February 2007 - slime-devel - mailman3.common-lisp.net

Re:_[slime-devel]_Can't_install_slime_due_to_variable_depth_errors-_how_to_debug??=
by public＠raburgess.com 17 Feb '07

17 Feb '07

Never mind my previous messages about nesting errors etc. - I am pretty sure that I have not set up the init file correctly to go with the particular version of Lisp (Franz/Allergo) that I will be using. Because I am running it on Windows apparently more steps are needed. Randy Burgess wrote: > I am brand new to Slime, not entirely new to Emacs. > Trying to install Slime 2.0 in either Emacs 22 on XP or Emacs 21 on Cygwin, > and with both I get the same two errors - "Lisp nesting exceeds > max-lisp-eval-depth" and "Variable binding exceeds | max-specpdl-size," both > related to tramp-file-name-handler.

1 0

Re:_[slime-devel]_Can't_install_slime_due_to_variable_depth_errors-_how_to_debug??=
by public＠raburgess.com 17 Feb '07

17 Feb '07

Harald Hanche-Olson wrote: > Try to set debug-on-error to t (M-x set-variable RET > debug-on-error RET t RET). When the error occurs, you > should get dumped in the emacs lisp debugger. I have already tried this and I do not get dumped into the debugger! That's what I meant when I said I don't know how to get a trace. I've used the debugger in other circumstances w/ set-debug-on-error, so I don't know why it's not getting triggered now - but it isn't. Does this effectively leave me without a hope of solving this problem?

2 1

[slime-devel] Can't install slime due to variable depth errors - how to debug?
by Randy Burgess 17 Feb '07

17 Feb '07

I am brand new to Slime, not entirely new to Emacs. Trying to install Slime 2.0 in either Emacs 22 on XP or Emacs 21 on Cygwin, and with both I get the same two errors - "Lisp nesting exceeds max-lisp-eval-depth" and "Variable binding exceeds max-specpdl-size," both related to tramp-file-name-handler. Attempting to increase values for these results in no improvement. Once values exceed 10000, Emacs predictably blows up. I have no idea how to set a backtrace for this error; setting up tramp error logging does nothing, probably because tramp isn't being started. No references that I can find in the Slime documentation for this problem. Here are the relevant init settings for Emacs 22 on XP - nothing unusual so far as I can tell: (add-to-list 'load-path "/L:/Free/Office/Contacts/Wanderlust/Hacking/lisp/slime/") ; your SLIME directory (setq inferior-lisp-program "C:/Program Files/acl80-express/allegro-ansi.exe") (require 'slime) (slime-setup) Help would be appreciated ...

2 1

[slime-devel] Re: ASDF-Binary-Locations and SLIME's source information
by Gary King 16 Feb '07

16 Feb '07

Hi Edgar, I've not noticed this behavior but will look into it later today. I'm also forwarding your question to the SLIME list. BTW, what Lisp are you using? thanks, On Feb 15, 2007, at 2:47 PM, Edgar Gonçalves wrote: > Hi! > > I've just noticed your ASDF-Binary-Locations project, and I started > using it right away. It works just as advertised, and I was pretty > happy with it, until I loaded up slime, and found out I cannot do a > \M-. in any symbol loaded after ASDF-Binary-Locations. The message > window warns me that there's "No source information available for: > "<symbol-name>. I suppose this is probably because SLIME is looking > for a source filename identical to the .fas from where the function > was loaded, but that was a wild guess. Am I doing something wrong, or > this is something I have to live with for choosing a different > location for the binaries? > > I hope my question won't cause you much trouble. Thanks in advance! > -- > Edgar Gonçalves > Instituto Superior Técnico, > INESC-ID - Software Engineering Group, > Portugal -- Gary Warren King, metabang.com Cell: (413) 885 9127 Fax: (206) 338-4052 gwkkwg on Skype * garethsan on AIM

1 0

[slime-devel] slime-edit-definition crash with cl-sql
by Jonathon McKitrick 12 Feb '07

