July 2004 - slime-devel - mailman3.common-lisp.net

[slime-devel] comma shortcut doesn't work always
by Edi Weitz 31 Jan '06

31 Jan '06

Another minor glitch that I encountered is that the comma shortcut in the REPL doesn't always work. I can consistently reproduce it (GNU Emacs 21.3.50 and also 21.3 release) like this: 1. Make sure point is at the beginning of a REPL input line. 2. Split Emacs frame into two windows, one of them being the REPL the other one containing something else. 3. Hide REPL window with C-x 0. 4. Now type M-x slime. 5. Type the comma. At this point I'm not asked for a shortcut command in the minibuffer but the comma is simply entered into the REPL instead. Anybody else seeing this? Cheers, Edi.

10 25

Re: [slime-devel] Re: CLISP SLIME implementation breaks in CVS
by Bill_Clementson＠peoplesoft.com 21 Jan '06

21 Jan '06

Luke Gorrie <luke(a)bluetail.com> writes: > Bill_Clementson(a)peoplesoft.com writes: > > > My previous SLIME installation was a CVS install from 4/18. I've just taken > > the latest CVS install and the CLISP SLIME implementation (under Windows) > > breaks. Here is the error: > > > > COERCE: #<MACRO #<COMPILED-CLOSURE SWANK::DEFSLIMEFUN>> cannot be coerced > > to type FUNCTION > > [Condition of type SIMPLE-TYPE-ERROR] > > I can reproduce this but I've no idea what is causing it. It /looks/ > like someone is trying to call defslimefun as a function rather than a > macro, but I don't see where it could happen -- the macro is declared > at the top of the only file that refers to it. > > I did fix a different but similar problem that I got under CLISP. It > seems like I had to swap the order of a couple of macros, one of which > expanded into into code that calls the other, but that surely can't be > the reason. > > > Backtrace: > > 0: EVAL frame for form (SWANK:CREATE-SWANK-SERVER 4007) > > Mine actually gives an empty backtrace (!). I fixed the Elisp-side to > handle that incase it helps someone more CLISP-savvy find the problem. Has anybody had any luck in finding the source of this problem? CLISP still breaks for me in the latest CVS. I have had a look at all of the DEFSLIMEFUN's that were modified since 4/18 (the previous CVS version that I had which was working ok with CLISP) and haven't found any obvious clues. Any suggestions on how I might debug this issue as it occurs at startup during the slime-connect processing in Emacs (after CLISP has started up successfully)? - Bill

5 6

[slime-devel] RFC: SLIME homepage suggestion
by Thomas Schilling 12 Aug '04

12 Aug '04

is it better now? http://www.inf.tu-dresden.de/~s3815210/slime-hp-suggestion/ -- ,, \../ / <<< The LISP Effect |_\\ _==__ __ | |bb| | _________________________________________________

5 12

[slime-devel] Finding source of functions compiled with slime-compile-defun.
by Peter Seibel 09 Aug '04

09 Aug '04

Again, having not looked into it in any great detail, is there any way to communicate source location back to Lisp so it can find the source of functions compiled with C-c C-c? -Peter -- Peter Seibel peter(a)javamonkey.com Lisp is the red pill. -- John Fraser, comp.lang.lisp

3 4

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 03 Aug '04

03 Aug '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.488 slime/ChangeLog:1.489 --- slime/ChangeLog:1.488 Mon Jul 26 04:50:19 2004 +++ slime/ChangeLog Wed Jul 28 14:15:30 2004 @@ -1,3 +1,18 @@ +2004-07-28 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> + + * slime.el (slime-eval-feature-conditional): Treat uppercase + operators NOT, AND, OR correctly. + (sldb-find-buffer): Remove killed buffers. + (sldb-quit): Raise an error if the RPC returns. + (slime-expected-failures): Delete unused function. + (complete-symbol): Test completion of + swank::compile-file. LispWorks has extra completions for + cl::compile-file. + (arglist): Test arglist of method cl:class-name. Add enough + regexpery to pass the test in most implementations. + + * swank-sbcl.lisp (list-callers, list-callees): Implemented. + 2004-07-26 Luke Gorrie <luke(a)bluetail.com> * slime.el (slime-first-change-hook): Add `save-match-data' to

3 5

[slime-devel] swank/xref and slime-load-file
by Jonathan Meeks 02 Aug '04

02 Aug '04

I think there is a bug in slime-load-file. I have noticed that when a file has its *.lisp extension in the parameter to slime-load-file, swank/xref is not able to find its callees (with slime-list-callees). When called interactively, the extension is stripped and slime-list-callees works. So, shouldn't the extension be stripped by slime-load-file when called non-interactively, as well? --Jonathan Meeks

