I'm giving up on this for now.  There are other weird bugs in the code,
e.g. this simple piece of code fails:

>>> x = urllib.urlopen("https://mail.google.com").read()
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/local/google/home/guido/python/py3kd/Lib/io.py", line 463,
in read
    return self.readall()
  File "/usr/local/google/home/guido/python/py3kd/Lib/io.py", line 473,
in readall
    data = self.read(DEFAULT_BUFFER_SIZE)
  File "/usr/local/google/home/guido/python/py3kd/Lib/io.py", line 466,
in read
    del b[n:]
TypeError: slice indices must be integers or None or have an __index__
method
[64813 refs]
>>>

The debugger tells me that 'n' contains a bytes object instead of an
int.  This appears due to the confused signature of _sslobj.read()
(which can behave as either read() or readinto()).  Its docstring is
wrong too.  We'll deal with this after 3.0a2 is released.