bug in java::lang::Class::getDeclaredMethods

Tue Dec 21 17:18:00 GMT 1999

Per> Sure enough, Class::getDeclaredMethod is setting the offset to
Per> the method's offset in the *result* array, which makes no sense.
Per> Instead, the offset should be the offset of the corresponding
Per> _Jv_Method in the declaringClass's `methods' array:
I've incorporated this into my working tree.
I hope to commit it in the next couple days.
Per> The problem is it looks like the fix causes gij to crash *earlier*
Per> than it did without the fix! Either my fix is wrong, or there is
Per> some *other* bug which depends on this bug in getDeclaredMethods!
Per> Any ideas?
Offhand, no.
I'm actively working on reflection right now, so I might be looking at
this soon. Right now I'm working on Method.invoke, which involves
getting Class.getMethod and friends working.
Tom