12 Feb '07

Hi everyone, I posted some time ago about a crash I get only on my Linux box. There are some other details I need to add. First, the environment: SBCL 1.0.2 SLIME cvs from a few days ago Emacs GTK snapshot for ubuntu/debian CL-SQL 3.6.7 If any function I am trying to jump to is found within the custom reader section of my CL-SQL code, slime turns into a runaway process, exhausts memory, breaks the connection, and shuts down. Any function in the same file *outside* the custom reader boundary works fine... slime-edit-definition jumps to it correctly. (clsql:locally-enable-sql-reader-syntax) . . . (clsql:locally-disable-sql-reader-syntax) What else can I do on my end to help track this down? -- Jonathon McKitrick jmckitrick(a)reedlarkeygroup.com

1 0

[slime-devel] slime-space and CUA mode
by David Reitter 10 Feb '07

10 Feb '07

Hi, it appears that you need a (put 'slime-space 'delete-selection t) after the definition of `slime-space' to make it collaborate with `delete-selection-mode' and `CUA-mode'. (See also the recent discussion regarding this on emacs-pretest-bugs.) Note that the above is not enough to make Undo behave correctly: after you have overwritten the region with a space, undoing the operation will only delete the space. It seems that Undo needs to understand that inserting the space is an atomic operation. I don't understand how undo works, so I can't offer a fix. - David -- http://aquamacs.org -- Aquamacs: Emacs on Mac OS X

1 0

[slime-devel] slime-defun-if-undefined
by Lennart Staflin 09 Feb '07

09 Feb '07

>From slime.el: (defmacro slime-defun-if-undefined (name &rest rest) (unless (fboundp name) `(defun ,name ,@rest))) Anyone see a problem with that definition? It assumes that slime hasn't been loaded before compiling slime.el. It bit me. I tend to just do a cvs up and byte-recompile-directory when upgrading slime with whatever emacs I have running. -- //Lennart Staflin

2 1

[slime-devel] ARGLIST: In a call to "Loop..Across":
by Ury Marshak 08 Feb '07

08 Feb '07

Hello, When using autodoc mode with the following form (defclass ajax-processor () ( ;; )) and moving the point around the comment I get the message: ARGLIST: In a call to "Loop..Across": NIL is not of type VECTOR. SLIME cvs, GNU Emacs 22 cvs Actually I seem to be getting this message even when typing just a pair of parentheses () and moving inside them. Happens in all Slime mode buffers: files, slime-scratch, REPL. Thanks, Ury

1 0

[slime-devel] cutting up slime.el
by Attila Lendvai 08 Feb '07

08 Feb '07

dear list, have anybody ever tought of cutting up slime.el into more files? especially cutting out the framework'ey and compatibility bits into a file that can be loaded at the head of slime.el. but also other sections could be cut out, like fuzzy completion, sldb, etc. i may not be aware of something, but i think it's not reasonable that slime.el is a single 450k file. my actual problem is that Helmut moved some annoying compatibility definitions out of the way (to the end of the file) but i'm using some of them in the darcs repo while loading. i've moved these back to the head of the file there, but such changes are moving the branches further away. if they were in slime-compatibility.el or a bigger slime-utils.el then it could be loaded easily at the top of slime.el. any opinions? -- - attila "- The truth is that I've been too considerate, and so became unintentionally cruel... - I understand. - No, you don't understand! We don't speak the same language!" (Ingmar Bergman - Smultronstället)

7 12

[slime-devel] communication between elisp and common lisp
by Jong W Kim 07 Feb '07

