slime-devel

slime-devel@common-lisp.net

4231 discussions

[slime-devel] Daily ChangeLog diff
by Helmut Eller 09 Oct '10

09 Oct '10

Index: slime/ChangeLog diff -u slime/ChangeLog:1.2148 slime/ChangeLog:1.2149 --- slime/ChangeLog:1.2148 Thu Oct 7 13:15:07 2010 +++ slime/ChangeLog Fri Oct 8 05:03:24 2010 @@ -1,3 +1,19 @@ +2010-10-08 Christophe Rhodes <csr21(a)cantab.net> + + Pass more detailed source location information to + swank:compile-string-for-emacs. Motivated by R's source location + needs (where srcrefs are line/column based). + + * slime.el (slime-compile-string): Generate :line location format + from start-offset, passing it through to + swank:compile-file-for-emacs. + (find-definition.2): Use new-style position. + + * swank.lisp (compile-string-for-emacs): Use only the offset from + the position argument. + + * swank-backend.lisp (swank-compile-string): Fix docstring typo. + 2010-10-07 Nikodemus Siivola <nikodemus(a)random-state.net> Hanging on to a *log-output* from a previous image is bad on SBCL

1 0

[slime-devel] contrib for repl images
by Christophe Rhodes 08 Oct '10

08 Oct '10

Hi, Attached are files which I propose to add to contrib/ to allow image display in the repl. The Common Lisp support is entirely non-existent, but I have prototype code using the slime support to do manipulations of graphical objects in R: <http://doc.gold.ac.uk/~mas01cr/tmp/swankr-images.png> displays an example. I'd like to put this or something like it in; I'm asking for review of this approach and implementation. Cheers, Christophe

4 4

[slime-devel] Daily ChangeLog diff
by Helmut Eller 08 Oct '10

08 Oct '10

Index: slime/ChangeLog diff -u slime/ChangeLog:1.2147 slime/ChangeLog:1.2148 --- slime/ChangeLog:1.2147 Sat Oct 2 07:16:54 2010 +++ slime/ChangeLog Thu Oct 7 13:15:07 2010 @@ -1,3 +1,13 @@ +2010-10-07 Nikodemus Siivola <nikodemus(a)random-state.net> + + Hanging on to a *log-output* from a previous image is bad on SBCL + at least: it is initialized to an FD-STREAM, which cannot be + reused from one image to another. + + Deal with this by using sb-ext:*save-hooks* to clear the stream. + + * swank-sbcl.lisp (deinit-log-output): New function. + 2010-10-02 Helmut Eller <heller(a)common-lisp.net> Rename slime-macro/compiler-macro-expand-1 => slime-expand-1

1 0

[slime-devel] Slime presentation marker bug
by Nathan Bird 04 Oct '10

04 Oct '10

I put a test case here that documents it: http://paste.lisp.org/+2GLO The issue is that when it starts a presentation it saves a marker, but then that marker gets moved when printing text and so when it ends the presentation the start and end markers are the same point, the one immediately after the text we are trying to present. I'm not sure where to go from here. I don't know how to get the old marker to stay in place any more than it is now. There is a comment: ;; We use markers because text can also be inserted before this presentation. ;; (Output arrives while we are writing presentations within REPL results.) Perhaps that is more relevant to printing with the target :repl-result vs target null? I tried changing the presentation start function (it's on the paste) to save the number if target is null and that looks like it is mostly working for me, can anyone else comment on the idea? -- Nathan Bird nathan(a)acceleration.net http://www.acceleration.net/ Custom Programming, Design, Hosting, and Broadband.

