为什么在Eclipse中使用类型推理编译的这个java 8示例？

Question

I am reading the newly released Java 8 in Action and found there is a piece of code pasted from Chapter 5 not compiling:

    List<Integer> numbers1 = Arrays.asList(1, 2, 3);
    List<Integer> numbers2 = Arrays.asList(3, 4);
    List<int[]> pairs =
    numbers1.stream()
    .flatMap((Integer i) -> numbers2.stream()
    .map(j -> new int[]{i, j})
    )
    .collect(toList());

Eclipse says: "Type mismatch: cannot convert from List<Object> to List<int[]> "

And after comparing with what the author gave on Github, the following compiled:

    List<Integer> numbers1 = Arrays.asList(1, 2, 3);
    List<Integer> numbers2 = Arrays.asList(3, 4);
    List<int[]> pairs =
    numbers1.stream()
    .flatMap((Integer i) -> numbers2.stream()
    .map((Integer j) -> new int[]{i, j})
    )
    .collect(toList());

The only change is from "j" to "(Integer j)".

But isn't the first version completely equivalent to the second with the syntax sugar provided by Java 8? Why does Java refuse to compile it?

Thanks

BTW:

java -version
java version "1.8.0_20"
Java(TM) SE Runtime Environment (build 1.8.0_20-b26)
Java HotSpot(TM) Client VM (build 25.20-b23, mixed mode)

Answer 1

First, correcting your terminology: when you say syntax sugar , what you really are asking about is type inference , that when asked to infer a type for j in the inner lambda, that the compiler fails to come up with the right type.

Second, correcting your data: The error messages you cite are not coming from the JDK compiler; they're coming from Eclipse.

This is just an Eclipse bug. The reference compiler ( javac from Oracle JDK) handles your first example just fine.