March 2005 - slime-devel - mailman3.common-lisp.net

[slime-devel] Missing features for LW
by Edi Weitz 20 Mar '05

20 Mar '05

See attached patch. Cheers, Edi.

2 2

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

18 Mar '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.639 slime/ChangeLog:1.641 --- slime/ChangeLog:1.639 Wed Mar 16 23:11:24 2005 +++ slime/ChangeLog Fri Mar 18 23:33:59 2005 @@ -1,3 +1,30 @@ +2005-03-18 Helmut Eller <heller(a)common-lisp.net> + + * swank-source-path-parser.lisp (make-source-recording-readtable): + Ignore non-ascii chars. + + * swank-sbcl.lisp (swank-compile-string): Re-implemented. This + time with temp-files and proper source-location tracking. + (install-debug-source-patch, debug-source-for-info-advice): Patch + SBCL's debug-source-for-info so that we can dump our own bits of + debug info. + (function-source-location, code-location-source-path): Rewritten + to handle C-c C-c functions. Also use the source-path to locate + the position. + (locate-compiler-note): Renamed from resolve-note-location. + (temp-file-name, call/temp-file): New utilities. + (file-source-location, lisp-source-location) + (temp-file-source-location, source-file-source-location) + (string-source-position, code-location-debug-source-info) + (code-location-debug-source-name, code-location-debug-source-created,) + (code-location-debug-fun-fun, code-location-from-emacs-buffer-p) + (function-from-emacs-buffer-p, function-debug-source-info) + (info-from-emacs-buffer-p, code-location-has-debug-block-info-p) + (stream-source-position): Lots of new helper functions. + (with-debootstrapping): Moved upwards so that it can be used for + source location searching. + (source-location-for-emacs): Deleted + 2005-03-16 Helmut Eller <heller(a)common-lisp.net> * slime/swank.lisp (*macroexpand-printer-bindings*): New user

1 0

[slime-devel] Fancy, yes, pleasant, no: slime-complete-symbol*-fancy-bit
by Steven E. Harris 18 Mar '05

18 Mar '05

I have my '[' key bound to `insert-parentheses' so that any open-paren I type gets a symmetric closing paren inserted. Consider what happens now when I type, say, [ user-home TAB Just before hitting tab, the buffer contains the following text, with extra spaces inserted to show the point's location: ( u s e r - h o m e ) | point Hitting tab completes the symbol, but decides, since this looks like a function call position and there are no arguments for this function, to insert a closing paren: (user-homedir-pathname)) Now I have an extra closing paren hanging out there. That's `slime-complete-symbol*-fancy-bit' doing its job. A corrective patch is simple: However, this may not play nicely for those who do not use `insert-parentheses' in a situation such as this: Existing code: (foo 1 2) Decide to add a third argument (spaced out again): ( f o o 1 2 ( b a ) | point Hit tab at the indicated point position and wind up with: (foo 1 2 (bar) Now the originally balanced parens are not balanced, because my patch disabled the auto-insert-paren behavior. Is there a way to satisfy both use cases without adding another configuration variable? -- Steven E. Harris

2 2

[slime-devel] Jump to source code
by hww 18 Mar '05

18 Mar '05

I think that jumping to source code from error message of compiler and interpretetor is very important feature. I use slime with 'sbcl' and created the file 'test.l" with code: (+ 1 2) (+ 3 "4") The last line has an error Next I do (in emacs) M-x slime CL-USER>(load "test.l") Course emacs teels me abou error. Then via debugging frame and commands 'M-n', 'M-p' or 'v' I can position to source code for the all frames. But no way to point at the 'test.l' file. Why? Or how i can from eror message jump to position of this error in a source code? BTW. I checked also 'lisp' and 'clisp' - same result!

4 10

Re: [slime-devel] Jump to source code
by hww 18 Mar '05

18 Mar '05

Nikodemus Siivola wrote: >> Maybe sombody know about a good Lisp (Scheme) IDE similar LispWorks >> or Allegro? > > > If you had not decided that Slime is not what you want, I'd say > "Slime"... ... What? Everybody can say same. As you can see me also: "Slime"...

1 0

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

16 Mar '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.638 slime/ChangeLog:1.639 --- slime/ChangeLog:1.638 Sun Mar 13 20:41:00 2005 +++ slime/ChangeLog Wed Mar 16 23:11:24 2005 @@ -1,3 +1,20 @@ +2005-03-16 Helmut Eller <heller(a)common-lisp.net> + + * slime/swank.lisp (*macroexpand-printer-bindings*): New user + variable. + (apply-macro-expander): Use it. + (call-with-bindings): Bind variables in reverse order. Thit makes + it easer to cons or push a new binding at the front the list. + (with-bindings): New macro. + + * slime.el (slime-run-when-idle): New function to hide + Emacs/XEmacs differences. + (slime-process-available-input): Use it. + + * swank-loader.lisp (unique-directory-name): Rewritten to avoid + the rather irritating warning that (warn "Don't know ...") is + unreachable. + 2005-03-13 Luke Gorrie <luke(a)synap.se> * slime.el (slime-dispatch-event): Use `slime-busy-p' to control

