September 2008 - slime-devel - mailman3.common-lisp.net

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 18 Sep '08

18 Sep '08

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1509 slime/ChangeLog:1.1517 --- slime/ChangeLog:1.1509 Tue Sep 16 14:15:14 2008 +++ slime/ChangeLog Wed Sep 17 19:14:45 2008 @@ -1,3 +1,71 @@ +2008-09-18 Helmut Eller <heller(a)common-lisp.net> + + * slime.el (slime-eval-with-transcript): Accept some more arguments + so that we can also use it for compile-file. + (slime-eval-with-transcript-cont): New. Insert prompt. + (slime-compile-file): Use slime-eval-with-transcript. + (slime-repl-show-maximum-output): Update window point. + (slime-repl-insert-prompt): Don't use insert-before-markers. + (slime-repl-emit): No longer use slime-with-output-end-mark. + +2008-09-17 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank-sbcl.lisp (make-definition-source-location): Fix typo + introduced with crlf-related commit on 2008-09-16. + +2008-09-17 Helmut Eller <heller(a)common-lisp.net> + + * slime.el (slime-repl-popup-on-output): New variable. + (slime-repl-emit): Honor slime-repl-popup-on-output. + (slime-eval-with-transcript): Use slime-repl-popup-on-output. + Also remove the function argument, as it was only used once + and that was slime-message. + +2008-09-17 Helmut Eller <heller(a)common-lisp.net> + + (*pre-reply-hook*): Add 'force-user-output. + +2008-09-17 Helmut Eller <heller(a)common-lisp.net> + + * swank.lisp (send-user-output): Lifted from make-output-function. + Make this a top-level function for easier redefinition. + +2008-09-17 Helmut Eller <heller(a)common-lisp.net> + + * slime.el (slime-test-find-top-level-restart): New function. + [def-slime-test] (interrupt-at-toplevel, interrupt-in-debugger): + Use it. + +2008-09-16 Helmut Eller <heller(a)common-lisp.net> + + * slime.el (slime-connection): Optionally select a new default + connection. + (slime-auto-select-connection): New variable. + (slime-auto-select-connection): New function. + +2008-09-16 Helmut Eller <heller(a)common-lisp.net> + + Adjust positions in files with CRLF style end-on-line markers. + + * slime.el (slime-eol-conversion-fixup): New function. + (slime-goto-location-position): Use it. Also add a new position + type :offset, so that we don't adjust offsets in strings that were + sent over the wire (which uses LF eol-convention). + + * swank-abcl.lisp + * swank-allegro.lisp + * swank-clisp.lisp + * swank-cmucl.lisp + * swank-corman.lisp + * swank-ecl.lisp + * swank-lispworks.lisp + * swank-openmcl.lisp + * swank-sbcl.lisp + * swank-scl.lisp: Create :offset style positions where needed. + + * swank-lispworks.lisp (skip-comments): New function. + (dspec-stream-position): Use it. + 2008-09-16 Tobias C. Rittweiler <tcr(a)freebits.de> * slime.el (slime-end-of-list): `backward-down-list' was used

1 0

[slime-devel] bogus Error: "Local socket address already in use"
by Roman Budzianowski 18 Sep '08

18 Sep '08

I just updated slime. I get the following error: CL-USER(1): (swank:create-server :port 6666) Error: "Local socket address already in use" (errno 98) occured while creating a passive socket on 127.0.0.1 port 6666 . [condition type: SOCKET-ERROR] Restart actions (select using :continue): 0: Return to Top Level (an "abort" restart). 1: Abort entirely from this (lisp) process. Two things: 1. I am running a reverse ssh tunnel on port 6666 2. After taking the default restart, the connection actually works. I had no problem with a previous version. This is in allegro. Roman

2 3

[slime-devel] Re: Buffered slime output
by Tayssir John Gabbour 17 Sep '08

17 Sep '08

Hi, I have that exact same problem with SBCL 1.0.20 and today's Slime from CVS. All output seems to be buffered. All best, Tayssir Raymond Toy wrote: > Until yesterday, I had been using a version from 2008-07-08. This > worked pretty well. Yesterday I updated to the latest CVS version. > > Everything seems to be working fine, but now the output appears to > be buffered strangely. Any function that produces a lot of output > seems to have it buffered. When I enter something else into the > repl, the missing output then comes out. This is a bit annoying.

