martin.v.loewis wrote:
Author: martin.v.loewis Date: Tue Aug 22 23:41:27 2006 New Revision: 51492
Modified: python/branches/int_unification/Include/boolobject.h python/branches/int_unification/Include/intobject.h python/branches/int_unification/Include/longobject.h python/branches/int_unification/Modules/_sre.c python/branches/int_unification/Objects/abstract.c python/branches/int_unification/Objects/boolobject.c python/branches/int_unification/Objects/exceptions.c python/branches/int_unification/Objects/intobject.c python/branches/int_unification/Objects/listobject.c python/branches/int_unification/Objects/longobject.c python/branches/int_unification/Python/bltinmodule.c python/branches/int_unification/Python/marshal.c python/branches/int_unification/Python/pythonrun.c Log: Drop the int type.
Wouldn't it be better to add the long implementation on top of the int implementation rather than replacing int with long ? Integers are by far the most used objects in Python and thus need to be as fast as possible. OTOH, longs are not used a lot in everyday Python programming. Furthermore, ints are fixed sized objects whereas longs are variable sized objects. As stated above, you rarely need the extended range that longs have to offer, but you still have to pay for it in every single int object you have in the Python process.
From a performance perspective I think it would be fair to put the burden on the long-type usage of integers and keep the int-type usage as optimized as it currently is.
-- Marc-Andre Lemburg eGenix.com Professional Python Services directly from the Source (#1, Aug 23 2006)
Python/Zope Consulting and Support ... http://www.egenix.com/ mxODBC.Zope.Database.Adapter ... http://zope.egenix.com/ mxODBC, mxDateTime, mxTextTools ... http://python.egenix.com/
::: Try mxODBC.Zope.DA for Windows,Linux,Solaris,FreeBSD for free ! ::::