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

[slime-devel] slime-repl-insert-prompt question
by Alan Ruttenberg 17 Sep '05

17 Sep '05

Is there a reason why the protocol should handle inserting the result of a repl evaluation with dedicated code vs just letting the lisp side print it? I ask because now that there are nested presentations, there are advantages to having the lisp side do the printing. In particular a list result which has presented objects in could have the elements of the list work as individual presentations in addition to the whole list being available for re-use, whereas in the current situation you can only get hold of the complete result. -Alan

2 1

[slime-devel] SBCL memory overcommitment after startup when using SLIME
by Aleksandar Bakic 17 Sep '05

17 Sep '05

Hi, I first thought this was a SBCL issue, but it appears to be SLIME's. Below is the scenario: 1. Run Emacs 2. Run SBCL from Emacs 3. slime-connect to SBCL 4. Open a file in lisp mode 5. In a terminal, run "top -d 1" and make sure you see SBCL and how much virtual memory it is using 6. Write in the file "(defun foo ())" and compile foo by pressing C-c C-c 7. You should notice that top shows a few MBs more virtual memory used by SBCL The virtual memory use does not increase if you type (compile 'foo) in REPL. SBCL man page describes memory overcommitment at startup, but not afterwards. Can you reproduce the above increase? Let me know if you need more information. Thanks, Alex __________________________________ Yahoo! Mail - PC Magazine Editors' Choice 2005 http://mail.yahoo.com

2 1

Re: [slime-devel] SLIME48: a Swank back end in Scheme48
by Helmut Eller 17 Sep '05

17 Sep '05

* Brian Mastenbrook [2005-09-16 13:55+0200] writes: > On Sep 16, 2005, at 6:14 AM, Helmut Eller wrote: > >> Sadly, the goals shifted a bit since then. > > I don't think very many people would agree that this is "sadly". If > this is the way you feel, however, I'm sure that there are enough > people who prefer using SLIME on other Common Lisps to sustain a fork > Sadly, the goals shifted a bit since then. > of the project. Sure. That's why Taylor Campbell should at least consider making a fork. >> I'm not that excited about supporting every Frankenstein Lisp on the >> planet, just because we can. And frankly, who wants to use a Lisp >> which doesn't even have docstrings? > > Taylor's goal in this project was to make a good Emacs IDE for > Scheme48, not to make SLIME support "every Frankenstein Lisp". SLIME > was just the tool he chose to make this happen. Scheme48 is a Frankenstein Lisp by my definition. If SLIME doesn't need to support Scheme48, why does he even bother us/me then? >> If you want to add your code to our CVS repo, it should be clear that >> Scheme is only a second class citizen here, and that we have limited >> motivation to make/keep the protocol working with the Scheme server. > > Why does that need to be? Why can't Taylor just maintain the Scheme48 > support to the best level possible, just as is done with the other > backends? Because he wrote his own server (swank) not only the implementation specific part (swank-backend). His server also uses a rather different approach to threading than we do. >> On the technical side, Scheme has continuations but SLIME assumes that >> the Lisp is stack oriented. I haven't thought too much about it, but >> there will likely be problems if somebody calls a continuation >> multiple times. Also, interrupting doesn't seem to work with your >> backend, or not in the way I expected it. > > I'm not sure what those problems would be. Can you clarify? Try (define k (call-with-current-continuation (lambda (c) c))) (k 1) and type C-x C-e at the end of each line.

2 2

[slime-devel] Daily ChangeLog diff
by lgorrie＠common-lisp.net 16 Sep '05

16 Sep '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.780 slime/ChangeLog:1.781 --- slime/ChangeLog:1.780 Thu Sep 15 10:31:50 2005 +++ slime/ChangeLog Fri Sep 16 05:43:08 2005 @@ -1,3 +1,7 @@ +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. + 2005-09-15 Helmut Eller <heller(a)common-lisp.net> * swank-clisp.lisp (compute-backtrace): Include only "function

1 0

[slime-devel] .slime/fasl/ directory names
by Lynn Quam 16 Sep '05

16 Sep '05

I discovered that the architecture dependent directory names in $HOME/.slime/fasl/ contain "non-standard" characters for Unix file names. For example: /homedir/quam/.slime/fasl/cmu-snapshot 2005-09 (19b)-linux-x86 I understand why this is the case, but it can be somewhat annoying because of the difficulty in properly "escaping" those names in the various Unix shells. For example, find . -name '*.x86f' | xargs grep '(defun ' grep: ./cmu-snapshot: No such file or directory grep: 2005-09: No such file or directory Admittedly, this is not a good example, but consider: find . -name '*.x86f' | xargs rm This wouldn't be a problem with if we had something like the Scheme-Shell (scsh), but ...