2 10

[slime-devel] SLIME development newbie questions
by Tamas K Papp 03 Oct '10

03 Oct '10

Hi, I decided that it would be worthwhile to learn some Emacs Lisp, mainly in order to understand the internals of SLIME better -- maybe I could get to the point that I could extend and customize it myself. This is why I am posting here instead of an emacs group. I am looking for help on how to get started, especially with the following: 1. Is there a SLIME-like IDE for Emacs Lisp? If not, what's the closest? I know that Emacs itself provides some functionality, but it does not look nearly as comfy as SLIME for CL. Eg I couldn't figure out the equivalent of M-. and similar. 2. Is there a document describing the architecture of SLIME? 3. How should I start? What's the best way to start poking around in SLIME? Tamas

6 6

[slime-devel] Daily ChangeLog diff
by Helmut Eller 03 Oct '10

03 Oct '10

Index: slime/ChangeLog diff -u slime/ChangeLog:1.2146 slime/ChangeLog:1.2147 --- slime/ChangeLog:1.2146 Fri Oct 1 05:30:43 2010 +++ slime/ChangeLog Sat Oct 2 07:16:54 2010 @@ -1,3 +1,18 @@ +2010-10-02 Helmut Eller <heller(a)common-lisp.net> + + Rename slime-macro/compiler-macro-expand-1 => slime-expand-1 + + * slime.el (slime-expand-1): Renamed from + slime-macro/compiler-macro-expand-1. + (slime-expand-1-inplace): Renamed from + slime-macro/compiler-macro-expand-1-inplace. + + * swank.lisp (swank-expand-1): Renamed from + swank-macro/compiler-macro-expand-1. + (swank-expand): Renamed from swank-macro/compiler-macro-expand. + (expand-1): Renamed from macro/compiler-macro-expand-1. + (expand): Renamed from macro/compiler-macro-expand. + 2010-10-01 Helmut Eller <heller(a)common-lisp.net> Don't use indent-sexp for macroexpanded code.

1 0

[slime-devel] Daily ChangeLog diff
by Helmut Eller 02 Oct '10

02 Oct '10

Index: slime/ChangeLog diff -u slime/ChangeLog:1.2144 slime/ChangeLog:1.2146 --- slime/ChangeLog:1.2144 Wed Sep 22 15:17:35 2010 +++ slime/ChangeLog Fri Oct 1 05:30:43 2010 @@ -1,3 +1,22 @@ +2010-10-01 Helmut Eller <heller(a)common-lisp.net> + + Don't use indent-sexp for macroexpanded code. + indent-sexp is potentially slow and usually redundant. + + * slime.el (slime-initialize-macroexpansion-buffer): Don't use + indent-sexp. + (slime-eval-macroexpand-inplace): Avoid indent-sexp but + use slime-insert-indented. + +2010-10-01 Helmut Eller <heller(a)common-lisp.net> + + Save result in kill ring for M-- C-x C-e. + For C-u C-x C-e set mark before inserting. + + * slime.el (slime-eval-save): New function. + (slime-eval-print): Set mark before inserting. + (slime-interactive-eval): Use it. + 2010-09-22 Stas Boukarev <stassats(a)gmail.com> * swank.lisp (eval-for-emacs): Send (:abort condition) where

1 0

[slime-devel] slime-interactive-eval into kill ring
by Tamas K Papp 01 Oct '10

01 Oct '10

Hi, C-x C-e (slime-interactive-eval) prints the value(s) in the minibuffer. Prefixed with C-u, it is inserted in the current buffer. Is there a way to insert it somewhere else? Ideally, it would be very handy to have a way to just save the result in the kill ring, then I could yank it later. Thanks, Tamas

2 5

[slime-devel] suggestion: make applying indent-sexp to results of macroexpansion optional
by michaelawells 01 Oct '10

