cPickle cannot be imported.  This was necessary because my last mass
checkin broke cPickle and I don't feel like debugging it right now;
but it seems a good idea in general not to require cPickle when
pickle.py is also there.  A few unrelated fixes for issues while
debigging various test failures.

setup.py: disable building of cPickle until I've fixed it

Objects/...

  genobject.c: disallow raising string exceptions

Lib/...

  Cookie.py: fix doctest not to fail if cPickle is missing
  ctypes/macholib/dyld.py: fix relative imports
  sqlite3/__init__.py: fix relative import
  xml/dom/__init__.py: fix relative import

Lib/test/...

  regrtest.py: reduce list of skipped items on darwin

  test_generators.py: don't test string exceptions; test throw() errors
  test_traceback.py: don't test string exceptions

  pickletester.py: don't fail if cPickle is missing
  test_datetime.py: don't fail if cPickle is missing
  test_descr.py: don't fail if cPickle is missing (still some other failures)
  test_exceptions.py: don't fail if cPickle is missing
  test_re.py: don't fail if cPickle is missing

  test_array.py: use pickle, not cPickle
  test_bool.py: don't fail if cPickle is missing
  test_deque.py: use pickle, not cPickle
  test_logging.py: use pickle, not cPickle

number of tests, all because of the codecs/_multibytecodecs issue described
here (it's not a Py3K issue, just something Py3K discovers):
http://mail.python.org/pipermail/python-dev/2006-April/064051.html

Hye-Shik Chang promised to look for a fix, so no need to fix it here. The
tests that are expected to break are:

test_codecencodings_cn
test_codecencodings_hk
test_codecencodings_jp
test_codecencodings_kr
test_codecencodings_tw
test_codecs
test_multibytecodec

This merge fixes an actual test failure (test_weakref) in this branch,
though, so I believe merging is the right thing to do anyway.