Neal Gafter's of the closures implementation in Java has given us enough playground to fool around with. Of late. I have been around with a couple of idioms in functional programming trying to implement it in Java. Many of them have already been tried using anonymous inner classes and the likes. Many of them work too but at the cost of high. The following is an attempt to get a clear implementation of infinite streams in Java. Infinite streams give you the illusion that it can contain infinite number of objects. The real kludge behind infinite streams is lazy evaluation introduces the term
delayed evaluation which enables us to represent very large (even infinite) sequences as streams
Functional languages like Haskell and Miranda employs laziness as the default paradigm of evaluation while languages like Scheme implement the same concepts as library functions (delay and force). Dominik Gruntz infinite streams in Java using the functor paradigm and inner classes. The obvious problem is verbosity resulting from the accidental complexity that they lend to the implementation. In this post. I attempt the same using Neal's closures prototype. So without further ado.. The Stream InterfaceHere's the contract for lazy evaluation..
class StreamTest { interface Stream<E> { E car(); Stream<E> cdr(); E get(int index); <R> Stream<R> map(Unary<? super E. R> f); Stream<E> filter(Unary<? super E. Boolean> f); } //..}
class StreamTest { interface Stream<E> { //. as above } static class LazyStream<E> implements Stream<E> { private E car; private {=>Stream<E>} cdr; // constructor public LazyStream(E car. {=>Stream<E>} cdr) { this car = car; this cdr = cdr; } // accessors public E car() { return car; } public Stream<E> cdr() { return cdr invoke(); } // access at position public E get(int index) { Stream<E> stream = this; while (index-- > 0) { stream = stream cdr(); } return stream car(); } // map over the stream public <R> Stream<R> map(Unary<? super E. R> f) { return cons(f invoke(car). {=>cdr() map(f)}); } // filter the stream public Stream<E> filter(Unary<? super E. Boolean> f) { if (car() != null) { if (f invoke(car()) == true) { return cons(car(). {=>cdr() filter(f)}); } else { return cdr() filter(f); } } return null; } // factory method cons public static <E> LazyStream<E> cons(E val. {=>Stream<E>} c) { return new LazyStream<E>(val c); } }}
class StreamTest { //. all above stuff //. and the tests // helper function generating sequence of natural numbers static LazyStream<Integer> integersFrom(final int start) { return LazyStream cons(start. {=>integersFrom(start+1)}); } // helper function for generating fibonacci sequence static LazyStream<Integer> fibonacci(final int a final int b) { return LazyStream cons(a. {=>fibonacci(b a+b)}); } public static void main(String[] args) { // natural numbers Stream<Integer> integers = integersFrom(0); Stream<Integer> s = integers; for(int i=0; i<20; i++) { System out print(s car() + " "); s = s cdr(); } System out println("..."); // a map example over the stream Stream<Integer> t = integers; Stream<Integer> u = t map(new Unary<Integer. Integer>({Integer i=>i*i})); for(int i=0; i<20; i++) { System out print(u car() + " "); u = u cdr(); } System out println("..."); // a filter over stream Stream<Integer> x = integers; Stream<Integer> y = x filter(new Unary<Integer. Boolean>({Integer i=>i%2==0})); for(int i=0; i<20; i++) { System out print(y car() + " "); y = y cdr(); } System out println("..."); }}
Closures in Java will surely bring in a new paradigm of programming within the developers. The amount of excitement that the prototype has already generated is phenomenal. It'll be too bad if they do not appear in Java 7. Update: Ricky Clarkson points out in the Comments that
And here's my 'answer': http://pastebin com/f5e4fd1abIn short because {A=>B} can be read as {? super A=>? extends B} you don't need to add it yourself. All I did was delete Unary and replace it with straight {A=>B}. Perhaps I missed something but the code compiles and runs fine. If I missed something add a test case that fails and I'll try again.
Silly me ! It just blew off me that {? super E=>? extends R} is the same as {E=>R}. Thanks for reminding me. I would have required the indirection in case I had an extends on the left hand side. I am not changing the post - just adding an Update on the changes. Thanks for the comment.
Good post debasish. But can you please help me out most of my programs in closures won't run in windowsXP?I always getC:\closures\test\tools\javac\closures>java -Xbootclasspath/p:c:/closures/lib/javac jar StreamTestException in thread "main" java lang. NoClassDefFoundError: javax/lang/function/OOC:\closures\test\tools\javac\closures>I could manage only very few closure examples to run. ThanksPrashant jalasutramhttp://prashantjalasutram blogspot com/
Prashant:What command are you using to compile? If you're not specifying the classpath on that command what value does %CLASSPATH% have?When you compile some classes are created. For me a javax/ directory appears in the same directory my class file appears in (assuming no package statement in the source). You'll need to make sure that the directory above javax/ is on the classpath. I think this is only a prototype issue and that in a release the types will be generated by the VM as needed much as array types are.
Ricky,I cannot see any folders getting created when it compiles successfully. Command i am using:C:\closures\test\tools\javac\closures>javac -J-Xbootclasspath/p:c:/closures/lib/javac jar -source 7 Demo javaand then i try to run but fail almost all the times likejava -Xbootclasspath/p:c:/closures/lib/javac jar DemoThanksPrashant
Ricky,Thanks a lot for your gr8 tip and yes it worked finally and i am very happy that i can try a lot of examples now. It worked when i added "-d." which allowed as you suggested to create a new directory and added OO class so finally my javac looks likejavac -d. -J-Xbootclasspath/p:c:/closures/lib/javac jar -source 7 * javaand running in XP does not change any thing. Debasish thanks a lot for allowing to act as mediator pattern between me and ricky to solve this :-)ThanksPrashant Jalasutram
Hi Prashant -It is good to find that ur problems have been fixed. I just now logged in and found the trail from Prashant. Thanks Ricky for all the help. Closures indeed provide great power of abstractions. I will be extremely disappointed if we miss it out in Java 7. Cheers.
情報サイトには、魅力的なお仕事をたくさん掲載しています。お仕事をお探しの皆さまにとって、より使いやすく便利なサイトにするべく、アデコ情報サイトをリニューアルしました。
「することによって絶対的な美を得られるわけではありません。『自分は変わった』という事実を物理的に確認することで、気になって仕方がなかった自分 の体に対するコンプレックスから解放される。そこではじめて心を研ぎ澄まし、自分の内面を磨いていくことができるようになるのです。そうして人は美しく なっていく。外見だけ磨こうとする人は美しくなれない、というのが私の持論です」
広島,岡山/四国(香川,徳島,愛媛,高知) -あなぶき不産ナビ四国４県、岡山の不動産、広島の不動産など不動産情報検索（マンション・一戸建て・土地・収益物件等）サイトです。穴吹不動産流通株式会社"
Forex Groups - Tips on Trading
Related article:
http://debasishg.blogspot.com/2007/11/infinite-streams-using-java-closures.html
comments | Add comment | Report as Spam
|
