February 2004 - slime-devel - mailman3.common-lisp.net

[slime-devel] Interactive compile loses DSPEC info
by Alain.Picard＠memetrics.com 03 Mar '04

03 Mar '04

Another one: I'm using latest CVS slime + Lispworks on Linux. If I compile and load my system from scratch, I can use M-. to jump around function definitions. but if I go to a function, hit C-c C-c to recompile the function, jump to a different emacs buffer and try M-. on the newly compiled function, I now get: make-single-line (:error The file #P"/tmp/filecLlfIH.lisp" does not exist.) (:location (:file /home/kooks/XOS/Bullseye/src/Common/generic-utilities.lisp) (:function-name MAKE-SINGLE-LINE)) in the *XREF buffer. Cheers,

3 7

[slime-devel] thread manipulation
by Marco Baringer 29 Feb '04

29 Feb '04

i've just commited a patch which allows you to do "stuff" with the threads in slime-list-threads. Killing threads: (bound to k in the slime-list-thread menu), simply kill-thread the thread at point. (Only implemented on OpenMCL). Attaching a slime-repl to a thread: (bound to a in the slime-list-thread menu) This will interuppt the thread, start a (fore ground) server in it and attach to that server in a new repl buffer. Debugging a thread: (bound to d in the slime-list-thread menu) This will interuppt the thread and then immediatly signal a simple-error putting you in an sldb-buffer. The backtrace will contain a bunch of extra frames intreduced by the simple-error, but all the restarts and frames are there. This has only been tested on OpenMCL and has, more than once, rendered my original (started via terminal) lisp unusable. -- Marco Ring the bells that still can ring. Forget the perfect offering. There is a crack in everything. That's how the light gets in. -Leonard Cohen

2 7

[slime-devel] What does "pipelined request" mean?
by Peter Seibel 29 Feb '04

29 Feb '04

I've got a busted connection to CLISP (along with several other connections to other Lisps which seem to be working) Whenever I type anything in the CLISP REPL I get a message from emacs of the form: ; pipelined request... (swank:listener-eval " ") What does this mean? -Peter -- Peter Seibel peter(a)javamonkey.com Lisp is the red pill. -- John Fraser, comp.lang.lisp

2 3

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 29 Feb '04

29 Feb '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.272 slime/ChangeLog:1.273 --- slime/ChangeLog:1.272 Fri Feb 27 07:32:06 2004 +++ slime/ChangeLog Sat Feb 28 04:13:26 2004 @@ -1,3 +1,21 @@ +2004-02-28 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> + + * slime.el (slime-compilation-finished-hook): New hook variable. + (slime-compilation-finished): Call it. + (slime-maybe-show-xrefs-for-notes): New function. + (slime-make-default-connection): Use the current connection. + (slime-connection-at-point): New function. + (slime-goto-connection, slime-connection-list-make-default): Use + it. + (slime-draw-connection-list): Minor cleanups. + + Define selectors for t and c for thread and connection list. + + * swank.lisp: (*initial-pprint-dispatch-table*) + (*arglist-pprint-dispatch-table*): Workaround for bug in + CLISP. Don't supply nil as argument to copy-pprint-dispatch. + (print-cons-argument): Insert a space after the car. + 2004-02-27 Marco Baringer <mb(a)bese.it> * slime.el (slime-read-port-and-connect,

1 0

[slime-devel] Re: Compiler Notes -- some suggestions
by Helmut Eller 28 Feb '04

28 Feb '04

Jan Rychter <jan(a)rychter.com> writes: > >>>>> "Helmut" == Helmut Eller <e9626484(a)stud3.tuwien.ac.at> writes: > Helmut> Jan Rychter <jan(a)rychter.com> writes: > >> But perhaps I'm missing something? > > Helmut> The primary reason for this browser is that some warnings don't > Helmut> have a source location, i.e., they don't belong to a particular > Helmut> source form but to the entire file or compilation unit. The > Helmut> other reason is that SBCL produces about hundred redefinition > Helmut> style warnings when I recompile swank-sbcl.lisp and I found it > Helmut> difficult to find the few (non-style) warnings. > > Hmm, I think I understand now. Thanks. > > >> All in all, I'd suggest making this buffer pop up optionally, I'll > >> just disable that option then, I guess. And it would be vastly more > >> useful if the keys were electric and the active extents began at the > >> beginning of the line (or if the electric keys jumped to the > >> appropriate place in the buffer). > > Helmut> OK. I'll remove this feature and only use it in my private > Helmut> setup. > > If people find it useful, perhaps it's a good candidate for a defcustom? Will make it customizable, disabled by default. > I just wanted to ask, because I intend to do some usability work on > slime (making sure that windows open and close properly, that > keybindings are consistent, etc) -- I've been wondering whether I > understand how this feature is supposed to work and whether I should try > to fix things in it. Yeah, some things could be improved. Helmut.

