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

[slime-devel] slime.texi nit - s/compliments/complements.
by John Paul Wallington 09 Sep '04

09 Sep '04

Here's a typo in the manual: --- slime.texi 04 Sep 2004 23:08:09 +0100 1.28 +++ slime.texi 05 Sep 2004 00:40:42 +0100 @@ -169,7 +169,7 @@ @SLIME{} extends Emacs with new support for interactive programming in Common Lisp. The features are centred around @code{slime-mode}, an Emacs -minor-mode that compliments the standard @code{lisp-mode}. While +minor-mode that complements the standard @code{lisp-mode}. While @code{lisp-mode} supports editing Lisp source files, @code{slime-mode} adds support for interacting with a running Common Lisp process for compilation, debugging, documentation lookup, and so on.

2 1

[slime-devel] SWANK-IO-PACAKGE being used amidst (slime-eval FORM ":cl")
by Sean Champ 09 Sep '04

09 Sep '04

Hello, While trying to get slime to tell the repl to evaluate some common lisp code, I had tried to use the emacs form: (slime-eval '(format t "ping ~d" 12) ":cl-user") but, when I use this, sldb pops-up, with a message about: Error in KERNEL:%COERCE-TO-FUNCTION: the function SWANK-IO-PACKAGE::FORMAT is undefined. So, I am just wondering why the package named SWANK-IO-PACKAGE is being used, as the package in which the REPL will evaluate the slime-eval'd FORM - instead of slime-eval using the :cl-user package, which was mentioned, also, in the call to slime-eval. Thank you -- Sean

2 1

[slime-devel] xemacs asking me "Enter recursive edit?" when sldb comes up
by Sean Champ 09 Sep '04

09 Sep '04

Hello again, I'm using the debianized slime 1.0, point-of-fact-ly. When sldb pops up, now, I am seeing XEmacs asking me (always, whenever sldb pops up) "Enter recursive edit?" in the minibuffer. When I say "yes" to that minibuffer-query, things go well. When I say "no" to that minibuffer-query, then things do not seem to go well - at least, not until after I C-g , then C-] to clear the resultant emacs error-message. So, I was wondering if there would be a way to tell {whatever is asking me whether to "Enter recursive edit?"} to always "yes" - that being, to always do so. Thank you -- Sean

2 2

[slime-devel] OS X: OpenMCL 0.14.2-p1 and Slime-1.0 Error
by Ian J Cottee 09 Sep '04

09 Sep '04

(found the solution to this but reporting it in case there's a better way to handle the situation in future versions of SLIME) I've been using Slime with CLISP on Windows (just as a learning environment) and love it so I thought I'd get an environment working with my iMac as well. I tried to get CLISP working but it threw me an error so I thought I'd try again and install OpenMCL instead. However, it also threw me an error (reproduced below) which I eventually tracked down to my ~/.slime directory causing me problems. Apparantly running slime originally with CLISP caused a problem. When I removed ~/.slime and started again everything worked fine. For the record (in case anybody else is googling like me), the error I got was as follows: Welcome to OpenMCL Version (Beta: Darwin) 0.14.2-p1! ? > Error in process listener(1): Incompatible lambda list in #<STANDARD-METHOD\ SWANK-COMPILE-STRING (T)> for #<STANDARD-GENERIC-FUNCTION SWANK-COMPILE-STRING\ #x6381456> > While executing: CCL::CHECK-DEFMETHOD-CONGRUENCY > Type :GO to continue, :POP to abort. > If continued: Remove 0 methods from the generic-function and change its lambd\ a list. Type :? for other options. 1 > Hope this is of some use Ian Cottee

1 0

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 08 Sep '04

08 Sep '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.514 slime/ChangeLog:1.515 --- slime/ChangeLog:1.514 Sat Sep 4 00:12:53 2004 +++ slime/ChangeLog Wed Sep 8 18:16:42 2004 @@ -1,3 +1,7 @@ +2004-09-08 Martin Simmons <martin(a)xanalys.com> + + * swank-lispworks.lisp: Implement call-with-compilation-hooks. + 2004-09-03 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> * NEWS: Summarize changes since August.

