c(++)python?

[Quinn Dunkan]

On Sat, 27 Jan 2001 02:20:54 -0500, Tim Peters <tim.one at home.com> wrote:
>Given the number of odd platforms Python is ported to, C++ would still be a
>major gamble.  If history repeats, around 2010 we'll insist that people
>finally use a 1999-flavor ANSI C compiler (although there's some doubt that
>vendors will actually implement all of C99).

As someone who uses python on a platform with no C++ compiler, and which is
unlikely to ever have a C++ compiler, I'm quite glad python is in plain old C.
Consider that anyone who wants any level of backward compatibility for their
OS is going to have to port or write a C compiler (and some minimal posix
emulation).  Writing a C compiler is not trivial, but it's presumably well
understood by now, and a lot easier than writing one for a complex,
higher-level-but-must-be-performant language.  Writing a C++ compiler is a lot
more work than most people working on new platforms are willing to go to for
mere backward compatibility :)

>still-if-you-reimplement-python-in-c++-please-post-the-patch-
>    to-sourceforge<wink>-ly y'rs  - tim

Not that I'm arguing against python/c++ interoperability, anything that can
improve that is great.  I don't currently and may never write in java, but I'm
glad jpython exists, because someday I might have to and then I'll be glad to
have python around :)