2 3

[slime-devel] Fix for compiler note placement in DOS newline format files
by Jeff Workman 17 Sep '08

17 Sep '08

The problem is that Emacs doesn't include \r characters in buffers when loading DOS \r\n format files, but \r characters are counted by the compiler and SLIME in finding note positions. So for each line before a compiler note in a DOS format file, note position is moved forward one char. I've only tested this in SBCL on Linux and LispWorks on Windows. Here's code to fix it in slime.el. Maybe someone more knowledgeable can confirm that checking for "dos" appearing in buffer-file-coding-system is the right way to detect DOS newline format? (defun emacs-pos-from-dos-file-pos (file-pos) (let ((fixed-file-pos file-pos) (emacs-pos 1)) (while (< emacs-pos fixed-file-pos) (when (= (char-after emacs-pos) 10) (decf fixed-file-pos)) (incf emacs-pos)) emacs-pos)) (defun slime-compile-note-fix-position-for-dos-file (note) (let ((i 0) (new-note nil)) (while (< i (length note)) (setq new-note (append new-note (list (nth i note)))) (if (eql (nth i note) :location) (let* ((location (nth (+ i 1) note)) (fixed-location (mapcar (lambda (entry) (if (and (listp entry) (eql (first entry) :position)) (append (list :position (emacs-pos-from-dos-file-pos (second entry))) (nthcdr 2 entry)) entry)) location))) (setq new-note (append new-note (list fixed-location)))) (setq new-note (append new-note (list (nth (+ i 1) note))))) (incf i 2)) new-note)) (defun slime-compilation-finished (compilation-result buffer &optional emacs-snapshot) (with-struct (slime-compilation-result. notes durations) compilation-result (let* ((coding-system (buffer-local-value 'buffer-file-coding-system buffer)) (fixed-notes (if (and coding-system (string-match ".*dos.*" (symbol-name coding-system))) (mapcar #'slime-compile-note-fix-position-for-dos-file notes) notes))) (with-current-buffer buffer (setf slime-compilation-just-finished t) (setf (slime-compilation-result.notes compilation-result) fixed-notes) (setf slime-last-compilation-result compilation-result) (slime-show-note-counts fixed-notes (reduce #'+ durations)) (when slime-highlight-compiler-notes (slime-highlight-notes fixed-notes))) (run-hook-with-args 'slime-compilation-finished-hook fixed-notes emacs-snapshot))))

2 1

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 17 Sep '08

17 Sep '08

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1508 slime/ChangeLog:1.1509 --- slime/ChangeLog:1.1508 Mon Sep 15 17:11:19 2008 +++ slime/ChangeLog Tue Sep 16 14:15:14 2008 @@ -1,3 +1,9 @@ +2008-09-16 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime.el (slime-end-of-list): `backward-down-list' was used + there which is defined by paredit.el. + Use `(down-list -1)' instead. + 2008-09-15 Helmut Eller <heller(a)common-lisp.net> * swank-lispworks.lisp (describe-symbol-for-emacs): Revert last

1 0

[slime-devel] Re: CMUCL Timers
by Madhu 16 Sep '08

16 Sep '08

* Raymond Toy <48CBFC17.5010103(a)gmail.com> : Wrote on Sat, 13 Sep 2008 13:44:55 -0400: | Madhu wrote: |> I'd like to see *use-dedicated-output-stream* supported: It is useful |> and sometimes necessary for output to not to go through the whole of |> slime repl's infrasturcture. | | What to you mean supported? It seems to be supported. I was remembering this thread http://article.gmane.org/gmane.lisp.slime.devel/6507/ (I believe the REPL code was added without considering this variable and mode of behaviour and so was incompatible, and now slime has to discourage the use of the mode in effect deprecating it. Had it been supported, the alternative design would have been a framework for the repl machinery in which both modes of operation coexisted without problems to either) | Do you have any quick examples of using serve-timer? Not at hand, but I could reconstruct them if needed -- Madhu

2 1

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 16 Sep '08

16 Sep '08

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1505 slime/ChangeLog:1.1508 --- slime/ChangeLog:1.1505 Sun Sep 14 13:10:34 2008 +++ slime/ChangeLog Mon Sep 15 17:11:19 2008 @@ -1,3 +1,45 @@ +2008-09-15 Helmut Eller <heller(a)common-lisp.net> + + * swank-lispworks.lisp (describe-symbol-for-emacs): Revert last + change. + +2008-09-15 Helmut Eller <heller(a)common-lisp.net> + + * swank.lisp (sldb-loop): Send a :sldb-return event to ourselfes + to inform the debug session at the lower level. + (wait-for-event): Drop the report-interrupt argument. No longer + needed. + (event-match-p): Add an OR pattern operator. Used to wait for + different events simultaneously. + + (read-packet): Use peek-char to detect EOF. read-sequence wouldn't + work. + + * slime.el (slime-test-interrupt-in-debugger): Call sldb-quit and + sldb-continue in the right buffer. + + * swank-backend.lisp (wait-for-input): + * swank-cmucl.lisp (wait-for-input): + * swank-clisp.lisp (wait-for-input): Use the idiom + "(when (check-slime-interrupts) (return :interrupt))". + +2008-09-15 Helmut Eller <heller(a)common-lisp.net> + + Interrupt related hacking. + + * swank-backend.lisp (*pending-slime-interrupts*): Should be + thread-local. Leave global value unbound. + + * swank.lisp (with-interrupts-enabled%): New helper macro. + (with-slime-interrupts, without-slime-interrupts): Use it. + (call-with-connection): Bind *pending-slime-interrupts* here. + (wait-for-event): Add a report-interrupt argument. Currently used + by the debugger to detect when a nested debugger session, which + was triggered by an interrupt in wait-for-event, returns. Doesn't + work well, though. + + * slime.el (slime-test-interrupt-in-debugger): New test. + 2008-09-14 Helmut Eller <heller(a)common-lisp.net> Introduce a WAIT-FOR-INPUT backend function.

