How can I pass an integer by reference in Python?
I want to modify the value of a variable that I am passing to the function. I have read that everything in Python is pass by value, but there has to be an easy trick. For example, in Java you could pass the reference types of Integer, Long, etc.
How can I pass an integer into a function by reference?
What are the best practices?
It doesn't quite work that way in Python. Python passes references to objects. Inside your function you have an object -- You're free to mutate that object (if possible). However, integers are immutable. One workaround is to pass the integer in a container which can be mutated:
def change(x):
x[0] = 3
x = [1]
change(x)
print x
This is ugly/clumsy at best, but you're not going to do any better in Python. The reason is because in Python, assignment (=) takes whatever object is the result of the right hand side and binds it to whatever is on the left hand side *(or passes it to the appropriate function).
Understanding this, we can see why there is no way to change the value of an immutable object inside a function -- you can't change any of its attributes because it's immutable, and you can't just assign the "variable" a new value because then you're actually creating a new object (which is distinct from the old one) and giving it the name that the old object had in the local namespace.
Usually the workaround is to simply return the object that you want:
def multiply_by_2(x):
return 2*x
x = 1
x = multiply_by_2(x)
*In the first example case above, 3 actually gets passed to x.__setitem__.
Most cases where you would need to pass by reference are where you need to return more than one value back to the caller. A "best practice" is to use multiple return values, which is much easier to do in Python than in languages like Java.
Here's a simple example:
def RectToPolar(x, y):
r = (x ** 2 + y ** 2) ** 0.5
theta = math.atan2(y, x)
return r, theta # return 2 things at once
r, theta = RectToPolar(3, 4) # assign 2 things at once
Not exactly passing a value directly, but using it as if it was passed.
x = 7
def my_method():
nonlocal x
x += 1
my_method()
print(x) # 8
Caveats:
nonlocal was introduced in python 3
If the enclosing scope is the global one, use global instead of nonlocal.
Maybe it's not pythonic way, but you can do this
import ctypes
def incr(a):
a += 1
x = ctypes.c_int(1) # create c-var
incr(ctypes.ctypes.byref(x)) # passing by ref
Really, the best practice is to step back and ask whether you really need to do this. Why do you want to modify the value of a variable that you're passing in to the function?
If you need to do it for a quick hack, the quickest way is to pass a list holding the integer, and stick a [0] around every use of it, as mgilson's answer demonstrates.
If you need to do it for something more significant, write a class that has an int as an attribute, so you can just set it. Of course this forces you to come up with a good name for the class, and for the attribute—if you can't think of anything, go back and read the sentence again a few times, and then use the list.
More generally, if you're trying to port some Java idiom directly to Python, you're doing it wrong. Even when there is something directly corresponding (as with static/#staticmethod), you still don't want to use it in most Python programs just because you'd use it in Java.
Maybe slightly more self-documenting than the list-of-length-1 trick is the old empty type trick:
def inc_i(v):
v.i += 1
x = type('', (), {})()
x.i = 7
inc_i(x)
print(x.i)
A numpy single-element array is mutable and yet for most purposes, it can be evaluated as if it was a numerical python variable. Therefore, it's a more convenient by-reference number container than a single-element list.
import numpy as np
def triple_var_by_ref(x):
x[0]=x[0]*3
a=np.array([2])
triple_var_by_ref(a)
print(a+1)
output:
7
The correct answer, is to use a class and put the value inside the class, this lets you pass by reference exactly as you desire.
class Thing:
def __init__(self,a):
self.a = a
def dosomething(ref)
ref.a += 1
t = Thing(3)
dosomething(t)
print("T is now",t.a)
In Python, every value is a reference (a pointer to an object), just like non-primitives in Java. Also, like Java, Python only has pass by value. So, semantically, they are pretty much the same.
Since you mention Java in your question, I would like to see how you achieve what you want in Java. If you can show it in Java, I can show you how to do it exactly equivalently in Python.
class PassByReference:
def Change(self, var):
self.a = var
print(self.a)
s=PassByReference()
s.Change(5)
class Obj:
def __init__(self,a):
self.value = a
def sum(self, a):
self.value += a
a = Obj(1)
b = a
a.sum(1)
print(a.value, b.value)// 2 2
In Python, everything is passed by value, but if you want to modify some state, you can change the value of an integer inside a list or object that's passed to a method.
integers are immutable in python and once they are created we cannot change their value by using assignment operator to a variable we are making it to point to some other address not the previous address.
In python a function can return multiple values we can make use of it:
def swap(a,b):
return b,a
a,b=22,55
a,b=swap(a,b)
print(a,b)
To change the reference a variable is pointing to we can wrap immutable data types(int, long, float, complex, str, bytes, truple, frozenset) inside of mutable data types (bytearray, list, set, dict).
