January 2009 - slime-devel - mailman3.common-lisp.net

[slime-devel] No REPL with SBCL and Emacs Snapshot 23
by Jonathan Lee 29 Jan '09

29 Jan '09

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Due to a problem I was having with SBCL suddenly taking 100% processor time, I recently upgraded from the existing Ubuntu SBCL package to version 1.0.23 and the latest version of SLIME from the CVS repository, running on Emacs Snapshot (GTK) 23. Unfortunately, now SLIME can connect to SBCL but does not create the REPL buffer. SLIME creates the *slime-events* * buffer and the *inferior-lisp* * buffer. In the inferior-lisp buffer, a message states that Swank has started and lists the port number. Here is the complete *inferior-lisp* buffer contents: (progn (load "/home/jonathan/.emacs.d/slime/swank-loader.lisp" :verbose t) (funcall (read-from-string "swank-loader:init")) (funcall (read-from-string "swank:start-server") "/tmp/slime.12111" :coding-system "iso-latin-1-unix")) This is SBCL 1.0.23, an implementation of ANSI Common Lisp. More information about SBCL is available at <http://www.sbcl.org/>. SBCL is free software, provided as is, with absolutely no warranty. It is mostly in the public domain; some portions are provided under BSD-style licenses. See the CREDITS and COPYING files in the distribution for more information. * ; loading #P"/home/jonathan/.emacs.d/slime/swank-loader.lisp" ; loading #P"/home/jonathan/.slime/fasl/2009-01-27/sbcl-1.0.23-linux-x86-64/swank-backend.fasl" ; loading #P"/home/jonathan/.slime/fasl/2009-01-27/sbcl-1.0.23-linux-x86-64/swank-source-path-parser.fasl" ; loading #P"/home/jonathan/.slime/fasl/2009-01-27/sbcl-1.0.23-linux-x86-64/swank-source-file-cache.fasl" ; loading #P"/home/jonathan/.slime/fasl/2009-01-27/sbcl-1.0.23-linux-x86-64/swank-sbcl.fasl" STYLE-WARNING: redefining EMACS-INSPECT (#<BUILT-IN-CLASS T>) in DEFMETHOD ; loading #P"/home/jonathan/.slime/fasl/2009-01-27/sbcl-1.0.23-linux-x86-64/swank-gray.fasl" STYLE-WARNING: Implicitly creating new generic function STREAM-READ-CHAR-WILL-HANG-P. ; loading #P"/home/jonathan/.slime/fasl/2009-01-27/sbcl-1.0.23-linux-x86-64/swank.fasl" WARNING: These Swank interfaces are unimplemented: (CALLS-WHO DISASSEMBLE-FRAME SLDB-BREAK-AT-START SLDB-BREAK-ON-RETURN WHO-SPECIALIZES) ;; Swank started at port: 44287. 44287 * This is what I get in the *Messages* buffer: Polling "/tmp/slime.12111".. (Abort with `M-x slime-abort-connection'.) [7 times] Connecting to Swank on port 44287.. [2 times] Connected. Hacks and glory await! This appears to suggest that SLIME thinks it has succeeded in creating the REPL. However, it does not appear in the buffer menu or the complete buffer list. Any assistance with this would be appreciated. Jonathan - -=-=- ... "Abstract math seems to be the lubricant for hammering square pegs into round holes." - Unknown -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFJgS689lVwGqYpXfERAjLoAJ0ZjaD3djT8svcNsoIQ2mmBM/3I4QCghoUX 2rni3L8R4ywQqqi5M9NikZE= =bg8C -----END PGP SIGNATURE-----

3 4

[slime-devel] wielding slime-load-system in .emacs
by david thompson 28 Jan '09

28 Jan '09

I don't spend much time meditating on the internals of slime and thought this felt about right... (add-hook 'slime-connected-hook (lambda () (slime-load-system 'foo))) However, slime doesn't like it and complains: The variable SWANK-IO-PACKAGE::FOO is unbound. What's the 'right' way to do it? Thanks for your time and assistance, Alan