3 3

[slime-devel] Did someone say they were working on a GF browser?
by Peter Seibel 31 Jul '04

31 Jul '04

I don't know if I saw it here, on c.l.l., or in my dreams but I seem to recall someone saying they were working on adding a generic function/method browser to SLIME. I.e. something by which you could see all the methods defined on a gf and then remove particular ones easily, etc. If so, how's it coming along? -Peter -- Peter Seibel peter(a)javamonkey.com Lisp is the red pill. -- John Fraser, comp.lang.lisp

6 15

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 31 Jul '04

31 Jul '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.490 slime/ChangeLog:1.492 --- slime/ChangeLog:1.490 Thu Jul 29 20:48:36 2004 +++ slime/ChangeLog Fri Jul 30 14:41:25 2004 @@ -1,3 +1,16 @@ +2004-07-30 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> + + * slime.el (slime-connection): Say "No default connection + selected" if there are open connections but no default connection. + (slime-tree-indent-item): Point wasn't updated correctly if the + last line was empty. Use insert-before-markers instead of insert + to do it properly. + (slime-draw-connection-list): Don't break if there is no default + connection. + + * swank-cmucl.lisp (call-with-debugging-environment): Only handle + DI::UNHANDLED-CONDITION not all DI:DEBUG-CONDITIONs. + 2004-07-30 Luke Gorrie <luke(a)bluetail.com> * slime.el (slime-choose-overlay-region): Tweaked the

1 0

[slime-devel] Compilation Problem with CVS-Slime
by R. Mattes 30 Jul '04

30 Jul '04

Hello everyone, i just checked out the latest CVS version of Slime, but when i try to load it in emacs (21.3.1) i get the following error: debugger invoked on a SIMPLE-ERROR in thread 17146: There is no applicable method for the generic function #<STANDARD-GENERIC-FUNCTION SWANK-BACKEND:SPAWN (0)> when called with arguments (# :NAME "Swank"). You can type HELP for debugger help, or (SB-EXT:QUIT) to exit from SBCL. restarts (invokable by number or by possibly-abbreviated name): 0: [ABORT ] Reduce debugger level (leaving debugger, returning to toplevel). 1: [TOPLEVEL] Restart at toplevel READ/EVAL/PRINT loop. ((NO-APPLICABLE-METHOD (T)) #<unavailable argument> #<unavailable argument> #<STANDARD-GENERIC-FUNCTION SWANK-BACKEND:SPAWN (0)> (#<FUNCTION "CLOSURE" {94EE2CD}> :NAME "Swank")) Any idea what's going wrong here? TIA Ralf Mattes

2 1

[slime-devel] no package name on a mode line after 2004-07-19?
by Chisheng Huang 30 Jul '04

30 Jul '04

Slime retrieved from CVS on or before 2004-07-19 would print a package name on the mode line of a buffer editing a Lisp file. After 2004-07-19, Slime stops printing a package name on a mode line. Is this change intentional? It's not documented in the change log. Printing a package name on a mode line is quite useful. Best, -cph

3 3

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 30 Jul '04

30 Jul '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.489 slime/ChangeLog:1.490 --- slime/ChangeLog:1.489 Wed Jul 28 14:15:30 2004 +++ slime/ChangeLog Thu Jul 29 20:48:36 2004 @@ -1,3 +1,17 @@ +2004-07-30 Luke Gorrie <luke(a)bluetail.com> + + * slime.el (slime-choose-overlay-region): Tweaked the + multiline-annotation-avoidance code to work with forms not + starting with an open-paren, e.g. `(..) or #'(..). + (slime-update-modeline-package): New configurable. Non-nil (the + default) means update the Lisp package in the modeline using an + idle timer. + (slime-repl-send-input): Make the `slime-repl-old-input' property + cover the whole input (including newline) so that pressing RET on + the end of an input line works. + Use a unique integer as the value of this property to distinguish + adjacent inputs. + 2004-07-28 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> * slime.el (slime-eval-feature-conditional): Treat uppercase

1 0

[slime-devel] Old REPL input
by Edi Weitz 30 Jul '04

30 Jul '04

I cannot find the old thread but I'm pretty sure this used to work at some point but doesn't work anymore: Start a new SLIME and in the REPL enter and evaluate a simple expression like, say, (+ 3 4) CL-USER> (+ 3 4) 7 CL-USER> Now move point back up until you're behind the closing parentheses and hit the return key. I expected this to move me back to current prompt with "(+ 3 4)" copied there ready to be edited (or evaluated by another return key) but I only get "no input at point." Did I miss a purposeful change or is this is a bug? Thanks, Edi.