#var is an instance of dictionary type
def change(var,key,new_value):
var[key]=new_value
var =dict()
var['a']=33
change(var,'a',2625)
print(var['a'])
Are parameters passed by reference or by value? How do I pass by reference so that the code below outputs 'Changed' instead of 'Original'?
class PassByReference:
def __init__(self):
self.variable = 'Original'
self.change(self.variable)
print(self.variable)
def change(self, var):
var = 'Changed'
See also: Why can a function modify some arguments as perceived by the caller, but not others?
Arguments are passed by assignment. The rationale behind this is twofold:
the parameter passed in is actually a reference to an object (but the reference is passed by value)
some data types are mutable, but others aren't
So:
If you pass a mutable object into a method, the method gets a reference to that same object and you can mutate it to your heart's delight, but if you rebind the reference in the method, the outer scope will know nothing about it, and after you're done, the outer reference will still point at the original object.
If you pass an immutable object to a method, you still can't rebind the outer reference, and you can't even mutate the object.
To make it even more clear, let's have some examples.
List - a mutable type
Let's try to modify the list that was passed to a method:
def try_to_change_list_contents(the_list):
print('got', the_list)
the_list.append('four')
print('changed to', the_list)
outer_list = ['one', 'two', 'three']
print('before, outer_list =', outer_list)
try_to_change_list_contents(outer_list)
print('after, outer_list =', outer_list)
Output:
before, outer_list = ['one', 'two', 'three']
got ['one', 'two', 'three']
changed to ['one', 'two', 'three', 'four']
after, outer_list = ['one', 'two', 'three', 'four']
Since the parameter passed in is a reference to outer_list, not a copy of it, we can use the mutating list methods to change it and have the changes reflected in the outer scope.
Now let's see what happens when we try to change the reference that was passed in as a parameter:
def try_to_change_list_reference(the_list):
print('got', the_list)
the_list = ['and', 'we', 'can', 'not', 'lie']
print('set to', the_list)
outer_list = ['we', 'like', 'proper', 'English']
print('before, outer_list =', outer_list)
try_to_change_list_reference(outer_list)
print('after, outer_list =', outer_list)
Output:
before, outer_list = ['we', 'like', 'proper', 'English']
got ['we', 'like', 'proper', 'English']
set to ['and', 'we', 'can', 'not', 'lie']
after, outer_list = ['we', 'like', 'proper', 'English']
Since the the_list parameter was passed by value, assigning a new list to it had no effect that the code outside the method could see. The the_list was a copy of the outer_list reference, and we had the_list point to a new list, but there was no way to change where outer_list pointed.
String - an immutable type
It's immutable, so there's nothing we can do to change the contents of the string
Now, let's try to change the reference
def try_to_change_string_reference(the_string):
print('got', the_string)
the_string = 'In a kingdom by the sea'
print('set to', the_string)
outer_string = 'It was many and many a year ago'
print('before, outer_string =', outer_string)
try_to_change_string_reference(outer_string)
print('after, outer_string =', outer_string)
Output:
before, outer_string = It was many and many a year ago
got It was many and many a year ago
set to In a kingdom by the sea
after, outer_string = It was many and many a year ago
Again, since the the_string parameter was passed by value, assigning a new string to it had no effect that the code outside the method could see. The the_string was a copy of the outer_string reference, and we had the_string point to a new string, but there was no way to change where outer_string pointed.
I hope this clears things up a little.
EDIT: It's been noted that this doesn't answer the question that #David originally asked, "Is there something I can do to pass the variable by actual reference?". Let's work on that.
How do we get around this?
As #Andrea's answer shows, you could return the new value. This doesn't change the way things are passed in, but does let you get the information you want back out:
def return_a_whole_new_string(the_string):
new_string = something_to_do_with_the_old_string(the_string)
return new_string
# then you could call it like
my_string = return_a_whole_new_string(my_string)
If you really wanted to avoid using a return value, you could create a class to hold your value and pass it into the function or use an existing class, like a list:
def use_a_wrapper_to_simulate_pass_by_reference(stuff_to_change):
new_string = something_to_do_with_the_old_string(stuff_to_change[0])
stuff_to_change[0] = new_string
# then you could call it like
wrapper = [my_string]
use_a_wrapper_to_simulate_pass_by_reference(wrapper)
do_something_with(wrapper[0])
Although this seems a little cumbersome.
