July 2009 - slime-devel - mailman3.common-lisp.net

[slime-devel] Slime and ECL
by Matthew Mondor 14 Jul '09

14 Jul '09

Hello, I was unsure to which of the lists this belonged best so I posted it to both slime-devel and ecls-list (if I understand, swank which is part of slime handles the implementation-specifics)... Is slime known to be broken with ECL? I initially tried with ECL 8.12.0 and first had an unknown utf8-unix external format error (so I adapted my ~/.emacs to temporarily not use utf-8 mode, which helped). Then there was another error related to missing ext:+ecl-version-number+ constant (which I could declare in ~/.eclrc, which also helped). However, there then was another error: "Wrong number of arguments passed to function #<compiled-function SI:IHS-TOP>". I then tried using ECL 9.6.0 (or .1? the ecl-9.6.0 archive had a directory named ecl-9.6.1), built with unicode support. The utf8 related error still occurs when utf8 mode is enabled, the +ecl-version-number+ constant is already defined this time, yet the ihs-top function being called with the wrong number of arguments still occurs. Backtrace: > LAMBDA > swank-backend:call-with-debugger-hook > swank:swank-debugger-hook > lambda > swank-backend:call-with-debugger-hook > lambda > swank::call-with-bindings > swank::call-with-connection > swank::simple-serve-requests > swank::serve-requests > swank::serve-connection > swank::serve > swank::setup-server > swank:start-server > si:bytecodes [Evaluation of: (progn (load "/home/mmondor/emacs-scripts/slime/swank-loader.lisp" :verbose t) (funcall (read-from-string "swank-loader:init")) (funcall (read-from-string "swank:start-server") "/tmp/slime.9787" :coding-system "iso-latin-1-unix"))] > si:bytecodes [Evaluation of: (si:top-level)] My experience with slime and ecl is limited, I used to use it with SBCL only previously. The reason why my initial tries were with ECL 8.12 was because I got it working properly with threads, when the bootstrap ecl_min appears to segfault and its signal handler stuck in an endless loop when I try to build 9.6.0/1 with threading support (it builds fine non-threaded). But this is another matter which I didn't spend time to debug yet. :) Thanks, -- Matt

3 4

[slime-devel] Daily ChangeLog diff
by Helmut Eller 13 Jul '09

13 Jul '09

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1810 slime/ChangeLog:1.1811 --- slime/ChangeLog:1.1810 Sat Jul 11 14:25:59 2009 +++ slime/ChangeLog Sun Jul 12 04:01:28 2009 @@ -1,3 +1,8 @@ +2009-07-12 Helmut Eller <heller(a)common-lisp.net> + + * slime.el (slime-editing-keys): Don't override M-*. The default + binding should do what Matthias wants. + 2009-07-11 Matthias Koeppe <mkoeppe(a)math.ucdavis.edu> * swank-backend.lisp (definterface): Fix typo in error message. Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.223 slime/contrib/ChangeLog:1.225 --- slime/contrib/ChangeLog:1.223 Sat Jul 11 15:20:18 2009 +++ slime/contrib/ChangeLog Sun Jul 12 04:01:19 2009 @@ -1,3 +1,15 @@ +2009-07-12 Helmut Eller <heller(a)common-lisp.net> + + * swank-kawa.scm (swank-writer): Use set! for now, as the svn + version has problems to resolve setLength. + (<faked-frame>:init): Don't forget the self slot. + +2009-07-12 Helmut Eller <heller(a)common-lisp.net> + + * swank-mit-scheme.scm (swank:create-repl): Implement it. + * swank-kawa.scm, swank-goo.scm, swank.rb (frame-source-location): + Renamed from frame-source-location-for-emacs. + 2009-07-11 Stas Boukarev <stassats(a)gmail.com> * slime-parse.el (slime-make-form-spec-from-string):