4 6

[slime-devel] question on auto-indenting
by rif 30 Jul '04

30 Jul '04

I'm having trouble with auto-indenting in the slime-1.0alpha tarball release. For instance, I have a macro fixfor, which takes an &body argument, and I'd want this to be indented just 2 spaces. It works fine in the REPL, but doesn't seem to work in files at all, whether they use the package that contains the macro or not. Looking at (swank:print-indentation-lossage), I see both fixfor and :fixfor, even though I don't define :fixfor anywhere as far as I know. Furthermore, neither of these are there when I first start CL --- I guess it must add them when it tries to indent the function or something? In any case, fixfor is only defined in a single package, although it's exported from that package, and that package is used by other packages. fixfor is just one example among many, more generally, I don't feel like autoindenting is "working" for me. Possibly relevant facts: swank:*configure-emacs-indentation* is true. In .emacs, my swank:*communiation-style* is :fd-handler. Any help is appreciated. Cheers, rif

2 3

[slime-devel] slime-space ceases to work
by Edi Weitz 29 Jul '04

29 Jul '04

I'm currently working on a mildly complicated web server app using CMUCL 19a-pre3, multi-threading, TBNL, CLSQL, and of course SLIME from CVS. Every now and then slime-space ceases to work, i.e. after I've worked with the system for a while pressing space doesn't produce an arglist in the minibuffer anymore. I've not managed to isolate this problem yet but it occurs rather often and it's really annoying. Is anyone else seeing this? Thanks, Edi.

3 3

[slime-devel] How do I make slime-interactive-eval insert the result in the current buffer?
by Pascal J.Bourguignon 28 Jul '04

28 Jul '04

Hello, I like to have the result of my expressions inserted in the buffer where the expression is. Like with C-u C-x C-e in emacs-lisp-mode. How should I proceed to get this behavior with slime? -- __Pascal Bourguignon__ http://www.informatimago.com/ There is no worse tyranny than to force a man to pay for what he does not want merely because you think it would be good for him. -- Robert Heinlein

1 0

[slime-devel] My personal Slime Feature wishlist
by Thomas Schilling 28 Jul '04

28 Jul '04

I just put together a list of features I'd like to have in SLIME. I think it's better to post it here than letting it get dusty in my home-dir. Consider it as a base of discussion or a possible roadmap for, say, SLIME 2.0 than as a list of must-haves. It's all in the spirit of making SLIME+Lisp the superior tool to accomplish everything :) Of course, I'm willing to contribute, too. They are ordered in terms of (by me) estimated efford. * Nicer Homepage How about this?: http://www.inf.tu-dresden.de/~s3815210/slime-hp-suggestion/ * Class Browser Open a window where the class under the point is shown with all its direct superclasses and its direct subclasses or all of both. - `s' shows/hides the class' direct slots - `m' shows/hides all methods specializing on the class * Source Outline Mode Provide an outline mode for lisp source. That is the user can collapse/expand single lisp expressions. - The keybindings should be the same as in the text outline mode. - What do we show of collapsed expressions? Should a `defun' include the argument list? Should it be specific to the expression being collapsed? Should every expression be collapsible? - A block of comments may be considered a) as a single sexp b) as belonging to the next sexp (if it's on a separate line) or to the provious sexp (if it's on the same line) I favour b) since it would make a full-collapsed file more concise. * Refactoring Aids Provide commands for refactoring operations. Examples for useful Lisp-refactorings are: - extract/inline function/method - move slot to super-/subclass - inline temp/replace temp with query - rename function/class/slot/variable/... - ... - We have to work out which refactorings from [1] are applicable to Lisp (and make sence) and which additional refactorings would be useful for Lisp. - Some refactorings are really hard to implement. Namely all that require a code walker - Note that refactorings are considered save modifications on _working_ code in order to ease implementation of new features. It might still be worth considering how much refactoring is possible without assuming that the code is already compiled and DTRT. * Program Slicing Display (only) all the code that can effect a specified object. For example if invoked on a slot name display all expressions that may change this slot's value _and_ show all expressions that effect these former statements. - How much effort would it take to implement this? - Is it really useful for lisp? (Since lisp does indeed have side-effects, I think it would be of a certain use. So maybe the question should be: How much useful can it be for lisp?) - Can we cover all cases? - What additional information would be needed? Footnotes: [1] "Refactoring - Improving the Design of Existing Code" by Martin Fowler et. al., Addison Wesley, 2000 Regards, -ts -- ,, \../ / <<< The LISP Effect |_\\ _==__ __ | |bb| | _________________________________________________