The problem comes from a misunderstanding of what variables are in Python. If you're used to most traditional languages, you have a mental model of what happens in the following sequence:
a = 1
a = 2
You believe that a is a memory location that stores the value 1, then is updated to store the value 2. That's not how things work in Python. Rather, a starts as a reference to an object with the value 1, then gets reassigned as a reference to an object with the value 2. Those two objects may continue to coexist even though a doesn't refer to the first one anymore; in fact they may be shared by any number of other references within the program.
When you call a function with a parameter, a new reference is created that refers to the object passed in. This is separate from the reference that was used in the function call, so there's no way to update that reference and make it refer to a new object. In your example:
def __init__(self):
self.variable = 'Original'
self.Change(self.variable)
def Change(self, var):
var = 'Changed'
self.variable is a reference to the string object 'Original'. When you call Change you create a second reference var to the object. Inside the function you reassign the reference var to a different string object 'Changed', but the reference self.variable is separate and does not change.
The only way around this is to pass a mutable object. Because both references refer to the same object, any changes to the object are reflected in both places.
def __init__(self):
self.variable = ['Original']
self.Change(self.variable)
def Change(self, var):
var[0] = 'Changed'
I found the other answers rather long and complicated, so I created this simple diagram to explain the way Python treats variables and parameters.
It is neither pass-by-value or pass-by-reference - it is call-by-object. See this, by Fredrik Lundh:
Call By Object
Here is a significant quote:
"...variables [names] are not objects; they cannot be denoted by other variables or referred to by objects."
In your example, when the Change method is called--a namespace is created for it; and var becomes a name, within that namespace, for the string object 'Original'. That object then has a name in two namespaces. Next, var = 'Changed' binds var to a new string object, and thus the method's namespace forgets about 'Original'. Finally, that namespace is forgotten, and the string 'Changed' along with it.
Think of stuff being passed by assignment instead of by reference/by value. That way, it is always clear, what is happening as long as you understand what happens during the normal assignment.
So, when passing a list to a function/method, the list is assigned to the parameter name. Appending to the list will result in the list being modified. Reassigning the list inside the function will not change the original list, since:
a = [1, 2, 3]
b = a
b.append(4)
b = ['a', 'b']
print a, b # prints [1, 2, 3, 4] ['a', 'b']
Since immutable types cannot be modified, they seem like being passed by value - passing an int into a function means assigning the int to the function's parameter. You can only ever reassign that, but it won't change the original variables value.
There are no variables in Python
The key to understanding parameter passing is to stop thinking about "variables". There are names and objects in Python and together they
appear like variables, but it is useful to always distinguish the three.
Python has names and objects.
Assignment binds a name to an object.
Passing an argument into a function also binds a name (the parameter name of the function) to an object.
That is all there is to it. Mutability is irrelevant to this question.
Example:
a = 1
This binds the name a to an object of type integer that holds the value 1.
b = x
This binds the name b to the same object that the name x is currently bound to.
Afterward, the name b has nothing to do with the name x anymore.
See sections 3.1 and 4.2 in the Python 3 language reference.
How to read the example in the question
In the code shown in the question, the statement self.Change(self.variable) binds the name var (in the scope of function Change) to the object that holds the value 'Original' and the assignment var = 'Changed' (in the body of function Change) assigns that same name again: to some other object (that happens to hold a string as well but could have been something else entirely).
How to pass by reference
So if the thing you want to change is a mutable object, there is no problem, as everything is effectively passed by reference.
If it is an immutable object (e.g. a bool, number, string), the way to go is to wrap it in a mutable object.
The quick-and-dirty solution for this is a one-element list (instead of self.variable, pass [self.variable] and in the function modify var[0]).
The more pythonic approach would be to introduce a trivial, one-attribute class. The function receives an instance of the class and manipulates the attribute.
Effbot (aka Fredrik Lundh) has described Python's variable passing style as call-by-object: http://effbot.org/zone/call-by-object.htm
Objects are allocated on the heap and pointers to them can be passed around anywhere.
When you make an assignment such as x = 1000, a dictionary entry is created that maps the string "x" in the current namespace to a pointer to the integer object containing one thousand.
When you update "x" with x = 2000, a new integer object is created and the dictionary is updated to point at the new object. The old one thousand object is unchanged (and may or may not be alive depending on whether anything else refers to the object).
When you do a new assignment such as y = x, a new dictionary entry "y" is created that points to the same object as the entry for "x".
Objects like strings and integers are immutable. This simply means that there are no methods that can change the object after it has been created. For example, once the integer object one-thousand is created, it will never change. Math is done by creating new integer objects.
Objects like lists are mutable. This means that the contents of the object can be changed by anything pointing to the object. For example, x = []; y = x; x.append(10); print y will print [10]. The empty list was created. Both "x" and "y" point to the same list. The append method mutates (updates) the list object (like adding a record to a database) and the result is visible to both "x" and "y" (just as a database update would be visible to every connection to that database).
