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

[slime-devel] Daily ChangeLog diff
by Helmut Eller 03 Feb '11

03 Feb '11

Index: slime/ChangeLog diff -u slime/ChangeLog:1.2171 slime/ChangeLog:1.2174 --- slime/ChangeLog:1.2171 Fri Jan 28 14:19:30 2011 +++ slime/ChangeLog Wed Feb 2 06:12:33 2011 @@ -1,3 +1,18 @@ +2011-02-02 Stas Boukarev <stassats(a)gmail.com> + + * swank.lisp (eval-in-emacs): Return unreadable results from Emacs + as an unreadable-object, not as a string. + +2011-02-02 Stas Boukarev <stassats(a)gmail.com> + + * slime.el (slime-eval-for-lisp): Return value as a string, + because it can be unreadable, e.g. #<buffer foo> + * swank.lisp (eval-in-emacs): Prevent reader errors. + +2011-02-02 Stas Boukarev <stassats(a)gmail.com> + + * swank.lisp (eval-in-emacs): Export it. + 2011-01-28 Stas Boukarev <stassats(a)gmail.com> * slime.el (slime-check-location-filename-sanity): Guard

1 0

[slime-devel] SLIME altering lisp-mode indentation
by Jeff Massung 02 Feb '11

02 Feb '11

Hello all, For some reason, SLIME overrides the default lisp-mode indentation and I'm unsure of how to fix this (I'm still learning Emacs). The indentation I'm speaking of is in regards to IF statements. If I don't ever load SLIME, my IF statements indent like so: IF ----THEN --ELSE Which is very much how I like them to indent. After SLIME is loaded, they are: IF ----THEN ----ELSE Does anyone know how I can fix this? Thanks! Jeff M.

2 1

Re: [slime-devel] CVS slime
by Helmut Eller 02 Feb '11

02 Feb '11

* CVS User sboukarev [2011-02-02 10:13] writes: > Update of /project/slime/cvsroot/slime > In directory cl-net:/tmp/cvs-serv14498 > > Modified Files: > ChangeLog slime.el swank.lisp > Log Message: > * slime.el (slime-eval-for-lisp): Return value as a string, > because it can be unreadable, e.g. #<buffer foo> > * swank.lisp (eval-in-emacs): Prevent reader errors. Please don't do this. The principle should be: let it crash. Don't sweep problems under the rug with "defensive" programming. Helmut

1 0

[slime-devel] binding slime-complete-symbol elsewhere
by C. Hever 30 Jan '11

30 Jan '11

I want to bind C-c C-t to 'slime-complete-symbol in *inferior-lisp*. I don't know which hook I should use to add (lambda () (local-set-key (kbd "C-c C-t") 'slime-complete-symbol)) or whatever else it is I need to do to get it to work. Any ideas?

2 1

[slime-devel] Daily ChangeLog diff
by Helmut Eller 29 Jan '11

29 Jan '11

Index: slime/ChangeLog diff -u slime/ChangeLog:1.2170 slime/ChangeLog:1.2171 --- slime/ChangeLog:1.2170 Wed Jan 26 02:17:50 2011 +++ slime/ChangeLog Fri Jan 28 14:19:30 2011 @@ -1,3 +1,8 @@ +2011-01-28 Stas Boukarev <stassats(a)gmail.com> + + * slime.el (slime-check-location-filename-sanity): Guard + against target-filename being NIL. + 2011-01-26 Helmut Eller <heller(a)common-lisp.net> Allow tail-merging in call-with-bindings.

1 0

[slime-devel] slime-compile-and-load-buffer
by Darren Hoo 27 Jan '11

27 Jan '11

Greetings, I've got this trivial function (defun slime-compile-and-load-buffer () "Compile the current buffer" (interactive) (slime-compile-region (point-min) (point-max))) Personally I rebind C-c C-k to slime-compile-and-load-buffer and C-c C-l to slime-eval-buffer. These have some benefits over slime-compile-and-load-file and slime-load-file Thanks to the great mechanism of slime-compile-string. 1. when working with swank running remotely, it is no longer neccessary to setup pathname translations. 2. compiling lisp files localy or remotely won't clutter local working directory with a lot fasl files. Regards, Darren.

1 0

[slime-devel] Daily ChangeLog diff
by Helmut Eller 26 Jan '11

26 Jan '11

Index: slime/ChangeLog diff -u slime/ChangeLog:1.2169 slime/ChangeLog:1.2170 --- slime/ChangeLog:1.2169 Thu Jan 20 18:34:21 2011 +++ slime/ChangeLog Wed Jan 26 02:17:50 2011 @@ -1,3 +1,10 @@ +2011-01-26 Helmut Eller <heller(a)common-lisp.net> + + Allow tail-merging in call-with-bindings. + + * swank.lisp (call-with-bindings): Don't use progv if alist is + empty alist is empty. + 2011-01-20 Stas Boukarev <stassats(a)gmail.com> * swank-ecl.lisp (+TAGS+): change

