Final variables without explicit type: Intersection types issue.

[Neal Gafter]

On Sun, Mar 22, 2009 at 8:15 AM, Florian Weimer <fw at deneb.enyo.de> wrote:
> * Neal Gafter:
> > I would suggest this kind of type inference only be allowed for local
> > variables.   That makes the spec and implementation fairly simple.
>
> Makes sense.
>
> Does the definition of LocalVariableTypeTable need updating?  I assume
> the intersection type would show up in it.  (I can't find its
> specification, unfortunately.  It seems this wasn't merged into the
> JVM spec.)

The spec you're looking for is
http://java.sun.com/docs/books/vmspec/2nd-edition/ClassFileFormat-final-draft.pdf

Yes, section 4.4.4 would have to add a syntax for intersection types,
and the implementation of the debugging APIs would need to be extended
to handle it.  No VM changes are required, though, as the VM just
ignores these attributes.