Hope that clarifies the issue for you.
Technically, Python always uses pass by reference values. I am going to repeat my other answer to support my statement.
Python always uses pass-by-reference values. There isn't any exception. Any variable assignment means copying the reference value. No exception. Any variable is the name bound to the reference value. Always.
You can think about a reference value as the address of the target object. The address is automatically dereferenced when used. This way, working with the reference value, it seems you work directly with the target object. But there always is a reference in between, one step more to jump to the target.
Here is the example that proves that Python uses passing by reference:
If the argument was passed by value, the outer lst could not be modified. The green are the target objects (the black is the value stored inside, the red is the object type), the yellow is the memory with the reference value inside -- drawn as the arrow. The blue solid arrow is the reference value that was passed to the function (via the dashed blue arrow path). The ugly dark yellow is the internal dictionary. (It actually could be drawn also as a green ellipse. The colour and the shape only says it is internal.)
You can use the id() built-in function to learn what the reference value is (that is, the address of the target object).
In compiled languages, a variable is a memory space that is able to capture the value of the type. In Python, a variable is a name (captured internally as a string) bound to the reference variable that holds the reference value to the target object. The name of the variable is the key in the internal dictionary, the value part of that dictionary item stores the reference value to the target.
Reference values are hidden in Python. There isn't any explicit user type for storing the reference value. However, you can use a list element (or element in any other suitable container type) as the reference variable, because all containers do store the elements also as references to the target objects. In other words, elements are actually not contained inside the container -- only the references to elements are.
A simple trick I normally use is to just wrap it in a list:
def Change(self, var):
var[0] = 'Changed'
variable = ['Original']
self.Change(variable)
print variable[0]
(Yeah I know this can be inconvenient, but sometimes it is simple enough to do this.)
(edit - Blair has updated his enormously popular answer so that it is now accurate)
I think it is important to note that the current post with the most votes (by Blair Conrad), while being correct with respect to its result, is misleading and is borderline incorrect based on its definitions. While there are many languages (like C) that allow the user to either pass by reference or pass by value, Python is not one of them.
David Cournapeau's answer points to the real answer and explains why the behavior in Blair Conrad's post seems to be correct while the definitions are not.
To the extent that Python is pass by value, all languages are pass by value since some piece of data (be it a "value" or a "reference") must be sent. However, that does not mean that Python is pass by value in the sense that a C programmer would think of it.
If you want the behavior, Blair Conrad's answer is fine. But if you want to know the nuts and bolts of why Python is neither pass by value or pass by reference, read David Cournapeau's answer.
You got some really good answers here.
x = [ 2, 4, 4, 5, 5 ]
print x # 2, 4, 4, 5, 5
def go( li ) :
li = [ 5, 6, 7, 8 ] # re-assigning what li POINTS TO, does not
# change the value of the ORIGINAL variable x
go( x )
print x # 2, 4, 4, 5, 5 [ STILL! ]
raw_input( 'press any key to continue' )
Python’s pass-by-assignment scheme isn’t quite the same as C++’s reference parameters option, but it turns out to be very similar to the argument-passing model of the C language (and others) in practice:
Immutable arguments are effectively passed “by value.” Objects such as integers and strings are passed by object reference instead of by copying, but because you can’t change immutable objects in place anyhow, the effect is much like making a copy.
Mutable arguments are effectively passed “by pointer.” Objects such as lists
and dictionaries are also passed by object reference, which is similar to the way C
passes arrays as pointers—mutable objects can be changed in place in the function,
much like C arrays.
In this case the variable titled var in the method Change is assigned a reference to self.variable, and you immediately assign a string to var. It's no longer pointing to self.variable. The following code snippet shows what would happen if you modify the data structure pointed to by var and self.variable, in this case a list:
>>> class PassByReference:
... def __init__(self):
... self.variable = ['Original']
... self.change(self.variable)
... print self.variable
...
... def change(self, var):
... var.append('Changed')
...
>>> q = PassByReference()
['Original', 'Changed']
>>>
I'm sure someone else could clarify this further.
There are a lot of insights in answers here, but I think an additional point is not clearly mentioned here explicitly. Quoting from Python documentation What are the rules for local and global variables in Python?
In Python, variables that are only referenced inside a function are implicitly global. If a variable is assigned a new value anywhere within the function’s body, it’s assumed to be a local. If a variable is ever assigned a new value inside the function, the variable is implicitly local, and you need to explicitly declare it as ‘global’.
