slime-devel November 2005

slime-devel@common-lisp.net

35 participants
41 discussions

[slime-devel] source path finding breaks in the presence of #.
by Andreas Fuchs 07 Nov '05

07 Nov '05

Hi all, I encountered a construct like the following in cl-irc. It makes the source path finder in M-. break: ------------- C-c C-k, then M-. A RET ------------ (cl:defpackage :swank-bork (:use :cl)) (cl:in-package :swank-bork) (eval-when (:compile-toplevel :load-toplevel :execute) (defvar *foo* '(a b c))) (defmacro define-foos (args) `(progn ,@(mapcar (lambda (arg) `(defun ,arg (,arg) ,arg)) args))) (define-foos #.*foo*) ------------- snip ------------ trying to M-. A or B or C will fail with "Error: The variable *FOO* is unbound." With Juho Snellman's help, I tracked this down to swank::read-source-form. My current suspicion on why this is happening is that it READs the form and doesn't bind *package* to the package that the stream uses. This causes evaluation of the #. form to yield bad results. Unfortunately, I have no idea how to fix this, or if this is fix-worthy at all. I guess read-source-form would have to try and find a top-level in-package form somewhere, possibly expanding macros along the way and digging into top-levelness preserving forms like eval-when and progn. That would be a lot of work for this one error, in my eyes. Thanks for any advice (or fixes (-8), -- Andreas Fuchs, <asf(a)boinkor.net>, asf(a)jabber.at, antifuchs

3 3

[slime-devel] non-blocking read from REPL
by Daniel Patru 07 Nov '05

07 Nov '05

This is a clarification of my previous question.. My question is: How do I do a non-blocking read in the slime REPL? I tried (and (listen) (read-line)) and (read-char-no-hang) but keep getting the message: ; pipelined request... (swank:listener-eval "<input>") in the minibuffer. E.g. CL-USER> (progn (sleep 3) (and (listen) (read-line))) 2 2 CL-USER> NIL I typed the first 2, the minibuffer flashed the message ; pipelined request... (swank:listener-eval "2 ) then returned the 2, CL-USER>, and NIL. In constrast, typing (read-line) without the (listen) works: CL-USER> (read-line) 2 "2" NIL Similar things happen when I try (read-char-no-hang) in place of (and (listen) (read-line)). Both of these work (function as a non-blocking read) in a REPL run from the shell (no emacs, no slime.) It seems that any new input in the slime REPL during a running command is buffered as a new command instead of as input to the currently running command. I tried a blocking read-line after the non-blocking read-line and typed: 1<newline>2<newline>3<newline>4<newline> and got: CL-USER> (progn (sleep 10) (and (listen) (read-line)) (print (list (read-line)))) 1 2 3 4 3 CL-USER> 4 CL-USER> 2 CL-USER> 1 CL-USER> 5 ("5") ("5") CL-USER> I typed the last 5 after the sleep period. It seems that the 1, 2, 3, and 4 were processed at toplevel while the last 5 was used as input to unblock the previous read-line. Is it even possible to do a non-blocking read in the slime REPL, or is this one of the things I have to give up if I want to use slime? Thanks, Dan

2 1

[slime-devel] Daily ChangeLog diff
by lgorrie＠common-lisp.net 06 Nov '05

06 Nov '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.805 slime/ChangeLog:1.806 --- slime/ChangeLog:1.805 Fri Nov 4 10:07:43 2005 +++ slime/ChangeLog Sun Nov 6 10:09:47 2005 @@ -1,3 +1,11 @@ +2005-11-06 Juho Snellman <jsnell(a)iki.fi> + + * swank-sbcl.lisp (find-definitions, make-source-location-specification + make-definition-source-location, source-hint-snippet): As of + SBCL 0.9.6.25 SB-INTROSPECT has better support for finding + source locations. Use as much of it in swank-sbcl as possible. + (Original version left reader-conditionalized for older SBCLs). + 2005-11-04 Helmut Eller <heller(a)common-lisp.net> * swank.lisp (connection-info): Docfix.

