[issue4561] Optimize new io library

[Raymond Hettinger]

Raymond Hettinger <rhettinger at users.sourceforge.net> added the comment:

I'm getting caught-up with the IO changes in 3.0 and am a bit confused.
 The PEP says, "programmers who don't want to muck about in the new I/O
world can expect that the open() factory method will produce an object
backwards-compatible with old-style file objects."  So, I would have
expected that the old implementation could have remained in-place and
the resultant object registered as matching the appropriate IO ABC.  If
that had been done, the performance would be unchanged.  Does anyone
know why the entire old implementation had to be thrown-out in cases
where the API was unchanged?  Is there anything about New IO that is
fundamentally different so that the old implementation had to be tossed
in all cases?

----------
nosy: +rhettinger

_______________________________________
Python tracker <report at bugs.python.org>
<http://bugs.python.org/issue4561>
_______________________________________