Though a bit surprising at first, a moment’s consideration explains this. On one hand, requiring global for assigned variables provides a bar against unintended side-effects. On the other hand, if global was required for all global references, you’d be using global all the time. You’d have to declare as global every reference to a built-in function or to a component of an imported module. This clutter would defeat the usefulness of the global declaration for identifying side-effects.
Even when passing a mutable object to a function this still applies. And to me it clearly explains the reason for the difference in behavior between assigning to the object and operating on the object in the function.
def test(l):
print "Received", l, id(l)
l = [0, 0, 0]
print "Changed to", l, id(l) # New local object created, breaking link to global l
l = [1, 2, 3]
print "Original", l, id(l)
test(l)
print "After", l, id(l)
gives:
Original [1, 2, 3] 4454645632
Received [1, 2, 3] 4454645632
Changed to [0, 0, 0] 4474591928
After [1, 2, 3] 4454645632
The assignment to an global variable that is not declared global therefore creates a new local object and breaks the link to the original object.
As you can state you need to have a mutable object, but let me suggest you to check over the global variables as they can help you or even solve this kind of issue!
http://docs.python.org/3/faq/programming.html#what-are-the-rules-for-local-and-global-variables-in-python
example:
>>> def x(y):
... global z
... z = y
...
>>> x
<function x at 0x00000000020E1730>
>>> y
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
NameError: name 'y' is not defined
>>> z
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
NameError: name 'z' is not defined
>>> x(2)
>>> x
<function x at 0x00000000020E1730>
>>> y
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
NameError: name 'y' is not defined
>>> z
2
Here is the simple (I hope) explanation of the concept pass by object used in Python.
Whenever you pass an object to the function, the object itself is passed (object in Python is actually what you'd call a value in other programming languages) not the reference to this object. In other words, when you call:
def change_me(list):
list = [1, 2, 3]
my_list = [0, 1]
change_me(my_list)
The actual object - [0, 1] (which would be called a value in other programming languages) is being passed. So in fact the function change_me will try to do something like:
[0, 1] = [1, 2, 3]
which obviously will not change the object passed to the function. If the function looked like this:
def change_me(list):
list.append(2)
Then the call would result in:
[0, 1].append(2)
which obviously will change the object. This answer explains it well.
Aside from all the great explanations on how this stuff works in Python, I don't see a simple suggestion for the problem. As you seem to do create objects and instances, the Pythonic way of handling instance variables and changing them is the following:
class PassByReference:
def __init__(self):
self.variable = 'Original'
self.Change()
print self.variable
def Change(self):
self.variable = 'Changed'
In instance methods, you normally refer to self to access instance attributes. It is normal to set instance attributes in __init__ and read or change them in instance methods. That is also why you pass self as the first argument to def Change.
Another solution would be to create a static method like this:
class PassByReference:
def __init__(self):
self.variable = 'Original'
self.variable = PassByReference.Change(self.variable)
print self.variable
#staticmethod
def Change(var):
var = 'Changed'
return var
I used the following method to quickly convert some Fortran code to Python. True, it's not pass by reference as the original question was posed, but it is a simple workaround in some cases.
a = 0
b = 0
c = 0
def myfunc(a, b, c):
a = 1
b = 2
c = 3
return a, b, c
a, b, c = myfunc(a, b, c)
print a, b, c
There is a little trick to pass an object by reference, even though the language doesn't make it possible. It works in Java too; it's the list with one item. ;-)
class PassByReference:
def __init__(self, name):
self.name = name
def changeRef(ref):
ref[0] = PassByReference('Michael')
obj = PassByReference('Peter')
print obj.name
p = [obj] # A pointer to obj! ;-)
changeRef(p)
print p[0].name # p->name
It's an ugly hack, but it works. ;-P
Since it seems to be nowhere mentioned an approach to simulate references as known from e.g. C++ is to use an "update" function and pass that instead of the actual variable (or rather, "name"):
def need_to_modify(update):
update(42) # set new value 42
# other code
def call_it():
value = 21
def update_value(new_value):
nonlocal value
value = new_value
need_to_modify(update_value)
print(value) # prints 42
This is mostly useful for "out-only references" or in a situation with multiple threads / processes (by making the update function thread / multiprocessing safe).
Obviously the above does not allow reading the value, only updating it.
Given the way Python handles values and references to them, the only way you can reference an arbitrary instance attribute is by name:
class PassByReferenceIsh:
def __init__(self):
self.variable = 'Original'
self.change('variable')
print self.variable
def change(self, var):
self.__dict__[var] = 'Changed'
In real code you would, of course, add error checking on the dict lookup.
