Python 2.0

Paul Boddie paulb at infercor.no
Thu Jun 3 11:25:22 CEST 1999


Graham Matthews wrote:

[...]

> But I am not sure I see what the connection is between how JPython does
> memory management and garbage collection in Python in general.

[...]

> Donn Cave (donn at u.washington.edu) wrote:
> : Well, I think there's something to be said for it.  With Python, we're
> : free from the obnoxious requirement imposed by the storage model of
> : languages like C, that forces the programmer to account for each scrap
> : of malloc'd memory lest there be leaks.
> : ...
> : Why should I enjoy this freedom with memory resources, but welcome exactly
> : the same constraints when an object has other termination requirements?
> 
> Because other kinds of objects are different to memory? Also I am not
> sure why you think GC can't handle other kinds of objects as well. Maybe
> Java doesn't do a good job of resource finalisation, but that doesn't
> mean it can't be done.

So, we have established that it can indeed be done, but reaching back into the
depths of time and pulling out the original proposal - being that Python should
have its primary (or only) implementation running in the Java environment - we
can see that regardless of the existence of satisfactory GC schemes, the typical
Java implementation surely falls short of what many people require.

Therefore, whilst there may well be a need for more sophisticated GC in Python
2.0, we must conclude that the Java environment may not be the appropriate place
in which such technologies are to be found.

Paul




More information about the Python-list mailing list