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

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 27 Aug '04

27 Aug '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.509 slime/ChangeLog:1.510 --- slime/ChangeLog:1.509 Mon Aug 23 11:57:45 2004 +++ slime/ChangeLog Sat Aug 28 04:27:08 2004 @@ -1,3 +1,9 @@ +2004-08-27 Peter Seibel <peter(a)javamonkey.com> + + * swank-backend.lisp (swank-compile-string): Add directory + argument which is used by Allegro backend to improve source + recording for definitions compiled with C-c C-c. + 2004-08-23 John Paul Wallington <jpw(a)gnu.org> * slime.el (slime-pretty-package-name): Fix last cond clause.

1 0

[slime-devel] Better solution to redefinition warnings in Allegro
by Peter Seibel 27 Aug '04

27 Aug '04

Since my last attempt (a couple weeks ago) was totally bogus I tried again. Here's my latest attempt. I wanted to check whether there was anything about this version that was as busted as my previous attempt before I made the corresponding changes to the other backends. The basic idea is that we pass a new argument to swank-compile-string containing the directory of the file of the buffer from which we are compiling (when there is one). In the Allegro backend we bind *DEFAULT-PATHNAME-DEFAULTS* to this directory so that the source recording mechanism thinks the file it is recording is in that directory. Then you can add something like this to one's .swank.lisp to make Allegro ignore the :<position> part of the recorded location. #+allegro (setf excl::*redefinition-pathname-comparison-hook* (list #'(lambda (old new obj type) (declare (ignore obj type)) (cond ((string= (namestring old) (let ((str (namestring new))) (subseq str 0 (position #\: str :from-end t))))))))) Here's the patch; please let me know what you think. -Peter Index: slime.el =================================================================== RCS file: /project/slime/cvsroot/slime/slime.el,v retrieving revision 1.392 diff -u -r1.392 slime.el --- slime.el 16 Aug 2004 21:41:35 -0000 1.392 +++ slime.el 17 Aug 2004 15:32:21 -0000 @@ -3147,7 +3147,7 @@ (defun slime-compile-string (string start-offset) (slime-eval-async - `(swank:compile-string-for-emacs ,string ,(buffer-name) ,start-offset) + `(swank:compile-string-for-emacs ,string ,(buffer-name) ,start-offset ,(if (buffer-file-name) (file-name-directory (buffer-file-name)))) (slime-compilation-finished-continuation))) (defvar slime-hide-style-warning-count-if-zero t) @@ -7157,7 +7157,7 @@ ("swank::emacs-connected" "(swank::emacs-connected stream)") ("swank::compile-string-for-emacs" - "(swank::compile-string-for-emacs string buffer position)") + "(swank::compile-string-for-emacs string buffer position directory)") ("swank::connection.socket-io" "(swank::connection.socket-io \$struct\\(ure\$?\\|object\\|instance\\))") ("cl:lisp-implementation-type" Index: swank-allegro.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank-allegro.lisp,v retrieving revision 1.49 diff -u -r1.49 swank-allegro.lisp --- swank-allegro.lisp 4 Aug 2004 17:17:55 -0000 1.49 +++ swank-allegro.lisp 17 Aug 2004 15:32:21 -0000 @@ -239,14 +239,15 @@ (when binary-filename (delete-file binary-filename)))))) -(defimplementation swank-compile-string (string &key buffer position) +(defimplementation swank-compile-string (string &key buffer position (directory *default-pathname-defaults*)) ;; We store the source buffer in excl::*source-pathname* as a string ;; of the form <buffername>:<start-offset>. Quite ugly encoding, but ;; the fasl file is corrupted if we use some other datatype. (with-compilation-hooks () (let ((*buffer-name* buffer) (*buffer-start-position* position) - (*buffer-string* string)) + (*buffer-string* string) + (*default-pathname-defaults* (merge-pathnames (pathname directory)))) (compile-from-temp-file (format nil "~S ~S~%~A" `(in-package ,(package-name *package*)) Index: swank-backend.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank-backend.lisp,v retrieving revision 1.63 diff -u -r1.63 swank-backend.lisp --- swank-backend.lisp 30 Jul 2004 21:44:05 -0000 1.63 +++ swank-backend.lisp 17 Aug 2004 15:32:21 -0000 @@ -191,14 +191,17 @@ (declare (ignore ignore)) `(call-with-compilation-hooks (lambda () (progn ,@body)))) -(definterface swank-compile-string (string &key buffer position) +(definterface swank-compile-string (string &key buffer position directory) "Compile source from STRING. During compilation, compiler conditions must be trapped and resignalled as COMPILER-CONDITIONs. If supplied, BUFFER and POSITION specify the source location in Emacs. Additionally, if POSITION is supplied, it must be added to source -positions reported in compiler conditions.") +positions reported in compiler conditions. + +And if DIRECTORY is specified it is bound as the value of +*DEFAULT-PATHNAME-DEFAULTS*.") (definterface operate-on-system (system-name operation-name &rest keyword-args) "Perform OPERATION-NAME on SYSTEM-NAME using ASDF. Index: swank.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank.lisp,v retrieving revision 1.222 diff -u -r1.222 swank.lisp --- swank.lisp 13 Aug 2004 16:14:13 -0000 1.222 +++ swank.lisp 17 Aug 2004 15:32:22 -0000 @@ -1505,13 +1505,13 @@ Record compiler notes signalled as `compiler-condition's." (swank-compiler (lambda () (swank-compile-file filename load-p)))) -(defslimefun compile-string-for-emacs (string buffer position) +(defslimefun compile-string-for-emacs (string buffer position directory) "Compile STRING (exerpted from BUFFER at POSITION). Record compiler notes signalled as `compiler-condition's." (with-buffer-syntax () (swank-compiler (lambda () - (swank-compile-string string :buffer buffer :position position))))) + (swank-compile-string string :buffer buffer :position position :directory directory))))) (defslimefun operate-on-system-for-emacs (system-name operation &rest keywords) "Compile and load SYSTEM using ASDF. -- Peter Seibel peter(a)javamonkey.com Lisp is the red pill. -- John Fraser, comp.lang.lisp