4 9

[slime-devel] "query-replace" strangeness.
by Iain Little 26 Jul '04

26 Jul '04

It seems that cvs slime is interfering with "query-replace" (also checked "replace-string"). (Using cmucl18e) As a simple example (putting file contents in triple-double-quotes): I start with: """ bug """ Then after doing a query-replace to replace "bug" with "buggy" I get: """ buggyug """ This only happens when slime is running. (It works fine both before slime starts up, and after quitting.) I have also noticed strange capitilisation of file contents in a way not at all related to the replace query on more complicated examples, but I couldn't. """ strange BUG """ Then doing query-replace "bug" with "buggy": """ sTRANGE BUGGYUG """ As a final twist, everything seems to work fine when the buffer has been modified since last being saved. This leads me to suspect that the problem is caused by the cache stuff, somehow... Iain

2 1

[slime-devel] Function signatures incomplete
by Iain Little 24 Jul '04

24 Jul '04

I noticed that in cvs slime the entire function signature (when typing a function/macro into a lisp-mode buffer) doesn't always get shown anymore; it seems to cut off after the first line. For example, the signature for MAKE-ARRAY cuts off after "(element-type t)" in cvs slime (cmucl18e). I checked with slime-1.0alpha, and it works as expected there. Cheers, Iain

1 0

[slime-devel] Emacs error when quitting Slime
by Ivan Boldyrev 24 Jul '04

24 Jul '04

Greetings! When I quit Slime with comma and 'quit', Emacs shows error message: error in process sentinel: Connection closed. Backtrace: Debugger entered--Lisp error: (error "Connection closed.") signal(error ("Connection closed.")) error("Connection closed.") (cond ((null conn) (error "Not connected.")) ((not ...) (error "Connection closed.")) (t conn)) (let ((conn ...)) (cond (... ...) (... ...) (t conn))) slime-connection() (eq (slime-connection) slime-default-connection) (and (slime-connected-p) (eq (slime-connection) slime-default-connection)) (if (and (slime-connected-p) (eq ... slime-default-connection)) (progn (setq slime-state-name name) (force-mode-line-update))) (when (and (slime-connected-p) (eq ... slime-default-connection)) (setq slime-state-name name) (force-mode-line-update)) slime-set-state("[not connected]") slime-net-sentinel(#<process SLIME Lisp> "exited abnormally with code 256\n") It is not fatal, but annoying. I tried to hack the code myself, but I still do not understand full code, so I am afraid to break something. -- Ivan Boldyrev Violets are red, Roses are blue. // I'm schizophrenic, And so am I.

3 5

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 24 Jul '04

24 Jul '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.485 slime/ChangeLog:1.486 --- slime/ChangeLog:1.485 Thu Jul 22 09:55:26 2004 +++ slime/ChangeLog Fri Jul 23 09:36:36 2004 @@ -1,3 +1,13 @@ +2004-07-23 Luke Gorrie <luke(a)bluetail.com> + + * slime.el (slime-set-state): Show the message in the modeline in + the case where we aren't connected. Otherwise the "not connected" + status is ignored. + (slime-net-sentinel): Close the connection before changing the + status message. The old behaviour of this combined with the old + behaviour of `slime-set-state' could generally cause spurious + errors after a connection was closed. + 2004-07-22 Luke Gorrie <luke(a)bluetail.com> * swank.lisp (carefully-find-package): Return *BUFFER-PACKAGE* if

1 0

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 23 Jul '04

23 Jul '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.483 slime/ChangeLog:1.485 --- slime/ChangeLog:1.483 Wed Jul 21 12:56:54 2004 +++ slime/ChangeLog Thu Jul 22 09:55:26 2004 @@ -1,3 +1,13 @@ +2004-07-22 Luke Gorrie <luke(a)bluetail.com> + + * swank.lisp (carefully-find-package): Return *BUFFER-PACKAGE* if + no other package can be found. This is reverting a previous change + that broke completion in buffers with no known package. + + * slime.el (slime-maybe-start-lisp): Check that *inferior-lisp* + exists /and/ has a running process. Fixes a startup problem if + your inferior-lisp has died and you want to restart SLIME. + 2004-07-21 Luke Gorrie <luke(a)bluetail.com> * slime.el (slime-sync-package-and-default-directory): Sync