Since your example happens to be object-oriented, you could make the following change to achieve a similar result:
class PassByReference:
def __init__(self):
self.variable = 'Original'
self.change('variable')
print(self.variable)
def change(self, var):
setattr(self, var, 'Changed')
# o.variable will equal 'Changed'
o = PassByReference()
assert o.variable == 'Changed'
Since dictionaries are passed by reference, you can use a dict variable to store any referenced values inside it.
# returns the result of adding numbers `a` and `b`
def AddNumbers(a, b, ref): # using a dict for reference
result = a + b
ref['multi'] = a * b # reference the multi. ref['multi'] is number
ref['msg'] = "The result: " + str(result) + " was nice!"
return result
number1 = 5
number2 = 10
ref = {} # init a dict like that so it can save all the referenced values. this is because all dictionaries are passed by reference, while strings and numbers do not.
sum = AddNumbers(number1, number2, ref)
print("sum: ", sum) # the returned value
print("multi: ", ref['multi']) # a referenced value
print("msg: ", ref['msg']) # a referenced value
You can merely use an empty class as an instance to store reference objects because internally object attributes are stored in an instance dictionary. See the example.
class RefsObj(object):
"A class which helps to create references to variables."
pass
...
# an example of usage
def change_ref_var(ref_obj):
ref_obj.val = 24
ref_obj = RefsObj()
ref_obj.val = 1
print(ref_obj.val) # or print ref_obj.val for python2
change_ref_var(ref_obj)
print(ref_obj.val)
While pass by reference is nothing that fits well into Python and should be rarely used, there are some workarounds that actually can work to get the object currently assigned to a local variable or even reassign a local variable from inside of a called function.
The basic idea is to have a function that can do that access and can be passed as object into other functions or stored in a class.
One way is to use global (for global variables) or nonlocal (for local variables in a function) in a wrapper function.
def change(wrapper):
wrapper(7)
x = 5
def setter(val):
global x
x = val
print(x)
The same idea works for reading and deleting a variable.
For just reading, there is even a shorter way of just using lambda: x which returns a callable that when called returns the current value of x. This is somewhat like "call by name" used in languages in the distant past.
Passing 3 wrappers to access a variable is a bit unwieldy so those can be wrapped into a class that has a proxy attribute:
class ByRef:
def __init__(self, r, w, d):
self._read = r
self._write = w
self._delete = d
def set(self, val):
self._write(val)
def get(self):
return self._read()
def remove(self):
self._delete()
wrapped = property(get, set, remove)
# Left as an exercise for the reader: define set, get, remove as local functions using global / nonlocal
r = ByRef(get, set, remove)
r.wrapped = 15
Pythons "reflection" support makes it possible to get a object that is capable of reassigning a name/variable in a given scope without defining functions explicitly in that scope:
class ByRef:
def __init__(self, locs, name):
self._locs = locs
self._name = name
def set(self, val):
self._locs[self._name] = val
def get(self):
return self._locs[self._name]
def remove(self):
del self._locs[self._name]
wrapped = property(get, set, remove)
def change(x):
x.wrapped = 7
def test_me():
x = 6
print(x)
change(ByRef(locals(), "x"))
print(x)
Here the ByRef class wraps a dictionary access. So attribute access to wrapped is translated to a item access in the passed dictionary. By passing the result of the builtin locals and the name of a local variable, this ends up accessing a local variable. The Python documentation as of 3.5 advises that changing the dictionary might not work, but it seems to work for me.
Pass-by-reference in Python is quite different from the concept of pass by reference in C++/Java.
Java and C#: primitive types (including string) pass by value (copy). A reference type is passed by reference (address copy), so all changes made in the parameter in the called function are visible to the caller.
C++: Both pass-by-reference or pass-by-value are allowed. If a parameter is passed by reference, you can either modify it or not depending upon whether the parameter was passed as const or not. However, const or not, the parameter maintains the reference to the object and reference cannot be assigned to point to a different object within the called function.
Python:
Python is “pass-by-object-reference”, of which it is often said: “Object references are passed by value.” (read here). Both the caller and the function refer to the same object, but the parameter in the function is a new variable which is just holding a copy of the object in the caller. Like C++, a parameter can be either modified or not in function. This depends upon the type of object passed. For example, an immutable object type cannot be modified in the called function whereas a mutable object can be either updated or re-initialized.
A crucial difference between updating or reassigning/re-initializing the mutable variable is that updated value gets reflected back in the called function whereas the reinitialized value does not. The scope of any assignment of new object to a mutable variable is local to the function in the python. Examples provided by #blair-conrad are great to understand this.
I am new to Python, started yesterday (though I have been programming for 45 years).