7 7

[slime-devel] Daily ChangeLog diff
by lgorrie＠common-lisp.net 15 Sep '05

15 Sep '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.778 slime/ChangeLog:1.780 --- slime/ChangeLog:1.778 Wed Sep 14 22:18:39 2005 +++ slime/ChangeLog Thu Sep 15 10:31:50 2005 @@ -1,3 +1,47 @@ +2005-09-15 Helmut Eller <heller(a)common-lisp.net> + + * swank-clisp.lisp (compute-backtrace): Include only "function + frames" in the backtrace. I hope that makes some sense. + (sldb-backtrace, function-frame-p): New functions. + (*sldb-backtrace*, call-with-debugging-environment, nth-frame): + Compute and remember the backtrace when entering the debugger. + (arglist): If the function has a function-lambda-expression, fetch + the arglist from there. + (find-encoding): Use strings instead of 'charset:foo symbols to + avoid compile time problems if the charset is not available. + Suggested by Vaucher Laurent. + + * swank.lisp (eval-in-emacs): Fix a race condition which occurred + with sigio. + (*echo-area-prefix*): New variable. + + * slime.el (slime-process-available-input): Simplify it a bit and + make it easier to debug read errors. + (slime-net-close): Don't kill the buffer if the new optional arg + `debug' is true. + (slime-run-when-idle): Accept arguments for the function. + (slime-init-connection-state): Close over the proc variable. It + was lost when the async evaluation returned. + (slime-output-buffer, slime-connection-output-buffer): Make + slime-output-buffer faster by keeping the buffer in a connection + variable. + (slime-restart-inferior-lisp-aux, slime-quit-lisp): Disable the + process filter to avoid errors in XEmacs. + +2005-09-14 Alan Ruttenberg <alanr-l(a)mumble.net> + + * slime.el (slime-menu-choices-for-presentation), (slime-presentation-menu) + Fix loss after refactoring. xemacs can't handle lambda forms in + the menu spec given to x-popup-menu, only symbols, so save the + actions in a hash table keyed by a gensym, give x-popup-menu the + gensym and then call the gensym. Haven't checked that it actually + works in xemacs because my xemacs is hosed in os x Tiger. Could + someone let me know... + + * swank.lisp (inspect-factor-more-action) + rename (inspect-show-more-action) Prompt before reading how many + more. Would be nicer to prompt in the minibuffer... + 2005-09-14 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> * slime.el (slime-presentation-expression): Remove handling of @@ -52,7 +96,7 @@ * swank.lisp: Simplify the object <-> presentation-id mapping. (save-presented-object): Remove the optional `id' arg. - (lookup-presented-object): Id's should be fixnums not some cons + (lookup-presented-object): Id should be a fixnum not some cons with fuzzy/non-documented meaning. Use the secondary return value to test for absence of the id. Update callers accordingly. (*not-present*): Deleted.

1 0

[slime-devel] Segmentation Fault with Slime, Uffi and CMUCL
by Stefan Kamphausen 15 Sep '05

