On Sep 14, 2010, at 4:59 PM, Alan wrote:
> I think you could just keep a vector of (color,shape) pairs as an
> atom, and reify a subclass of JPanel whose paint() method closes
> around that atom, calling (.draw shape color). Then as long as you
> call repaint every time you modify the atom, you should be done. Seems
> like it's hard to see a library making this much easier, as there's
> not much incidental complexity to begin with. And alpha will be
> trivial, since java.awt.Color supports alpha values - you won't even
> have to know about it.
Thanks Alan. Following is one of my current versions, which does something
similar to what you suggest, although it differs in some details. It also
differs somewhat from what I originally requested insofar as it just sticks
size limits and the panel object in vars, rather than allowing them to be set
dynamically in a window-creating function call (although that could easily be
fixed in various ways). It also has a single draw-shape function that takes a
shape name as its first arg, rather than having separate functions for
rectangles and ovals -- just because that also made things a little more
concise. And I haven't added line-drawing to this yet. All of this could be
fixed easily, but I think that this is enough for me to ask my question about
minimality more concretely.
The code below is still more than I would like, and I'm wondering if there's a
more concise way to do this (again, without additional libraries). One way to
reframe my question is to imagine that you're in front of a class (as I will be
in a couple of hours :-), and you want to say "And here's how you can draw a
circle in a graphics window..." and (for pedagogical reasons into which I won't
digress here) you don't want to use any additional libraries or open a file
with pre-written code that does magic that you haven't explained. You want to
start with an empty file and literally type in the code needed to draw shapes,
from scratch. You want to provide functionality sufficient to do the kind or
random-shape-drawing example that I have at the bottom of the code. How little
code can you get away with writing to do this?
The sources of "more code than I would like" in my code below include:
- All that stuff in the initialization value of my panel var. Is that really
the simplest way to create a graphics window and store the reference that
allows me to draw to it?
- The code to store and redraw all of my shapes. I know I could clean up my
representation a bit (and probably will do so to add lines, since they will
have a different number of parameters), but is there some different kind of
java graphics canvas that buffers automatically so that I don't have to do this
explicitly myself?
- The "case" in render-shape, which is there because the various more concise
approaches that I thought of all founder on the fact that I can't seem to treat
methods like functions, passing them and applying them, etc. Is there a more
elegant approach to this?
Any advice would be appreciated.
Thanks!
;;; simple shape drawing code
(def max-x 500)
(def max-y 500)
(def shapes (atom ()))
(defn render-shape [g [shape-name x y h w color]]
(.setColor g color)
(case shape-name
rect (.fillRect g x y h w)
oval (.fillOval g x y h w)))
(def panel
(let [jp (proxy [javax.swing.JPanel]
[]
(getPreferredSize [] (java.awt.Dimension. max-x max-y))
(paint [g] (doall (map #(render-shape g %) @shapes))))]
(doto (new javax.swing.JFrame "My graphics window")
(.setSize max-x max-y)
(.add jp)
(.setVisible true))
jp))
(defn draw-shape [shape-name x y h w color]
(swap! shapes concat (list (list shape-name x y h w color)))
(.paint panel (.getGraphics panel)))
;; test it out by drawing a bunch of random shapes
(dotimes [_ 20]
(draw-shape (rand-nth '(rect oval))
(rand-int 400) (rand-int 200) (rand-int 400) (rand-int 400)
(new java.awt.Color
(rand-int 256) (rand-int 256) (rand-int 256) (rand-int 256))))
--
Lee Spector, Professor of Computer Science
School of Cognitive Science, Hampshire College
893 West Street, Amherst, MA 01002-3359
[email protected], http://hampshire.edu/lspector/
Phone: 413-559-5352, Fax: 413-559-5438
Check out Genetic Programming and Evolvable Machines:
http://www.springer.com/10710 - http://gpemjournal.blogspot.com/
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to [email protected]
Note that posts from new members are moderated - please be patient with your
first post.
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