3 8

[slime-devel] minor nit -- request
by Sundar Narasimhan 27 Aug '04

27 Aug '04

When I eval in frame .. if I mistype or stimulate another debug entry somehow, it's somewhat annoying when I pop back to the previous debugger level, the cursor moves to the 0'th frame.. wouldn't it be nicer if it stayed at the frame level where I was? (or am I missing the common idiom here?)

2 1

[slime-devel] slime 1.0 plan, party
by Luke Gorrie 23 Aug '04

23 Aug '04

Ahoy, The new plan is to release slime 1.0 on saturday the 4th of september. The release party will be in Cesky Krumlov (*) in the Czech republic (near the Austrian border) on the same day. See you all there! :-) (*) http://www.virtourist.com/europe/krumlov/ P.S., from today my email address is luke(a)member.fsf.org

1 0

[slime-devel] slime-pretty-package-name buglet
by John Paul Wallington 23 Aug '04

23 Aug '04

The byte-compiler noticed this: Index: slime.el =================================================================== RCS file: /project/slime/cvsroot/slime/slime.el,v retrieving revision 1.392 diff -u -r1.392 slime.el --- slime.el 16 Aug 2004 21:41:35 -0000 1.392 +++ slime.el 23 Aug 2004 00:09:40 -0000 @@ -434,7 +434,7 @@ "Return a pretty version of a package name designator (as a string)." (cond ((string-match "^:\$.*\$$" name) (match-string 1 name)) ((string-match "^\"\$.*\$\"$" name) (match-string 1 name)) - ((t name)))) + (t name))) (when slime-update-modeline-package (run-with-idle-timer 0.2 0.2 'slime-update-modeline-package)) 2004-08-23 John Paul Wallington <jpw(a)gnu.org> * slime.el (slime-pretty-package-name): Fix last cond clause.