01 Oct '10

In some cases, "indent-sexp" can be very slow, particularly when it is used on complex "macroexpanded" expressions. In my case, it takes 43 seconds to apply "indent-sexp" to a buffer of ~6000 lines. (I measured this using the elp package. The expression produced by SBCL and Allegro is already nicely formatted, so I would prefer not to display the result as formatted, rather than wait. So, I suggest adding an option to slime.el called "slime-indent-sexp-after-macroexpand", which would allow "indent-sexp" to be disabled. --- slime.el 22 Sep 2010 19:17:35 -0000 1.1340 +++ slime.el 24 Sep 2010 04:23:29 -0000 @@ -150,6 +150,13 @@ :prefix "slime-" :group 'slime) +(defcustom slime-indent-sexp-after-macroexpand t + "Perform 'indent-sexp' before presenting results of macroexpansion. +If disabled, macroexpansion will be presented as formated by the Lisp +in use." + :type 'boolean + :group 'slime-ui) + (defcustom slime-truncate-lines t "Set `truncate-lines' in popup buffers. This applies to buffers that present lines as rows of data, such as @@ -4942,7 +4949,7 @@ (erase-buffer) (insert expansion) (goto-char (point-min)) - (indent-sexp) + (when slime-indent-sexp-after-macroexpand (indent-sexp)) (font-lock-fontify-buffer))) (defun slime-create-macroexpansion-buffer () @@ -4976,8 +4983,8 @@ (goto-char start) (delete-region start end) (insert expansion) - (goto-char start) - (indent-sexp) + (goto-char start) + (when slime-indent-sexp-after-macroexpand (indent-sexp)) (goto-char point)))))))) (defun slime-macroexpand-1 (&optional repeatedly)

3 2

[slime-devel] Lisp connection closed unexpectedly
by J. Bromley 27 Sep '10

27 Sep '10

I am having a problem with SLIME getting disconnected from the Lisp process when requiring certain libraries. I am using clbuild to manage my Lisp implementation and SLIME itself. The problem is that when I do (require 'cl-opengl) in SLIME, I get the message "Lisp connection closed unexpectedly: connection broken by remote peer" in the mini-buffer area. I set swank:*log-events* to t and tried, and unfortunately that was not terribly helpful. Here are the lines generated by the require cl-opengl line: ===== dispatch-event: (:EMACS-REX (SWANK:OPERATOR-ARGLIST "require" "COMMON-LISP-USER") "COMMON-LISP-USER" :REPL-THREAD 15) send-event: #<SB-THREAD:THREAD "repl-thread" RUNNING {10030E2AE1}> (:EMACS-REX (SWANK:OPERATOR-ARGLIST "require" "COMMON-LISP-USER") "COMMON-LISP-USER" 15) wait-for-event: (COMMON-LISP:OR (:EMACS-REX . SWANK::_) (:EMACS-CHANNEL-SEND . SWANK::_)) NIL dispatch-event: (:RETURN #<SB-THREAD:THREAD "repl-thread" RUNNING {10030E2AE1}> (:OK "(require MODULE-NAME &OPTIONAL PATHNAMES)") 15) encode-message decode-message dispatch-event: (:EMACS-REX (SWANK:LISTENER-EVAL "(require 'cl-opengl) ") "COMMON-LISP-USER" :REPL-THREAD 16) send-event: #<SB-THREAD:THREAD "repl-thread" RUNNING {10030E2AE1}> (:EMACS-REX (SWANK:LISTENER-EVAL "(require 'cl-opengl) ") "COMMON-LISP-USER" 16) /Users/jay/Applications/clbuild/clbuild: line 450: 11820 Trace/BPT trap ${lisp} $common_options $eval "$require_asdf" $eval "$set_central_registry" $eval "$EXTRA_CLIM_FEATURES" "$@" Process inferior-lisp exited abnormally with code 133 ===== I am using the following software: * Mac OS X 10.6.4 * SBCL 1.0.42.52 (checked out by clbuild and built) * GNU Emacs 23.1.50.1 (x86_64-apple-darwin10.2.0, NS apple-appkit-1038.25) from MacPorts * SLIME from CVS via clbuild * cl-opengl from git via clbuild I have looked at line 450 of clbuild and it is just the call to launch SBCL. I find no more information about the Trace/BPT trap. I have also tried to find the meaning of the exit code, but have been unsuccessful. If I run clbuild lisp (which runs SBCL without SLIME) the require works fine. Obviously, SLIME provides a much nicer working environment, so it would be great to get it working. I would greatly appreciate any suggestions on possible solutions or about how to debug this further. Thanks and regards, -- Jay Bromley | jbromley at gmail dot com

2 1

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

slime-devel