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

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 25 Nov '07

25 Nov '07

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1245 slime/ChangeLog:1.1250 --- slime/ChangeLog:1.1245 Fri Nov 23 03:26:31 2007 +++ slime/ChangeLog Sat Nov 24 03:19:25 2007 @@ -1,3 +1,47 @@ +2007-11-24 Helmut Eller <heller(a)common-lisp.net> + + Drop remaining dependencies on nregex. + + * swank-lispworks.lisp (unmangle-unfun): Use sys::setf-symbol-p + instead of regexp matching. + + * swank-loader.lisp (*sysdep-files*): Don't include nregex. + +2007-11-24 Helmut Eller <heller(a)common-lisp.net> + + Mirror *modules* in Emacs. + + * slime.el (slime-lisp-modules): New connction variable. + (slime-set-connection-info): Load requested modules. + (slime-required-modules): New variable. + (slime-require, slime-load-contribs): New functions. + + * swank.lisp (connection-info): Include *modules*. + (swank-require): Accept a list of modules as argument. + +2007-11-24 Helmut Eller <heller(a)common-lisp.net> + + * swank.lisp (parse-package): The old version didn't pass the + test-suite. Now use the reader directly instead of emulating it + half-heartedly. + + * slime.el (slime-search-buffer-package): Don't remove double + quotes or "#:", swank:parse-package takes care of that. + +2007-11-24 Helmut Eller <heller(a)common-lisp.net> + + * swank.lisp (apropos-symbols): Use simple search instead of + regexps. + (make-apropos-matcher): Used to be make-regexp-matcher. + + (*sldb-printer-bindings*): Set *print-right-margin* to + most-positive-fixnum. This prints each frame in the backtrace in a + single long line. But is suboptimal for other purposes, like + eval-in-frame. + + (setup-server): Initialize multiprocessing here, so that is also + done for create-server. + 2007-11-23 Tobias C. Rittweiler <tcr(a)freebits.de> * swank.lisp (swank-require): Fix typo (:key was used instead of Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.65 slime/contrib/ChangeLog:1.66 --- slime/contrib/ChangeLog:1.65 Tue Nov 20 16:30:51 2007 +++ slime/contrib/ChangeLog Sat Nov 24 03:15:20 2007 @@ -1,3 +1,7 @@ +2007-11-24 Helmut Eller <heller(a)common-lisp.net> + + * slime-fuzzy.el: Use slime-require instead of a connected-hook. + 2007-11-20 Helmut Eller <heller(a)common-lisp.net> * swank-fancy-inspector.lisp (inspect-for-emacs function t): Don't

1 0

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 24 Nov '07

24 Nov '07

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1243 slime/ChangeLog:1.1245 --- slime/ChangeLog:1.1243 Thu Nov 22 17:33:39 2007 +++ slime/ChangeLog Fri Nov 23 03:26:31 2007 @@ -1,3 +1,8 @@ +2007-11-23 Tobias C. Rittweiler <tcr(a)freebits.de> + + * swank.lisp (swank-require): Fix typo (:key was used instead of + :test.) Reported by Stelian Ionescu. + 2007-11-22 Helmut Eller <heller(a)common-lisp.net> * swank.lisp (swank-require): Don't search the file if the

1 0

[slime-devel] patch to make ecl work again
by Geo Carncross 23 Nov '07

23 Nov '07

Hello, ecl+slime doesn't seem to have worked for a while, so I cobbled together a minimal threads implementation that seems to work, and I can do basic stuff again. I'm not certain what the policy is on contributions/patches, nor how most people read here, so I'm including it both as an attachment and as a link. http://geocar.googlepages.com/swank-ecl-threads.patch Is there anything I should do in the future?

2 3

[slime-devel] ecl patch to get arglist of compiled functions
by Geo Carncross 23 Nov '07

23 Nov '07

It occurs to me that ecl actually puts the arglist in the documentation strings of compiled functions. This makes slime feel a *lot* more usable under ecl (esp: considering how its compiler works), even if it otherwise seems a bit hackish. http://geocar.googlepages.com/swank-ecl-more-arglist.patch I also noticed that (streamp slime-output-stream) is true on sbcl and false on the current (0.9i) release of ecl. This already seems to have been fixed in ecl's cvs, so if anyone else wanted to use slime+ecl they really should use ecl-cvs or wait until ecl 1.0.

1 0

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 23 Nov '07

23 Nov '07

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1242 slime/ChangeLog:1.1243 --- slime/ChangeLog:1.1242 Mon Nov 19 15:18:29 2007 +++ slime/ChangeLog Thu Nov 22 17:33:39 2007 @@ -1,3 +1,10 @@ +2007-11-22 Helmut Eller <heller(a)common-lisp.net> + + * swank.lisp (swank-require): Don't search the file if the + module-name is present in *modules*. That should avoid problems if + swank is included in a core file and moved to a different + location. Reported by John Wiegley. + 2007-11-19 Tobias C. Rittweiler <tcr(a)freebits.de> * slime.el (slime-repl-mode-map, slime-repl-read-mode),

1 0

[slime-devel] Slime behavior different (SBCL vs CMUCL)?
by Jeff Cunningham 22 Nov '07

