This is from the SF tracker as well; for some reason the
content of test_index.py was lost and an empty file was
checked in instead.

Backed out an old patch from 2000.

- update the refcount information (late, but not a bad thing to do...)
- clarify that PyGen_New() steals a reference

for setup.py "upload" command.

Fix the hit and miss style of testing for sets and dicts.

interpolate PY_FORMAT_SIZE_T instead of #if'ing on
MS_WIN64.

otherwise, test/test_ctypes.py fails.  What about ctypes.macholib?

with title().

This was discovered by test_email failing with -Qnew.

Makes test_xdrlib pass.

itertools.tee->instance->attribute->itertools.tee and
itertools.tee->teedataobject->itertools.tee cycles, which can be found now
that itertools.tee and its teedataobject participate in GC, remain findable
and cleanable. The test won't fail when they aren't, but at least the
frequent hunt-refleaks runs would spot the rise in refleaks.

doesn't get cleaned up and thus leaks.

so it passes w/ -Qnew.

with or without -Qnew.