1 0

[slime-devel] slime 1.0 : debian package now available
by Sean Champ 07 Sep '04

07 Sep '04

Hello, I wanted to mention that I've done the stuff that would debianize slime. It's done-up for Slime 1.0, even. The debian package is available here: http://commonwerx.org/software/contrib/slime/deb/ ...specficially, as http://commonwerx.org/software/contrib/slime/deb/slime_1.0-1_all.deb Now, some notes, about the package: - I'll be working on making a formal apt-style repository for it, sometime, i figure (once i can get some repository-making stuff worked-up with Common Lisp, and maybe not with PERL, I think) - The package was built on debian/sid - I'm not sure how well it'll work on other Debian releases - like, on sarge or on woody -, though I figure that it should be fine, regardless. If it's not, I can work on packaging it for other Debianish releases. - the package, it does include the postinst/prerm stuff with common-lisp-controller, for swank.asd, and it does include some stuff for installing the slime emacs code, with the debian policies about emacsen - so, hopefully it'll fit all snugly, with everything that it needs to. If there are any problems about the package or the packaging, please just mention it, and I'll get to resolving it, momentarily. Now, some notes taht are not, so much, about the package, but which are probably worth mentioning, here, at about the same time: - The current URL, for it, is "kind of temporary" (until i get around to configuring a vhost to use specifically for "software stuff") - but, the URLs mentioned here, they do work, and they should continue to work, even after any-such-updates - I'm not 100% sure about using the "all" architecture for the package - I I figure that it should work, but if it causes a problem, I'd be glad to hear of it - the *.diff.gz there - available under the first URL given, above - the diff should contain just the contents of the ./debian/ directory; I've no problem, about the prospect of those ./debian/* files being added to the slime upstream cvs code. (I've no worries about it, regardless - whether or not it would be seen as if those ./debian/* files - from the diff.gz - as if they would worth adding to the main slime codebase) - I'd be willing to be an official debian package maintainer for slime. I'm not sure how that would work, though - I've heard something about being "sponsored" for it, at least. I notice that this does go off-topic from the list, though - but for to mention, if anyone might be in a spot for it, that I'm willing to take it up. I think that's all of it. Slime is really really keen; I'm glad to be able to use it, seriously. So, thank you. -- Sean Champ schamp(a)commonwerx.org

1 0

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

04 Sep '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.513 slime/ChangeLog:1.514 --- slime/ChangeLog:1.513 Thu Sep 2 08:00:38 2004 +++ slime/ChangeLog Sat Sep 4 00:12:53 2004 @@ -1,3 +1,22 @@ +2004-09-03 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> + + * NEWS: Summarize changes since August. + + * slime.el: Add some docstrings. + + * mkdist.sh: Add PROBLEMS file. We're no longer alpha. + + * swank.lisp: Remove debugging code in comment. + + * swank-sbcl.lisp: Delete dead code. + + * swank-lispworks.lisp (defimplementation): define-dspec-alias + seems to more apropriate than define-form-parser. + + * swank-cmucl.lisp (print-frame): Catch errors during printing. + + * README: Fix typo. + 2004-09-02 Wolfgang Mederle <wolfgang.mederle(a)stud.uni-muenchen.de> * swank-loader.lisp (*lisp-name*): Replace / with - in CMUCL