2 1

[slime-devel] Binding of M-* wrong
by Matthias Koeppe 12 Jul '09

12 Jul '09

The current SLIME binding of M-* is slime-edit-definition, but it should be slime-pop-find-definition-stack, in analogy to the standard Emacs binding of M-* to pop-tag-mark. This seems to be the change that caused it, probably just a typo. 2009-01-16 Helmut Eller <heller(a)common-lisp.net> * slime.el (slime-keys): Put C-c C-i and M-* back. Could someone please fix it? Thanks, Matthias -- Matthias Koeppe -- http://www.math.ucdavis.edu/~mkoeppe

2 1

[slime-devel] Daily ChangeLog diff
by Helmut Eller 12 Jul '09

12 Jul '09

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1808 slime/ChangeLog:1.1810 --- slime/ChangeLog:1.1808 Mon Jul 6 08:32:12 2009 +++ slime/ChangeLog Sat Jul 11 14:25:59 2009 @@ -1,3 +1,12 @@ +2009-07-11 Matthias Koeppe <mkoeppe(a)math.ucdavis.edu> + + * swank-backend.lisp (definterface): Fix typo in error message. + +2009-07-11 Stas Boukarev <stassats(a)gmail.com> + + * doc/Makefile (clean): Remove bash extensions. Patch by + Teemu Likonen. + 2009-07-06 Tobias C. Rittweiler <tcr(a)freebits.de> * swank-ecl.lisp (find-external-format): Copied from @@ -79,7 +88,7 @@ 2009-06-25 Geo Carncross <geocar(a)gmail.com> - * swank-ecl.lisp: Profiling support by Marko Kocić + * swank-ecl.lisp: Profiling support by Marko Koci? 2009-06-27 Helmut Eller <heller(a)common-lisp.net> Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.222 slime/contrib/ChangeLog:1.223 --- slime/contrib/ChangeLog:1.222 Sat Jul 4 05:54:16 2009 +++ slime/contrib/ChangeLog Sat Jul 11 15:20:18 2009 @@ -1,3 +1,9 @@ +2009-07-11 Stas Boukarev <stassats(a)gmail.com> + + * slime-parse.el (slime-make-form-spec-from-string): + set syntax table to lisp-mode-syntax-table in a temporary buffer. + Necessary for some parsing functions to work correctly. + 2009-07-04 Tobias C. Rittweiler <tcr(a)freebits.de> Fix font-lock magic for s-code.lisp in the series package.

1 0

[slime-devel] Who uses slime-complete-form (C-c C-s)?
by Tobias C. Rittweiler 11 Jul '09

11 Jul '09

I'd like to remove C-c C-s. This would remove a good chunk of code, and would make some other code more straightforward. Disclaimer: I do use C-c C-s myself to automatically insert the situation keywords for EVAL-WHEN forms but I could live without that. -T.

6 5

[slime-devel] Bug in C-c C-s
by Matthias Koeppe 11 Jul '09

11 Jul '09

I type (find 1 '(1 2 3) C-c C-s and get an error: (wrong-type-argument arrayp nil) from somewhere within slime-make-form-spec-from-string. Thanks, Matthias -- Matthias Koeppe -- http://www.math.ucdavis.edu/~mkoeppe

2 1

[slime-devel] connection.repl-results
by Tobias C. Rittweiler 11 Jul '09

11 Jul '09

What's the purpose of CONNECTION.REPL-RESULTS? Or asked differently: Shouldn't SEND-REPL-RESULTS-TO-EMACS use it? -T.