1 0

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 28 Feb '04

28 Feb '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.271 slime/ChangeLog:1.272 --- slime/ChangeLog:1.271 Thu Feb 26 17:31:13 2004 +++ slime/ChangeLog Fri Feb 27 07:32:06 2004 @@ -1,3 +1,25 @@ +2004-02-27 Marco Baringer <mb(a)bese.it> + + * slime.el (slime-read-port-and-connect, + slime-read-port-and-connect-to-running-swank): Refactor + slime-read-port-and-connect into two functions so that + slime-thread-attach can use the logic in + slime-read-port-and-connect. + (slime-thread-control-mode-map): Added key bindings for + slime-thread-kill, slime-thread-attach, slime-thread-debug and + slime-list-threads. + (slime-thread-kill, slime-thread-attach, slime-thread-debug): New + functions. + + * swank-backend.lisp (kill-thread): Added to swank interface. + + * swank-openmcl.lisp (kill-thread): Implement. + + * swank.lisp (start-server): Add optional background argument, + defaults to *swank-background*. + (lookup-thread-by-id): New function. + (debug-thread): New function. + 2004-02-26 Peter Seibel <peter(a)javamonkey.com> * slime.el (slime-draw-connection-list): Use text-properties to

1 0

[slime-devel] CLISP breakage
by Peter Seibel 27 Feb '04

27 Feb '04

So I just noticed that due to a bug in CLISP's implementation of COPY-PPRINT-DISPATCH swank.lisp won't load. I sent a bug report to clisp-devel; in the meantime here's a patch that will get CLISP users back on their feet, though depending on how you read the spec, this may or may not really be equivalent. (I assume this use of copy-pprint-dispatch went in recently since I was having no problems with CLISP until today.) -Peter Index: swank.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank.lisp,v retrieving revision 1.127 diff -u -r1.127 swank.lisp --- swank.lisp 26 Feb 2004 07:13:01 -0000 1.127 +++ swank.lisp 27 Feb 2004 20:31:12 -0000 @@ -697,7 +697,7 @@ ;;; cover cases like (&key (function #'cons) (quote 'quote)). Too ;;; much code for such a minor feature? -(defvar *initial-pprint-dispatch-table* (copy-pprint-dispatch nil)) +(defvar *initial-pprint-dispatch-table* (copy-pprint-dispatch)) (defun print-cons-argument (stream object) (pprint-logical-block (stream object :prefix "(" :suffix ")") @@ -710,7 +710,7 @@ (princ object stream))) (defvar *arglist-pprint-dispatch-table* - (let ((table (copy-pprint-dispatch nil))) + (let ((table (copy-pprint-dispatch))) (set-pprint-dispatch 'cons #'print-cons-argument 0 table) (set-pprint-dispatch 'symbol #'print-symbol-argument 0 table) table)) -- Peter Seibel peter(a)javamonkey.com Lisp is the red pill. -- John Fraser, comp.lang.lisp

1 0

[slime-devel] Compiler Notes -- some suggestions
by Jan Rychter 27 Feb '04

27 Feb '04

I've been thinking about the Compiler Notes buffer that pops up after compilation. I have to admit I don't find it particularly useful as it is now, for a number of reasons: -- it shows very little information by default: -+ All (21) |[+] Notes (8) `[+] Warnings (13) -- it obscures my REPL, -- I have to do "C-x o C-x k" to kill it (or at least "C-x o q"), which I consider to be a major usability flaw, -- in the buffer, keys are not electric, and the usual "n" or "p" don't get me anywhere, -- the active extents where pressing RET would take me to the code do not begin at the beginning of the line, so even if I do C-n several times, I still have to C-f to the appropriate part of the tree where I can press RET, -- overall, there seems to be no information in the tree that I wouldn't normally get by just pressing M-n or M-p to navigate the compiler notes. But perhaps I'm missing something? I toyed with the idea of making this buffer more useful by adding electric "n" and "p" commands, which would also adjust the source buffer. That would be similar to how M-n and M-p keys work in the source buffer. But then I discovered that the data structure used is a tree, so jumping across leafs isn't all that easy. And I can't really understand why it is a tree? Only to provide the "All", "Warnings", and "Notes" branches, or for some other reason? Do we have nested warnings? All in all, I'd suggest making this buffer pop up optionally, I'll just disable that option then, I guess. And it would be vastly more useful if the keys were electric and the active extents began at the beginning of the line (or if the electric keys jumped to the appropriate place in the buffer). --J.

