[C++-sig] Re: Why Python for C++ programmers
dave at boost-consulting.com
Tue Jan 7 23:05:14 CET 2003
Daryl P McDaniel <python at mc2research.org> writes:
> The Python / C++ relationship is not adversarial; it is
> complementary. I think that this is a very important point to get
> across to a C++ audience.
Yes, it was something I wanted to convey, but I thought I was failing.
That's why I posted the question. I've long since solved those
problems in my paper, but I had to rewrite it three times to get there
> One can easily prototype in Python then convert to C++. (There are
> significant benefits to prototyping in something other than your production
> language.) By coupling Python with application specific extension libraries,
> in C++, one can produce an application with very good performance while
> retaining the benefits (as expressed in the original messages) of Python.
> For us, the greatest benefit comes when Python is embedded in our product
> (Low-level management firmware for Mainframes). Through the facilities
> provided by Boost.Python, all "important" C++ objects are available in the
> Python environment. This allows one to easily interract with the system, and
> the system objects, for experimentation, development, and debugging. Where
> appropriate, portions of the final application that are not time critical can
> be written in Python. The breadth of Python library support usually results
> in small, easily maintained, modules.
All good points.
> As one who has been programming in C++ daily for about 20 years, my
> initial reaction (in a seminar environment) to the phrase "C++ is
> hard, Python is easy" would be fairly negative. C++ is harder than
> Python to use correctly and fully. That which you know is always
> easier than that which you don't.
I daresay I know C++ better than I know Python. I still think C++ is
harder. Maybe I'm just being too optimistic about Python? <wink>
dave at boost-consulting.com * http://www.boost-consulting.com
Boost support, enhancements, training, and commercial distribution
More information about the Cplusplus-sig