1 0

[slime-devel] SLDB hangs with CLISP/Cygwin
by Steven E. Harris 16 Mar '05

16 Mar '05

I am having trouble expanding frames to see local variables in SLDB using SLIME in XEmacs 21.4.13 with CLISP 2.33.1 on Cygwin. If I define a function with a break expression, (defun foo (n) (break) n) then call that function, (foo 1) SLDB presents the backtrace: ,---- | Break | [Condition of type SIMPLE-CONDITION] | | Restarts: | 0: [CONTINUE] Return from BREAK loop | 1: [ABORT] Abort handling SLIME request. | | Backtrace: | 0: frame binding variables (~ = dynamically): | | ~ SYSTEM::*FASOUTPUT-STREAM* <--> NIL | 1: EVAL frame for form (BREAK) | 2: EVAL frame for form (PROGN (BREAK) N) | 3: APPLY frame for call (FOO '1) | 4: EVAL frame for form (FOO 1) | 5: EVAL frame for form (SWANK:INTERACTIVE-EVAL "(foo 1)") | 6: EVAL frame for form (SWANK:START-SERVER "/tmp/seh/slime.800" :EXTERNAL-FORMAT :ISO-LATIN-1-UNIX) `---- Moving the cursor down to, say, the "1:" line and pressing enter to expand the frame causes XEmacs to hang. The minibuffer reports: ; pipelined request... (swank:frame-locals-for-emacs 1) Interrupting XEmacs with C-g produces the following backtrace: 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) (... ... ...) (... ...)) (let (... ...) (while t ...))) (apply (function funcall) (catch tag (slime-rex ... ... ... ...) (let ... ...))) (let* ((tag ...) (slime-stack-eval-tags ...)) (apply (function funcall) (catch tag ... ...))) slime-eval((swank:frame-locals-for-emacs 1)) sldb-frame-locals(1) (let* ((i 0) (G43782 ...) (var nil)) (while (consp G43782) (setq var ...) (destructuring-bind ... var ... ...) (setq i ...) (setq G43782 ...)) nil) (catch (quote --cl-block-nil--) (let* (... ... ...) (while ... ... ... ... ...) nil)) (cl-block-wrapper (catch (quote --cl-block-nil--) (let* ... ... nil))) (block nil (let* (... ... ...) (while ... ... ... ... ...) nil)) (loop for i from 0 for var in (sldb-frame-locals frame) do (destructuring-bind (&key name id value) var (slime-propertize-region ... ... ... ...) (insert "\n"))) sldb-insert-locals(1 " ") (progn (sldb-insert-frame frame t) (insert indent1 (in-sldb-face section "Locals:") "\n") (sldb-insert-locals frame-number indent2) (when sldb-show-catch-tags (let ... ...)) (unless sldb-enable-styled-backtrace (terpri)) (point)) (prog1 (progn (sldb-insert-frame frame t) (insert indent1 ... "\n") (sldb-insert-locals frame-number indent2) (when sldb-show-catch-tags ...) (unless sldb-enable-styled-backtrace ...) (point)) (add-text-properties G43777 (point) (backquote ...))) (let ((G43777 ...)) (prog1 (progn ... ... ... ... ... ...) (add-text-properties G43777 ... ...))) (slime-propertize-region (backquote (frame ... details-visible-p t)) (sldb-insert-frame frame t) (insert indent1 (in-sldb-face section "Locals:") "\n") (sldb-insert-locals frame-number indent2) (when sldb-show-catch-tags (let ... ...)) (unless sldb-enable-styled-backtrace (terpri)) (point)) (let* ((props ...) (frame ...) (frame-number ...) (standard-output ...) (indent1 " ") (indent2 " ")) (delete-region start end) (slime-propertize-region (backquote ...) (sldb-insert-frame frame t) (insert indent1 ... "\n") (sldb-insert-locals frame-number indent2) (when sldb-show-catch-tags ...) (unless sldb-enable-styled-backtrace ...) (point))) (save-excursion (goto-char start) (let* (... ... ... ... ... ...) (delete-region start end) (slime-propertize-region ... ... ... ... ... ... ...))) (let* ((G43776 ...) (start ...) (end ...)) (save-excursion (goto-char start) (let* ... ... ...))) (multiple-value-bind (start end) (sldb-frame-region) (save-excursion (goto-char start) (let* ... ... ...))) sldb-show-frame-details() (if (or on (not ...)) (sldb-show-frame-details) (sldb-hide-frame-details)) (let ((inhibit-read-only t) (column ...)) (if (or on ...) (sldb-show-frame-details) (sldb-hide-frame-details)) (move-to-column column)) sldb-toggle-details() funcall(sldb-toggle-details) (if fn (funcall fn)) (let ((fn ...)) (if fn (funcall fn))) (lambda nil "Invoke the action at point." (interactive) (let (...) (if fn ...)))() call-interactively(sldb-default-action) If I press "c" continue from that backtrace, focus returns to the SLDB buffer, from which I can still invoke restarts. That is, the Lisp connection is still alive and synchronized with SLIME. Recent activity in the *slime-events* buffer is: ,---- | (:emacs-rex | (swank:interactive-eval "(foo 1)") | nil t 32) | (:debug 0 1 | ("Break" " [Condition of type SIMPLE-CONDITION]" nil nil) | (("CONTINUE" "Return from BREAK loop") | ("ABORT" "Abort handling SLIME request.")) | ((0 "frame binding variables (~ = dynamically):\n | ~ SYSTEM::*FASOUTPUT-STREAM* <--> NIL") | (1 "EVAL frame for form (BREAK)") | (2 "EVAL frame for form (PROGN (BREAK) N)") | (3 "APPLY frame for call (FOO '1)") | (4 "EVAL frame for form (FOO 1)") | (5 "EVAL frame for form (SWANK:INTERACTIVE-EVAL \"(foo 1)\")") | (6 "EVAL frame for form (SWANK:START-SERVER \"/tmp/seh/slime.800\" :EXTERNAL-FORMAT :ISO-LATIN-1-UNIX)")) | (32)) | (:debug-activate 0 1) | (:emacs-rex | (swank:frame-locals-for-emacs 1) | nil 0 33) | (:return | (:ok | ((:name "N" :id 0 :value "1"))) | 33) `---- -- Steven E. Harris