4 4

[slime-devel] Daily ChangeLog diff
by Helmut Eller 28 Jan '09

28 Jan '09

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1671 slime/ChangeLog:1.1672 --- slime/ChangeLog:1.1671 Fri Jan 23 10:05:03 2009 +++ slime/ChangeLog Tue Jan 27 14:56:14 2009 @@ -1,3 +1,13 @@ +2009-01-27 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank-backend.lisp (with-symbol): New function, to be used with #+. + + * swank-sbcl.lisp: Use WITH-SYMBOL and get rid of SBCL-WITH-SYMBOL. + + * swank-openmcl.lisp (macroexpand-all): Implement it. + + Patch by Stas Boukarev. + 2009-01-23 Tobias C. Rittweiler <tcr(a)freebits.de> * slime.el (slime-editing-keys): New variable; splitted from @@ -9,7 +19,7 @@ Fix C-u C-c C-c in SLDB. - * slime.el (sldb-recompile-frame-source): Bind + * slime.el (sldb-recompile-frame-source): sind `slime-compilation-policy' at the right place. 2009-01-15 Martin Simmons <martin(a)lispworks.com> Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.168 slime/contrib/ChangeLog:1.169 --- slime/contrib/ChangeLog:1.168 Fri Jan 23 10:07:14 2009 +++ slime/contrib/ChangeLog Tue Jan 27 15:13:52 2009 @@ -1,3 +1,10 @@ +2009-01-27 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime-repl.el ([shortcut] quit): Quit the connection before + killing the REPL buffer; otherwise the default connection is + selected rather than the connection of the REPL buffer. + Reported by Stas Boukarev. + 2009-01-23 Tobias C. Rittweiler <tcr(a)freebits.de> * slime-repl.el (slime-repl-mode-map): Use the key bindings from

1 0

[slime-devel] value of slime-buffer-connection and multiple connections
by Stas Boukarev 27 Jan '09

27 Jan '09

When I do ",quit" or M-x slime-quit-lisp it calls `slime-connection' to determine what connection to kill, which in its turn returns value of `slime-buffer-connection'. But if the repl is not the defualt this value is wrong, namely, it returns connection of the default repl and the wrong connection gets killed. -- With Best Regards, Stas.

2 1

[slime-devel] macroexpand-all for ccl
by Stas Boukarev 27 Jan '09

27 Jan '09

Here it is: diff -u -r1.157 swank-openmcl.lisp --- swank-openmcl.lisp 16 Jan 2009 15:49:40 -0000 1.157 +++ swank-openmcl.lisp 26 Jan 2009 20:31:48 -0000 @@ -925,6 +925,9 @@ (< (symbol-value s) 255)) (setf (gethash (symbol-value s) *value2tag*) s))) +(defimplementation macroexpand-all (form) + (ccl:macroexpand-all form)) + ;;;; Inspection (defimplementation describe-primitive-type (thing) -- With Best Regards, Stas.

2 3

[slime-devel] no repl
by Walter C. Pelissero 26 Jan '09

26 Jan '09

As of today's CVS slime doesn't seem to create the interaction buffer. The inferior lisp looks alright but, at the "happy hacking" message, no repl appears (the buffer itself is missing). Here is the *inferior-lisp* buffer: (progn (load "/usr/home/wcp/emacs/slime/swank-loader.lisp" :verbose t) (funcall (read-from-string "swank-loader:init")) (funcall (read-from-string "swank:start-server") "/tmp/slime.21159" :coding-system "iso-latin-1-unix")) ; Loading #P"/usr/home/wcp/.cmucl-init.lisp". ;; Loading #P"/usr/home/wcp/lisp/asdf/asdf.x86f". ;; Loading #P"/usr/home/wcp/lisp/executable.x86f". ;; Loading #P"/usr/home/wcp/lisp/asdfa.x86f". ; Converted PARSE-UNIX-NAMESTRING. ; Converted UNPARSE-UNIX-FILE. ; [GC threshold exceeded with 12,014,984 bytes in use. Commencing GC.] ; [GC completed with 1,693,416 bytes retained and 10,321,568 bytes freed.] ; [GC will next occur when at least 13,693,416 bytes are in use.] ; Converted MAKE-PATHNAME. CMU Common Lisp 19e (19E), running on zaphod.home.loc With core: /usr/home/local/cmucl-19e/lib/cmucl/lib/lisp.core Dumped on: Sun, 2008-05-04 16:41:05+02:00 on daland See <http://www.cons.org/cmucl/> for support information. Loaded subsystems: Python 1.1, target Intel x86 CLOS based on Gerd's PCL 2004/04/14 03:32:47 * ; Loading #P"/usr/home/wcp/emacs/slime/swank-loader.lisp". ; Loading #P"/usr/home/wcp/.slime/fasl/2009-01-23/cmu-19e_(19e)-unix-x86/swank-backend.x86f". ; Loading #P"/usr/home/wcp/.slime/fasl/2009-01-23/cmu-19e_(19e)-unix-x86/swank-source-path-parser.x86f". ; Loading #P"/usr/home/wcp/.slime/fasl/2009-01-23/cmu-19e_(19e)-unix-x86/swank-source-file-cache.x86f". ; Loading #P"/usr/home/wcp/.slime/fasl/2009-01-23/cmu-19e_(19e)-unix-x86/swank-cmucl.x86f". ; Loading #P"/usr/home/wcp/.slime/fasl/2009-01-23/cmu-19e_(19e)-unix-x86/swank.x86f". ; Compiling LAMBDA (OBJECT): ; Compiling Top-Level Form: ; Compiling LAMBDA (OBJECT): ; Compiling Top-Level Form: ; Compiling LAMBDA (OBJECT): ; Compiling Top-Level Form: Warning: These Swank interfaces are unimplemented: (CALLS-WHO FIND-SOURCE-LOCATION RESTART-FRAME SLDB-STEP-INTO SLDB-STEP-NEXT SLDB-STEP-OUT) ; Loading #P"/usr/home/wcp/.swank.lisp". ;; Swank started at port: 50722. 50722 * A this point the inferior-lisp buffer is not replaced by the REPL. The relevant Emacs messages are: Polling "/tmp/slime.21159".. (Abort with `M-x slime-abort-connection'.) [6 times] Connecting to Swank on port 50722.. [2 times] Connected. Let the hacking commence! /tmp/slime.21159 is missing, though. -- walter pelissero http://www.pelissero.de

