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

[slime-devel] *break-on-signals* in SBCL breaks CVS slime
by Alain Picard 23 Sep '09

23 Sep '09

In the CVS version as of Wed Sep 23 11:04:18 EST 2009, I get this behaviour: I type CL-USER> (setf *break-on-signals* t) T CL-USER> (just hit <cr> here) and the world blows up. This is with current SLIME and SBCL 1.0.30 on an x86_64 GNU/Linux box. The *inferior-lisp* buffer shows: READ: "(:emacs-rex (swank:listener-eval \" \") \"COMMON-LISP-USER\" :repl-thread 13) " close-connection: invalid number of arguments: 4 ... wait-for-event: (:SLDB-RETURN 2) T ;; Event history end. ;; Backtrace: (0 ((LAMBDA (SWANK-BACKEND::DEBUGGER-LOOP-FN)) #<FUNCTION (LAMBDA #) {10035078F9}>)) (1 (SWANK::SAFE-BACKTRACE)) (2 ((FLET #:LAMBDA1493) #<SB-INT:SIMPLE-PROGRAM-ERROR {1003A0D281}>)) (3 (SIGNAL #<SB-INT:SIMPLE-PROGRAM-ERROR {1003A0D281}>)[:EXTERNAL]) (4 (ERROR SB-INT:SIMPLE-PROGRAM-ERROR)[:EXTERNAL]) (5 (SB-KERNEL::INVALID-ARG-COUNT-ERROR-HANDLER ..)) (6 (SB-KERNEL:INTERNAL-ERROR #.(SB-SYS:INT-SAP #X7FFFF46B7940) #<unavailable argument>)) (7 ("foreign function: #x4222F0")) (8 ("foreign function: #x40D11C")) (9 ("foreign function: #x413B75")) (10 ("foreign function: #x411CEA")) (11 (SWANK::EVAL-FOR-EMACS (SWANK:LISTENER-EVAL "\n") "COMMON-LISP-USER" :REPL-THREAD)[:EXTERNAL]) (12 (SWANK::SLDB-LOOP 1)) (13 ((LAMBDA (SWANK-BACKEND::DEBUGGER-LOOP-FN)) #<FUNCTION (LAMBDA #) {10035FE359}>)) (14 (SWANK::DEBUG-IN-EMACS #<SIMPLE-CONDITION {10035727D1}>)) (15 (SWANK:INVOKE-SLIME-DEBUGGER #<SIMPLE-CONDITION {10035727D1}>)) (16 ((LAMBDA (SWANK-BACKEND::HOOK SWANK-BACKEND::FUN)) #<FUNCTION SWANK:SWANK-DEBUGGER-HOOK> #<CLOSURE (LAMBDA #) {10035727F9}>)) (17 (SWANK:SWANK-DEBUGGER-HOOK #<SIMPLE-CONDITION {10035727D1}> #<unavailable argument>)) (18 (INVOKE-DEBUGGER #<SIMPLE-CONDITION {10035727D1}>)) (19 (SB-INT:%BREAK BREAK "~A~%BREAK was entered because of *BREAK-ON-SIGNALS* ~\n (now rebound to NIL)." #<TIMEOUT {10035721E1}>)) (20 (BREAK "~A~%BREAK was entered because of *BREAK-ON-SIGNALS* ~\n (now rebound to NIL)." #<TIMEOUT {10035721E1}>)) (21 (SIGNAL #<TIMEOUT {10035721E1}>)[:EXTERNAL]) (22 (CERROR "Continue" TIMEOUT)[:EXTERNAL]) (23 ((FLET SB-THREAD::WITH-RECURSIVE-LOCK-THUNK))) (24 ((FLET #:WITHOUT-INTERRUPTS-BODY-[CALL-WITH-RECURSIVE-LOCK]291))) (25 ((FLET #:WITHOUT-INTERRUPTS-BODY-[MAKE-CANCELLABLE-INTERRUPTOR]141))) (26 ((FLET #:WITHOUT-INTERRUPTS-BODY-[WITH-SYSTEM-MUTEX-THUNK]441))) (27 ((FLET #:WITHOUT-INTERRUPTS-BODY-[WITHOUT-INTERRUPTS-BODY-[INVOKE-INTERRUPTION]11]19))) (28 ((FLET #:WITHOUT-INTERRUPTS-BODY-[INVOKE-INTERRUPTION]11))) (29 (SB-SYS:INVOKE-INTERRUPTION #<CLOSURE (FLET SB-UNIX::INTERRUPTION) {7FFFF46B8A89}>)) (30 (SB-SYS:INVOKE-INTERRUPTION #<CLOSURE (FLET SB-UNIX::INTERRUPTION) {7FFFF46B8A89}>)[:EXTERNAL]) (31 ((FLET SB-UNIX::RUN-HANDLER) 13 #.(SB-SYS:INT-SAP #X7FFFF46B8EB0) #.(SB-SYS:INT-SAP #X7FFFF46B8D80))) (32 ((FLET SB-UNIX::RUN-HANDLER) 13 #.(SB-SYS:INT-SAP #X7FFFF46B8EB0) #.(SB-SYS:INT-SAP #X7FFFF46B8D80))[:OPTIONAL]) (33 ((FLET SB-UNIX::RUN-HANDLER))[:EXTERNAL]) (34 ("foreign function: #x4222F0")) (35 ("foreign function: #x40D0E0")) (36 ("foreign function: #x413DAB")) (37 ("foreign function: #x413FBF")) (38 ("bogus stack frame")) (39 ((FLET #:WITHOUT-INTERRUPTS-BODY-[CONDITION-WAIT]183))) (40 (SB-THREAD:CONDITION-WAIT #S(SB-THREAD:WAITQUEUE :NAME NIL :DATA #<SB-THREAD:THREAD "reader-thread" RUNNING {1002FD0171}>) #S(SB-THREAD:MUTEX :NAME NIL :%OWNER NIL :STATE 0))) (41 (SB-THREAD:CONDITION-WAIT #S(SB-THREAD:WAITQUEUE :NAME NIL :DATA #<SB-THREAD:THREAD "reader-thread" RUNNING {1002FD0171}>) #S(SB-THREAD:MUTEX :NAME NIL :%OWNER NIL :STATE 0))[:EXTERNAL]) (42 ((FLET #:FORM-FUN-[WITH-MUTEX-THUNK]2460))) (43 ((FLET SB-THREAD::WITH-MUTEX-THUNK))) (44 ((FLET #:WITHOUT-INTERRUPTS-BODY-[CALL-WITH-MUTEX]267))) (45 (SB-THREAD::CALL-WITH-MUTEX #<CLOSURE (FLET SB-THREAD::WITH-MUTEX-THUNK) {7FFFF46B9839}> #S(SB-THREAD:MUTEX :NAME NIL :%OWNER NIL :STATE 0) #<SB-THREAD:THREAD "control-thread" RUNNING {1002FCFCD1}> T)) (46 ((LAMBDA (SWANK-BACKEND::TEST &OPTIONAL SWANK-BACKEND::TIMEOUT)) #<FUNCTION (LAMBDA #) {100335A569}> NIL)) (47 (SWANK::DISPATCH-LOOP #<SWANK::CONNECTION {1002FCFB61}>)) (48 (SWANK::CONTROL-THREAD #<SWANK::CONNECTION {1002FCFB61}>)) (49 ((FLET #:WITHOUT-INTERRUPTS-BODY-[BLOCK382]387))) (50 ((FLET SB-THREAD::WITH-MUTEX-THUNK))) (51 ((FLET #:WITHOUT-INTERRUPTS-BODY-[CALL-WITH-MUTEX]267))) (52 (SB-THREAD::CALL-WITH-MUTEX ..)) (53 ((LAMBDA ()))) (54 ("foreign function: #x4222F0")) (55 ("foreign function: #x419237")) ;; Connection to Emacs lost. [ ;; condition: invalid number of arguments: 4 ;; type: SB-INT:SIMPLE-PROGRAM-ERROR ;; encoding: NIL style: :SPAWN dedicated: NIL] and the slime connection is broken. This makes is essentially impossible to use slime for real programming/debugging. Cheers, --ap -- Please read about why Top Posting is evil at: http://en.wikipedia.org/wiki/Top-posting and http://www.dickalba.demon.co.uk/usenet/guide/faq_topp.html Please read about why HTML in email is evil at: http://www.birdhouse.org/etc/evilmail.html

