May 2006 - slime-devel - mailman3.common-lisp.net

[slime-devel] [Patch] setting up highlight-edits-mode in a proper way
by Tobias C. Rittweiler 18 May '06

18 May '06

2006-05-18 Tobias Rittweiler <PUT-MY-ADDRESS-HERE> * slime.el (slime-use-autodoc-mode): Fix typo in docstring. * slime.el (slime-use-highlight-edits-mode): New variable, analogous to SLIME-USE-AUTODOC-MODE. (slime-setup, slime-lisp-mode-hook): Make above variable work. Also, activates the HIGHLIGHT-EDITS-MODE in proper way (thus avoiding the nasty "Toggling ... off; better pass an explicit argument." message.) * slime.el: Fix typo in comment about communication protocol. --- /home/tcr/src/from-upstream/slime/slime.el 2006-04-15 15:46:14.000000000 +0200 +++ slime.el 2006-05-11 18:09:30.000000000 +0200 @@ -65,7 +65,10 @@ (require 'easymenu) (defvar slime-use-autodoc-mode nil - "When non-nil always enabled slime-autodoc-mode in slime-mode.") + "When non-nil always enable slime-autodoc-mode in slime-mode.") + +(defvar slime-use-highlight-edits-mode nil + "When non-nil always enable slime-highlight-edits-mode in slime-mode") (defun* slime-setup (&key autodoc typeout-frame highlight-edits) "Setup Emacs so that lisp-mode buffers always use SLIME." @@ -73,15 +76,16 @@ (when typeout-frame (add-hook 'slime-connected-hook 'slime-ensure-typeout-frame)) (setq slime-use-autodoc-mode autodoc) - (when highlight-edits - (add-hook 'slime-mode-hook 'slime-highlight-edits-mode))) + (setq slime-use-highlight-edits-mode highlight-edits)) (defun slime-lisp-mode-hook () (slime-mode 1) (set (make-local-variable 'lisp-indent-function) 'common-lisp-indent-function) (when slime-use-autodoc-mode - (slime-autodoc-mode 1))) + (slime-autodoc-mode 1)) + (when slime-use-highlight-edits-mode + (slime-highlight-edits-mode 1))) (eval-and-compile (defvar slime-path @@ -2281,8 +2294,8 @@ This is set only in buffers bound to specific packages.")) ;;; `slime-rex' is the RPC primitive which is used to implement both -;;; `slime-eval' and `slime-eval-async'. You can use it directly you -;;; need to but the others are usually more convenient. +;;; `slime-eval' and `slime-eval-async'. You can use it directly if +;;; you need to, but the others are usually more convenient. (defmacro* slime-rex ((&rest saved-vars) (sexp &optional

1 0

[slime-devel] Search definition
by Levente Mészáros 18 May '06

18 May '06

Hi, I wrote a slime extension to search definitions (variables,.functions, generic functions, etc.) based on fuzzy completions and edit definiton. It's supposed to work like apropos, but the result is interactive. Here is the idea (although not yet fully implemented) + hitting search definitons key brings up minibuffer + while editing in the minibuffer you can press TAB which shows fuzzy competion list for the current content of the minibuffer + completion list has a current selection + up/down arrows move the current selection while leaving the focus in the minibuffer, so that you can edit and hit TAB again to update the completion window + hitting enter goes to the definiton in the source files currently selected and closes minibuffer and completion window and leaves the source buffer open + hitting space opens the definition while leaving the focus in the minibuffer (minibuffer, completion window and definition window is visible at the same time) + hitting space on several definitons will show the sources of the definitons on top of each other + ESC leaves search definition - completion window shows specialized versions of generic functions in place, so that it can be selected directly (tree) - switch between two modes, where changing the selection immediately shows the definition of the current selection, so you can just move up and down (C-SPACE?) - close all sources that were not open before starting find definition except the one that was opened by pressing enter Key bindings are local to search definitions. Any ides, suggestions? levy -- There's no perfectoin

2 1

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 17 May '06

17 May '06

Index: slime/ChangeLog diff -u slime/ChangeLog:1.895 slime/ChangeLog:1.896 --- slime/ChangeLog:1.895 Sun May 14 09:25:01 2006 +++ slime/ChangeLog Tue May 16 14:04:49 2006 @@ -1,3 +1,9 @@ +2006-05-16 Marco Baringer <mb(a)bese.it> + + * slime.el (slime-repl-return-behaviour): New variable which + controls slime-repl-return's heaviour. + (slime-repl-return): Respect slime-repl-return-behaviour. + 2006-05-14 Marco Baringer <mb(a)bese.it> * slime.el (slime-macroexpansion-minor-mode-map): Rebind 'undo' to

1 0

[slime-devel] How to erase input lines in REPL?
by doug arro 17 May '06

17 May '06

My configuration is slime 2.0 + sbcl 0.9.12. When I enter something like this at the REPL prompt: CL-USER>line1 to be erased line 2 to be erased How do I erase the lines without using the backspace key and get back to the REPL prompt? Doug --------------------------------- Blab-away for as little as 1¢/min. Make PC-to-Phone Calls using Yahoo! Messenger with Voice.