1 0

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 27 Feb '04

27 Feb '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.264 slime/ChangeLog:1.268 --- slime/ChangeLog:1.264 Wed Feb 25 02:29:59 2004 +++ slime/ChangeLog Thu Feb 26 02:19:07 2004 @@ -1,3 +1,40 @@ +2004-02-26 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> + + * swank-backend.lisp (:swank): export connection-info. + + * swank-allegro.lisp (lisp-implementation-type-name): Implement + it. + + * swank-sbcl.lisp (compile-file-for-emacs): Load the fasl file + regardless of f-p. + + * swank.lisp (swank-pprint): Bind *package* to *buffer-package*. + Reported by Alan Picard. + + * swank-lispworks.lisp (dspec-buffer-position): Renamed from + dspec-buffer-buffer-position. Handle dspecs of the form (defmacro + foo). Reported by Alan Picard. + (arglist-string): Handle unknown arglists properly. + +2004-02-25 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> + + * swank-cmucl.lisp (arglist-string): Delay the call to + di::function-debug-function until it is actually needed. + (compile-file-for-emacs): Load the fasl file irrespective of + COMILE-FILE's third return value. + + * swank.lisp (connection-info): New function. + (open-streams): Don't send the :check-protocol-version message. Now + handled with CONNECTION-INFO. + + * slime.el (slime-symbol-at-point): Don't skip backwards across + whitespace when we are at the first character of a symbol. To + handle this case: skip symbol constituents forward before skipping + whitespace backwards. Reported by Jan Richter. + (slime-connection-close-hook, slime-next-connection) + (slime-make-default-connection): Remove extra call to format. + (slime-init-connection-state): Use only a single RPC instead of 4. + 2004-02-25 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> * slime.el (slime-with-chosen-connection): Bind @@ -5,8 +42,10 @@ slime-buffer-connection is a buffer local variable not a dynamic variable. (slime-find-connection-by-type-name) - (slime-read-lisp-implementation-type-name): Where lost during the + (slime-read-lisp-implementation-type-name): Were lost during the merge. + (sldb-fetch-more-frames): Use (goto-char (point-max)) instead of + end-of-buffer. 2004-02-25 Peter Seibel <peter(a)javamonkey.com>

3 2

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 27 Feb '04

27 Feb '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.268 slime/ChangeLog:1.271 --- slime/ChangeLog:1.268 Thu Feb 26 02:19:07 2004 +++ slime/ChangeLog Thu Feb 26 17:31:13 2004 @@ -1,3 +1,21 @@ +2004-02-26 Peter Seibel <peter(a)javamonkey.com> + + * slime.el (slime-draw-connection-list): Use text-properties to + associate the connections each line of the connections list + buffer. + +2004-02-26 Peter Seibel <peter(a)javamonkey.com> + + * slime.el (slime-list-connections): Make the buffer created by + this function do a bit more: Can use it to switch to different + connections and change the default. + +2004-02-26 Marco Baringer <mb(a)bese.it> + + * swank-openmcl.lisp (ccl::force-break-in-listener): Pass a + condition object to invoke-debugger. + Patch by Bryan O'Connor <bryan-slime(a)lunch.org> + 2004-02-26 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> * swank-backend.lisp (:swank): export connection-info.

1 0

[slime-devel] Enhanced slime-list-connections
by Peter Seibel 26 Feb '04

26 Feb '04

Okay, so you gave me commit privileges and I'm taking advantage of them. I just checked in a change that makes slime-list-connections a bit more useful: now you can use it to switch to the REPL (hit enter on the line you want) for a particular connection and to change the default connection (hit "d") on the appropriate line. "q" still gets you out of the buffer. If someone wants to look over my change and let me know if I violated any SLIME norms that'd be fine. -Peter -- Peter Seibel peter(a)javamonkey.com Lisp is the red pill. -- John Fraser, comp.lang.lisp

2 2

[slime-devel] slime-xref-mode
by bryan o'connor 26 Feb '04

26 Feb '04

[mac os x 10.3 / emacs 21.3.50.1 (carbon) / cvs slime / openmcl 0.14.1] when slime-edit-fdefinition pops up a xref buffer for selecting a location, the xref keymap doesn't seem to be active. for example, " " is bound to slime-space instead of slime-goto-xref. manually calling slime-goto-xref from the minibuffer does the right thing. it did work at least a week or two ago. unfortunately i don't know enough about emacs keymaps to debug this one. ...bryan

2 2

[slime-devel] swank-openmcl bug + patch
by bryan o'connor 26 Feb '04

26 Feb '04

the redefinition of ccl::force-break-in-listener in swank-openmcl.lisp relied on the more-featured-but-not-ansi pre-0.14.1 behavior of invoke-debugger. the following patch accomplishes the same thing by explicitly creating the condition as the argument to invoke-debugger. it should work for pre-0.14.1 versions, too. ...bryan Index: swank-openmcl.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank-openmcl.lisp,v retrieving revision 1.64 diff -u -r1.64 swank-openmcl.lisp --- swank-openmcl.lisp 24 Feb 2004 23:31:34 -0000 1.64 +++ swank-openmcl.lisp 26 Feb 2004 18:22:16 -0000 @@ -114,7 +114,8 @@ (setq *swank-debugger-stack-frame* p) (return-from find-frame)) (setq previous-f (ccl::lfun-name lfun))))) - (restart-case (invoke-debugger) + (restart-case (invoke-debugger + (make-condition 'simple-condition :format-control "")) (continue () :report (lambda (stream) (write-string "Resume interrupted evaluation" stream)) t)) ))))))

2 1

[slime-devel] slime-edit-fdefinition blows up
by Alain.Picard＠memetrics.com 26 Feb '04

