According to Groovy doc, Example 40.2, a new class for a task should be created so (simply copy that to the build.gradle):
task hello(type: GreetingTask)
class GreetingTask extends DefaultTask {
#TaskAction
def greet() {
println 'hello from GreetingTask'
}
}
But if you really try to do it, the class declaration is marked as an erroneous because no abstract methods are written. About ten of them should be here... Is it really impossible to do a task class in a cheap way as in documentation? What is the problem with documentation? Or rather with some settings on my IntelliJ?
You are using the wrong DefaultTask class. The one from Gradle is not abstract.
When I create the mentioned task in the empty Gradle/Groovy project, no errors is shown.
In the context of my project, the error IS shown. But the task runs OK! It again seems as some hallucination of IntelliJ.
Related
I am trying to serialize a object in GWT using SerializationFactory, but I am not able to get it working. Here is the sample code of my POC:
import com.google.gwt.user.client.rpc.SerializationException;
import com.google.gwt.user.client.rpc.SerializationStreamFactory;
import com.google.gwt.user.client.rpc.SerializationStreamReader;
import com.google.gwt.user.client.rpc.SerializationStreamWriter;
...........
Some code here....
.........
......
SerializationStreamFactory factory = (SerializationStreamFactory) GWT.create(MyClass.class);
SerializationStreamWriter writer = factory.createStreamWriter();
try {
writer.writeObject(new MyClass("anirudh"));
String value = writer.toString();
SerializationStreamReader reader = factory.createStreamReader(value);
MyClass myObj = (MyClass) reader.readObject();
System.out.println(myObj.getName());
} catch (SerializationException e) {
e.printStackTrace();
}
It gave me the following exception
Caused by: java.lang.RuntimeException: Deferred binding failed for 'com.anirudh..client.MyClass' (did you forget to inherit a required module?)
also in my code the class whose object I am trying to serialize implements IsSerializable
MyClass implements IsSerializable
I don't want to use GWT Auto-Bean framework because it does not fit my use case. Also I am not using GWT-RPC framework and right now I am quite adamant about using SerializationStreamFactory :D because I seriously want to know how this thing works.
Can anyone share a working example of SerializationStreamFactory or help me out pointing any mistake(s) I did.
Thanks in advance
SerializationStreamFactory factory = (SerializationStreamFactory) GWT.create(MyClass.class);
What are you expecting this line to do? GWT will attempt to find a replace-with or generate-with rule that matches this class (either when-type-assignable or when-type-is), or failing that will attempt to invoke a zero-arg constructor on MyClass, effectively new MyClass(). Is this what you are expecting?
The selected exception you've pasted suggests that MyClass may not be on the source path that GWT has been given to compile from, but the full error log will provide more information.
It looks as though you are trying to mimic the generated RPC code, where a *Async rpc interface would be implemented by code that extends from com.google.gwt.user.client.rpc.impl.RemoteServiceProxy (which implements SerializationStreamFactory). That base implementation is extended further to initialize several fields such as the com.google.gwt.user.client.rpc.impl.Serializer instance, actually responsible for serializing and deserializing object streams.
Serializers are created (by default) from the base class of com.google.gwt.user.client.rpc.impl.SerializerBase, through the rebind class com.google.gwt.user.rebind.rpc.TypeSerializerCreator. If you've build your own generator for MyClass, you should be kicking this off to get the work done as ProxyCreator already should be doing.
Remember when building your own serialization/deserialization mechanism that you need to decide which types can be marshalled within this system - if you open it to all types, then you will need to generate FieldSerializer types for all possible objects on the source path. This will greatly expand the size of your compiled code.
If your main goal is learning how this 'magic' works, dig into the generators and associated code that live in the com.google.gwt.user.rebind.rpc package. There are other libraries that leverage these ideas such as the gwt-atmosphere project (see https://github.com/Atmosphere/atmosphere to get started). Also review the generated code that GWT creates when it builds a 'tradition' RPC interface.
IntelliJ IDEA 10.5 (probably this matters).
I am new to Scala, so I started in an akward way. I created one file with two classes -- empty MainApp and another class, HelloWorld with method main.
I compiled it and executed -- IntelliJ automatically detected HelloWorld as main class. It was OK.
Then, I moved main method to MainApp, and deleted (then empty) HelloWorld class. When I tried to run it, IntelliJ sticked to HelloWorld nevertheless. So I reconfigured project and selected MainApp as main class.
I tried to run it with such result:
MainApp main method should be static
I am completely puzzled. First of all, Scala does not have static methods. Second of all, why it does not compile now, when it compiled before (with HelloWorld class). I though that only requirement is having one main method.
Thank you in advance for your help.
Please note: I know I can start a new project from scratch to avoid the problem altogether, but I would like to learn something, i.e. get to know what is going on, and fixing this project.
static methods in Java roughly correspond to singleton methods in Scala. You should have
object MainApp {
def main(args : Array[String]) = ...
}
in your code, not class MainApp.
In a Groovy script is it possible to do a conditional import statement?
if (test){
import this.package.class
} else {
import that.package.class
}
The background to this is wanting to use something on MacOS 10.5 which only has JDK1.5 so one specific class is unavailable, but I have found someone who has written a back-port for it.