1 0

[slime-devel] return-from-frame for CMUCL
by Jan Rychter 04 Sep '04

04 Sep '04

The enclosed tiny patch enables support for return-from-frame ("R" in sldb or DEBUG-RETURN in CMUCL debugger) in CMUCL. You need the latest (09/2004) CMUCL snapshot and optimization set to (debug 3) to use it. --J. ChangeLog entry: 2004-09-05 Jan Rychter <jan(a)rychter.com> * swank-cmucl.lisp (return-from-frame): Add it. ------------------------------------------------------------------------- Index: swank-cmucl.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank-cmucl.lisp,v retrieving revision 1.113 diff -u -3 -p -u -r1.113 swank-cmucl.lisp --- swank-cmucl.lisp 28 Aug 2004 02:27:08 -0000 1.113 +++ swank-cmucl.lisp 2 Sep 2004 16:35:30 -0000 @@ -1511,6 +1511,14 @@ A utility for debugging DEBUG-FUNCTION-A (defimplementation frame-catch-tags (index) (mapcar #'car (di:frame-catches (nth-frame index)))) +;; This requires a custom patch to CMUCL at the moment, hopefully this +;; will be incorporated into CMUCL soon. +(defimplementation return-from-frame (index form) + (let* ((frame (nth-frame index)) + (probe (di::find-debug-tag-for-frame frame))) + (cond (probe (throw (car probe) (eval-in-frame form index))) + (t (format nil "Cannot return from frame: ~S" frame))))) + (defun set-step-breakpoints (frame) (when (di:debug-block-elsewhere-p (di:code-location-debug-block (di:frame-code-location frame)))

1 0

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 02 Sep '04

02 Sep '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.511 slime/ChangeLog:1.513 --- slime/ChangeLog:1.511 Tue Aug 31 00:34:24 2004 +++ slime/ChangeLog Thu Sep 2 08:00:38 2004 @@ -1,3 +1,32 @@ +2004-09-02 Wolfgang Mederle <wolfgang.mederle(a)stud.uni-muenchen.de> + + * swank-loader.lisp (*lisp-name*): Replace / with - in CMUCL + version strings. + +2004-09-01 John Paul Wallington <jpw(a)gnu.org> + + * slime.el (slime-oneliner): Don't use free variable. + (slime-recenter-window, slime-set-connection-info) + (slime-pprint-event, slime-compiler-notes-quit) + (slime-apropos-summary): Likewise. + (slime-connect): Tidy up handshake `message' call. + +2004-09-01 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> + + * slime.el (slime-repl-push-directory): Fix interactive spec. + (sldb-reference-properties): Take a the reference object as + argument instead of its parts. Fix callers accordingly. + (slime-fuzzy-choices-buffer): Remove assignment to unused variable + slime-fuzzy-target-mtime. + (slime-ed): Replace call to new-frame with make-frame. + (sldb-find-buffer): Cleanup. + (sldb-highlight-sexp): Fix regexp. It's now almost a full line. + + * swank.lisp (completion-set, tokenize-symbol-designator) + (tokenize-completion, fuzzy-completion-set) + (briefly-describe-symbol-for-emacs): Remove simple-base-string + declarations. + 2004-08-30 Helmut Eller <e9626484(a)stud3.tuwien.ac.at> * PROBLEMS: We require SBCL 0.8.13. 0.8.12 is no longer

1 0

[slime-devel] Bug in directory creation
by Wolfgang Mederle 02 Sep '04

02 Sep '04

Hi, I encountered a bug in a fresh CVS slime with the latest cmucl binary for Darwin: CL-USER> (lisp-implementation-version) "19a+ Darwin/Exp 2004-07-25-090" ^ That's the problem. I'd like to suggest this change in swank-loader.lisp: ----- 8< ------------ --- swank-loader.lisp.~1.31.~ Fri Aug 6 02:43:35 2004 +++ swank-loader.lisp Thu Sep 2 02:32:22 2004 @@ -41,7 +41,7 @@ ))) (defparameter *lisp-name* - #+cmu (format nil "cmu-~A" (lisp-implementation-version)) + #+cmu (format nil "cmu-~A" (substitute #\- #\/ (lisp-implementation-version))) #+sbcl (format nil "sbcl-~A" (lisp-implementation-version)) #+openmcl "openmcl" #+lispworks (format nil "lispworks-~A" (lisp-implementation-version)) ------------- >8 ------------ (I hope, the above format is correct, it's the first time I try to send a patch.) Regards, Wolfgang -- Wolfgang Mederle wolfgang.mederle(a)stud.uni-muenchen.de * GPG key id E8EF7E7F <URL:http://www.mederle.de/> ICQ# 1435333 * AIM therealmadearl

2 1