I came here because I was writing a function where I wanted to have two so-called out-parameters. If it would have been only one out-parameter, I wouldn't get hung up right now on checking how reference/value works in Python. I would just have used the return value of the function instead. But since I needed two such out-parameters I felt I needed to sort it out.
In this post I am going to show how I solved my situation. Perhaps others coming here can find it valuable, even though it is not exactly an answer to the topic question. Experienced Python programmers of course already know about the solution I used, but it was new to me.
From the answers here I could quickly see that Python works a bit like JavaScript in this regard, and that you need to use workarounds if you want the reference functionality.
But then I found something neat in Python that I don't think I have seen in other languages before, namely that you can return more than one value from a function, in a simple comma-separated way, like this:
def somefunction(p):
a = p + 1
b = p + 2
c = -p
return a, b, c
and that you can handle that on the calling side similarly, like this
x, y, z = somefunction(w)
That was good enough for me and I was satisfied. There isn't any need to use some workaround.
In other languages you can of course also return many values, but then usually in the from of an object, and you need to adjust the calling side accordingly.
The Python way of doing it was nice and simple.
If you want to mimic by reference even more, you could do as follows:
def somefunction(a, b, c):
a = a * 2
b = b + a
c = a * b * c
return a, b, c
x = 3
y = 5
z = 10
print(F"Before : {x}, {y}, {z}")
x, y, z = somefunction(x, y, z)
print(F"After : {x}, {y}, {z}")
which gives this result
Before : 3, 5, 10
After : 6, 11, 660
Alternatively, you could use ctypes which would look something like this:
import ctypes
def f(a):
a.value = 2398 ## Resign the value in a function
a = ctypes.c_int(0)
print("pre f", a)
f(a)
print("post f", a)
As a is a c int and not a Python integer and apparently passed by reference. However, you have to be careful as strange things could happen, and it is therefore not advised.
Use dataclasses. Also, it allows you to apply type restrictions (aka "type hints").
from dataclasses import dataclass
#dataclass
class Holder:
obj: your_type # Need any type? Use "obj: object" then.
def foo(ref: Holder):
ref.obj = do_something()
I agree with folks that in most cases you'd better consider not to use it.
And yet, when we're talking about contexts, it's worth to know that way.
You can design an explicit context class though. When prototyping, I prefer dataclasses, just because it's easy to serialize them back and forth.
There are already many great answers (or let's say opinions) about this and I've read them, but I want to mention a missing one. The one from Python's documentation in the FAQ section. I don't know the date of publishing this page, but this should be our true reference:
Remember that arguments are passed by assignment in Python. Since
assignment just creates references to objects, there’s no alias
between an argument name in the caller and callee, and so no
call-by-reference per se.
If you have:
a = SOMETHING
def fn(arg):
pass
and you call it like fn(a), you're doing exactly what you do in assignment. So this happens:
arg = a
An additional reference to SOMETHING is created. Variables are just symbols/names/references. They don't "hold" anything.
Why does the following behave unexpectedly in Python?
>>> a = 256
>>> b = 256
>>> a is b
True # This is an expected result
>>> a = 257
>>> b = 257
>>> a is b
False # What happened here? Why is this False?
>>> 257 is 257
True # Yet the literal numbers compare properly
I am using Python 2.5.2. Trying some different versions of Python, it appears that Python 2.3.3 shows the above behaviour between 99 and 100.
Based on the above, I can hypothesize that Python is internally implemented such that "small" integers are stored in a different way than larger integers and the is operator can tell the difference. Why the leaky abstraction? What is a better way of comparing two arbitrary objects to see whether they are the same when I don't know in advance whether they are numbers or not?
Take a look at this:
>>> a = 256
>>> b = 256
>>> id(a) == id(b)
True
>>> a = 257
>>> b = 257
>>> id(a) == id(b)
False
Here's what I found in the documentation for "Plain Integer Objects":
The current implementation keeps an array of integer objects for all integers between -5 and 256. When you create an int in that range you actually just get back a reference to the existing object.
So, integers 256 are identical, but 257 are not. This is a CPython implementation detail, and not guaranteed for other Python implementations.
Python's “is” operator behaves unexpectedly with integers?
In summary - let me emphasize: Do not use is to compare integers.
This isn't behavior you should have any expectations about.
Instead, use == and != to compare for equality and inequality, respectively. For example:
>>> a = 1000
>>> a == 1000 # Test integers like this,
True
>>> a != 5000 # or this!
True
>>> a is 1000 # Don't do this! - Don't use `is` to test integers!!
False
Explanation
To know this, you need to know the following.
First, what does is do? It is a comparison operator. From the documentation:
The operators is and is not test for object identity: x is y is true
if and only if x and y are the same object. x is not y yields the
inverse truth value.