15 Sep '05

Dear Slime-Developers, in a project that involves heavy usage of UFFI (loading a library written in C++ with a C-header file using extern "C") I am encountering repeatable segmentation faults when working under SLIME. An example session would be: -------------------------------------------------- M-x slime ,load-system cl-the-uffi-using-system (using tab completion and a symlink in my central registry) ,cd /path/to/another/lib/im/currently/workin/on and the echo area tells me: Lisp connection closed unexpectedly: exited abnormally with code 256 -------------------------------------------------- The debugger output I get after C-g'ing is pasted below A few more notes: * I am using CVS HEAD slime, XEmacs 21.4 (patch 17) and CMU Common Lisp CVS 19b 19b-release-20050628-3 + minimal debian patches (19B) on Gentoo/x86. SLIME ist not emerged but manually kept in sync with CVS. * I can't tell for sure that there are no problems with the C-lib. But I'd expect those problems to show up when calling the functions not after loading the system and doing a simple directory change. * I have double checked the UFFI-function definitions and the C-header file. That should be correct. * Currently I can't change the C-library at all, but when the author will be back from his holiday we can change everything there. * Unfortunately I can't provide the source, it's an inhouse project of my employer (although we released one part of it under BSD license recently :-). * I can't crash it easily with SBCL 0.9.4. * And, well, it works when running CMUCL from the shell. That's why I am writing to this list in hope of some enlightenment. Of course I'd be happy if you'd say: "ah, this is a bug, I just uploaded a patch", but any pointer as to where and how I might get more information will be appreciated. Kind Regards Stefan Kamphausen PS: Funny enough, I had some problems sending to this list due to greylisting and just at that moment martin wrote his somewhat similar message. ---------------------------------------------------------------------- Debugger output: Debugger entered--Lisp error: (quit) accept-process-output(nil 0 10000) (while t (accept-process-output nil 0 10000)) (let ((debug-on-quit t) (inhibit-quit nil)) (while t (accept-process-output nil 0 10000))) (catch tag (slime-rex (tag sexp) (sexp package) ((:ok value) (unless (member tag slime-stack-eval-tags) (error "tag = %S eval-tags = %S sexp = %S" tag slime-stack-eval-tags sexp)) (throw tag (list (function identity) value))) ((:abort) (throw tag (list (function error) "Synchronous Lisp Evaluation aborted.")))) (let ((debug-on-quit t) (inhibit-quit nil)) (while t (accept-process-output nil 0 10000)))) (apply (function funcall) (catch tag (slime-rex (tag sexp) (sexp package) ((:ok value) (unless (member tag slime-stack-eval-tags) (error "tag = %S eval-tags = %S sexp = %S" tag slime-stack-eval-tags sexp)) (throw tag (list (function identity) value))) ((:abort) (throw tag (list (function error) "Synchronous Lisp Evaluation aborted.")))) (let ((debug-on-quit t) (inhibit-quit nil)) (while t (accept-process-output nil 0 10000))))) (let* ((tag (gensym "slime-result-")) (slime-stack-eval-tags (cons tag slime-stack-eval-tags))) (apply (function funcall) (catch tag (slime-rex (tag sexp) (sexp package) ((:ok value) (unless (member tag slime-stack-eval-tags) (error "tag = %S eval-tags = %S sexp = %S" tag slime-stack-eval-tags sexp)) (throw tag (list (function identity) value))) ((:abort) (throw tag (list (function error) "Synchronous Lisp Evaluation aborted.")))) (let ((debug-on-quit t) (inhibit-quit nil)) (while t (accept-process-output nil 0 10000)))))) slime-eval((swank:set-default-directory "/home/ska/prj/lisp/cl-meta-worker/")) (slime-from-lisp-filename (slime-eval (backquote (swank:set-default-directory (\, (slime-to-lisp-filename directory)))))) (message "default-directory: %s" (slime-from-lisp-filename (slime-eval (backquote (swank:set-default-directory (\, (slime-to-lisp-filename directory))))))) (lambda (directory) "Make DIRECTORY become Lisp's current directory." (interactive (list (read-directory-name "Directory: " nil nil t))) (message "default-directory: %s" (slime-from-lisp-filename (slime-eval (backquote (swank:set-default-directory (\, (slime-to-lisp-filename directory))))))) (with-current-buffer (slime-output-buffer) (setq default-directory (expand-file-name directory)) (when (boundp (quote header-line-format)) (slime-repl-update-banner))))("~/prj/lisp/cl-meta-worker/") call-interactively(slime-set-default-directory) (let ((shortcut (slime-lookup-shortcut (completing-read "Command: " (slime-bogus-completion-alist (slime-list-all-repl-shortcuts)) nil t nil (quote slime-repl-shortcut-history))))) (call-interactively (slime-repl-shortcut\.handler shortcut))) (if (save-excursion (goto-char slime-repl-input-start-mark) (looking-at " *$")) (let ((shortcut (slime-lookup-shortcut (completing-read "Command: " (slime-bogus-completion-alist (slime-list-all-repl-shortcuts)) nil t nil (quote slime-repl-shortcut-history))))) (call-interactively (slime-repl-shortcut\.handler shortcut))) (insert (string slime-repl-shortcut-dispatch-char))) (lambda nil (interactive) (if (save-excursion (goto-char slime-repl-input-start-mark) (looking-at " *$")) (let ((shortcut (slime-lookup-shortcut (completing-read "Command: " (slime-bogus-completion-alist (slime-list-all-repl-shortcuts)) nil t nil (quote slime-repl-shortcut-history))))) (call-interactively (slime-repl-shortcut\.handler shortcut))) (insert (string slime-repl-shortcut-dispatch-char))))() call-interactively(slime-handle-repl-shortcut) -- Stefan Kamphausen --- http://www.skamphausen.de a blessed +42 regexp of confusion (weapon in hand) You hit. The format string crumbles and turns to dust. 5 GB Mailbox, 50 FreeSMS http://www.gmx.net/de/go/promail +++ GMX - die erste Adresse f�r Mail, Message, More +++