26 Feb '04

Dear all, On slime + Lispworks: Sometimes slime-edit-fdefinition can get pretty confused; placing point after DEFINE-TOOLTIP in the following form and hitting M-. (define-tooltip data-manipulation-node "Some generic info on import/exports" :br "You can put " (:strong "arbitrary") "html code in here.") emacs asks me "Enter recursive edit? (Y/N)" Hitting Y, and opening a few key frames, I get the following backtraces. Which ECASE expression are we falling out of? Thanks. ================================================================ DEFMACRO fell through ECASE expression. Wanted one of (DEFUN METHOD LISPWORKS:TOP-LEVEL-FORM). [Condition of type CONDITIONS:CASE-FAILURE] Restarts: 0: [ABORT] Return to SLIME toplevel. 1: [ABORT] Quit process. Backtrace: 0: GET-CALL-FRAME (FRAME-POINTER PREV-FRAME) Locals: DBG::FRAME-POINTER = (6920 . #<pumpkin, tag 0 C9590030>) DBG::PREV-FRAME = #<unbound> [No catch-tags] 1: SWANK-DEBUGGER-HOOK (CONDITION HOOK) 2: DEBUG1 (&OPTIONAL DATUM &REST ARGUMENTS) 3: INVOKE-DEBUGGER (&REST X) 4: CONDITIONS-ERROR (INVISIBLEP T DATUM ARGUMENTS) 5: MAKE-DSPEC-LOCATION (DSPEC LOCATION &OPTIONAL TMPFILE BUFFER POSITION) Locals: SWANK::DSPEC = (DEFMACRO DEFINE-TOOLTIP) SWANK::LOCATION = #P"XOS:UI;USER-INTERFACE.LISP.NEWEST" SWANK::TMPFILE = NIL SWANK::BUFFER = NIL POSITION = NIL [No catch-tags] 6: DSPEC-SOURCE-LOCATIONS (DSPEC) Locals: SWANK::DSPEC = DEFINE-TOOLTIP SWANK::LOCATIONS = (((DEFMACRO DEFINE-TOOLTIP) #P"XOS:UI;USER-INTERFACE.LISP.NEWEST")) SWANK::DSPEC = (DEFMACRO DEFINE-TOOLTIP) SWANK::LOCATION = #P"XOS:UI;USER-INTERFACE.LISP.NEWEST" DBG::|tail-| = NIL DBG::|accumulator-| = (NIL) DBG::|aux-var-| = (NIL) [No catch-tags] 7: %EVAL-NOHOOK (EXP) 8: PROCESS-TOP-LEVEL (*CURRENT-FORM* ENV CONT &OPTIONAL MODE) 9: EVAL-STRING (STRING BUFFER-PACKAGE ID) Locals: STRING = "(swank:find-function-locations \"define-tooltip\")" SWANK::BUFFER-PACKAGE = "xos" SWANK::ID = 62 *DEBUGGER-HOOK* = #<function SWANK::SWANK-DEBUGGER-HOOK 21F8C54A> SWANK::OK = NIL SWANK::RESULT = NIL SWANK::*BUFFER-PACKAGE* = #<PACKAGE XOS> SWANK::*SWANK-STATE-STACK* = (:EVAL-STRING :HANDLE-REQUEST) [No catch-tags] 10: FUNCALL (FUNCTION &REST ARGS) 11: (SUBFUNCTION 2 HANDLE-REQUEST) NIL 12: CALL-WITH-REDIRECTED-IO (CONNECTION FUNCTION) 13: HANDLE-REQUEST NIL 14: (SUBFUNCTION PROCESS-SG-FUNCTION INITIALIZE-PROCESS-STACK) (IGNORED) 15: %%FIRST-CALL-TO-STACK NIL --more--

2 1

[slime-devel] Macroexpand in "wrong" package?
by Alain.Picard＠memetrics.com 26 Feb '04

26 Feb '04

I'm using latest CVS slime + Lispworks on Linux. Even after doing a SLIME-REPL-SET-PACKAGE to the correct package, macroexpanding a form produces an expansion which is being printed from the CL-USER package; e.g. C-c <RET> with point at start of (define-action-button modify-traffic-button :display-name "Modify Rates" :operation modify-traffic-op) Yields this: (XOS::DEFINE-ACTION-BUTTON-WIDGET XOS::MODIFY-TRAFFIC-BUTTON "Modify Rates" XOS::MODIFY-TRAFFIC-OP T NIL (LAMBDA (XOS::OPERAND) (XOS::INVOKE-OPERATION-SCRIPT 'XOS::MODIFY-TRAFFIC-OP XOS::OPERAND NIL))) I _know_ I'm in the darn XOS package already, so I'd prefer to see the expansion you get if you type (pprint (macroexpand '(define-action-button modify-traffic-button :display-name "Modify Rates" :operation modify-traffic-op))) in the REPL; namely: (PROGN (WHEN (GET 'MODIFY-TRAFFIC-BUTTON 'HTML-WIDGET-FACTORY) (CLEAR-EXISTING-WIDGETS 'MODIFY-TRAFFIC-BUTTON)) (SETF (GET 'MODIFY-TRAFFIC-BUTTON 'HTML-WIDGET-FACTORY) (LAMBDA () (MAKE-INSTANCE 'ACTION-BUTTON-WIDGET :NAME "Modify Rates" :OPERATION 'MODIFY-TRAFFIC-OP :DISABLED-MESSAGE (CANONICALIZE-TO-FUNCTION NIL) :ENABLED-P (CANONICALIZE-TO-FUNCTION T) :JSCRIPT-CODE (LAMBDA (OPERAND) (INVOKE-OPERATION-SCRIPT 'MODIFY-TRAFFIC-OP OPERAND NIL))))) 'MODIFY-TRAFFIC-BUTTON) Any way to get this behaviour?