07 Feb '07

Hello, I am trying to have elisp and common lisp communicate with each other inside Emacs. I am using OpenMCL with SLIME on a PowerBook G4. I have two files. One is "jong.el" and the other is "jong.lisp". Here is the code. in the file of "jong.el" (defun man-hello (astring) (if (equal "hello-el" astring) (slime-repl-send-input "hello-cl")) (progn (goto-char (point-max)) (insert ":hello-cl") (slime-repl-return))) in the file of "jong.lisp" (defvar aa nil "a test variable.") (defun ehello-world () (format t "~%Connecting in ehello-world...") (format t "~%MAN:hello-el") (setq aa (read-line )) (if (string= aa "hello-cl") (format t "...conected.") (format t "connection not established with emacs."))) I am loading the "jong.lisp" file into OpenMCL. Then, by pressing M-x load-file, I am loading the "jong.el". The result is: Connecting in ehello-world... MAN:hello-el and it hangs until I press a key. If I press a key, I get the message below: connection not established with emacs. I used "slime-repl-send-input". Could you please give some advice on this? Thanks, Jong Kim

2 1

[slime-devel] Slime startup error with SBCL 1.0.0.10
by Vamsee Kanakala 06 Feb '07

06 Feb '07

Hi, I am using the latest SBCL built from sources (1.0.0.10) and doing M-x slime gives me the following error: ; compiling file "/home/vamsee/lisp/slime/swank-sbcl.lisp" (written 27 NOV 2006 02:22:50 PM): (/usr/local/lib/sbcl/sb-bsd-sockets/constants.lisp /usr/local/lib/sbcl/sb-bsd-sockets/constants.fasl /usr/local/lib/sbcl/sb-bsd-sockets/constants.fasl /usr/local/lib/sbcl/sb-bsd-sockets/foo.c /usr/local/lib/sbcl/sb-bsd-sockets/a.out /usr/local/lib/sbcl/sb-bsd-sockets/constants.lisp-temp) ; compilation aborted after 0:00:00 ;; Error while loading: /home/vamsee/.slime/fasl/2006-11-26/sbcl-1.0.0.10-linux-x86/swank-sbcl.fasl ;; Condition: The value NIL ;; is not of type ;; (OR (VECTOR CHARACTER) (VECTOR NIL) BASE-STRING PATHNAME ;; FILE-STREAM). ;; Aborting. ; ; compilation unit aborted ; caught 3 fatal ERROR conditions * I am using the latest slime too. Any help appreciated. Thanks, Vamsee.

6 10

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 06 Feb '07

06 Feb '07

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1066 slime/ChangeLog:1.1067 --- slime/ChangeLog:1.1066 Sun Feb 4 17:29:39 2007 +++ slime/ChangeLog Mon Feb 5 11:25:46 2007 @@ -1,3 +1,8 @@ +2007-02-05 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> + + * slime.el (sldb-insert-locals): Repair presentation markup of + frame locals. + 2007-02-04 Antonio Menezes Leitao <aml(a)gia.ist.utl.pt> * swank-lispworks.lisp (dspec-file-position): Bind

1 0

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 05 Feb '07

05 Feb '07

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1064 slime/ChangeLog:1.1066 --- slime/ChangeLog:1.1064 Sat Feb 3 08:34:06 2007 +++ slime/ChangeLog Sun Feb 4 17:29:39 2007 @@ -1,3 +1,16 @@ +2007-02-04 Antonio Menezes Leitao <aml(a)gia.ist.utl.pt> + + * swank-lispworks.lisp (dspec-file-position): Bind + *compile-file-pathname*, *compile-file-truename*, *load-pathname* + and *load-truename* in dspec-file-position. + +2007-02-04 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> + + * slime.el (slime-write-string): When writing a :repl-result, + update the slime-output-end marker for the purpose of asynchronous + output (when *use-dedicated-output-stream* is true). + Reported by Madhu <enometh(a)meer.net>. + 2007-02-03 Marco Baringer <mb(a)bese.it> * slime.el (slime-delete-swank-port-file): Fix typo in

1 0

[slime-devel] Small fix for swank-lispworks
by Antonio Menezes Leitao 04 Feb '07

04 Feb '07

Hi, The file swank-lispworks contains a function named dspec-file-position that, in recent LispWorks versions, reads forms from a file. The problem is that the file might contain #. forms that might refer to *compile-file-pathname* or *compile-file-truename*, or *load-pathname* or *load-truename*. In order to "emulate" the compilation or loading process, this function must also provide values to those variables so that the forms can be properly read. My quick fix is: (defun dspec-file-position (file dspec) (let ((*compile-file-pathname* (pathname file)) (*compile-file-truename* (truename file))) (let ((*load-pathname* *compile-file-pathname*) (*load-truename* *compile-file-truename*)) (with-open-file (stream file) (let ((pos #-(or lispworks4.1 lispworks4.2) (dspec-stream-position stream dspec))) (if pos (list :position (1+ pos) t) (dspec-buffer-position dspec 1))))))) Best regards, António Leitão.

