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

[slime-devel] window-configuration
by Madhu 28 Sep '08

28 Sep '08

I would request that all the slime window-configuration/restoration code be moved to a contrib module. By default slime could have a simple `bury-buffer on q' mechanism to get rid of all windows. The fancy window configuration/restoration should work only if the contrib is loaded. Emacs has many window and frame related variables that one can customize, and display-functions that one can advise to make it work the way one wants. The current code (I suspect) makes particular assumptions that are no doubt valid for the developers but these are invalid in general. I won't waste my breath on scenarios any more but slime always always gets it wrong: getting back to the buffer I want usually takes 2-3 window/buffer rearrange commands. Moving the window configuration stuff out of slime would also be a step towards making much touted `simplicity' goals of slime a reality in the code rather the "mailing list concept" -- Madhu

3 2

[slime-devel] bug: debugger slime and acl: slime creating unclosed new processes
by Madhu 28 Sep '08

28 Sep '08

With allegro, [acl express-8.1 on linux, with default settings, start up slime with C-u M-x slime], Everytime the debugger is entered when slime evaluates an expression, a new process called "worker" is created. This is not closed even on quitting the debugger. The processes are blocked on receive-if. Over a session I noticed 1000s of stale worker processes (I enter the debugger a lot:) Can you reproduce this? - inspect *mp::all-processes*, - evaluate (break) in *slime-scratch* (say) and return from debugger, a few times - inspect *mp::all-processes* again. I see items like #<MULTIPROCESSING:PROCESS worker(8) receive-if @ #x717377da> Which are never repead. -- Madhu

2 1

[slime-devel] patch swank-allegro.lisp
by Madhu 28 Sep '08

28 Sep '08

diff --git a/swank-allegro.lisp b/swank-allegro.lisp index 1ef352d..7f58a33 100644 --- a/swank-allegro.lisp +++ b/swank-allegro.lisp @@ -597,7 +597,7 @@ (ecase type ((:unsigned-word :unsigned-byte :unsigned-natural :unsigned-long :unsigned-half-long - :unsigned-3byte) + :unsigned-3byte :func) (label-value-line name (inspect::component-ref-v object access type))) ((:lisp :value) (label-value-line name (inspect::component-ref object access)))

2 1

[slime-devel] Installation guide in spanish
by Felip Alàez Nadal 27 Sep '08

27 Sep '08

Hello: I've written a little installation guide for slime on Windows XP in spanish. Maybe It could be useful for somebody. It's at my blog http://lispinspain.blogspot.com/ Bye. -- Felip Alàez Nadal

1 0

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 27 Sep '08

27 Sep '08

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1543 slime/ChangeLog:1.1545 --- slime/ChangeLog:1.1543 Wed Sep 24 05:14:13 2008 +++ slime/ChangeLog Fri Sep 26 19:14:10 2008 @@ -1,3 +1,16 @@ +2008-09-27 Tobias C. Rittweiler <tcr(a)freebits.de> + + Improve ECL's arglist support somewhat. + + * swank-ecl.lisp (grovel-docstring-for-arglist): New function. + (arglist): Use it. Now also try to find an arglist for special + operators, and macros. + +2008-09-26 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime.el (slime-cycle-connections): Do not make the new + connection buffer-local if we're currently in a REPL buffer. + 2008-09-24 Knut Olav Bøhmer <knut-olav.bohmer(a)telenor.com> * slime.el (slime-cycle-connections): New command.

