On Wed, Feb 07, 2007 at 11:57:10PM +0900, David Cournapeau wrote:
tasks. For once, the current implementation relies
heavily on dynamic
memory allocation, which makes it hard to control what happens when
memory wise. Whether this is inherent to the language itself, or its
implementation, I don't know, thi is beyond my knowledge.
Well, the language itself is highly dynamic, you can rebind pretty
much anything to any type of object at any time, so the runtime has to
be pretty flexible.
Could be intereseting to ask to the pypy's folks
about the possibility
of having a real time capability to python (pypy is an implementation
of python in python, with the goal to have a flexible implementation,
this making changin many key points of python "easy".
It's not just flexibility, they also want to make things much more
optimization-friendly. It's definitely an important project in the
python world, but frankly a lot of what they are doing is way beyond
me :-)
I suspect if "realtime python" is even possible, there would have to
be some restrictions to get things realtime safe... which would still
probably be pretty useful :-)
Maybe something like the "restricted python" they're using to
implement a lot of pypy?
http://codespeak.net/pypy/dist/pypy/doc/coding-guide.html#restricted-python
--
Paul Winkler
http://www.slinkp.com