En Tue, 28 Apr 2009 05:09:25 -0300, Benjamin Liu <liupingp...@gmail.com>
escribió:
I use guppy-pe to identify this issue in my program. The resources links
are all embedded in source code already.
How do you know there is actually a memory leak here?
Can you show the leak without resorting to guppy?
If I understand correctly your example, this program:
def i(**params):
print params
def test4():
params = { "a":"a_value", "b":"b_value" }
i(**params)
while True:
test4()
should consume unlimited memory - is that what you see?
On Windows, using 2.6.2, after half an hour it still consumes the same
amount of memory.
(Note that, even if there is a real bug in 2.5, it's too late in the
development cycle - 2.5 gets only security fixes now)
--
Gabriel Genellina
--
http://mail.python.org/mailman/listinfo/python-list