7 12

[slime-devel] XREF failure on ACL 7.0
by Julian Stecklina 15 May '06

15 May '06

Hello, using stuff like C-c C-w resulted in an error with Allegro CL 7.0 because in fspec-definition-locations in swank-allegro.lisp the destructuring-bind was given a list without position element. A quick fix is: Index: swank-allegro.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank-allegro.lisp,v retrieving revision 1.87 diff -u -r1.87 swank-allegro.lisp --- swank-allegro.lisp 4 May 2006 14:38:07 -0000 1.87 +++ swank-allegro.lisp 15 May 2006 20:39:32 -0000 @@ -402,8 +402,10 @@ (defun fspec-definition-locations (fspec) (cond ((and (listp fspec) - (eql (car fspec) :top-level-form)) - (destructuring-bind (top-level-form file position) fspec + (eql (car fspec) :top-level-form) + (= (length fspec) 3)) + (destructuring-bind (top-level-form file position) fspec + (declare (ignore top-level-form)) (list (list (list nil fspec) (make-location (list :buffer file) Regards, -- Julian Stecklina "I object to doing things that computers can do." - Olin Shivers

1 0

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 15 May '06

15 May '06

Index: slime/ChangeLog diff -u slime/ChangeLog:1.893 slime/ChangeLog:1.895 --- slime/ChangeLog:1.893 Fri May 12 08:22:01 2006 +++ slime/ChangeLog Sun May 14 09:25:01 2006 @@ -1,3 +1,10 @@ +2006-05-14 Marco Baringer <mb(a)bese.it> + + * slime.el (slime-macroexpansion-minor-mode-map): Rebind 'undo' to + set buffer-read-only temporarily to t. + (slime-repl-return): Only send repl input if point is past a + complete form. + 2006-05-12 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> * swank.lisp (update-indentation-information): Fix for problem

1 0

[slime-devel] Trying to wrap my head around slime-eval
by Brad Beveridge 14 May '06

14 May '06

Hi all, I'm looking at the slime source code and I am pretty puzzled by slime-eval. I think that I understand slime-rex, it sends an sexpr to Swank and pushes a destructure-case continuation onto the global continuation assoc list. At sometime later Swank returns the evaluation of the sexpr and calls either the :ok clause or the :abort clause. The part that I am having trouble understanding is what parts of the code are closed over by the continuation that is saved, and where execution runs normally. Looking at slime-eval, there needs to be two paths of execution, call them A and B. Path A is synchronous, slime-rex returns without throwing the tag. There is a while loop waiting on process-output, and the result of that while is passed to (apply #'funcall <result-of-while>) What is the usual return here though? Path B is a continuation. The asynchronous return from Swank causes either :ok or :abort to be thrown up to the (catch tag...) form, and then (apply #'funcall....) will work on that. Basically this all bends my mind, but is very cool at the same time. If somebody could please elaborate on what I have said, and correct me I would really appreciate it. There is probably a lot that I am not getting just now :) For example, why THROW? Couldn't the destructure-case sections just call/return the appropriate bits. Also, are there any ELisp/CommonLisp differences in this function? Cheers Brad

3 3

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 13 May '06

13 May '06

Index: slime/ChangeLog diff -u slime/ChangeLog:1.892 slime/ChangeLog:1.893 --- slime/ChangeLog:1.892 Thu May 4 10:38:17 2006 +++ slime/ChangeLog Fri May 12 08:22:01 2006 @@ -1,3 +1,10 @@ +2006-05-12 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> + + * swank.lisp (update-indentation-information): Fix for problem + with Allegro CL 8.0: If I type M-x slime-update-indentation, + Allegro CL starts growing until it hits a STORAGE-CONDITION or + even segfaults. + 2006-05-04 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> * swank-allegro.lisp (fspec-definition-locations): Handle

1 0

Re: [slime-devel] Re: compile-file, slime and line number?
by Bill Clementson 13 May '06

13 May '06

> > To me, it looks like it's a problem with the bes-i0 function; however, > > you will probably get better advice on a CLM mailing list. > > Yes, that's where i eventually found the solution. I'm going to start > with asking there first next time with regards to clm. > > > Instead of using compile-file at the REPL in SLIME, try just doing C-c > > M-k when you are in the mus.lisp source file. That will compile the > > file and produce a list of compile errors in a separate > > *compiler-notes* buffer. In the *compiler-notes* buffer, you can place > > your cursor on the particular error message and press Enter. That will > > take you to the place in the mus.lisp source file where the > > compilation error occured. > > That's information I was looking for. I'm used to working in emacs > with compile so I was hoping that sort of functionality was in slime. > Thanks for the help. You're welcome. Incidentally, if you're interested in CLM, you might want to have a look at a blog posting I made a while back: http://bc.tech.coop/blog/050501.html It discusses a course at Stanford University that is using CLM & SLIME. On the links from that entry, you can get to info on the course material, computer music, CLM, Common Music, a SLIME tutorial and other goodies. If you want to learn more about using SLIME, it is worth your while to read the SLIME manual too. As with most technical documentation, it makes more sense once you've used the software a bit, but it is quite readable. ;-) Of course, the "definitive" SLIME tutorial is the movie that Marco Barringer made. I comment and link to that here: http://bc.tech.coop/blog/050728.html - Bill

1 0

[slime-devel] compile-file, slime and line number?
by Burton Samograd 12 May '06

12 May '06

Hi, I'm getting back into actual lisp development and I'm trying to install the common lisp music (clm) package using sbcl. I've gotten a compile error (in mus.lisp for those that are interested) that I'd like to figure out how to fix it, but I can't really find it since the error doesn't seem to provide line numbers for the error. Is there a way to get the line number that the error occured on printed? TIA -- burton samograd kruhft .at. gmail kruhft.blogspot.com www.myspace.com/kruhft metashell.blogspot.com

2 8