September 2008 - slime-devel - mailman3.common-lisp.net

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 15 Sep '08

15 Sep '08

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1504 slime/ChangeLog:1.1505 --- slime/ChangeLog:1.1504 Fri Sep 12 14:55:42 2008 +++ slime/ChangeLog Sun Sep 14 13:10:34 2008 @@ -1,3 +1,31 @@ +2008-09-14 Helmut Eller <heller(a)common-lisp.net> + + Introduce a WAIT-FOR-INPUT backend function. + CMUCL's blocking input functions READ-CHAR etc. + are hard to use with interrupts. In the backend + we have a more realistic chance to get interrupts working. + + * swank-backend.lisp (wait-for-input): New function. + + * swank-cmucl.lisp, swank-clisp.lisp (wait-for-input): Implement + it. + + * swank.lisp (wait-for-event/event-loop): Use WAIT-FOR-INPUT and + rescan the event-queue if WAIT-FOR-INPUT was interrupted. + (reader-event): Deleted. Merged into wait-for-event/event-loop + resp. dispatch-loop. + (decode-message): Drop the timeout argument. + (*events-enqueued*): A counter to quickly detect new events after + a wait. + (call-with-connection): If the argument is already the current + connection, don't rebind anything. + (without-slime-interrupts, with-slime-interrupts): Don't rebind + *pending-slime-interrupts*. Just to be save. + + * slime.el (sldb-maybe-kill-buffer): New function, to handle + the case when the debugger was interrupted in WAIT-FOR-INPUT and + we want to return to the previous debug level. + 2008-09-12 Helmut Eller <heller(a)common-lisp.net> For Lispworks, parse the $LWHOME/lwdoc file.

1 0

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 14 Sep '08

14 Sep '08

Index: slime/contrib/ChangeLog diff -u slime/contrib/ChangeLog:1.129 slime/contrib/ChangeLog:1.130 --- slime/contrib/ChangeLog:1.129 Wed Sep 10 19:51:17 2008 +++ slime/contrib/ChangeLog Sat Sep 13 06:39:02 2008 @@ -1,3 +1,16 @@ +2008-09-13 Tobias C. Rittweiler <tcr(a)freebits.de> + + * slime-parse.el (slime-has-symbol-syntax-p): New. + (slime-parse-symbol-name-at-point): New; works on top of + `slime-parse-sexp-at-point'. + (slime-enclosing-form-specs): Use it. + + * slime-enclosing-context.el (slime-find-bound-names): Use + `slime-parse-symbol-name-at-point'. + (slime-find-bound-functions): Ditto. + (def-slime-test enclosing-context.1): New test case. Thanks to + John Pallister for reporting this bug. + 2008-09-11 Tobias C. Rittweiler <tcr(a)freebits.de> * swank-asdf.lisp (operate-on-system-for-emacs): Adapted to recent

1 0

[slime-devel] Buffered slime output
by Raymond Toy 13 Sep '08

13 Sep '08

Until yesterday, I had been using a version from 2008-07-08. This worked pretty well. Yesterday I updated to the latest CVS version. Everything seems to be working fine, but now the output appears to be buffered strangely. Any function that produces a lot of output seems to have it buffered. When I enter something else into the repl, the missing output then comes out. This is a bit annoying. So I set *use-dedicated-output-stream* to T. This works much better, but now the output is interleaved with the output of the repl. That is, the result of the repl is printed before the output of the function. I can live with that, but it's kind of annoying too. Is there something I've missed to cause this behavior? Oh, I'm also running this with the latest snapshot of cmucl (no unicode) Ray

3 3

[slime-devel] Slime-Inspector, and SLDB
by Tobias C. Rittweiler 13 Sep '08

13 Sep '08

`q' in the in Slime inspector, and in SLDB still unconditionally restore the window configuration. Either they should be written on top of slime-popup-buffers, or at least their restoration code should be tweaked. -T.

1 0

[slime-devel] swank-compilation-result
by Helmut Eller 13 Sep '08

13 Sep '08

Currently swank-compilation-result is declared as: (defstruct (:swank-compilation-result (:type list) :named (:conc-name swank-compilation-result.) (:constructor make-swank-compilation-result ())) notes ; results ; one result is of type (MEMBER T NIL :COMPLAINED) durations ; ) All the slots are lists. Can we change that, so that one compilation-result contains only one element of those lists? If a command runs multiple compilations it could return a list of those simpler compilation-results. Helmut.

1 0

[slime-devel] slime-autodoc fails on `(flet ((, nil ())) (cond (, nil)))
by John Pallister 13 Sep '08

13 Sep '08

Hello list, I have noticed a bug when using slime-autodoc. I've duplicated it on GNU Emacs 23.0 with Lispworks 5.1.1 on Windows, and GNU Emacs 21.4 with SBCL 1.0.15 on Linux. If I enter the form (either in the REPL or in a buffer): `(flet ((,nil ())) (cond (,nil))) ^----^ then when I move the point into the indicated sexp, I get a "comma not inside a backquote" error from the Lisp. The following forms don't trigger the error: `(flet ((nil ())) (cond (,nil))) `(flet ((,nil ())) ,nil (cond (,nil))) The offending event is: (:emacs-rex (swank:format-arglist-for-echo-area ",nil" :operator nil :highlight nil :print-right-margin 1000 :print-lines 1) "COMMON-LISP-USER" :repl-thread 156) I had a bit of a look with the debugger and got as far as SLIME-AUTODOC-LOCAL-ARGLIST calling SLIME-FIND-BOUND-FUNCTIONS before giving up. Hopefully these clues will help someone more familiar with SLIME's internal workings. Cheers, John :^P -- John Pallister john(a)synchromesh.com

