[resteasy-dev] Lambda question

[Ron Sigal]

When I run

>     @Test
>     public void test2()
>     {
>        MapFunction<String, Integer> fn = str -> Integer.valueOf(str);
>        Class<?>[] typeArgs = 
> TypeResolver.resolveRawArguments(MapFunction.class, fn.getClass()); // 
> Using Jonathan Halterman's TypeResolver
>        System.out.println("typeArgs.length: " + typeArgs.length);
>        for (Class<?> clazz : typeArgs)
>        {
>           System.out.println(clazz);
>        }
>     }
with openjdk 8, I get

> typeArgs.length: 2
> class java.lang.String
> class java.lang.Integer
but when I run it with IBM jdk 8, I get

> typeArgs.length: 2
> class org.jboss.resteasy.test.contextresolver.TypeResolver$Unknown
> class org.jboss.resteasy.test.contextresolver.TypeResolver$Unknown

I see that IBM jdk 8 has sun.reflect.ConstantPool, but that doesn't seem 
to be enough.

I think we should revisit this issue when JDK 9 is released. It's 
probably too late to make a request, right?

-Ron


On 04/07/2017 02:54 AM, Rostislav Svoboda wrote:
> https://github.com/jhalterman/typetools#on-lambda-support
>
> Lambda type argument resolution is currently supported for:
>      Oracle JDK 8
>      Open JDK 8
>
> IBM JDK 8 will need some attention too.
>
> Rostislav
>
> ----- Original Message -----
>> Hi Steven,
>>
>> That's really interesting. It works for both Oracle JDK 1.8.0_91 and JDK
>> 9 build 163 from java.net. !!
>>
>> One problem I see is that Jonathan Halterman's TypeResolver class has
>> the line
>>
>>> String constantPoolName = JAVA_VERSION < 9 ?
>>> "sun.reflect.ConstantPool" : "jdk.internal.reflect.ConstantPool";
>> so I don't think we could use it for any JDK < 9. Once JDK 9 stabilizes
>> and Resteasy is releasing JDK 9 code, maybe it would be OK.
>>
>> Even better would be if Class.getGenericInterfaces() were to work in JDK
>> 9, but, as of build 163, it doesn't.
>>
>> In any case, you've helped to verify that the solution requires external
>> tricks. Let's hope that your "right approach" is followed.
>>
>> Thank you very much,
>>
>> Ron
>>
>> On 04/03/2017 01:49 PM, Steven Schlansker wrote:
>>>> On Apr 3, 2017, at 10:11 AM, Ron Sigal <rsigal at redhat.com> wrote:
>>>> ...
>>>>
>>>> So, it seems that, for lambdas, Java doesn't keep the implementation type
>>>> and value of the type variable at run time. Does anyone have any ideas or
>>>> tricks?
>>> I did not have time to play with it myself, but I did run across this:
>>> https://stackoverflow.com/questions/23863716/java-how-to-resolve-generic-type-of-lambda-parameter
>>> maybe that helps?
>>>
>>> It sounds like the "right approach" would be to request Java stash the type
>>> information better for lambdas...
>>>
>> --
>> My company's smarter than your company (unless you work for Red Hat)
>>
>> _______________________________________________
>> resteasy-dev mailing list
>> resteasy-dev at lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/resteasy-dev
>>

-- 
My company's smarter than your company (unless you work for Red Hat)