2 1

[slime-devel] slime-apropos-package doesn't work with readtable-case = :invert
by Harald Hanche-Olsen 04 Feb '07

04 Feb '07

The subject says it all, I think. I run my lisps with (setf (readtable-case *readtable*) :invert) and if I try C-c C-d C-p cl-user <RET> I get the error: No such package: "CL-USER" The reason being as follows: slime-apropos-package calls slime-read-package-name which does a completing-read on the result of (swank:list-all-package-names t) which simply returns all package names as strings: In particular, they are all typically upper case. So the package name is passed to swank:apropos-list-for-emacs which calls (find-package (string-to-package-designator package)), and swank::string-to-package-designator passes the string through read-from-string, ending with a symbol whose name is all lower case. But :|cl-user| does not designated a package. Surely, there is some inconsistency, but I can't tell which part can be corrected without introducing another inconsistency. - Harald

3 6

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 04 Feb '07

04 Feb '07

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1063 slime/ChangeLog:1.1064 --- slime/ChangeLog:1.1063 Fri Feb 2 06:55:16 2007 +++ slime/ChangeLog Sat Feb 3 08:34:06 2007 @@ -1,8 +1,17 @@ +2007-02-03 Marco Baringer <mb(a)bese.it> + + * slime.el (slime-delete-swank-port-file): Fix typo in + warning message. + 2007-02-02 Marco Baringer <mb(a)bese.it> - * slime.el (slime-inferior-connect): Warn, as oposed to bailing - out with an error, when deleting the port file fails. Patch by: - Samium Gromoff <_deepfire(a)feelingofgreen.ru> + Warn, as oposed to bailing out with an error, when deleting the + port file fails. Patch by: Samium Gromoff + <_deepfire(a)feelingofgreen.ru> + + * slime.el (slime-delete-swank-port-file): New function. + (slime-inferior-connect): Use slime-delete-swank-port-file. + (slime-read-port-and-connect): Use slime-delete-swank-port-file. 2007-01-31 Marco Baringer <mb(a)bese.it>

1 0

[slime-devel] fork ?
by jan 03 Feb '07

03 Feb '07

I think Attila's enthusiasm for improving slime is great, however, I also think that some of the unspoken rules of open source have been broken here. These conventions are documented in ESR's 'Homesteading the Nooshere', but since that essay is rather long winded and philosophical I'll summarize here. 1. Attila, you're playing on Helmut's turf, you need to respect the ground rules set and also accept that Helmut has final say. 2. If you really, really can't abide by point 1 then you should create a proper fork. This means choosing a new name and creating a new project with its own mailing list. On your project website you should also try and justify the fork and the inevitable drain on developer resources. Note that forking due to trivialities is highly unlikely to attract other developers. Once you've forked, it's inappropriate to send suggestions to the slime mailing list. If you wish for slime to continue to improve, then release your fork under an open source license, but allow slime developers to port the changes they want. It's especially bad form to re-suggest changes you couldn't push through before the fork. http://article.gmane.org/gmane.lisp.slime.devel/5542 http://article.gmane.org/gmane.lisp.slime.devel/5788 http://article.gmane.org/gmane.lisp.slime.devel/5953 So I think Helmut's position is justified and not an overreaction (although an ultimatum is probably not the most effective course of action). My advice to Attila would be firstly to apologize, then either retract the fork or commit to it, the current situation cannot continue. just my 2c -- jan

