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

[cells-devel] Celtk: Making a widget or an item invisible - how?
by Frank Goenninger 03 Jan '07

03 Jan '07

Hi again - I have the following model: (defmodel workarea (ctk::widget ctk::window) ((visiblep :accessor visiblep :initform (c- in :undefined) :initarg :visiblep)) (:default-initargs :id (gensym "WORKAREA-"))) This workarea is meant to work as a manager for kids that will be displayed if visiblep (yeah, it's not visible-p ;-) is t and will be invisible when visiblep is nil. Question now is how that can be done... Via an observer that walks all kids and configures them as invisible? If so, then I have the following questions: 1. Tk-related: How to configure arbitrary objects as invisible ? 2. Cells-related: How to walk a family object's kids to call a function with the paths of the kids ? Thx! Frank

1 0

Fwd: [cells-devel] Happy New Year, and where is CVS?
by Thomas F. Burdick 03 Jan '07

03 Jan '07

---------- Forwarded message ---------- From: Thomas F. Burdick <tburdick(a)gmail.com> Date: Jan 3, 2007 4:02 PM Subject: Re: [cells-devel] Happy New Year, and where is CVS? To: Dmitri Hrapof <hrapof(a)common-lisp.ru> Cc: cells-devel(a)common-lisp.net Bonne année, frohes neues Jahr, etc. http://common-lisp.net/cgi-bin/viewcvs.cgi/cells/?cvsroot=cells

1 0

[cells-devel] Happy New Year, and where is CVS?
by Dmitri Hrapof 03 Jan '07

03 Jan '07

Thanks, Dmitri

1 0

[cells-devel] Celtk: Window is small despite given height and width
by Frank Goenninger 02 Jan '07

02 Jan '07

Hi - me again ;-) Anybody seen this: I am having a Celtk-based window and am giving height and width upon definition: (defmodel tibrv-console (window) () (:default-initargs :id :tibrv-console :title$ (conc$ "*** G&C::TIBCO Rendezvous Console - " *tibrv- console-version* " ***") :width (c-in 1200) :height (c-in 600) :kids (c? (the-kids ...........) All the kids are using direct placing as geometry control. This results in having only a small window despite the given height and width. Just looking for a solution other than this one: (defmethod initialize-instance :after ((self tibrv-console) &key &allow-other-keys) (tk-format '(:pre-make-tk self) "wm geometry ~A ~Ax~A" (^path) (^width) (^height))) Am I missing something ? All the best - Frank

1 0

Re: [cells-devel] Celtk: Button as kid of a canvas ...
by Ken Tilton 31 Dec '06

31 Dec '06

Frank Goenninger wrote: > > Hm - I think it's actually a problem of Celtk - it (oh, I shouldn't > use "it" - it's not Lispy enough ;-) not being complete in terms of > providing a Window as a container for arbitrary objects in a canvas. I saw that capability (widget in canvas) mentioned, just never had occasion to use it so it never got Celtk-ized. And at this point you may really have to fork Celtk to take it forward in re use of Tk widgets since I am concentrating on Cello for widgets. Maybe pick a new name and start a c-l.net project? > > ACL 8.0 - on Mac. Sh*t - this is a case where Slime really sucks. And > Franz, too: I've been asking for the full IDE on OS X for years now! <g> Given the number of Powerbooks I see at Lisp conferences, they might be missing an opportunity. But, hey, stop whining, how hard would it be to whip up a find-definitions dialog using Celtk? I once did a killer inspector using Cello. Do the classic incremental thing, just get ACL to pop up a list of definitions using as much standard mop as possible. Costanza has done some work on a portable MOP, maybe contrib to that as you go (since moppery will be needed to replicate the ACL IDE). Needless to say, a Slime-killer is something I would love to do just to piss off the Yobbos, and if I end up switching to LW for enough work it might just happen. Not yet, tho. kt kt

2 1

Re: [cells-devel] Celtk: Button as kid of a canvas ...
by Frank Goenninger 31 Dec '06

31 Dec '06

Thanks, Thomas. That did the trick. Frank Am 29.12.2006 um 14:16 schrieb Thomas F. Burdick: > In Ltk, you'd do this by creating a canvas-window item on the canvas, > then packing your button into that. So the button is a child of a > child of the canvas. I'm guessing you can achieve the same effect in > Celtk by making a canvas-window in the kids of the canvas, and your > button in its kids. > > On 12/28/06, Frank Goenninger <frgo(a)mac.com> wrote: >> Hi all: >> >> being too lazy for searching in the code of Celtk I wonder if there >> is a way to have a button as a kid of a canvas - same of for an entry >> widget ... >> >> I figure that the kid would be configured differently compared to >> being a kid of a frame... >> >> Anyone done that before? >> >> Thx for any pointers... >> >> Best wishes - Frank >> >> _______________________________________________ >> cells-devel site list >> cells-devel(a)common-lisp.net >> http://common-lisp.net/mailman/listinfo/cells-devel >>

2 2

[cells-devel] Re: Celtk: Button as kid of a canvas ...
by Thomas F. Burdick 31 Dec '06

