October 2009 - cells-devel - mailman3.common-lisp.net

[cells-devel] A few patches for making cells (3.0 from cvs) run on debian unstable using sbcl.
by Peder Chr. Nørgaard 14 Dec '09

14 Dec '09

Hello cells-devel list. I am investigating cells and cells-gtk3 for a project that I am planning. For a starter, I needed to get the software running on my Debian unstable system. It needed a few patches that you may wish to take into your development. The version of cells that I use is taken with cvs -z3 -d :pserver:anonymous:anonymous@common-lisp.net:/project/cells/cvsroot co cells That is the most actual one, yes? Well, the sbcl on Debian unstable (1.0.18) does not have nor need the EXCL package. So it must be configured out: diff -u -r cells/defpackage.lisp cl/cells/defpackage.lisp --- cells/defpackage.lisp 2008-10-12 03:21:07.000000000 +0200 +++ MDM/cl/cells/defpackage.lisp 2008-11-28 20:08:38.000000000 +0100 @@ -23,7 +23,7 @@ (in-package :common-lisp-user) (defpackage :cells - (:use #:common-lisp #:excl #:utils-kt) + (:use #:common-lisp #-sbcl #:excl #:utils-kt) (:import-from ;; MOP #+allegro #:excl diff -u -r cells/utils-kt/defpackage.lisp cl/cells/utils-kt/defpackage.lisp --- cells/utils-kt/defpackage.lisp 2008-10-12 03:21:10.000000000 +0200 +++ MDM/cl/cells/utils-kt/defpackage.lisp 2008-11-28 20:04:08.000000000 +0100 @@ -35,7 +35,8 @@ (defpackage :utils-kt (:nicknames #:ukt) - (:use #:common-lisp #:excl + (:use #:common-lisp + #-sbcl #:excl #+(or allegro lispworks clisp) #:clos #+cmu #:mop #+sbcl #:sb-mop The sbcl does not have the MOP package - it is called SB-MOP. Much of the code knows that already, but I had to patch one file: diff -u -r cells/utils-kt/detritus.lisp cl/cells/utils-kt/detritus.lisp --- cells/utils-kt/detritus.lisp 2008-10-12 03:21:10.000000000 +0200 +++ MDM/cl/cells/utils-kt/detritus.lisp 2008-11-28 20:07:57.000000000 +0100 @@ -33,7 +33,7 @@ (let ((cc (find-class c))) (when cc (finalize-inheritance cc)) - (mop::class-prototype cc))) + (#+sbcl sb-mop:class-prototype #-sbcl mop::class-prototype cc))) (defun brk (&rest args) After that, I tried to run the test suite. Worked fine, except that it is a little bit too Window'ish. I did this: diff -u -r cells/cells-test/test.lisp cl/cells/cells-test/test.lisp --- cells/cells-test/test.lisp 2008-10-12 03:21:09.000000000 +0200 +++ MDM/cl/cells/cells-test/test.lisp 2008-12-01 22:25:26.000000000 +0100 @@ -69,7 +69,7 @@ (defun test-cells () - (dribble "c:/0algebra/cells-test.txt") + (dribble "/tmp/cells-test.txt") (progn ;prof:with-profiling (:type :time) (time (progn but that is not what you want to go in the main sources. Maybe something based on the features - (dribble #+unix "/tmp/cells-test.txt" #-unix "c:/0algebra/cells-test.txt") - or perhaps just a file name parameter to TEST-CELLS with the above as the default value? Hope this is of a little help to you. I am sure to come back for assistance for my project later. best regards -- Peder Chr. Nørgaard e-mail: pcn(a)pogt.dk Gefionsvej 19 DK-8230 Åbyhøj tel: +45 87 44 11 99 Denmark mob: +45 30 91 84 31

6 5

[cells-devel] Great idea!
by Kenneth Tilton 31 Oct '09

31 Oct '09

http://coherence-lang.org/Onward09.pdf (thx to Paul Tarvydas for the heads up) Edwards acknowledges Trellis as a temporal predecessor tho he was unaware of it. Mentions Garnet, too. FRP, too. Not sure I like re-running rules to achieve correctness, tho it would be nice not having to record dependencies. Worth noting: not yet implemented. kt -- http://thelaughingstockatpngs.com/ http://www.facebook.com/pages/The-Laughingstock/115923141782?ref=nf

1 0

[cells-devel] mk-row / mk-stack
by Andy Chambers 24 Oct '09

24 Oct '09

Plaese find below an alternative version of def-mk-inline which means that if the "title" of a row or stack is nil, an unlabelled frame will be created. I modified this because I kept feeling like I was repeating myself with code like this..... (bif (title (maybe-get-title)) (mk-row (title) ....) (mk-row () ....)) With the definition below, I don't have to repeat myself. (mk-row ((maybe-get-title)) ...) What do you guys think? (defmacro def-mk-inline (name (unlabelled labelled)) `(defmacro ,name ((&rest initargs) &rest kids) (cond ((evenp (length initargs)) `(make-instance ',',unlabelled :fm-parent *parent* ,@initargs :kids (c? (the-kids ,@kids)))) ((null (car initargs)) `(make-instance ',',unlabelled :fm-parent *parent* ,@(cdr initargs) :kids (c? (the-kids ,@kids)))) (t `(make-instance ',',labelled :fm-parent *parent* :text ,(car initargs) ,@(cdr initargs) :kids (c? (the-kids ,@kids))))))) -- ---- Andy Chambers Formedix Ltd

1 0