Proposal: Elvis and Other Null-Safe Operators

Mark Mahieu markmahieu at googlemail.com
Sun Apr 5 04:40:55 PDT 2009


Hi Stephen,

Revisiting an email from a couple of weeks ago...

On 19 Mar 2009, at 17:18, Stephen Colebourne wrote:

> Neal Gafter wrote:
>> On Thu, Mar 19, 2009 at 1:26 AM, Jacek Kolodziejczyk
>> <jacek.p.kolodziejczyk at gmail.com> wrote:
>>
>>> In my opinion the most obtrusive part in the current state of  
>>> proposals
>>> is assumption that the invocation on null object should always  
>>> return null.
>>> If I understand it correctly the following would break at runtime,
>>> because null cannot be assigned to boolean:
>>>
>>>   List list = null;
>>>   boolean b = list?.isEmpty();
>>>
>>
>> That would be a compile-time error because of the incompatibility
>> between null and boolean.
>>
> The intention is that you can write a combination to handle this with
> the submitted Coin proposal:
>
> List list = null;
> boolean b = list?.isEmpty() ?: true;
>
> Stephen
>

I don't understand how that combination is supposed to work, given  
this part of the proposal:

"A null-safe method invocation expression e1?.name(args) ... the type  
of the result is the same as the type of e1.name(args).  It is an  
error if this is not a reference type."

Doesn't that mean that the compiler should reject the expression  
'list?.isEmpty()' ?


Mark




More information about the coin-dev mailing list