2 2

[slime-devel] SLIME/Allegro communication-style
by Lynn Quam 15 Sep '05

15 Sep '05

Is it possible to run SLIME with Allegro using communication-style = :sigio rather than :spawn? I have a custom REPL loop which handles tcl/tk events in addition to input from *standard-input*. This work fine with SLIME and CMUCL with communication-style = :sigio. I have encountered difficulties implementing event-handling for tcl/tk with SLIME using communication-style = :spawn. One problem is that (method stream-listen slime-input-stream) on Allegro just checks if the buffer contains characters. It does not attempt to ask the socket is there are characters available. This does not appear to be a problem in CMUCL.

2 1

[slime-devel] Problem with EUC-JP encoding on Clisp-2.35 Win32 native (+fix)
by VAUCHER Laurent 15 Sep '05

15 Sep '05

Hi. My current configuration is Windoze/Emacs-21.3/Clisp-2.35 native (not the cygwin port, which is much slower) and it seems that a recent fix in swank-clisp.lisp in function find-encoding is not "portable". In fact, it is compatible with Clisp-2.35-Unix-flavor (MacOSX and Cygwin included) but not with the native W32 version which lacks some encodings (only 85 vs 121 for the unix version, to be accurate). I include a patch that solves the problem for me (by not using the inexistant symbol directly) ant should also work for those using the japanese encoding on Clisp versions supporting it. Laurent. --- swank-clisp.lisp.bak 2005-09-14 09:16:12.523260100 +0200 +++ swank-clisp.lisp 2005-09-14 09:19:14.141146000 +0200 @@ -123,7 +123,7 @@ :line-terminator :unix)) (:utf-8-unix (ext:make-encoding :charset 'charset:utf-8 :line-terminator :unix)) - (:euc-jp-unix (ext:make-encoding :charset 'charset:euc-jp + (:euc-jp-unix (ext:make-encoding :charset (find-symbol "EUC-JP" :charset) :line-terminator :unix)))) (defimplementation accept-connection (socket

2 1

[slime-devel] Daily ChangeLog diff
by lgorrie＠common-lisp.net 14 Sep '05

14 Sep '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.777 slime/ChangeLog:1.778 --- slime/ChangeLog:1.777 Tue Sep 13 07:37:16 2005 +++ slime/ChangeLog Wed Sep 14 22:18:39 2005 @@ -1,3 +1,8 @@ +2005-09-14 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> + + * slime.el (slime-presentation-expression): Remove handling of + cons presentation-ids. + 2005-09-13 Alan Ruttenberg <alanr-l(a)mumble.net> * slime.el (defcustom slime-ed-use-dedicated-frame ... vs defvar

1 0