Python different memory management behaviours when run in Python shell and *.py file

Question

Example 1:

➜  /tmp  cat t.py 
a = 250000000000
b = 250000000000
print id(a), id(b), id(a) == id(b)
➜  /tmp  python t.py 
140450848587992 140450848587992 True  #(Why is True?)

Example 2:

➜  /tmp  python
Python 2.7.10 (default, Oct 23 2015, 19:19:21) 
[GCC 4.2.1 Compatible Apple LLVM 7.0.0 (clang-700.0.59.5)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> a = 250000000000
>>> b = 250000000000
>>> print id(a), id(b), id(a) == id(b)
140443481339400 140443481339208 False #(I think it should be False)

I know Python has a small integer cache pool(from -5 to 256), so two big integers should have different id all the time.

How to explain the different behaviours of big integer when run in Python shell and *.py file ?

ForceBru · Accepted Answer · 2016-04-14 15:52:42Z

3

If I were the Python interpreter reading the .py file (as in your first example), I'd allocate memory for this number only once and then make a and b point to this location.

For example, the interpreter reads this file and sees that two variables are assigned the same value and thinks: "Do I want to waste memory by allocating two portions for the same value? No, I don't. Instead, I'd better allocate one single chunk and store this value in it". So, there's only one copy of this value and that's why these variables have identical id's.

In the second case Python has to allocate memory when you do the assignment, so there are two pieces of memory with the same data.

To sum up, in the first case the interpreter knows much more about the code (the whole code is given), so it can optimize it (and generate .pyo files) while in the second one it simply can't do any optimization.

edited Apr 14, 2016 at 15:52

answered Apr 14, 2016 at 10:52

ForceBru

45k10 gold badges72 silver badges104 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

bruno desthuilliers · Accepted Answer · 2016-04-14 11:57:09Z

2

Actually the CPython runtime does indeed "interns" (cache) some immutable objects (ints up to a certain value, strings that could be legal python identifiers etc) as an optimisation. All of this behaviour is totally implementation dependant and should not in any case be relied upon, as you just noticed.

answered Apr 14, 2016 at 11:57

bruno desthuilliers

78.3k6 gold badges103 silver badges129 bronze badges

1 Comment

virusdefender Over a year ago

Thanks, now i know this behaviour is totally implementation dependant !

Collectives™ on Stack Overflow

Python different memory management behaviours when run in Python shell and *.py file

2 Answers 2

Comments

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related