2 1

[slime-devel] Daily ChangeLog diff
by Helmut Eller 22 Sep '09

22 Sep '09

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1861 slime/ChangeLog:1.1862 --- slime/ChangeLog:1.1861 Sun Sep 20 06:51:50 2009 +++ slime/ChangeLog Mon Sep 21 15:36:53 2009 @@ -1,3 +1,7 @@ +2009-09-21 Stas Boukarev <stassats(a)gmail.com> + + * doc/slime.texi (slime-sprof): document slime-sprof contrib. + 2009-09-20 Tobias C. Rittweiler <tcr(a)freebits.de> * swank-abcl.lisp (thread-description): Fix typo. Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.246 slime/contrib/ChangeLog:1.247 --- slime/contrib/ChangeLog:1.246 Thu Sep 17 10:56:22 2009 +++ slime/contrib/ChangeLog Mon Sep 21 15:08:28 2009 @@ -1,3 +1,9 @@ +2009-09-21 Stas Boukarev <stassats(a)gmail.com> + + * slime-sprof.el, swank-sprof.lisp: New contrib for + integration with SBCL's sb-sprof profiler, + adopted from Juho Snellman's code. + 2009-09-17 Stas Boukarev <stassats(a)gmail.com> * slime-repl.el (slime-repl-clear-buffer): Don't change cursor