4 3

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 03 Feb '07

03 Feb '07

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1062 slime/ChangeLog:1.1063 --- slime/ChangeLog:1.1062 Wed Jan 31 08:23:52 2007 +++ slime/ChangeLog Fri Feb 2 06:55:16 2007 @@ -1,3 +1,9 @@ +2007-02-02 Marco Baringer <mb(a)bese.it> + + * slime.el (slime-inferior-connect): Warn, as oposed to bailing + out with an error, when deleting the port file fails. Patch by: + Samium Gromoff <_deepfire(a)feelingofgreen.ru> + 2007-01-31 Marco Baringer <mb(a)bese.it> * slime.el (slime-repl-update-banner): Restore animation.

1 0

[slime-devel] [BUG] running under debian's emacs-snapshot
by Samium Gromoff 02 Feb '07

02 Feb '07

I'm not sure if it is a useful bugreport, but i'm trying to get my unicode workflow straight, which is why i tried running on emacs-snapshot. After starting slime i get this: <many many many> error in process filter: sldb-mode: Symbol's function definition is void: add-local-hook error in process filter: Symbol's function definition is void: add-local-hook </many many many> error in process filter: common-lisp-indent-function-1: Variable binding depth exceeds max-specpdl-size error in process filter: Variable binding depth exceeds max-specpdl-size slime is today's CVS, emacs is emacs-snapshot 20061218-1 regards, Samium Gromoff

1 0

[slime-devel] [PATCH] prevent failing delete-file from busting up the connection
by Samium Gromoff 02 Feb '07

02 Feb '07

This patch stop a failed port-information-file deletion from preventing slime to proceed with connection to swank. This helps to smooth the sbcl-suid workflow, when the setuid instance of sbcl creates a file which is then un-deleteable by slime running in an insufficiently privileged emacs process. regards, Samium Gromoff diff -urN slime-old/slime.el slime/slime.el --- slime-old/slime.el 2007-01-31 19:18:11.000000000 +0300 +++ slime/slime.el 2007-02-02 01:20:41.000000000 +0300 @@ -1596,7 +1596,9 @@ (defun slime-inferior-connect (process args) "Start a Swank server in the inferior Lisp and connect." (when (file-regular-p (slime-swank-port-file)) - (delete-file (slime-swank-port-file))) + (condition-case nil + (delete-file (slime-swank-port-file)) + (error nil))) (slime-start-swank-server process args) (slime-read-port-and-connect process nil)) @@ -1662,7 +1664,9 @@ (cond ((file-exists-p (slime-swank-port-file)) (let ((port (slime-read-swank-port)) (args (slime-inferior-lisp-args process))) - (delete-file (slime-swank-port-file)) + (condition-case nil + (delete-file (slime-swank-port-file)) + (error nil)) (let ((c (slime-connect slime-lisp-host port (plist-get args :coding-system)))) (slime-set-inferior-process c process))))

2 2

[slime-devel] Slime blows up with SBCL 1.0.2 on OS X/Intel
by Matt Pillsbury 02 Feb '07

