JEP 186: Collection Literals

[Tom Hawtin]

On 23/01/2014 20:53, Zhong Yu wrote:
> On Thu, Jan 23, 2014 at 12:11 PM, Tom Hawtin <tom.hawtin at oracle.com> wrote:
>> [These are my own opinions.]

> `Dog dog = [blah, blah];`? Such code is not a good way to communicate
> ideas between programmers.

Assuming Dog is a standard concrete class

     Dog dog = {blah, blah};

means

     Dog dog = new Dog(blah, blah);

If Dog is abstract *and* has a well known standard general 
implementation, it'll be one of those. I don't think the latter adds 
anything useful.

Any reasonable "Dog" is not a type appropriate to be considered a 
collection. Even so, assuming we have `Dog(Blah... blahs)` and 
`Dog[Blah... blahs]`, then

     Dog dog = [blah, blah];

means

     Dog dog = new Dog(blah, blah);

(Only without the ambiguity if the constructors clash.)

> Using words to express meanings is better than using symbols - if it
> doesn't become too verbose. List.of(x,y,z) means a list of x, y, z,
> simple and clear. The wordiness of Java programs makes them easier to
> understand.

It's wordiness that is the problem. "of" does not bring any extra 
meaning. We know it's a List because of the target type. You didn't go 
for `List.<Thing>of(x, y, z)`, where Thing is perhaps less locally 
obvious than List.

> I think we should almost *never* declare a weaker type on
> non-publicized variables, like
>      List<Thing> things = new ArrayList<>();
> instead, we should almost always declare the most specific type, like
>      ArrayList<Thing> things = new ArrayList<>();

There's nothing here that would stop you carrying on doing that with the 
less duplicative syntax.

     ArrayList<Thing> things = new(); // or {};

( I don't think that's an unreasonable opinion for java.util 
collections. The interfaces are wooly and implementations pick and 
choose which methods work (there's one 1.2 collection that has just the 
one working mutation method). Even then, many of the bundled collection 
implementations do not conform to the "interfaces" they "implement". 
However, since before 1.0 the Java way, to a greater or lesser extent, 
has been to declare an interface and assign the obvious implementation 
to it (usually with more appropriate types). )

Tom