3 12

[slime-devel] How to disable *Slime Note* buffer?
by Andrew Stebakov 16 Mar '05

16 Mar '05

Hi, When I type in a function say "(format" and press blank space there is a hint in the *Minibuf-0* and that's fine. When I type socket:socket-connect a new buffer *Slime Note* opens and shifts my windows layout. Is it possible to prevent *Slime Note* from opening so all the hints would go to minibuffer? Why different functions hints behave differently? Thanks, Andrew __________________________________ Do you Yahoo!? Make Yahoo! your home page http://www.yahoo.com/r/hs

2 1

[slime-devel] Need to issue M-x slime twice to connect
by Andrew Stebakov 16 Mar '05

16 Mar '05

Hi, I am using XEmacs on Windows with Slime 1.0. When I enter M-x slime it says: WARNING: These Swank interfaces are unimplemented: (ADD-FD-HANDLER ADD-SIGIO-HANDLER ALL-THREADS DISASSEMBLE-FRAME FIND-THREAD GETPID INTERRUPT-THREAD RECEIVE REMOVE-FD-HANDLERS REMOVE-SIGIO-HANDLERS SEND SLDB-STEP SPAWN THREAD-ID WHO-MACROEXPANDS WHO-SPECIALIZES) ;; Loaded file C:\Linux\slime-1.0\swank-loader.lisp T [2]> And in the minibuffer it says: Polling "C:\\DOCUME~1\\stebakov\\LOCALS~1\\Temp\\slime.8132".. (Abort with `M-x slime-abort-connection'.) Now, when I enter M-x slime once again it gives me a proper prompt in the REPL buffer. How this can be fixed? Thanks, Andrew __________________________________ Do you Yahoo!? Yahoo! Small Business - Try our new resources site! http://smallbusiness.yahoo.com/resources/

2 1

[slime-devel] Slime + ecb mode
by Andrew Stebakov 16 Mar '05

16 Mar '05

Hi, I started to walk through Peter Seibel's Practical Lisp examples. I found that I can't execute the following code from the editor using (C-x C-e) commands. (defvar *db* nil) (defun make-cd (title artist rating ripped) (list :title title :artist artist :rating rating :ripped ripped)) (defun add-record (cd) (push cd *db*)) (add-record (make-cd "Roses" "Kathy Mattea" 7 t)) (add-record (make-cd "Fly" "Dixie Chicks" 8 t)) (add-record (make-cd "Home" "Dixie Chicks" 9 t)) The reason for that (I think) is that when I add a second record it tries to put two lines in the mini-buffer, when I add a third record it breaks the whole layout of ecb and loses the connection with slime. Same happens if I do all add-records from REPL but only evaluate *db* from the editor. I found out that if I issue C-u C-x C-e than the output goes in REPL window which is fine. I need a help to understand what I am facing here. Is this a correct behavior or some bug? Is it an ECB or SLIME bug? I pasted and evaluated the same code from my XEmacs scratch buffer and it prints the output from adding records in one long line (as opposed to multiple lines with Slime). Any help is greatly appreciated! Andrew I am not sure __________________________________ Do you Yahoo!? Yahoo! Small Business - Try our new resources site! http://smallbusiness.yahoo.com/resources/

2 1