02 Feb '07

Resent-From: mtbp(a)rci.rutgers.edu From: mtbp(a)rci.rutgers.edu Subject: Slime blows up with SBCL 1.0.2/Mac OS X Date: February 1, 2007 7:13:59 PM EST Resent-To: slime-devel(a)common-lisp.net To: slime-devel(a)common-lisp.net Hi, when I try to start the version of slime in CVS tarfile from the website using M-x slime (with SBCL 1.0.2 on OS X/x86 as my CL implementation), I get the following rather intimidating error message: === begin scary thing (progn (load "/Library/Application Support/Emacs/slime/slime/swank- loader.lisp" :verbose t) (funcall (read-from-string "swank:start- server") "/tmp/slime.7829" :coding-system "iso-latin-1-unix")) This is SBCL 1.0.2, an implementation of ANSI Common Lisp. More information about SBCL is available at <http://www.sbcl.org/>. SBCL is free software, provided as is, with absolutely no warranty. It is mostly in the public domain; some portions are provided under BSD-style licenses. See the CREDITS and COPYING files in the distribution for more information. * ; loading #P"/Library/Application Support/Emacs/slime/slime/swank- loader.lisp" ; compiling file "/Library/Application Support/Emacs/slime/slime/ swank-backend.lisp" (written 19 DEC 2006 05:47:36 AM): Help! 11 nested errors. SB-KERNEL:*MAXIMUM-ERROR-DEPTH* exceeded. 0: (BACKTRACE 536870911 #<SYNONYM-STREAM :SYMBOL SB-SYS:*TTY* {100C9831}>) 1: ((LAMBDA NIL)) 2: ((LAMBDA NIL)) 3: (SB-IMPL::%WITH-STANDARD-IO-SYNTAX #<CLOSURE (LAMBDA NIL) {11CD7D6D}>) 4: (SB-IMPL::ERROR-ERROR) 5: (SB-IMPL::INFINITE-ERROR-PROTECTOR) 6: (ERROR TYPE-ERROR) Help! 11 nested errors. SB-KERNEL:*MAXIMUM-ERROR-DEPTH* exceeded. 0: (BACKTRACE 536870911 #<SYNONYM-STREAM :SYMBOL SB-SYS:*TTY* {11CE0E91}>) 1: ((LAMBDA NIL)) 2: ((LAMBDA NIL)) 3: (SB-IMPL::%WITH-STANDARD-IO-SYNTAX #<CLOSURE (LAMBDA NIL) {11CE0FAD}>) 4: (SB-IMPL::ERROR-ERROR) 5: (SB-IMPL::INFINITE-ERROR-PROTECTOR) ; ; compilation unit aborted ; caught 1 fatal ERROR condition ; compilation aborted after 0:00:01 6: * === end scary thing I have no idea what's going on here. I do know that SLIME seems to be working fine with CLISP, though. Many thanks, Matt

2 1

[slime-devel] Problem with SLIME-INPUT-STREAM and WITH-TIMEOUT (SBCL)
by Ariel Badichi 01 Feb '07

01 Feb '07

Hello, There seems to be a problem with SLIME-INPUT-STREAM in SLIME 2007-01-20 which causes the SLIME connection to break if the following steps are performed (using SBCL 1.0.1): 1. Evaluate (with-timeout 5 (read-line)) at the toplevel. 2. Wait; a timeout condition will be signalled. 3. Choose the continue restart. 4. Type something for read-line to return. Ariel

5 5

[slime-devel] slime-reindent-defun
by Gábor Melis 01 Feb '07

01 Feb '07

Consider this: (defmacro foo (x) "docstring" `(,x)) If I press C-U M-q in the docstring it ends up looking like this: (defmacro foo (x) "docstring" `(,x)) Strangely enough there is no problem with: (defmacro foo (x) "docstring" (list x)) Cheers, Gábor

2 1

[slime-devel] completion for directories that start with a space (Mac)
by Jean-Christophe Helary 01 Feb '07

01 Feb '07

I am just starting Practical Common Lisp (again) with the Fink (unstable) carbon emacs/slime and I don't get slime to do proper completion when the first character of the directory to complete is a space. Even if I enter "\ "before the stripped directory name I can't get slime to complete it. My only solution is to use the mouse on the completion table. Is there anything I missed ? Jean-Christophe Helary

1 0

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 01 Feb '07

01 Feb '07

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1061 slime/ChangeLog:1.1062 --- slime/ChangeLog:1.1061 Tue Jan 30 17:23:39 2007 +++ slime/ChangeLog Wed Jan 31 08:23:52 2007 @@ -1,3 +1,8 @@ +2007-01-31 Marco Baringer <mb(a)bese.it> + + * slime.el (slime-repl-update-banner): Restore animation. + (slime-startup-animation): restore. + 2007-01-30 Helmut Eller <heller(a)common-lisp.net> * slime.el (slime-complete-symbol-function): Restore old default.

1 0