1 0

[slime-devel] slime-region-for-defun-function
by Madhu 25 Sep '08

25 Sep '08

* 2008-03-13 Helmut Eller <heller(a)common-lisp.net> | * slime.el (slime-region-for-defun-function): Deleted. | (slime-region-for-defun-at-point): Use beginning-of-defun | and not beginning-of-sexp. This breaks multiple use cases when evaluating s-expressions from lisp file buffers. `beginning-of-sexp' is the right thing to use here. What motivated you to change this to `beginning-of-defun'? -- Madhu

3 6

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 25 Sep '08

25 Sep '08

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1540 slime/ChangeLog:1.1543 --- slime/ChangeLog:1.1540 Tue Sep 23 00:57:51 2008 +++ slime/ChangeLog Wed Sep 24 05:14:13 2008 @@ -1,3 +1,17 @@ +2008-09-24 Knut Olav Bøhmer <knut-olav.bohmer(a)telenor.com> + + * slime.el (slime-cycle-connections): New command. + +2008-09-24 Helmut Eller <heller(a)common-lisp.net> + + * slime.el (slime-prefix-map): New keymap. + (slime-define-key): Use it. Also drop unused :inferior arg. + (slime-inspector-mode-map): Bind C-c to slime-prefix-map. + (slime-define-both-key-bindings): New function. Factor of + slime-init-keymaps. + (slime-init-keymaps): Use it. + (slime-control-modified-char): Deleted. + 2008-09-23 Douglas Crosher <dcrosher(a)common-lisp.net> * swank-scl.lisp: update for Scieneer CL 1.3.8.

1 0

[slime-devel] add C-c C-z kind binding to inspector
by Gary King 24 Sep '08

24 Sep '08

The attached patch does two things. Regarding the name change of the inspector's buffer. Wouldn't it be more consistent if all the SLIME buffers used the same name template. E.g., *slime-repl*, *slime-inspector*, *slime-events*, etc. Currently, the case format used is all over the place. 2008-09-20 Gary King <gwking(a)metabang.com> * slime.el (slime-inspector-buffer, slime-inspector-mode-map): switch slime-inspector's buffer name to lowercase. Add binding to C-c C-z to switch to the output frame (usually the REPL) -- Gary Warren King, metabang.com Cell: (413) 559 8738 Fax: (206) 338-4052 gwkkwg on Skype * garethsan on AIM

2 1

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 24 Sep '08

24 Sep '08

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1539 slime/ChangeLog:1.1540 --- slime/ChangeLog:1.1539 Mon Sep 22 18:56:18 2008 +++ slime/ChangeLog Tue Sep 23 00:57:51 2008 @@ -1,3 +1,11 @@ +2008-09-23 Douglas Crosher <dcrosher(a)common-lisp.net> + + * swank-scl.lisp: update for Scieneer CL 1.3.8. + + * swank.lisp (ed-in-emacs): customize for the SCL. + + * swank.lisp (signal-interrupt): fix typo. + 2008-09-22 Nikodemus Siivola <nikodemus(a)random-state.net> * swank.lisp (guess-package): Return NIL if string designator is

1 0

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 23 Sep '08

23 Sep '08

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1536 slime/ChangeLog:1.1539 --- slime/ChangeLog:1.1536 Sun Sep 21 07:18:00 2008 +++ slime/ChangeLog Mon Sep 22 18:56:18 2008 @@ -1,3 +1,25 @@ +2008-09-22 Nikodemus Siivola <nikodemus(a)random-state.net> + + * swank.lisp (guess-package): Return NIL if string designator is + NIL: makes files without IN-PACKAGE forms more *BREAK-ON-SIGNALS* + friendly. + +2008-09-22 Helmut Eller <heller(a)common-lisp.net> + + * swank-sbcl.lisp (wait-for-input): Implement this in backend, + since read-char-no-hang doesn't work in fd-handlers. + (install-sigint-handler): Go through invoke-interruption and + with-interrupts to support nested interrupts. + + * slime.el (slime-lisp-implementations): Mention :env keyword + in docstring. + +2008-09-21 Helmut Eller <heller(a)common-lisp.net> + + * slime.el (slime-repl-input-end-mark): Deleted. It was always at + the end of buffer. Use point-max instead. + (slime-repl-eol): Removed. The usual end-of-line does the same. + 2008-09-21 Helmut Eller <heller(a)common-lisp.net> * slime.el (slime-eol-conversion-fixup): Return 0 (not nil) for

1 0