31 Dec '06

On 12/31/06, Ken Tilton <kentilton(a)gmail.com> wrote: > > Am 29.12.2006 um 14:16 schrieb Thomas F. Burdick: > > > >> In Ltk, you'd do this by creating a canvas-window item on the canvas, > >> then packing your button into that. So the button is a child of a > >> child of the canvas. I'm guessing you can achieve the same effect in > >> Celtk by making a canvas-window in the kids of the canvas, and your > >> button in its kids. > > Weird, right? Is this a Tk issue? A Cells issue? An issue arising out of > the Lisp API to Tk? Follow the bouncing API.... It's a Tk thing -- canvasses are their own weird little world where you manually place canvas-items instead of letting a layout manager place widgets. canvas-windows just bridge the gap. You could automatically make the canvas-window if you try to place a widget in a canvas, but it seems like it might just make things uglier.

1 0

[cells-devel] Celtk: Why can I place a label but not a rectangle with :parent-x and :parent-y ?
by Frank Goenninger 30 Dec '06

30 Dec '06

Hi all you Celtk users: In my quest to design a beautiful GUI for my app I try to place my widgets in the app window. Therefore I try the following (part of Celtk-based code): (defmodel app-window (window) () (:default-initargs :width (c-in 1000) :height (c-in 600) :kids (c? (the-kids (mk-rectangle :tk-fill "grey" :height 100 :width 200 :parent-y 100 :parent-x 100))))) ... which results in an error saying :PARENT-Y :PARENT-X :HEIGHT are invalid initargs to make-instance of class #<STANDARD-CLASS CTK:RECTANGLE>. The valid initargs are :DISABLEDWIDTH :ACTIVEWIDTH :WIDTH :TAGS :STATE :DISABLEDSTIPPLE :ACTIVESTIPPLE :STIPPLE :DISABLEDOUTLINESTIPPLE :ACTIVEOUTLINESTIPPLE :OUTLINESTIPPLE :DISABLEDOUTLINE :ACTIVEOUTLINE :OUTLINE :OFFSET :DISABLEDFILL :ACTIVEFILL :TK-FILL :DASHOFFSET :DISABLEDDASH :ACTIVEDASH :DASH :DECORATIONS :COORDS :COORDS-TWEAK :L-COORDS :ID-NO :USER-ERRORS :ON-KEY-UP :ON-KEY-DOWN :ON-COMMAND :TIMERS :TK-CLASS :MD-NAME :ID :VALUE :AWAKEN-ON-INIT-P :FM-PARENT. [Condition of type PROGRAM-ERROR] Hm - now, that means I can't really say I want a specific height for a rectangle?! Also, I can't place this rectangle... Trying this with a label works Ok. How can that be? Any pointers really appreciated. Frank

1 0

[cells-devel] Celtk: Button as kid of a canvas ...
by Frank Goenninger 28 Dec '06

28 Dec '06

Hi all: being too lazy for searching in the code of Celtk I wonder if there is a way to have a button as a kid of a canvas - same of for an entry widget ... I figure that the kid would be configured differently compared to being a kid of a frame... Anyone done that before? Thx for any pointers... Best wishes - Frank

1 0

Re: [cells-devel] CVS: Missing file variables.lisp ?!
by Attila Lendvai 13 Dec '06

13 Dec '06

ok, bring it back on list then. there are useful infos in these mails... (or not :) > We? Parallel project? You and Thomas? C4? Other_________? > > Is it open source? Has it changed much/been rewritten from scratch? What > are you using it for, if you can say? oh, i tought you heard about it, its at: http://common-lisp.net/project/computed-class/ it's rewritten from scratch. has a little different nomenclature, but i'm planning to rename "computed-state" to "cell" after talking with the others, to bring the two projects closer. of course you're welcome to contribute/use/join the list/etc... > I do not mind at all. Keeps you yobbo rugrats out of my hair. :) And NIH hehe :) > is certainly my guiding light in software development... I can write my > own as fast as I can learn someone else's nonsense, and then it does > /exactly/ what I want and I know it cold and I can change it and.... well, i'm not that NIH-oriented, a well written lib can be useful, but i agree that *sometimes* it's worth rolling your own. hm, now that i'm thinking about it, we've ended up not using clsql and cells... maybe i do have NIH, i'm just not aware of it... :) but i do try first with the existing libs. we had a headache with cells that some internal state got corrupted and from then we had to reload the lib to make it work again. all this in the middle of a big rewrite. then the next day we rolled a prototype that in a few hours that was doing all we needed (we need a simple engine cells-wise, but full clos integration on the other hand). we use it for keeping computed state (for caching purposes) in sync with the declared state inside the model. "the model" is an UML-like data model integrated with CLOS. you write something like UML (semantically) and get a working app with persistency and a web gui. we is a small startup, working on http://common-lisp.net/project/cl-wdim/ if you don't mind we may bring this back to the list. but i didn't want to shamelessly advertise the project on the cells list... -- - attila "- The truth is that I've been too considerate, and so became unintentionally cruel... - I understand. - No, you don't understand! We don't speak the same language!" (Ingmar Bergman - Smultronstället)

1 0