How robust is Python ?

[Cary O'Brien]

In article <mailman.978774489.6833.python-list at python.org>,
Moshe Zadka  <moshez at zadka.site.co.il> wrote:
>On Sat, 06 Jan 2001, rturpin at my-deja.com wrote:
>
>> In theory, that's an application flaw, rather than a
>> bug in the compiler or libraries. In practice, good
>> software engineering produces C/C++ programs that
>> very rarely dereference bad pointers. It's harder
>> and more work than development with Python.
>
>I have programmed in C/C++ and I know what I'm talking about. *Of course*
>you can have long running processes in C/C++: take things like Linux,
>which run for years with no problems. I'm just saying that making a C/C++
>app which runs without problems is much harder, so I would be less inclined
>to trust C in that role. Of course, other factors may decide the problem
>for me, such as in...
>

I have to agree with Moshe here.  I was on one project writing C
communications code using a real-time kernel [1].  We had a pretty
good team, but the system got to a point where we had to stop all new
feature development for nearly a month and do NOTHING but track down
pointer/stack problems.  OK, this was in an mmu-less environment, but
the fact is that this pretty good development team generated many,
many nasty pointer/stack bugs.

The point?  If you have to write something robust, you're better off
with a pointer-less language.

-- cary

[1] pSOS if anyone cares, running on a rack full of VME CPU cards.  Fun.


>> Keep in mind that Python *is* a C program.
>
>I know that. It's one of the few C programs which are being debugged daily
>by thousands of people around the world, with a very mature code base.
>I doubt many companies can expect to achieve this level of robustness
>for their apps.
>-- 
>Moshe Zadka <sig at zadka.site.co.il>
>This is a signature anti-virus. 
>Please stop the spread of signature viruses!
>