This issue seems to rear its ugly head on a periodic basis. Certainly I think about it on a periodic basis.
A fundamental problem with the approach a la:
for row in view:
vw2.append((row.a, row.b))
is that it is *extremely* inefficient. It hurts me to even look at it, thinking of all the accesses, assignments, copies, function call overhead, etc. that are going on underneath things (added to which is the overhead induced by the fact that it is a Python loop instead of a C loop).
Surely there must be a very fast implementation of the same thing in the underlying C++ code that likely uses memmoves and could (at some point one hopes) be made available via a view or storage method from Python?
--------------------------------------
Gary H. Merrill
Director and Principal Scientist, New Applications
Data Exploration Sciences
GlaxoSmithKline Inc.
(919) 483-8456
_____________________________________________ Metakit mailing list - [email protected] http://www.equi4.com/mailman/listinfo/metakit