1 0

[slime-devel] nightly build
by Jeff Bowman 23 Jul '04

23 Jul '04

I am behind a firewall that prevents cvs from coming in. I have downloaded the slime-1.0alpha.tar.gz file and am using it, but I noticed one of the problems I'm having is mentioned in the list as being fixed. So my question is: is there a nightly build mechanism in place where I can download a more current version of slime than the tarball, or is the tarball contents updated frequently so I can get the fix? [RE: Symbol Completion Broken (CLISP)] Thanks! This is a great tool! Jeff

2 1

[slime-devel] Symbol completion broken (CLISP)
by Steven E. Harris 22 Jul '04

22 Jul '04

Using CLISP with SLIME updated from CVS as of 21 June 2004, symbol completion coaxed by M-Tab (`slime-complete-symbol') no longer produces anything useful. If I type (with-o in a buffer in SLIME mode connected to a CLISP repl and hit M-Tab, SLIME reports in the minibuffer [Can't find completion for "with-o"] and refuses to either complete or suggest completions for the partial symbol. This used to work as of about two weeks ago, but I can't recall when the exact last time I tested this was. -- Steven E. Harris

2 2

[slime-devel] Uh, (require 'slime) errors out
by Peter Seibel 22 Jul '04

22 Jul '04

I can no longer (require 'slime) because I get an error "Not connected". I'm trying to track it down from the backtrace (see below) but since it just started happening I'm assuming this is due to some recent change and perhaps the author of said change will know right away what's wrong. -Peter P.S. I'm a CVS idiot--how do I see what changes have been made recently to a given file? P.P.S Here's the backtrace: Debugger entered--Lisp error: (error "Not connected.") signal(error ("Not connected.")) error("Not connected.") (cond ((null conn) (error "Not connected.")) ((not ...) (error "Connection closed.")) (t conn)) (let ((conn ...)) (cond (... ...) (... ...) (t conn))) slime-connection() (or process (slime-connection) (error "No connection")) (process-buffer (or process (slime-connection) (error "No connection"))) (set-buffer (process-buffer (or process ... ...))) (save-current-buffer (set-buffer (process-buffer ...)) slime-rex-continuations:connlocal) (with-current-buffer (process-buffer (or process ... ...)) slime-rex-continuations:connlocal) (slime-with-connection-buffer (process) slime-rex-continuations:connlocal) slime-rex-continuations() slime-busy-p() (null (slime-busy-p)) (and (not (sldb-get-default-buffer)) (null (slime-busy-p))) slime-at-top-level-p() (progn (slime-at-top-level-p)) (if (progn (slime-at-top-level-p)) (slime-print-check-ok check-name-14528) (cond (... ... ...) (t ... ...)) (when slime-test-debug-on-error (debug ...))) (let ((check-name-14528 "At the top level (no debugging or pending RPCs)")) (if (progn ...) (slime-print-check-ok check-name-14528) (cond ... ...) (when slime-test-debug-on-error ...))) (slime-check "At the top level (no debugging or pending RPCs)" (slime-at-top-level-p)) slime-check-top-level() eval-buffer(#<buffer *load*> nil "slime" nil t) load-with-code-conversion("/home/peter/cvs/slime/slime.el" "slime" nil t) require(slime) eval((require (quote slime))) eval-last-sexp-1(nil) eval-last-sexp(nil) call-interactively(eval-last-sexp) -- Peter Seibel peter(a)javamonkey.com Lisp is the red pill. -- John Fraser, comp.lang.lisp

5 5

[slime-devel] Slime cannot connect to ACL6.2 win32
by Thomas Schilling 22 Jul '04

22 Jul '04

The exact error message is: (stream/warning) failure to open network stream to host "localhost" for service "4005" Took a look at the changelog--Luke, something to confess? The relevant init.el part: ;; start up slime (global-set-key [(f5)] '(lambda () (interactive) (shell-command "c:/programme/acl62-demo/allegro-ansi.exe +B +cm -L ~/slime.lisp&") (sleep-for 3) (slime-connect "localhost" 4005) (cd "~/Eigene Dateien/"))) and ~/slime.lisp is: (load "~/Eigene Dateien/devel/lisp/slime/swank-loader.lisp") (swank::create-swank-server 4005 :spawn #'swank::simple-announce-function t) -- ,, \../ / <<< The LISP Effect |_\\ _==__ __ | |bb| | _________________________________________________

2 10