1 0

[slime-devel] Newbie problems w/slime
by Jeffrey Berger 15 Sep '08

15 Sep '08

Am attempting to set up an sbcl lisp environment in emacs and getting undefined symbol error for load-path. Here's my .emacs file: cat .emacs (setq inferior-lisp-program "/usr/local/bin/sbcl") (add-to-list ’load-path "/usr/share/emacs/22.1/site-lisp/slime") (require ’slime) (slime-setup) ... the debugger backtrace: Debugger entered--Lisp error: (void-variable ùload-path) (add-to-list ùload-path "/usr/share/emacs/22.1/site-lisp/slime") eval-buffer(#<buffer *load*> nil "/home/jeff/.emacs" nil t) ; Reading at buffer position 116 load-with-code-conversion("/home/jeff/.emacs" "/home/jeff/.emacs" t t) load("~/.emacs" t t) #[nil "... and the result of eval'ing load-path: ("/usr/share/emacs22/site-lisp/slime" "/usr/share/emacs22/site-lisp/slime/contrib" "/usr/share/emacs22/site-lisp/dictionaries-common" "/etc/emacs22" "/etc/emacs" "/usr/local/share/emacs/22.1/site-lisp" "/usr/local/share/emacs/site-lisp" "/usr/share/emacs/22.1/site-lisp" "/usr/share/emacs/22.1/site-lisp/dictionaries-common" "/usr/share/emacs/22.1/site-lisp/slime" "/usr/share/emacs/22.1/site-lisp/slime/contrib" "/usr/share/emacs/site-lisp" ...) What am I doing wrong? TIA

2 2

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 15 Sep '08

15 Sep '08

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1501 slime/ChangeLog:1.1504 --- slime/ChangeLog:1.1501 Thu Sep 11 08:32:01 2008 +++ slime/ChangeLog Fri Sep 12 14:55:42 2008 @@ -1,3 +1,58 @@ +2008-09-12 Helmut Eller <heller(a)common-lisp.net> + + For Lispworks, parse the $LWHOME/lwdoc file. + + * swank-lispworks.lisp (lwdoc, lookup-lwdoc, parse-lwdoc-record): + New functions. + (describe-symbol-for-emacs): Use it. + +2008-09-12 Tobias C. Rittweiler <tcr(a)freebits.de> + + In an SLDB buffer, `C-c C-c' will now recompile the source behind + a frame. In particular, `C-u C-c C-c' will recompile the frame + with high debug settings. + + * slime.el (sldb-recompile-frame-source): New function. + (sldb-mode-map): Bind `C-c C-c' to it. + + * slime.el (sldb-overlays, sldb-delete-overlays, slime-xref-cleanup): + Removed. Sldb-overlays weren't created anymore since 2008-08-17. + +2008-09-12 Tobias C. Rittweiler <tcr(a)freebits.de> + + New faces: `sldb-restartable-frame-line-face', + `sldb-non-restartable-frame-line-face'. + + The former is the face for frames that are surely restartable, the + latter for frames that are surely not restartable. If + restartability of a frame cannot be reliably determined, the face + `sldb-frame-line-face' is used. + + At the moment, determination of frame restartability is supported + by the SBCL backend only. + + * slime.el (sldb-frame.string): New. + (sldb-frame.number): New. + (sldb-frame.plist): New. + (sldb-prune-initial-frames): Use them. + (sldb-insert-frames): Ditto. + (sldb-compute-frame-face): New. + (sldb-insert-frame): Use `sldb-compute-frame-face' to insert + frames with one of the faces described above. + + * swank.lisp (defslimefun backtrace): Changed return value; each + frame is now accompanied with a PLIST which at the moment can + contain :RESTARTABLE NIL/T/:UNKNOWN depending on whether the frame + is restartable, or not. + + * swank-backend.lisp (defstruct swank-frame): New structure. + (compute-backtrace): Is now supposed to return a list of SWANK-FRAMEs. + (print-frame): Renamed to PRINT-SWANK-FRAME. + + * swank-sbcl.lisp, swank-cmucl.lisp, swank-lispworks.lisp, + * swank-allegro.lisp, swank-scl.lisp, swank-openmcl.lisp, + * swank-abcl.lisp, swank-clisp.lisp: Adapted to swank-backend changes. + 2008-09-11 Helmut Eller <heller(a)common-lisp.net> * doc/slime-refcard.tex: Fix typos.

3 3

[slime-devel] is there a way to capture / grab a local variable in SLDB?
by Gary King 15 Sep '08

15 Sep '08

I often find that I want to grab a local variable in an SLDB frame but don't know how to do this. Is there a way? thanks, -- Gary Warren King, metabang.com Cell: (413) 559 8738 Fax: (206) 338-4052 gwkkwg on Skype * garethsan on AIM

3 3