1 0

[slime-devel] Daily ChangeLog diff
by Helmut Eller 21 Sep '09

21 Sep '09

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1858 slime/ChangeLog:1.1861 --- slime/ChangeLog:1.1858 Fri Sep 18 17:29:59 2009 +++ slime/ChangeLog Sun Sep 20 06:51:50 2009 @@ -1,3 +1,43 @@ +2009-09-20 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank-abcl.lisp (thread-description): Fix typo. + (set-thread-description): Ditto. + +2009-09-20 Tobias C. Rittweiler <tcr(a)freebits.de> + + Generalize M-? (or M-_ respectively.) + + It will now list: + + - call sites for functions, + + - macroexpand sites for macros, + + - binding, setting, referencing sites for variables, + + - specializing methods for classes. + + * slime.el (slime-xref): Deal with :not-implemented. + (slime-xrefs): New. Makes RPC request to XREFS. + (slime-edit-callers): Renamed to `slime-edit-uses'. + (slime-edit-uses): Use slime-xrefs. + + * swank-backend.lisp (who-*): Add default implementation which + returns :not-implemented. + + * swank.lisp (xref-doit): Extracted from XREF. + (xref): Pass over :not-implemented to Emacs side. + (xrefs): New slime fun. To return results of multiple XREF + requests at once. + +2009-09-20 Mark Evenson <evenson(a)panix.com> + + Use *INVOKE-DEBUGGER-HOOK* introduced in ABCL by analogy to SBCL + (Tobias Rittweiler). + + * swank-abcl.lisp (sys::break): Conditionally redefine SYS::BREAK + only if SYS::INVOKE-DEUBBGER-HOOK is not present. + 2009-09-18 Tobias C. Rittweiler <tcr(a)freebits.de> * slime.el (slime-parent-bindings): Define M-? as

1 0

[slime-devel] Daily ChangeLog diff
by Helmut Eller 19 Sep '09

19 Sep '09

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1856 slime/ChangeLog:1.1858 --- slime/ChangeLog:1.1856 Thu Sep 17 11:51:52 2009 +++ slime/ChangeLog Fri Sep 18 17:29:59 2009 @@ -1,3 +1,27 @@ +2009-09-18 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime.el (slime-parent-bindings): Define M-? as + `slime-edit-callers', too. My previous choice of M-_ was warped + due to my German layout. + +2009-09-18 Tobias C. Rittweiler <tcr(a)freebits.de> + + New binding: M-_ (`slime-edit-callers'). + + Similiar to `slime-who-calls' but only creates an Xref buffer if + needed for disambiguation, and also pushes to the definition + stack. + + Spiritually like M-. but works "in the other direction". + + * slime.el (sllime-edit-callers): New function. Cross between + `slime-who-calls' and `slime-edit-definition'. + (slime-parent-bindings): Define M-_ as `slime-edit-callers'. + (slime-pop-xref-buffer): New helper. Extracted from + `slime-show-xrefs'. + (slime-show-xrefs): Use it. + (slime-xref): Let callers specify a continuation. + 2009-09-17 Tobias C. Rittweiler <tcr(a)freebits.de> * swank-abcl.lisp (sys::break): Fix typo.

1 0

[slime-devel] Daily ChangeLog diff
by Helmut Eller 18 Sep '09

18 Sep '09

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1855 slime/ChangeLog:1.1856 --- slime/ChangeLog:1.1855 Thu Sep 17 02:23:03 2009 +++ slime/ChangeLog Thu Sep 17 11:51:52 2009 @@ -1,3 +1,14 @@ +2009-09-17 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank-abcl.lisp (sys::break): Fix typo. + (slot-definition-documentation, slot-definition-type) + (class-prototype, generic-function-declarations) + (specializers-direct-methods, slot-boundp-using-class) + (slot-value-using-class): Add IGNORE declarations. Compiling + swank-abcl.lisp is now free of warnings. + (swank-compile-file): Load the compiled file even though warnings + were signalled during compilation. + 2009-09-17 Mark Evenson <evenson(a)panix.com> * swank-abcl.lisp (source-location): Emacs buffers start at 1, Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.245 slime/contrib/ChangeLog:1.246 --- slime/contrib/ChangeLog:1.245 Tue Sep 15 18:30:13 2009 +++ slime/contrib/ChangeLog Thu Sep 17 10:56:22 2009 @@ -1,3 +1,8 @@ +2009-09-17 Stas Boukarev <stassats(a)gmail.com> + + * slime-repl.el (slime-repl-clear-buffer): Don't change cursor + position if is already at the prompt. + 2009-09-16 Tobias C. Rittweiler <tcr(a)freebits.de> * slime-references.el: Largely refactored: decoupled code from

