question on exports to

[Alan Bateman]

On 02/06/2016 09:41, Andrew Dinn wrote:

> On 01/06/16 20:07, Alex Buckley wrote:
>> Core Reflection is still bound by the Java language's notion of
>> accessibility, so calling newInstance() on a Class or get() on a Field
>> or invoke() on a Method will perform an access check in SE 9 just like
>> in SE 8. The access check has two parts, as previously discussed, but by
>> policy the first part (X-reads-Y) is a "yes" for these reflective
>> operations ... that leaves second part where the underlying
>> class/field/method must be exported to the code calling
>> newInstance()/get()/invoke(). That's why setAccessible has been
>> @CallerSensitive since the module system implementation was merged into
>> JDK 9.
> That's not quite true at present since this exports test need not be
> passed when the caller module is java.base i.e. the current permissions
> are wider than you state.
>
> Alan Bateman explained to me that that this was merely a heuristic to
> enable the current Jigsaw implementation to work. If/when that heuristic
> is removed then is the intention that the condition for reflective
> access will reduce simply to the exports check or will there still be a
> need for some slightly wider condition? If the latter then how will the
> condition be widened?
There is special handling for java.base in java.base but that is not 
specified and lots of things would need to happen before that can go 
away. So I think a separate topic .

-Alan