5 5

[slime-devel] output partly goes to inferior-lisp
by darabi＠web.de 24 Jan '09

24 Jan '09

Hi, I recently switched to slime head with sbcl head and the output behaviour has changed (I used the slime package in Ubuntu hardy, before): part of the output of the application which went to *slime-repl* buffer now appears in *inferior-lisp*. Can I do something to have all output in the REPL? Many thanks Kambiz Darabi

1 0

[slime-devel] swank:*globally-redirect-io* behaviour
by Timur Sufiev 24 Jan '09

24 Jan '09

It seems that we have some misunderstaning of swank:*globally-redirect-io* variable. Consider code shown below: (setf swank:*globally-redirect-io* t) (setf swank:*use-dedicated-output-stream* nil) (swank:create-server :dont-close t) (threads:make-thread (lambda () (some-processing-function))) When slime is connected to this program it does not receive any output (which goes to any of *standard-output*, *trace-output*, *debug-io*). The expected behaviour is output being redirected to slime connected. -- Timur Sufiev

3 5

[slime-devel] Daily ChangeLog diff
by Helmut Eller 24 Jan '09

24 Jan '09

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1670 slime/ChangeLog:1.1671 --- slime/ChangeLog:1.1670 Sat Jan 17 20:27:28 2009 +++ slime/ChangeLog Fri Jan 23 10:05:03 2009 @@ -1,3 +1,10 @@ +2009-01-23 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime.el (slime-editing-keys): New variable; splitted from + `slime-keys'. Contains key bindings that are useful for buffers + where users can edit s-exprs, such as source buffers and the REPL. + (slime-keys): Use it. + 2009-01-17 Tobias C. Rittweiler <tcr(a)freebits.de> Fix C-u C-c C-c in SLDB. Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.167 slime/contrib/ChangeLog:1.168 --- slime/contrib/ChangeLog:1.167 Sun Jan 18 14:18:53 2009 +++ slime/contrib/ChangeLog Fri Jan 23 10:07:14 2009 @@ -1,3 +1,8 @@ +2009-01-23 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime-repl.el (slime-repl-mode-map): Use the key bindings from + `slime-editing-keys'. + 2009-01-18 Tobias C. Rittweiler <tcr(a)freebits.de> Local M-. and local arglist display didn't work for

1 0

[slime-devel] By name character display error
by Slawek Zak 23 Jan '09

23 Jan '09

Allegro CL Express 8.1 on Linux x86 enters sldb in slime 09-10-01 when a unicode character name is entered. Example: CL-USER> #\greek_small_letter_lambda Error: Recursive error while printing or signalling an error. This is sometimes caused by an error in a print-object method. Restart actions (select using :continue): 0: Return to sldb level 99. 1: Return to sldb level 98. 2: Return to sldb level 97. .......... 95: Return to sldb level 4. 96: Return to sldb level 3. 97: Return to sldb level 2. 98: Return to sldb level 1. 99: Retry SLIME REPL evaluation request. 100: Return to SLIME's top level. 101: Abort entirely from this (lisp) process. [Current process: repl-thread] [1] CL-USER(1): :res Warning: stream-unread-char: ignoring #\null (expected #\:) Relevant settings: (setq slime-net-coding-system 'utf-8-unix) (setq slime-lisp-implementations '((allegro ("/usr/local/acl81.express/alisp") :coding-system utf-8-unix) (sbcl ("/usr/local/sbcl/bin/sbcl") :env ("SBCL_HOME=/usr/local/sbcl/lib/sbcl") :coding-system utf-8-unix))) Regards, /S

1 0