There is no way to conditionally import a class, but you can achieve something similar by attempting to load the class and then load another class if that one is not found.
Here's just an example:
def someClass
try {
someClass = "org.apache.webdavlib.WebdavFile" as Class
} catch (Exception ex) {
someClass = "java.io.File" as Class
}
def someInstance = someClass.newInstance("~/project/temp.log")
assert "java.io.File" == someInstance.getClass().getName()
Jochen "blackdrag" Theodorou proposed the following on the groovy user list a while ago:
wsh = this.class.classLoader.loadClass("org.codehaus.groovy.scriptom.ActiveXObject").newInstance("WScript.Shell")
Then you do not need to use the import statement.
Here is the thread on the mailing list
No, conditional imports are not supported... Best I can think of atm would be to use reflection as you would need to in java...
An ast transform could also be used here to tag the class and wrap the code that uses the missing class with the required reflection code
I guess a class loader could do the trick, but will be complicated.
Have you considered to use a shadow class and jsut deploy different jars?
Something like
//jdk 1.5
somethingelse extends this {
}
.
//jdk 1.6
somtheingelse extends that {
}
=> compile both to two different jar files, which you deploy on one system but not the other...
not perfect, but could work
...wait: if your libraries just differ in the package name, then you don't need a shadow class. Can't you move the one or the other in the same package?
Using Eclipse (Helios), I could create a JUnit test file ClassATest.java of the source file ClassA.java by using New -> JUnit Test Case -> Class under test..., then choose all the methods of ClassA to be tested.
If later we add some more methods to ClassA, how do we easily reflect this addition in ClassATest ? (No copy/paste plz).
One solution is to use MoreUnit
With MoreUnit installed to Eclipse, one can right click onto the newly added method (and not yet unit tested), and choose "Generate Test"
Of course, if one always follows the writing-test-before-writing-method style, then this solution is not needed. However in reality sometimes you don't have a clear idea of what you would want to do, in that case you would have to code up some method, play with it, then rethink and code again until you are satisfied with the code and want to make it stable by adding unit test.
You should look into creating a JUnit test suite which will execute all tests within the classes you specify. Thus, adding new test cases is as simple as creating a new class and adding it to the #Suite.SuiteClasses list (as seen below).
Here's an example.
Example JUnit Test Suite Class:
#RunWith(Suite.class)
#Suite.SuiteClasses({
TestClassFoo.class
})
public class ExampleTestSuite {}
Example Test Case class:
public class TestClassFoo {
#Test
public void testFirstTestCase() {
// code up test case
}
}
I am writing a Scala class to inherit from a Java class, and I must override a method that takes a protected Java inner class as a parameter. The Java dependency comes as a jar without source code.
I have the exact same setup as found in https://issues.scala-lang.org/browse/SI-3120 except that I do not have the Java source code available, so scalac only knows about the Java dependency by looking at the byte code (in jar or class files).
This is basically what I'm trying to do:
// javapkg/JavaSuperClass.java
package javapkg;
public class JavaSuperClass {
protected class JavaInnerClass {
}
public void method(JavaInnerClass javaInnerclass) {
System.out.println("hello");
}
}
// scalapkg/ScalaSubClass.scala
package scalapkg
import javapkg.JavaSuperClass
class ScalaSubClass extends JavaSuperClass {
override def method(javaInnerClass: JavaSuperClass#JavaInnerClass) {
println("world")
}
}
I have Java Sun JDK Hotspot 1.6.0_24 and Scala 2.9.0.1 on Linux. This is what happens:
$ cd javapkg
$ javac JavaSuperClass.java
$ cd ../scalapkg
$ scalac -cp .. ScalaSubClass.scala
ScalaSubClass.scala:6: error: class JavaInnerClass in class JavaSuperClass cannot be accessed in javapkg.JavaSuperClass
Access to protected class JavaInnerClass not permitted because
prefix type javapkg.JavaSuperClass does not conform to
class ScalaSubClass in package scalapkg where the access take place
override def method(javaInnerclass: JavaSuperClass#JavaInnerClass) {
^
one error found
Note, if I change JavaSuperClass#JavaInnerClass to simply JavaInnerClass, I get this:
ScalaSubClass.scala:6: error: method method overrides nothing
override def method(javaInnerClass: JavaInnerClass) {
^
one error found
Note: I know this sounds very similar to the common "protected static inner class" Java-compatibility issue in Scala, but I believe this is unrelated because there are no statics anywhere in my example.
I feel like something is wrong, because when I put the same code into a mixed java/scala project in Eclipse, it seemed to compile fine (with the latter JavaInnerClass syntax); it's only when I compile the Scala code with only the Java byte code (and no Java source code) that I cannot get it to work. Am I just completely missing the correct syntax to refer to a Java inner class, is this a known defect, or should I file a compiler bug? I couldn't find anything about this exact use case in my searching.
This is an excellent article that discuss the topic.
EDIT-1
My bad, I answered to quickly. This actually may be a bug Mike, I'm trying to see if I can find a hack around. I'll let you know if I find one.
EDIT-2
I've tried different things but I can't find a way to make it work. Mike I'd suggest you to file a bug report.