22 Nov '07

Hi; When I first started using Slime I was using CMUCL. One of the features I really appreciated was that when running a program under Slime I could modify and recompile a defun and it would take effect the next time it was called by the running program. I subsequently switched to using SBCL, but when I did I seemed to lose this ability to recompile while running. The C-c C-c commands would queue up an not execute until the program finished. But recently I setup Hunchentoot daemon with Swank in a core image that when run allows me to slime-connect, and when I do so I find that I can recompile stuff and have it take effect without killing the running process. This makes me wonder if there isn't some way to do this under Slime after all. Clearly I don't fully understand the interactions between Slime, Swank and the inferior lisp. Is there another way I can launch SBCL from within Slime so it works the way CMUCL used to? Thanks. --Jeff

3 2

[slime-devel] Problem with `swank-require'
by John Wiegley 22 Nov '07

22 Nov '07

Around line 2283 in swank.lisp is a function, swank-require. It calls CL's `require' passing three arguments. Now, the third argument is determined by a call to `module-filename', just a few lines down, which triggers an error if it can't find the file. The reason this is a problem is that I bind swank into my SBCL core and then I move that core to other systems, which don't have the swank Lisp files on them. Although a regular call to require works fine in this case, swank generates an error because the .lisp file itself cannot be found. I think the solution is for module-filename to just return nil if it can't find anything, and leave it up to `require' to generate the error if need be. John

2 1

[slime-devel] [BUG] slime-apropos with invalid regexes
by Michael Weber 22 Nov '07

22 Nov '07

Hi, With a CVS snapshot of SLIME (19.11.), M-x slime-apropos * RET makes my sbcl spin with 100% CPU. Interrupting it in the REPL (C-c C-c) brings it back. "*" is, of course, not a valid regex, but I'd appreciate a better indication of the error. This is with SBCL 1.0.10.55 on OS X/Tiger. OpenMCL (Version 1.1- pre-070722 (DarwinX8664)) shows similar behavior. After interrupting, the following is printed at the REPL: ; in: ; LAMBDA (STRING &KEY (SLIME-NREGEX::START 0) ; (SLIME-NREGEX::END (LENGTH STRING))) ; (1- SLIME-NREGEX::START) ; --> - ; ==> ; SLIME-NREGEX::START ; ; note: deleting unreachable code ; (LET ((LENGTH SLIME-NREGEX::END)) ; (DO ((SLIME-NREGEX::MARKER SLIME-NREGEX::START (1+ SLIME- NREGEX::MARKER))) ; ((> SLIME-NREGEX::MARKER SLIME-NREGEX::END) NIL) ; (LET ((SLIME-NREGEX::INDEX SLIME-NREGEX::MARKER)) ; (IF (BLOCK SLIME-NREGEX::COMPARE #) (RETURN T))))) ; ; caught STYLE-WARNING: ; The variable LENGTH is defined but never used. ; (DO ((SLIME-NREGEX::START SLIME-NREGEX::INDEX (1- SLIME- NREGEX::START))) ; ((< SLIME-NREGEX::START SLIME-NREGEX::OINDEX) NIL) ; (LET ((SLIME-NREGEX::INDEX SLIME-NREGEX::START)) ; (BLOCK SLIME-NREGEX::COMPARE ; (SETF (CADR #) SLIME-NREGEX::INDEX) ; (RETURN-FROM SLIME-NREGEX::FINAL-RETURN T)))) ; --> BLOCK LET ; ==> ; SLIME-NREGEX::INDEX ; ; note: deleting unreachable code ; (LET ((SLIME-NREGEX::INDEX SLIME-NREGEX::START)) ; (BLOCK SLIME-NREGEX::COMPARE ; (SETF (CADR (AREF SLIM

2 1

Re: [slime-devel] Slime behavior different (SBCL vs CMUCL)?
by Jeff Cunningham 22 Nov '07

22 Nov '07

John Wiegley wrote: > On Nov 22, 2007, at 4:48 PM, Jeff Cunningham wrote: > >> What about when you aren't running Hunchentoot? Can you interact with >> a running program by re-compiling code without stopping it? (assuming >> threads are enabled). > > When Hunchentoot is not running, I don't run functions that don't > exit. I can imagine that in that case, it may be possible for SLIME > to hang waiting for a response from the inferior Lisp. I just built a > threaded SBCL yesterday, so until now I've only used the REPL for > commands that return immediately. > > John > I guess I'm not being very clear. Suppose the following scenario: I am testing a loop function that monitors external hardware status and responds to state changes. The loop runs continuously, but has been launched by me in the REPL by calling a defun. Now while this is functioning I determine that I need to change the value of a defparameter. Currently, I cannot re-evaluate the defparameter using C-c C-c without stopping the loop that is running in the REPL. When I used to use CMUCL I could do this. Does this make any sense? The reason why this is nice to do, is the loop requires half an hour of operation to get to the point where the parameter comes into play. Regards, Jeff

1 0

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 21 Nov '07

21 Nov '07

Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.64 slime/contrib/ChangeLog:1.65 --- slime/contrib/ChangeLog:1.64 Wed Oct 24 16:03:22 2007 +++ slime/contrib/ChangeLog Tue Nov 20 16:30:51 2007 @@ -1,3 +1,9 @@ +2007-11-20 Helmut Eller <heller(a)common-lisp.net> + + * swank-fancy-inspector.lisp (inspect-for-emacs function t): Don't + specialize the second argument, so that backend methods take + precedence. Reported by Maciej Katafiasz. + 2007-10-24 Tobias C. Rittweiler <tcr(a)freebits.de> * swank-arglist.lisp (decode-arglist): Fix incompatibility with

1 0