2 1

[slime-devel] [RfC] Marking frames restartable in SLDB
by Tobias C. Rittweiler 12 Sep '08

12 Sep '08

I'd like to mark restartable frames specially within a SLDB backtrace. One way would be to use the first column, and use "*" or "R" if a frame is restartable. Another way would be to place "Restartable" in the details section if one toggles a frame. Or perhaps both, or something else entirely. Thoughts? -T.

2 2

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 12 Sep '08

12 Sep '08

Index: slime/ChangeLog diff -u slime/ChangeLog:1.1498 slime/ChangeLog:1.1501 --- slime/ChangeLog:1.1498 Wed Sep 10 19:18:36 2008 +++ slime/ChangeLog Thu Sep 11 08:32:01 2008 @@ -1,3 +1,26 @@ +2008-09-11 Helmut Eller <heller(a)common-lisp.net> + + * doc/slime-refcard.tex: Fix typos. + +2008-09-11 Gary King <gwking(a)metabang.com> + + * swank-allegro.lisp (fspec-definition-locations): add declare + ignores to prevent warnings + + (emacs-inspect): remove first definition on function since it + was being overwritten by the next one. Wrap the method on t with + a excl:without-redefinition-warnings to prevent warning. + +2008-09-11 Helmut Eller <heller(a)common-lisp.net> + + * swank-cmucl.lisp (slime-output-stream): Remove last-flush-time + slot. + (sos/flush): Renamed from sos/misc-force-output. Don't try to be + clever: no timestamps and no line buffering. + (sos/write-char, sos/write-string): Renamed from sos/out + resp. sos/sout. Call output-fn outside without-interrupts. + (sos/reset-buffer): New function. + 2008-09-11 Tobias C. Rittweiler <tcr(a)freebits.de> * slime.el (slime-compilation-unit): Renamed to

1 0

[slime-devel] identifying 64-bit lisps in the lisp-version-string
by Gary King 11 Sep '08

11 Sep '08

I don't follow OpenMCL / CCL or Lispworks closely so maybe their is no longer a need to differentiate between 64-bit and non-64-bit fasls. If there is, then shouldn't the definition of lisp-version-string look more like the one below? thanks (diff -u this time! <smile>) Index: swank-loader.lisp =================================================================== RCS file: /project/slime/cvsroot/slime/swank-loader.lisp,v retrieving revision 1.88 diff -u -w -r1.88 swank-loader.lisp --- swank-loader.lisp 17 Aug 2008 08:31:17 -0000 1.88 +++ swank-loader.lisp 11 Sep 2008 13:18:52 -0000 @@ -58,10 +58,16 @@ :sparc64 :sparc :hppa64 :hppa)) (defun lisp-version-string () - #+(or openmcl cmu) (substitute-if #\_ (lambda (x) (find x " /")) + #+cmu (substitute-if #\_ (lambda (x) (find x " /")) (lisp-implementation-version)) - #+(or cormanlisp scl sbcl ecl) (lisp-implementation-version) - #+lispworks (lisp-implementation-version) + #+openmcl (format nil "~a~a" + (substitute-if #\_ (lambda (x) (find x " /")) + (lisp-implementation-version)) + #+ppc64-target "-64bit" + #-ppc64-target "") + #+lispworks (format nil "~A~@[~A~]" + (lisp-implementation-version) + (when (member :lispworks-64bit *features*) "-64bit")) #+allegro (format nil "~A~A~A" excl::*common-lisp-version-number* @@ -69,7 +75,9 @@ (if (member :64bit *features*) "-64bit" "")) #+clisp (let ((s (lisp-implementation-version))) (subseq s 0 (position #\space s))) - #+armedbear (lisp-implementation-version)) + #+(or armedbear cormanlisp ecl scl sbcl) + (lisp-implementation-version) + ) (defun unique-dir-name () "Return a name that can be used as a directory name that is -- Gary Warren King, metabang.com Cell: (413) 559 8738 Fax: (206) 338-4052 gwkkwg on Skype * garethsan on AIM

2 1

[slime-devel] just upgrading and am looking for previous behavior of command navigation
by Gary King 11 Sep '08

11 Sep '08

My apologies if this is a FAQ. Until yesterday, I was using a very old version of SLIME. I've upgraded and am missing a few things. The first is that I used to be able to type a little bit in the REPL and then use m-p and m-n to navigate to other commands that started with the same characters as what I had typed. E.g., if I typed > (in- [m-p] SLIME would find > (in-package cl-graph) This now longer is the default behavior. I expect that there is a way to turn it back on but my emacs foo is still very weak. Any help would be appreciated, -- Gary Warren King, metabang.com Cell: (413) 559 8738 Fax: (206) 338-4052 gwkkwg on Skype * garethsan on AIM

2 2