3 2

[slime-devel] BUG + PATCH: doc/Makefile uses Bash extensions without setting SHELL=/bin/bash
by Teemu Likonen 11 Jul '09

11 Jul '09

The "clean" target in doc/Makefile uses features of Bash shell, namely brace expansions: rm -f slime.{info,pdf,dvi,ps,html} This feature does not work with POSIX sh shell which the /bin/sh implements and which is the default shell used by "make". In some systems /bin/sh is a symlink to /bin/bash but not in all. For example, in Ubuntu GNU/Linux systems /bin/sh link points to /bin/dash which is very close to POSIX sh. Debian GNU/Linux will switch to /bin/sh -> /bin/dash in their next release. So, to fix the bug you should either do this: --8<---------------cut here---------------start------------->8--- diff --git i/doc/Makefile w/doc/Makefile index 9ea9ebf..38ff658 100644 --- i/doc/Makefile +++ w/doc/Makefile @@ -14,6 +14,9 @@ INSTALL_CMD=install -m 644 # Info files generated here. infofiles=slime.info +# Use Bash shell +SHELL=/bin/bash + TEXI = slime.texi contributors.texi all: slime.info slime.pdf html/index.html --8<---------------cut here---------------end--------------->8--- Or you should stop using Bash extensions in the Makefile: --8<---------------cut here---------------start------------->8--- diff --git i/doc/Makefile w/doc/Makefile index 9ea9ebf..1e68cf5 100644 --- i/doc/Makefile +++ w/doc/Makefile @@ -95,6 +95,8 @@ uninstall-info: clean: rm -f contributors.texi - rm -f slime.{aux,cp,cps,fn,fns,ky,kys,log,pg,tmp,toc,tp,vr,vrs} - rm -f slime.{info,pdf,dvi,ps,html} - rm -rf html{,.tgz} + rm -f slime.aux slime.cp slime.cps slime.fn slime.fns slime.ky + rm -f slime.kys slime.log slime.pg slime.tmp slime.toc slime.tp + rm -f slime.vr slime.vrs + rm -f slime.info slime.pdf slime.dvi slime.ps slime.html + rm -rf html html.tgz --8<---------------cut here---------------end--------------->8--- I suggest doing the latter.

2 1

[slime-devel] SWANK STREAMS clobbered with lispworks
by Madhu 11 Jul '09

11 Jul '09

On Lispworks, the output associated with an editor window goes to the Output Browser pane. So evaluating a line such as (format t "foo~%"), or (format *standard-output* "foo~%") with C-M-x in an editor pane would print "foo" in the Output Browser. However, when using SLIME, once emacs has connected to Lispworks over SWANK, the output starts goes to the terminal on which Lispworks was initially started. This is because the value of *STANDARD-OUTPUT* in the editor pane gets clobbered and set to a Synonym stream to *TERMINAL-IO*. This behaviour is observed on LWL with any version of slime, and is not desirable. How can one get the `output stream' back so output from the Lispworks IDE can be seen inside the IDE ? -- Madhu

6 18

[slime-devel] Daily ChangeLog diff
by Helmut Eller 07 Jul '09

07 Jul '09

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1806 slime/ChangeLog:1.1808 --- slime/ChangeLog:1.1806 Thu Jul 2 13:00:21 2009 +++ slime/ChangeLog Mon Jul 6 08:32:12 2009 @@ -1,3 +1,15 @@ +2009-07-06 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank-ecl.lisp (find-external-format): Copied from + swank-sbcl.lisp. A lot of backends seem to share the same + implementation, perhaps some refactoring is needed. + +2009-07-06 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank.lisp (make-repl-input-stream): Make sure to redirect + output from slime requests to the REPL buffer for + communication-style=NIL. + 2009-07-02 Stas Boukarev <stassats(a)gmail.com> * slime.el (slime-beginning-of-symbol): Limit searching to 2000 @@ -148,7 +160,7 @@ * swank-ecl.lisp: Support new environment changes in recent ECL/CVS patch largely from ECL maintainer. -2009-06-1 Tobias C. Rittweiler <tcr(a)freebits.de> +2009-06-11 Tobias C. Rittweiler <tcr(a)freebits.de> * slime.el ([test] font-lock-magic): Moved to fontifying-fu contrib.

1 0