JDK support for VM to read classes from modules in a module library

[Rémi Forax]

On 05/09/2012 06:52 PM, Mandy Chung wrote:
> Hi Paul,
>
> On 5/9/2012 8:44 AM, Paul Sandoz wrote:
>> Hi Mandy,
>>
>> Regarding the change to com/sun/jmx/mbeanserver/MXBeanMapping.java:
>>
>>   172             return Class.forName(className, false, 
>> MXBeanMapping.class.getClassLoader());
>>
>> Do you think it would make sense to add the following method to Class:
>>
>>    public static Class<?>  forName(String className, boolean 
>> initialize) {
>>      return forName0(className, initialize, 
>> ClassLoader.getCallerClassLoader());
>>    }
>>
>> there by removing the explicit reference to the class loader in 
>> MXBeanMapping.java. It intuitively feels like a best practice to 
>> reduce the cases of explicitly dealing with a class loader.
>
> That's a good observation and this new convenient method may be useful 
> (and less to type).   One thing I'm not sure about is how common 
> application code calls Class.forName(String, boolean, ClassLoader) 
> with the "initialize" parameter be false.  i.e. the case when we only 
> want a class to be loaded but not to call its static initializer.  I 
> think it'd be useful to find out how existing code uses Class.forName 
> as additional data to determine if it's worth adding this new 
> convenient method.  BTW, I have been talking with Jon to find out the 
> usage of "null" class loader from corpus.  I can also gather 
> Class.forName usage and send it to you.

In fact, this method is really useful, most of the tools that does 
bytecode transformation at runtime
(most O/R mappers, dynamic language runtimes, profilers, code coverage 
tools, etc)
are now required to compute the stack frame attribute which requires to 
compute the lower upper bound
of a set of types. For that, being able to crawle the type hierachy 
without loading the static initializer
is a must have.

>
>>
>>
>> The change to java/lang/management/PlatformComponent.java from:
>>
>>   366                        Class.forName(mxbeanInterfaceName, 
>> false, null);
>>
>> to:
>>
>>   367                        Class.forName(mxbeanInterfaceName);
>>
>> changes the initialization [*] flag from false to true. Is that 
>> intended?
>
> mxbeanInterfaceName is an interface which does not have <clinit>.   
> This change doesn't cause any
> difference.

interfaces can have a <clinit> if you initialize a static final field 
that store something which is not
a primitive type or a String, e.g. an ArrayList.

>
>>
>> Paul.
>>
>> [*] i must admit to never quite understanding what this flag means.
>> Java doc states that for initialize parameter "whether the class must 
>> be initialized". Is that a constraint or an action? I presume the later.
>
> Yes, the later - that indicates whether the class static initializer 
> will be invoked or not.
>
> Mandy

Rémi