2 1

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 21 Aug '04

21 Aug '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.507 slime/ChangeLog:1.508 --- slime/ChangeLog:1.507 Thu Aug 19 07:53:46 2004 +++ slime/ChangeLog Fri Aug 20 18:38:17 2004 @@ -1,3 +1,8 @@ +2004-08-21 Luke Gorrie <luke(a)bluetail.com> + + * swank.lisp (*global-debugger*): New configurable to globally + install swank-debugger-hook as *debugger-hook*. True by default. + 2004-08-19 Luke Gorrie <luke(a)bluetail.com> * doc/slime.texi: C-c C-c C-a, C-c C-u C-e.

1 0

[slime-devel] XREF for SBCL
by Nikodemus Siivola 21 Aug '04

21 Aug '04

The attached patch + new file hooks the PXREF to SBCL as well, sharing code with the Clisp backend. To make usage a bit less tedious I hooked PXREF:XREF-FILE into swank-compile-file. Clisp hasn't been tested after these changes, and neither has SBCL been subjected to much pounding, but "seems to work". To make the pxref more useful it might make sense to save the collected db on exits and load on startup... Also, an xref-op for asdf would not come amiss. (Just noting these out loud in case someone has nothing to do over the weekend.) Cheers, -- Nikodemus Schemer: "Buddha is small, clean, and serious." Lispnik: "Buddha is big, has hairy armpits, and laughs."

1 0

[slime-devel] Daily ChangeLog diff
by Luke Gorrie 20 Aug '04

20 Aug '04

Index: slime/ChangeLog diff -u slime/ChangeLog:1.506 slime/ChangeLog:1.507 --- slime/ChangeLog:1.506 Wed Aug 18 14:21:52 2004 +++ slime/ChangeLog Thu Aug 19 07:53:46 2004 @@ -1,3 +1,8 @@ +2004-08-19 Luke Gorrie <luke(a)bluetail.com> + + * doc/slime.texi: C-c C-c C-a, C-c C-u C-e. + Thanks Barry Fishman for reporting incorrect indexing. + 2004-08-18 Matthew Danish <mrd+nospam(a)cmu.edu> * slime/swank-allegro.lisp (swank-compile-string): Use #\; instead

1 0

[slime-devel] [PATCH} Problem with Restart node in slime.texi
by Barry Fishman 19 Aug '04

19 Aug '04

Version 4.7 makeinfo fails with case problems when referencing "Restarts" node as "restarts" in menus. Fix actually performed by Emacs texinfo mode by selecting "Texinfo/Update All Menus" from menubar. -- Barry Fishman

2 1

[slime-devel] Problem with SLIME + CMUCL + CLSQL + Oracle
by Ivan Toshkov 19 Aug '04

19 Aug '04

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I'm using recently updated version of SLIME, CMUCL 19a, CLSQL 3 and Oracle 8.1.7, everything on linux (Red Hat Enterprise Server 3). I do: CL-USER> (asdf:oos 'asdf:load-op :clsql-oracle) CL-USER> (use-package :clsql-user) CL-USER> (connect '("SID" "user" "pass") :database-type :oracle) At this point CMUCL starts using the processor on >90% and stays that way untill killed. I've tried the same dialog without SLIME and it works fine. Also, I've tried with SQLite instead of Oracle and it works under SLIME. Any ideas? Regards, Ivan -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) iD8DBQFBJGd/uUkHlsKSHdcRAg+NAJ9K7ZXwIpZK1IttTauTgjvxCP0BYwCZATTf q9AkKeWgtZWd/5Z+BWt2Hkw= =/RHQ -----END PGP SIGNATURE-----

2 4