Home Original page

Python vs. C++ Builder - speed of development

Andy Freeman anamax at earthlink.net
Sat Feb 1 22:20:21 EST 2003 
"Brandon Van Every" <vanevery at 3DProgrammer.com> wrote in message news:<gID_9.887$ek4.89394 at newsread2.prod.itd.earthlink.net>...
> Lisp sounds like a bad fit to my current problems.  I'm dealing with simple
> geometric relationships, not complicated ones.

Then what is taking you so long?

> If I had a lot of very
> complicated, freeform geometic relationships to deal with, Lisp would be
> more attractive.

Or, if you had a lot of simple relationships that were slightly different
from one another, both Python and Lisp[1] would let you exploit that similarity
without being forced to keep things as C++ simple as possible.  (I previously
commented on the difference between "as simple as possible" and "as simple
as appropriate".)

Martelli described one way to turn Python implementations into C++.  I've
been using another.

I annotated my Python classes with some meta-information.  I then wrote
a class walker that uses the meta-information to generate equivalent C++.
I wanted the C++ to be as self-checking as the Python and C++ is more
verbose, so said meta-information is significantly larger than the "working"
Python.  (The code walker, on the other hand, is very small.)  However,
the meta-information and the working python in total are much smaller
than the generated C++.  And, since I'm already writing Python faster than
I wrote C++, I'm ahead of the game.

FWIW, my application has lots of simple geometric relationships....

-andy


[1] I think that lisp is a better applications language than Python.
However, Python is close enough, or at least so much better than the
alternatives, that Python's social and glue language advantages are
often decisive.
More information about the Python-list mailing list