Backward compatibility environments (was Future division patch available)
Chris Barker
chrishbarker at home.net
Tue Jul 24 19:38:55 EDT 2001
Erik Max Francis wrote:
> Agreed. Let's try categorizing the Python environments which will have
> different ramifications for a backward-incompatible change:
>
> 1. First and easiest is code that is in active use which can be changed
> in place after an upgrade to Python 238 (what I'll use to refer to a
> version of Python which implements PEP 238 and/or requires PEP 238's
> semantics).
>
> 2. Code that is established and in long-term use which is not in active
> use but which still has resources that can be thrown at it to update it.
>
> 3. Code that is established and in longer-term use but which is not in
> a position to be fixed by the original engineers, either because it has
> been passed around repeatedly or because it is, say, at a client side
> that no longer has any contact with the original company (presumably
> because pre-238 Python has been so reliable that there was no need :-).
>
> 4. Code which may be in use and up to date but which is not trivially
> available in editable source form and/or from which compatibility
> warnings may not be easily found, such as stowed in a database or in
>
> 5. Code running in very special environments in which the ramifications
> of PEP 238 may not even make sense. What does always-float division
> mean on an architecture that does not support floats of any kind (say,
> on an architecture with no FPU and with limited space so that software
> emulation was deemed wasteful)?
You forgot:
6. Code taht is being curretnly maintained, but which needs to run on
both old and new versions of the interpreter.
-Chris
--
Christopher Barker,
Ph.D.
ChrisHBarker at home.net --- --- ---
http://members.home.net/barkerlohmann ---@@ -----@@ -----@@
------@@@ ------@@@ ------@@@
Oil Spill Modeling ------ @ ------ @ ------ @
Water Resources Engineering ------- --------- --------
Coastal and Fluvial Hydrodynamics --------------------------------------
------------------------------------------------------------------------
More information about the Python-list
mailing list