andrew cooke wrote:
Gabriel Rossetti wrote:
Ok, maybe I mis-stated my problem (or mis-understood your answers).. I
don' t want to share code as in have multiple processes access a
variable and have the same value, like it is done in threads, what I
want is to not have n copies of the code (classes, functions, etc)
loaded by n python interpreters. When you load Apache for instance, it
runs n processes but loads only one copy of parts of it's code (so/dll),
that's what I want to do. In C/C++ I would write a shared-lib (so/dll),
so once the system has loaded it, it doesn' t re-load it when another
process needs it.

i think i understand what you want.  the problem is that the kind of
language that python is really doesn't work well with that kind of
approach.  this is (and i may be wrong - i've not really thought much
about this before) largely because python places a very strong emphasis on
late binding.  that means that everything is very flexible - all kinds of
things can be changed "under the hood".  and that means that it is very
difficult to share things safely because almost any solution would end up
doing what you don't want (sharing variables, like with threads (even
though they are very deeply buried variables)) rather than what you do
want (just sharing the static part of the code to save space).

another way to see this is to see the connection with security.  this is
the flip side of your case - because things are so flexible it is very
hard to run some python code in a "sandbox" in a larger python program. there's some discussion on the dev group at the moment and it looks like
the solution is to only use function closures.  but even that doesn't work
without some patches that remove certain ways to dynamically alter the
code base.

andrew

Ok, I see, thank you. I thought this was possible since I have used *.pyd files before and from what I had read they were dynamic library files. I guess they must be written in C/C++ then. Thank you for the explanation.

Gabriel
--
http://mail.python.org/mailman/listinfo/python-list

Reply via email to