property-change point (quote slime-repl-presentation) object to)) (while (< point to) (setq point (next-single-property-change point ... object to)) (let* (... ...) (when ... ...))) (let ((point from)) (while (< point to) (setq point ...) (let* ... ...))) (progn (fset (quote handle-presentation) (function* ...)) (dolist (presentation ...) (handle-presentation presentation from)) (let (...) (while ... ... ...))) (unwind-protect (progn (fset ... ...) (dolist ... ...) (let ... ...)) (if --letf-bound--57870 (fset ... --letf-save--57869) (fmakunbound ...))) (let* ((--letf-bound--57870 ...) (--letf-save--57869 ...)) (unwind-protect (progn ... ... ...) (if --letf-bound--57870 ... ...))) (letf ((... ...)) (dolist (presentation ...) (handle-presentation presentation from)) (let (...) (while ... ... ...))) (letf* ((... ...)) (dolist (presentation ...) (handle-presentation presentation from)) (let (...) (while ... ... ...))) (flet ((handle-presentation ... ...)) (dolist (presentation ...) (handle-presentation presentation from)) (let (...) (while ... ... ...))) (catch (quote --cl-block-slime-for-each-presentation-in-region--) (flet (...) (dolist ... ...) (let ... ...))) (cl-block-wrapper (catch (quote --cl-block-slime-for-each-presentation-in-region--) (flet ... ... ...))) (block slime-for-each-presentation-in-region (flet (...) (dolist ... ...) (let ... ...))) (let* ((object ...)) (if --rest--51329 (signal ... ...)) (block slime-for-each-presentation-in-region (flet ... ... ...))) slime-for-each-presentation-in-region(5995 6010 (lambda (presentation from to whole-p) (cond (whole-p ...) (... ... ...)))) (let ((real-start ...) (real-end ...) (any-change nil)) (slime-for-each-presentation-in-region real-start real-end (lambda ... ...)) (when any-change (undo-boundary))) (let ((inhibit-modification-hooks t)) (let (... ... ...) (slime-for-each-presentation-in-region real-start real-end ...) (when any-change ...))) slime-after-change-function(5996 6009 5) replace-match("X-From-Line: ") nnml-save-mail((("cygwin" . 7482))) nnmail-check-duplication("<20050919142929.GJ20844(a)calimero.vinschen.de>" nnml-save-mail nnml-active-number) nnmail-process-unix-mail-format(nnml-save-mail nnml-active-number) nnmail-split-incoming("~/.emacs-mail-crash-box" nnml-save-mail nil nil nnml-active-number) #<compiled-function (file orig-file) "...(17)" [orig-file file nnmail-split-incoming nnml-save-mail nil nnmail-get-split-group (file :path "/var/mail/russe") nnml-active-number] 7>("~/.emacs-mail-crash-box" "~/.emacs-mail-crash-box") mail-source-callback(#<compiled-function (file orig-file) "...(17)" [orig-file file nnmail-split-incoming nnml-save-mail nil nnmail-get-split-group (file :path "/var/mail/russe") nnml-active-number] 7> "~/.emacs-mail-crash-box") mail-source-fetch((file :path "/var/mail/russe") #<compiled-function (file orig-file) "...(17)" [orig-file file nnmail-split-incoming nnml-save-mail nil nnmail-get-split-group (file :path "/var/mail/russe") nnml-active-number] 7>) nnmail-get-new-mail(nnml nnml-save-nov "~/Mail/" nil) nnml-request-scan(nil "private") gnus-request-scan(nil (nnml "private")) gnus-read-active-file-1((nnml "private") nil) gnus-read-active-file(nil nil) gnus-setup-news(nil nil nil) byte-code("..." [gnus-slave gnus-current-startup-file gnus-startup-file did-connect level dont-connect gnus-group-quit gnus-run-hooks gnus-startup-hook gnus-make-newsrc-file gnus-dribble-read-file bbb-login add-hook gnus-summary-mode-hook gnus-grouplens-mode gnus-request-create-group "queue" (nndraft "") "drafts" (nndraft "") gnus-setup-news nil gnus-setup-news-hook gnus-start-draft-setup gnus-group-list-groups gnus-group-first-unread-group gnus-configure-windows group gnus-group-set-mode-line gnus-started-hook gnus-use-dribble-file gnus-use-grouplens gnus-agent] 4) gnus-1(nil nil nil) gnus(nil) call-interactively(gnus) command-execute(gnus t) execute-extended-command(nil) call-interactively(execute-extended-command)

1 0

[slime-devel] problem with slime in latest cvs
by Matthew D Swank 19 Sep '05

19 Sep '05

For the last week or so slime give the following error compiling code: error in process filter: Symbol's value as variable is void: bridge-filter [<some obscene number> times] It doesn't seem to matter what lisp I am running ( clisp, sbcl, abcl). I am running slime on emacs 22.0.50 (a cvs build on 9/10/05). puzzled, Matt

3 5

[slime-devel] Daily ChangeLog diff
by lgorrie＠common-lisp.net 18 Sep '05

18 Sep '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.781 slime/ChangeLog:1.784 --- slime/ChangeLog:1.781 Fri Sep 16 05:43:08 2005 +++ slime/ChangeLog Sun Sep 18 23:16:51 2005 @@ -1,3 +1,20 @@ +2005-09-19 Helmut Eller <heller(a)common-lisp.net> + + * slime.el (slime48): New command. + +2005-09-19 Taylor Campbell <campbell(a)mumble.net> + + * swank-scheme48/: New backend. + +2005-09-18 Wolfgang Jenkner <wjenkner(a)inode.at> + + * bridge.el: cl is required at macro expansion time (because of + `block'). Reported by Matthew D Swank. + +2005-09-18 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> + + * swank.lisp: Move presentation menu protocol here from present.lisp. + 2005-09-15 Alan Ruttenberg <alanr-l(a)mumble.net> * slime.el (slime-repl-return) don't copy presentation to input if already in input area.

1 0