1 0

[slime-devel] Daily ChangeLog diff
by Helmut Eller 17 Sep '09

17 Sep '09

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1853 slime/ChangeLog:1.1855 --- slime/ChangeLog:1.1853 Mon Sep 14 01:51:51 2009 +++ slime/ChangeLog Thu Sep 17 02:23:03 2009 @@ -1,3 +1,12 @@ +2009-09-17 Mark Evenson <evenson(a)panix.com> + + * swank-abcl.lisp (source-location): Emacs buffers start at 1, + whereas CL files start at 0 (Tobias Rittweiler). + +2009-09-16 Stas Boukarev <stassats(a)gmail.com> + + * swank.lisp (documentation-symbol): Return more readable information. + 2009-09-14 Mark Evenson <evenson(a)panix.com> * swank-abcl.lisp (source-location): Fix typo.

1 0

[slime-devel] Daily ChangeLog diff
by Helmut Eller 16 Sep '09

16 Sep '09

Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.243 slime/contrib/ChangeLog:1.245 --- slime/contrib/ChangeLog:1.243 Sat Sep 12 23:15:03 2009 +++ slime/contrib/ChangeLog Tue Sep 15 18:30:13 2009 @@ -1,3 +1,14 @@ +2009-09-16 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime-references.el: Largely refactored: decoupled code from + SLDB; add references to the compilation log. + +2009-09-15 Stas Boukarev <stassats(a)gmail.com> + + * slime-autodoc.el (slime-fontify-string): setup *slime-fontify* + buffer without calling (lisp-mode) to avoid turning slime-mode there, + which may cause interference. + 2009-09-13 Stas Boukarev <stassats(a)gmail.com> * slime-autodoc.el (slime-fontify-string): do not call

1 0

[slime-devel] fuzzy completion on buffer names?
by Jeff Cunningham 16 Sep '09

16 Sep '09

There was a time when fuzzy completion was working on while typing buffer names. For example, I could just type "repl"and not have to type "*slime-r" to disambiguate various *slime... buffers. I don't know if I did something to turn this off or I need to do something now to turn it back on, but I miss it. I have searched for the setting without success. Any advice? Thanks Jeff

3 3

[slime-devel] Fontifying *SLIME Compilation*
by Leo 15 Sep '09

15 Sep '09

Hi there, Every time I C-c C-k on a file regardless of how small it may be, I am see a message lingering in the echo area for a few seconds, making me feel the compiliation is slow. Fontifying *SLIME Compilation*... (regexps.......................................................) Any idea how to fix this? I am running SLIME 2009-09-12 + ccl 1.3 on Emacs.app from 2009-08-10 on OSX leopard 10.5.8. Leo

2 1

[slime-devel] [patch] color slime-repl mode line
by Stanislaw Halik 15 Sep '09

15 Sep '09

When the REPL is started without lisp-mode windows, the arglist isn't fontified until something is „M-x slime-mode”'d because of this: [slime-autodoc.el] (defun slime-fontify-string (string) "Fontify STRING as `font-lock-mode' does in Lisp mode." (with-current-buffer (get-buffer-create " *slime-fontify*") (erase-buffer) (unless (eq major-mode 'lisp-mode) (lisp-mode) (slime-autodoc-mode -1) ...) ...)) …and „slime-autodoc-mode” sets „slime-echo-arglist-function” when arg satisfies „minusp”. So i made an ugly patch to fix it. Here it goes: diff --git a/contrib/slime-autodoc.el b/contrib/slime-autodoc.el index 524539d..e69a65c 100644 --- a/contrib/slime-autodoc.el +++ b/contrib/slime-autodoc.el @@ -184,7 +184,11 @@ Return DOCUMENTATION." (erase-buffer) (unless (eq major-mode 'lisp-mode) (lisp-mode) - (slime-autodoc-mode -1) + (let ((slime-echo-arglist-function slime-echo-arglist-function) + (slime-autodoc-mode slime-autodoc-mode) + (eldoc-idle-delay eldoc-idle-delay) + (eldoc-documentation-function eldoc-documentation-function)) + (slime-autodoc-mode -1)) (set (make-local-variable 'slime-highlight-suppressed-forms) nil)) (insert string) (let ((font-lock-verbose nil)) -- polscy Lispnicy, łączmy się -- #lisp-pl na freenode

3 2