Python 2.0

scott cotton scott at chronis.pobox.com
Tue Jun 1 18:21:34 CEST 1999


kaffe, the gnu java implementation, has some interesting GC
features which might be explored for python 2.0:

0601 12:11 bort:src/kaffe-1.0b4# kaffe --help
Unknown flag: --help
usage: kaffe [-options] class
Options are:
[...]
        -ss <size>              Maximum native stack size
        -mx <size>              Maximum heap size
        -ms <size>              Initial heap size
        -as <size>              Heap increment
        -verbosegc              Print message during garbage collection
        -noclassgc              Disable class garbage collection
[...]
        -noasyncgc *            Do not garbage collect asynchronously
        -oss <size> *           Maximum java stack size

0601 12:11 bort:src/kaffe-1.0b4#

Additionally, it's garbage collection takes place through an
interface which can be implemented any number of ways,
including using reference counting.

import gc; gc.refcounting(1)

or

gc.markandsweep(1)

or

class MyGC: pass

gc.setgc(MyGC)

seems like they could come in handy. 

memory management has no single ideal for any language, it
seems best to let the programmer manipulate it a bit.

in the land of python-2.0 anything is possible ;)

scott



On Tue, Jun 01, 1999 at 03:39:27PM +0000, Graham Matthews wrote:
| Paul Boddie (paulb at infercor.no) wrote:
| : As has been discussed before on this newsgroup/list, such resources get
| : overlooked in many a garbage collection scheme. I'm sure a perusal of various
|                 ----
| : news/list archives would provide a substantial amount of relevant literature.
| 
| That word "many" is your key Paul. Just because many a garbage collection
| scheme works that way doesn't mean all do, or all have to. It's entirely
| an implementation dependent thing.
| 
| You are creating a strawman argument that GC cannot properly free resources
| like files. But some GC do just that so what is your point?
| 
| graham
| 
| 
| -- 
|       This mountain is not visible from any inhabitable place. 
|        It has no name, and since it's discovery in 1856 by a 
|       British survey team, it has mysteriously resisted every
|     attempt to name it. The mountain is known simply and starkly
|      by it's original designation on the surveyor's map ... K2.
| 
| -- 
| http://www.python.org/mailman/listinfo/python-list




More information about the Python-list mailing list