3 4

[slime-devel] Weird one to do with pretty-printing
by Peter Seibel 26 Feb '04

26 Feb '04

Okay, this one is quite strange. First define this function: (defun foo () (write (macroexpand-1 '(with-open-file (in "foo.txt") (read in))) :pretty t)) Then invoke it several times. What I observe is that the body of the let moves to the right each time I invoke FOO, as if the pretty-printing stream is somehow preserving some information about what column it is indenting to from call to call. I believe I had observed this behavior before when I was screwing around with pretty-printing logical blocks. This weird shifting only happens in the SLIME repl, the same function invoked from the *inferior-lisp* buffer works as expected. I've observed this in both OpenMCL and in Allegro on GNU/Linux. -Peter -- Peter Seibel peter(a)javamonkey.com Lisp is the red pill. -- John Fraser, comp.lang.lisp

2 1

[slime-devel] slime-compile-and-load-file doesn't
by Jan Rychter 25 Feb '04

25 Feb '04

slime-compile-and-load-file compiles and displays the compiler notes, but doesn't load the file. I can't find anything around slime-compilation-finished-continuation that would actually load the file. I'm puzzled now -- doesn't anyone use C-c C-k, except me? :-) --J.

2 3

[slime-devel] Suggestion: improve sldb-fetch-more-frames
by Jan Rychter 25 Feb '04

25 Feb '04

I've just tried to do a cut&paste on a large backtrace in the sldb buffer. Unfortunately, that doesn't work too well, because when I mark the beginning and start going down in the backtrace, as soon as I pass the --more-- marker, the selection gets changed. I've tried adding a save-excursion to sldb-fetch-more-frames, but for some reason that doesn't quite do the trick. The beginning of my region then stays untouched, but after passing the --more-- marker, the end of my region jumps to the end of the backtrace. Which means, I find the current behavior annoying, but I don't know how to fix it :-( --J.

4 4

[slime-devel] Suggestion: slime-symbol-at-point
by Jan Rychter 25 Feb '04

25 Feb '04

I found it confusing that slime-symbol-at-point doesn't really return the symbol at point. An example: (defun install-package (source system packagename)... Placing the point on the first "i" in the function name and pressing C-c C-d returns the decription for defun, not install-package. The following fixes it, but I keep scratching my head to understand why the (skip-syntax-backward "-") was originally there -- I can't find a purpose for it. --J. diff -c /home/jwr/Mail/.software/slime/slime.el /home/jwr/Mail/.software/slime/slime.el.original --- /home/jwr/Mail/.software/slime/slime.el 2004-02-25 03:48:33.000000000 -0800 +++ /home/jwr/Mail/.software/slime/slime.el.original 2004-02-25 03:48:33.000000000 -0800 @@ -769,7 +769,6 @@ (defun slime-symbol-at-point () "Return the symbol at point, otherwise nil." (save-excursion - (skip-syntax-backward "-") (let ((string (thing-at-point 'symbol))) (if string (intern (substring-no-properties string)) nil))))

2 2

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 25 Feb '04

25 Feb '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.262 slime/ChangeLog:1.264 --- slime/ChangeLog:1.262 Mon Feb 23 02:21:31 2004 +++ slime/ChangeLog Wed Feb 25 02:29:59 2004 @@ -1,3 +1,26 @@ +2004-02-25 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> + + * slime.el (slime-with-chosen-connection): Bind + slime-dispatching-connection and not slime-buffer-connection. + slime-buffer-connection is a buffer local variable not a dynamic + variable. + (slime-find-connection-by-type-name) + (slime-read-lisp-implementation-type-name): Where lost during the + merge. + +2004-02-25 Peter Seibel <peter(a)javamonkey.com> + + * slime.el: Various bits of support for maintaining multiple SLIME + connections to different Lisp implementations simultaneously. + + * swank-backend.lisp (lisp-implementation-type-name): Add function to + return simple name of lisp implementation; used by new + multi-connection functionality in slime.el. + +2004-02-25 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> + + * swank.lisp (format-arglist): Use a special pprint-dispatch table. + 2004-02-22 Lawrence Mitchell <wence(a)gmx.li> * swank.lisp (format-arglist): Bind *PRINT-PRETTY* to NIL.