1 0

[slime-devel] Daily ChangeLog diff
by lgorrie＠common-lisp.net 04 Nov '05

04 Nov '05

Index: slime/ChangeLog diff -u slime/ChangeLog:1.804 slime/ChangeLog:1.805 --- slime/ChangeLog:1.804 Mon Oct 31 09:26:54 2005 +++ slime/ChangeLog Fri Nov 4 10:07:43 2005 @@ -1,3 +1,11 @@ +2005-11-04 Helmut Eller <heller(a)common-lisp.net> + + * swank.lisp (connection-info): Docfix. + + * slime.el (slime-set-connection-info): Generate a new connection + name only if the implementation-name and the inferior-lisp-name + are different. + 2005-10-31 Helmut Eller <heller(a)common-lisp.net> * slime.el (slime-start, slime-lookup-lisp-implementation)

1 0

[slime-devel] CVS slime and Lispworks no longer working
by Luke J Crook 04 Nov '05

04 Nov '05

I receive an error in Lispworks after downloading the latest version of slime from CVS. After entering M-x slime-connect in Emacs, Lispworks returns the following: Undefined function SWANK::INSTALL-DEBUGGER-GLOBALLY called with arguments (#<function SWANK:SWANK-DEBUGGER-HOOK 21593A82>). -Luke

2 2

[slime-devel] SLIME history file location
by Peter BARABAS 04 Nov '05

04 Nov '05

Hello, What do you think about putting .slime-history.eld in ~/.slime? Since we already have that directory, we could just use it for storing the history file. -- '(Yours parenthetically), peter barabas.

1 0

[slime-devel] Endless Loop in Lisp when ,loading twice
by Stefan Kamphausen 04 Nov '05

04 Nov '05

Hi, currently I am experiencing a quite strange behaviour with slime and (I think) clsql: I had two lisp-files with def-view-class'es for two different databases, plus some interface functions. Everything worked fine. However, I had to merge the databases into different tables in one database and merrged the two lisp-files, too. Since then I two strange things happen: 1. First time I load the whole project (,l at the REPL) everything is fine (except for repeated compilations of some files in clsql; I posted that to clsql-help-list). But when I try to do that again, the files get compiled and then lisp goes into an endless loop, growing in memory as long as I say sayoonara to the REPL. 2. When trying to go to a function defined in that file (M-.) I end up at the previous defun in that file. I use CVSHEAD slime, CMU Common Lisp CVS 19b 19b-release-20050628-3 + minimal debian patches (19B), XEmacs 21.4 (patch 17) "Jumbo Shrimp" XEmacs Lucid. Can anyone give my any pointers what may be wrong here? Regards and thanks, Stefan -- 10 GB Mailbox, 100 FreeSMS/Monat http://www.gmx.net/de/go/topmail +++ GMX - die erste Adresse f�r Mail, Message, More +++

2 3

[slime-devel] slime stopped working
by Jeff Cunningham 04 Nov '05

04 Nov '05

I had this problem once before and I can't for the life of me remember what I did to fix it. I just downloaded the latest CVS of slime (todays) and blew away allthe fasls like always. Then I started emacs and when I ran M-x slime I get the following: (load "/home/jcunningham/installed/slime/swank-loader.lisp" :verbose t) (swank:start-server "/tmp/slime.10888" :external-format :iso-latin-1-unix) Error in function UNIX::SIGSEGV-HANDLER: Segmentation Violation at #x100168E8. [Condition of type SIMPLE-ERROR] Restarts: 0: [ABORT] Skip remaining initializations. Debug (type H for help) (UNIX::SIGSEGV-HANDLER #<unused-arg> #<unused-arg> #.(SYSTEM:INT-SAP #x3FFFEC98)) Source: Error finding source: Error in function LISP::%ENUMERATE-SEARCH-LIST: Undefined search list: default 0] I'm running CMUCL 19b and it hasn't changed in awhile. I'm stumped. Does anyone have any ideas? --Jeff

2 2

[slime-devel] Source display not working in SLDB
by Jonathan Lee 04 Nov '05

04 Nov '05

When I am in the SLDB, the manual states that I can display the associated source code by hitting the <v> key. However, when I do this I just get a message in the minibuffer: Cannot find source for frame: #<address #??????> NIL #<address #??????> The center value is sometimes T. I am running CLISP 2.35 as my base LISP. Is this the problem or is the function broken for all Lisp implementations?

2 1

[slime-devel] input not getting through
by Daniel Patru 04 Nov '05

04 Nov '05

The following code: (defun prompt () (do ((line nil (and (listen) (read)))) ((eq line 'q)) (when line (print line)) (print '>) (sleep 5))) when run in directly in the clisp interpreter in a shell, will output a > every 5 seconds, and print out any symbols entered until q is entered. when run under slime, only > is printed every 5 seconds, the input is not printed out, instead a message in the minibuffer says: ; pipelined request... (swank:listener-eval "<input>") Anyway to make this work? (I'm trying to do some processing in between user input.) I'm running clisp 2.34, slime 1.2.1, emacs 21.3.1 and windows xp. Thanks, Dan

3 2

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

slime-devel November 2005