1 0

[slime-devel] Daily ChangeLog diff
by Helmut Eller 23 Jan '11

23 Jan '11

Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.430 slime/contrib/ChangeLog:1.431 --- slime/contrib/ChangeLog:1.430 Thu Jan 20 17:00:11 2011 +++ slime/contrib/ChangeLog Sat Jan 22 19:08:08 2011 @@ -1,3 +1,9 @@ +2011-01-22 Stas Boukarev <stassats(a)gmail.com> + + * slime-repl.el (slime-repl-shortcut-help): Don't make ? an alias + for help, ? is bound to minibuffer-completion-help, and you can't + enter it. + 2011-01-20 Helmut Eller <heller(a)common-lisp.net> * swank-mit-scheme.scm (swank:load-file): Print the result

1 0

[slime-devel] propose change to `slime-handle-repl-shortcut'
by MON KEY 22 Jan '11

22 Jan '11

`slime-handle-repl-shortcut' uses completing-read to help user find repl shortcuts. However, it currently doesn't handle the "?" alias properly and it is not possible to acctually get a "?" completion because `minibuffer-local-completion-map' already sits on that binding with `minibuffer-completion-help' which isn't really useful in the slime context (or any other FTM). Proposed changes put "?" in the `slime-repl-shortcut-history'. This allows user to get to "?" via `previous-history-element'. Changes `slime-handle-repl-shortcut' by binding the local `shortcut` var to the return value of `slime-complete-repl-shortcuts' which is cleaner IMHO. Adds a new function `slime-complete-repl-shortcuts'. The purpose of this function is to steal back some control from completing-read. This is done by dynamically bind `minibuffer-local-completion-map' to a copy of itself then dynamically binds the copied keymap to value of `overriding-local-map'. Within the scope, of this binding a local keybinding is established for the copied keymap which attaches an anonymous function to insert a "?" such that the user can actually type enter with "?" as the completed value. The cummulative affect of these changes is taht user will now have the benefit of actually being able to access `slime-repl-shortcut-help' which in turn will present a nice popup buffer per `slime-list-repl-short-cuts'. A patch against CVS is attached and as follows: --- slime-repl.el.~1.52.~ 2010-09-26 14:10:33.000000000 -0400 +++ slime-repl.el 2011-01-22 18:56:02.876713429 -0500 @@ -1191,25 +1191,32 @@ (defvar slime-repl-shortcut-table nil "A list of slime-repl-shortcuts") -(defvar slime-repl-shortcut-history '() +(defvar slime-repl-shortcut-history '("?") "History list of shortcut command names.") (defvar slime-within-repl-shortcut-handler-p nil "Bound to T if we're in a REPL shortcut handler invoked from the REPL.") +(defun slime-complete-repl-shortcuts () + (let* ((new-local-map (copy-keymap minibuffer-local-completion-map)) + (minibuffer-local-completion-map new-local-map) + (overriding-local-map minibuffer-local-completion-map)) + (define-key new-local-map "?" (function (lambda () (interactive) (insert 63)))) + (slime-lookup-shortcut + (completing-read "Slime REPL shortcut: " + (slime-bogus-completion-alist + (slime-list-all-repl-shortcuts)) + nil t nil + 'slime-repl-shortcut-history)))) + (defun slime-handle-repl-shortcut () (interactive) (if (> (point) slime-repl-input-start-mark) (insert (string slime-repl-shortcut-dispatch-char)) - (let ((shortcut (slime-lookup-shortcut - (completing-read "Command: " - (slime-bogus-completion-alist - (slime-list-all-repl-shortcuts)) - nil t nil - 'slime-repl-shortcut-history)))) - (with-struct (slime-repl-shortcut. handler) shortcut - (let ((slime-within-repl-shortcut-handler-p t)) - (call-interactively handler)))))) + (let ((shortcut (slime-complete-repl-shortcuts))) + (with-struct (slime-repl-shortcut. handler) shortcut + (let ((slime-within-repl-shortcut-handler-p t)) + (call-interactively handler)))))) (defun slime-list-all-repl-shortcuts () (loop for shortcut in slime-repl-shortcut-table

1 0

[slime-devel] Daily ChangeLog diff
by Helmut Eller 21 Jan '11

21 Jan '11

Index: slime/ChangeLog diff -u slime/ChangeLog:1.2168 slime/ChangeLog:1.2169 --- slime/ChangeLog:1.2168 Fri Dec 10 10:05:05 2010 +++ slime/ChangeLog Thu Jan 20 18:34:21 2011 @@ -1,3 +1,11 @@ +2011-01-20 Stas Boukarev <stassats(a)gmail.com> + + * swank-ecl.lisp (+TAGS+): change + (translate-logical-pathname #P"SYS:TAGS") to + (merge-pathnames "TAGS" (translate-logical-pathname "SYS:")) + because of case conversion the former results in a pathname with a + name "tags", which doesn't exist. + 2010-12-10 Stas Boukarev <stassats(a)gmail.com> * slime.el (slime-with-popup-buffer): Correct the docstring. Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.429 slime/contrib/ChangeLog:1.430 --- slime/contrib/ChangeLog:1.429 Wed Jan 12 10:01:07 2011 +++ slime/contrib/ChangeLog Thu Jan 20 17:00:11 2011 @@ -1,3 +1,10 @@ +2011-01-20 Helmut Eller <heller(a)common-lisp.net> + + * swank-mit-scheme.scm (swank:load-file): Print the result + instead of returning it which breaks the protocol. + + * swank-mit-scheme.scm: Require release 9. + 2011-01-12 Helmut Eller <heller(a)common-lisp.net> Some more MIT Scheme fixes.

1 0