Hi Blake, Also, I presume that a Closure program has immediate access to all Java > classes and methods without any work. Is that true?
Clojure embraces the JVM by providing a number of interoperability features <http://clojure-doc.org/articles/language/interop.html> that allow you to do almost anything (if not anything) you can do in Java. You can even create Java classes or interfaces with the gen-class <http://clojuredocs.org/clojure.core/gen-class> and gen-interface <http://clojuredocs.org/clojure.core/gen-interface> macros. Among these features there is a special *.* (dot) operator for accessing field or method members. Even though the result of using reads almost like a Clojure function call, it's actually a special interoperability construct. The split <http://docs.oracle.com/javase/8/docs/api/java/lang/String.html#split-java.lang.String-> method of a Java String would be called like this *(.split "Hello World!" " ")* for the single argument signature and like this *(.split "Hello World!" " " 1)* for the one with two arguments. It recently occurred to me that the author of Closure probably took that > problem into account when creating Closure. So, Closure would be able to > call Java methods as if they were Closure functions taking into account the > argument signature. Does Closure do that? The use of a method like indexOf <http://docs.oracle.com/javase/8/docs/api/java/lang/String.html#indexOf-int-> that accepts a single argument of either type* String* or* int*, is resolved automatically in Clojure. If the type of the argument doesn't match an existing method, an exception is thrown. *(.indexOf "Hello World!" "Wo")* *;;= 6* *(.indexOf "Hello World!" (int \!))* *;;= 11* *(.indexOf "Hello World!" 2.0)* *;; IllegalArgumentException No matching method found: indexOf for class java.lang.String * *;; clojure.lang.Reflector.invokeMatchingMethod (Reflector.java:80) * The *.instanceMember* syntax is special though, it is not treated as a regular Clojure function and is resolved at compile time. This means for example that you can't use partial <http://clojuredocs.org/clojure.core/partial> with instance methods. If you wanted to create a Clojure function equivalent to calling the *split *method of any arity you would have to wrap it up in a function with multiple arities. *(defn split [s]* * ([regex]* * (.split s regex))* * ([regex limit]* * (.split s regex limit)))* Some Clojure libraries are actually wrappers around Java libraries, but they provide a more idiomatic API for the Clojure programmer. Seesaw for example "*Seesaw turns the Horror of Swing into a friendly, well-documented, Clojure library*". I hope I understood your questions correctly and I was able to express myself clearly. Cheers! Juan On Friday, November 7, 2014 11:32:00 PM UTC-3, Blake McBride wrote: > > Greetings, > > I am an old lisp guy that is new to Clojure. At one time I wanted to use > ABCL as a scripting language for a large, web-based Java app. (ABCL is > Common Lisp written in for the JVM.) The integration and operation of it > was easy and nice. I did, however, run into an insurmountable problem. > > I created some clue code that, through reflection of a class, discovered > Java methods and created a CLOS mirror. Worked well. The problem I ran > into is that Java allows more than one method to have the same name > provided it has a different argument signature. Lisp can't do that without > a real lot of work (make generic functions evaluate the arguments and then > execute the applicable method). It appeared as a big job to me to have > Lisp automatically generate all that code for each Java method. I > eventually kind of gave up. > > It recently occurred to me that the author of Closure probably took that > problem into account when creating Closure. So, Closure would be able to > call Java methods as if they were Closure functions taking into account the > argument signature. Does Closure do that? > > Also, I presume that a Closure program has immediate access to all Java > classes and methods without any work. Is that true? > > Thanks. > > Blake McBride > > -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en --- You received this message because you are subscribed to the Google Groups "Clojure" group. To unsubscribe from this group and stop receiving emails from it, send an email to clojure+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