1 0

[slime-devel] PATCH: Some more support for running multiple simultaneous connections
by Peter Seibel 25 Feb '04

25 Feb '04

Here's a patch containing some code I wrote to support running multiple simultaneous SLIME connections to different Lisp implementations. I think I made about as good use of the existing infrastructure as I could but may have missed something. (Speaking of which, is there a reason certain buffer local vars (e.g. slime-connection-number) aren't defined with slime-def-connection-var?) Anyway, here it is: Index: ChangeLog =================================================================== RCS file: /project/slime/cvsroot/slime/ChangeLog,v retrieving revision 1.262 diff -u -r1.262 ChangeLog --- ChangeLog 23 Feb 2004 07:21:31 -0000 1.262 +++ ChangeLog 24 Feb 2004 19:17:11 -0000 @@ -1,3 +1,12 @@ +2004-02-24 Peter Seibel <peter(a)javamonkey.com> + + * slime.el: Various bits of support for maintaining multiple SLIME + connections to different Lisp implementations simultaneously. + + * swank.lisp (lisp-implementation-type-name): Add function to + return simple name of lisp implementation; used by new + multi-connection functionality in slime.el. + 2004-02-22 Lawrence Mitchell <wence(a)gmx.li> * swank.lisp (format-arglist): Bind *PRINT-PRETTY* to NIL. Index: slime.el =================================================================== RCS file: /project/slime/cvsroot/slime/slime.el,v retrieving revision 1.216 diff -u -r1.216 slime.el --- slime.el 23 Feb 2004 22:14:56 -0000 1.216 +++ slime.el 24 Feb 2004 19:17:13 -0000 @@ -633,6 +633,15 @@ (put 'destructure-case 'lisp-indent-function 1) + +(defmacro* slime-with-chosen-connection ((n) &rest body) + "Make the connection choosen by a universal argument current." + `(let ((slime-buffer-connection (slime-get-named-connection ,n slime-buffer-connection))) + ,@body)) + +(put 'slime-with-chosen-connection 'lisp-indent-function 1) + + (defmacro slime-define-keys (keymap &rest key-command) `(progn . ,(mapcar (lambda (k-c) `(define-key ,keymap . ,k-c)) key-command))) @@ -1229,8 +1238,9 @@ (when (eq process slime-default-connection) (when slime-net-processes (slime-select-connection (car slime-net-processes)) - (message (format "Default connection closed; switched to #%S" - (slime-connection-number)))))) + (message (format "Default connection closed; switched to #%S (%S)" + (slime-connection-number) + (slime-lisp-implementation-type-name)))))) (defun slime-connection-number (&optional connection) (slime-with-connection-buffer (connection) @@ -1249,7 +1259,18 @@ (length slime-net-processes)) slime-net-processes))) (slime-select-connection conn) - (message (format "Selected connection #%S" (slime-connection-number))))) + (message (format "Selected connection #%S (%s)" + (slime-connection-number) + (slime-lisp-implementation-type-name))))) + +(defun slime-make-default-connection () + "Make the current buffer connection the default connection." + (interactive) + (slime-select-connection slime-buffer-connection) + (message (format "Connection #%S (%s) now default SLIME connection." + (slime-connection-number) + (slime-lisp-implementation-type-name)))) + (put 'slime-with-connection-buffer 'lisp-indent-function 1) @@ -1299,6 +1320,9 @@ (slime-def-connection-var slime-lisp-implementation-type nil "The implementation type of the Lisp process.") +(slime-def-connection-var slime-lisp-implementation-type-name nil + "The short name for the implementation type of the Lisp process.") + (slime-def-connection-var slime-use-sigint-for-interrupt nil "If non-nil use a SIGINT for interrupting.") @@ -1398,6 +1422,8 @@ (setf (slime-pid) (slime-eval '(swank:getpid))) (setf (slime-lisp-implementation-type) (slime-eval '(cl:lisp-implementation-type))) + (setf (slime-lisp-implementation-type-name) + (slime-eval '(swank:lisp-implementation-type-name))) (setq slime-state-name "") (when-let (repl-buffer (slime-repl-buffer)) ;; REPL buffer already exists - update its local @@ -1766,6 +1792,41 @@ (defvar slime-repl-mode-map) +(defun slime-switch-to-repl (p) + "Switch to the REPL. With a prefix arg, prompt for a named REPL." + (interactive "p") + (cond + ((= p 1) + (slime-switch-to-output-buffer)) + ((or (= p 4) (= p 16)) + (switch-to-buffer + (slime-find-connection-buffer-by-type-name (slime-read-lisp-implementation-type-name))) + (if (= p 16) (slime-make-default-connection))))) + +(defun slime-find-connection-by-type-name (name) + (dolist (p slime-net-processes) + (let ((slime-dispatching-connection p)) + (slime-with-connection-buffer (p) + (when (string= (slime-lisp-implementation-type-name) name) + (return p)))))) + +(defun slime-find-connection-buffer-by-type-name (name) + (let* ((p (slime-find-connection-by-type-name name)) + (slime-dispatching-connection p)) + (slime-with-connection-buffer (p) + (slime-repl-buffer)))) + +(defun slime-read-lisp-implementation-type-name () + (let ((default (slime-lisp-implementation-type-name))) + (completing-read + (format "Name (default %s): " default) + (mapcar #'(lambda (p) (let ((slime-dispatching-connection p)) (slime-with-connection-buffer (p) (list (slime-lisp-implementation-type-name))))) slime-net-processes) + nil + t + nil + nil + default))) + (defun slime-repl-buffer (&optional create) "Get the REPL buffer for the current connection; optionally create." (funcall (if create #'get-buffer-create #'get-buffer) @@ -2219,15 +2280,20 @@ ;;; Compilation and the creation of compiler-note annotations -(defun slime-compile-and-load-file () +(defun slime-compile-and-load-file (n) "Compile and load the buffer's file and highlight compiler notes. Each source location that is the subject of a compiler note is underlined and annotated with the relevant information. The commands `slime-next-note' and `slime-previous-note' can be used to navigate between compiler notes and to display their full details." - (interactive) - (slime-compile-file t)) + (interactive "p") + (slime-with-chosen-connection (n) + (slime-compile-file t))) + +(defun slime-get-named-connection (n default) + "Get a named connection based on an interactive `p' argument." + (case n (1 default) (4 (slime-find-connection-by-type-name (slime-read-lisp-implementation-type-name))))) (defun slime-compile-file (&optional load) "Compile current buffer's file and highlight resulting compiler notes. @@ -2272,19 +2338,22 @@ (slime-compilation-finished-continuation t)) (message "Compiling system %s.." system-name)) -(defun slime-compile-defun () +(defun slime-compile-defun (n) "Compile the current toplevel form." - (interactive) - (slime-compile-string (slime-defun-at-point) - (save-excursion - (end-of-defun) - (beginning-of-defun) - (point)))) + (interactive "p") + (slime-with-chosen-connection (n) + (slime-compile-string + (slime-defun-at-point) + (save-excursion + (end-of-defun) + (beginning-of-defun) + (point))))) -(defun slime-compile-region (start end) +(defun slime-compile-region (n start end) "Compile the region." - (interactive "r") - (slime-compile-string (buffer-substring-no-properties start end) start)) + (interactive "p\nr") + (slime-with-chosen-connection (n) + (slime-compile-string (buffer-substring-no-properties start end) start))) (defun slime-compile-string (string start-offset) (slime-eval-async @@ -4639,18 +4708,21 @@ (kill-buffer "*SLIME connections*")) (slime-with-output-to-temp-buffer "*SLIME connections*" (let ((default (slime-connection))) - (insert " Nr Type Port Pid\n" - " -- ---- ---- ---\n") + (insert + (format "%s%2s %-7s %-17s %-7s %-s\n" " " "Nr" "Name" "Port" "Pid" "Type")) + (insert + (format "%s%2s %-7s %-17s %-7s %-s\n" " " "--" "----" "----" "---" "----")) (dolist (p slime-net-processes) (let ((slime-dispatching-connection p)) (insert (slime-with-connection-buffer (p) - (format "%s%2d %-20s %-17s %-5s\n" + (format "%s%2d %-7s %-17s %-7s %-s\n" (if (eq default p) "*" " ") (slime-connection-number) - (slime-lisp-implementation-type) + (slime-lisp-implementation-type-name) (or (process-id p) (process-contact p)) - (slime-pid))))))))) + (slime-pid) + (slime-lisp-implementation-type))))))))) ;;; Inspector Index: swank.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank.lisp,v retrieving revision 1.124 diff -u -r1.124 swank.lisp --- swank.lisp 23 Feb 2004 07:21:07 -0000 1.124 +++ swank.lisp 24 Feb 2004 19:17:14 -0000 @@ -1483,6 +1483,18 @@ (setq *thread-list* nil)) + + +;;;; Meta-info about Lisp + +(defslimefun lisp-implementation-type-name () + #+cmu "cmu" + #+sbcl "sbcl" + #+openmcl "openmcl" + #+lispworks "lispworks" + #+allegro "allegro" + #+clisp "clisp") + ;;; Local Variables: ;;; eval: (font-lock-add-keywords 'lisp-mode '(("(\\(defslimefun\\)\\s +\\(\\(\\w\\|\\s_\\)+\\)" (1 font-lock-keyword-face) (2 font-lock-function-name-face)))) ;;; End: -- Peter Seibel peter(a)javamonkey.com Lisp is the red pill. -- John Fraser, comp.lang.lisp

2 3

[slime-devel] Printing of function arglists
by lawrence mitchell 23 Feb '04

23 Feb '04

I notice, occasionally, that slime prints arglists as: (some-func #'foo &rest body) This is due to the pretty-printing of the arglist, it should be: (some-func (function foo) &rest body) I'd propose the following change, unless it's likely to have unwanted other effects. (This patch also fixes a typo in EVAL-IN-EMACS). Index: ChangeLog =================================================================== RCS file: /project/slime/cvsroot/slime/ChangeLog,v retrieving revision 1.261 diff -u -r1.261 ChangeLog --- ChangeLog 21 Feb 2004 16:45:25 -0000 1.261 +++ ChangeLog 22 Feb 2004 23:54:46 -0000 @@ -1,3 +1,8 @@ +2004-02-22 Lawrence Mitchell <wence(a)gmx.li> + + * swank.lisp (format-arglist): Bind *PRINT-PRETTY* to NIL. + (eval-in-emacs): Fix typo in docstring. + 2004-02-21 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> Add support for SERVE-EVENT based communication. Index: swank.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank.lisp,v retrieving revision 1.123 diff -u -r1.123 swank.lisp --- swank.lisp 21 Feb 2004 16:37:27 -0000 1.123 +++ swank.lisp 22 Feb 2004 23:47:25 -0000 @@ -691,7 +691,8 @@ (let ((symbol (find-symbol-or-lose function-name))) (values (funcall lambda-list-fn symbol)))) (cond (condition (format nil "(-- ~A)" condition)) - (t (let ((*print-case* :downcase)) + (t (let ((*print-case* :downcase) + (*print-pretty* nil)) (format nil "(~{~A~^ ~})" arglist)))))) @@ -795,7 +796,7 @@ ;;;; Evaluation (defun eval-in-emacs (form) - "Execute FROM in Emacs." + "Execute FORM in Emacs." (destructuring-bind (fn &rest args) form (send-to-emacs `(:%apply ,(string-downcase (string fn)) ,args)))) -- lawrence mitchell <wence(a)gmx.li>

3 5

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 23 Feb '04

23 Feb '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.261 slime/ChangeLog:1.262 --- slime/ChangeLog:1.261 Sat Feb 21 11:45:25 2004 +++ slime/ChangeLog Mon Feb 23 02:21:31 2004 @@ -1,3 +1,10 @@ +2004-02-22 Lawrence Mitchell <wence(a)gmx.li> + + * swank.lisp (format-arglist): Bind *PRINT-PRETTY* to NIL. + (eval-in-emacs): Fix typo in docstring. + + * swank-cmucl.lisp (arglist-string): Bind *PRINT-PRETTY* to NIL. + 2004-02-21 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> Add support for SERVE-EVENT based communication.

1 0

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 22 Feb '04

22 Feb '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.260 slime/ChangeLog:1.261 --- slime/ChangeLog:1.260 Sat Feb 21 02:37:36 2004 +++ slime/ChangeLog Sat Feb 21 11:45:25 2004 @@ -1,5 +1,31 @@ 2004-02-21 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> + Add support for SERVE-EVENT based communication. + + * swank-sbcl.lisp (add-sigio-handler, remove-sigio-handlers): + Renamed. + (add-fd-handler, remove-fd-handlers): Implement interface. + + * swank-cmucl.lisp (fcntl): New function. + (add-sigio-handler, remove-sigio-handlers): Renamed. + (add-fd-handler, remove-fd-handlers): Implement interface. + + * swank.lisp (create-connection): Add support for fd-handlers. + (install-fd-handler, deinstall-fd-handler): New functions. + + * swank-backend.lisp (add-sigio-handler): Renamed from + add-input-handler. + (remove-sigio-handlers): Renamed from remove-input-handlers. + (add-fd-handler, remove-fd-handlers): New interface functions. + + * slime.el (slime-batch-test): Use sit-for instead of + accept-process-output, so that we see something when swank gets + compiled. May be problematic in real batch mode. + (loop-interrupt-continue-interrupt-quit): Wait a second before + interrupting. The signal seems to arrive before the evaluation + request if don't wait => the endless loop is executed inside the + debugger and sldb-quit will not be processed with fd-handlers. + * swank.lisp (process-available-input): Move auxiliary function to toplevel. Test if the stream is open. (install-sigio-handler): Handle the first request after installing

1 0

[slime-devel] SLIME and OpenMCL 0.14.1: debugger question
by Constantine Vetoshev 21 Feb '04

21 Feb '04

Greetings all, I'm giving SLIME a spin, and I fetched a FAIRLY-STABLE version from CVS yesterday. M-x slime worked fine, and I had a REPL buffer. Unfortunately, whenever I drop into the OpenMCL debugger and choose a restart (like ABORT or ABORT-BREAK), all further attempts to interact with the REPL result in a "Lisp is not ready for requests from the REPL" message, and I have to restart SLIME. ABORT-BREAK actually gave me a CL-USER prompt, but with the same error message. Do I simply misunderstand what's going on? Is this an OpenMCL issue? Some kind of threading issue? I tried searching on gmane for this error, but it looks like people generally run into it when trying to do several things at once. I just evaluated (error "test error") on the Lisp prompt; I expect to just enter the debugger and choose a restart. Many thanks, CV

2 2