And so the following are equivalent.
>>> a is b
>>> id(a) == id(b)
From the documentation:
id
Return the “identity” of an object. This is an integer (or long
integer) which is guaranteed to be unique and constant for this object
during its lifetime. Two objects with non-overlapping lifetimes may
have the same id() value.
Note that the fact that the id of an object in CPython (the reference implementation of Python) is the location in memory is an implementation detail. Other implementations of Python (such as Jython or IronPython) could easily have a different implementation for id.
So what is the use-case for is? PEP8 describes:
Comparisons to singletons like None should always be done with is or
is not, never the equality operators.
The Question
You ask, and state, the following question (with code):
Why does the following behave unexpectedly in Python?
>>> a = 256
>>> b = 256
>>> a is b
True # This is an expected result
It is not an expected result. Why is it expected? It only means that the integers valued at 256 referenced by both a and b are the same instance of integer. Integers are immutable in Python, thus they cannot change. This should have no impact on any code. It should not be expected. It is merely an implementation detail.
But perhaps we should be glad that there is not a new separate instance in memory every time we state a value equals 256.
>>> a = 257
>>> b = 257
>>> a is b
False # What happened here? Why is this False?
Looks like we now have two separate instances of integers with the value of 257 in memory. Since integers are immutable, this wastes memory. Let's hope we're not wasting a lot of it. We're probably not. But this behavior is not guaranteed.
>>> 257 is 257
True # Yet the literal numbers compare properly
Well, this looks like your particular implementation of Python is trying to be smart and not creating redundantly valued integers in memory unless it has to. You seem to indicate you are using the referent implementation of Python, which is CPython. Good for CPython.
It might be even better if CPython could do this globally, if it could do so cheaply (as there would a cost in the lookup), perhaps another implementation might.
But as for impact on code, you should not care if an integer is a particular instance of an integer. You should only care what the value of that instance is, and you would use the normal comparison operators for that, i.e. ==.
What is does
is checks that the id of two objects are the same. In CPython, the id is the location in memory, but it could be some other uniquely identifying number in another implementation. To restate this with code:
>>> a is b
is the same as
>>> id(a) == id(b)
Why would we want to use is then?
This can be a very fast check relative to say, checking if two very long strings are equal in value. But since it applies to the uniqueness of the object, we thus have limited use-cases for it. In fact, we mostly want to use it to check for None, which is a singleton (a sole instance existing in one place in memory). We might create other singletons if there is potential to conflate them, which we might check with is, but these are relatively rare. Here's an example (will work in Python 2 and 3) e.g.
SENTINEL_SINGLETON = object() # this will only be created one time.
def foo(keyword_argument=None):
if keyword_argument is None:
print('no argument given to foo')
bar()
bar(keyword_argument)
bar('baz')
def bar(keyword_argument=SENTINEL_SINGLETON):
# SENTINEL_SINGLETON tells us if we were not passed anything
# as None is a legitimate potential argument we could get.
if keyword_argument is SENTINEL_SINGLETON:
print('no argument given to bar')
else:
print('argument to bar: {0}'.format(keyword_argument))
foo()
Which prints:
no argument given to foo
no argument given to bar
argument to bar: None
argument to bar: baz
And so we see, with is and a sentinel, we are able to differentiate between when bar is called with no arguments and when it is called with None. These are the primary use-cases for is - do not use it to test for equality of integers, strings, tuples, or other things like these.
I'm late but, you want some source with your answer? I'll try and word this in an introductory manner so more folks can follow along.
A good thing about CPython is that you can actually see the source for this. I'm going to use links for the 3.5 release, but finding the corresponding 2.x ones is trivial.
In CPython, the C-API function that handles creating a new int object is PyLong_FromLong(long v). The description for this function is:
The current implementation keeps an array of integer objects for all integers between -5 and 256, when you create an int in that range you actually just get back a reference to the existing object. So it should be possible to change the value of 1. I suspect the behaviour of Python in this case is undefined. :-)
(My italics)
Don't know about you but I see this and think: Let's find that array!
If you haven't fiddled with the C code implementing CPython you should; everything is pretty organized and readable. For our case, we need to look in the Objects subdirectory of the main source code directory tree.
PyLong_FromLong deals with long objects so it shouldn't be hard to deduce that we need to peek inside longobject.c. After looking inside you might think things are chaotic; they are, but fear not, the function we're looking for is chilling at line 230 waiting for us to check it out. It's a smallish function so the main body (excluding declarations) is easily pasted here:
PyObject *
PyLong_FromLong(long ival)
{
// omitting declarations
CHECK_SMALL_INT(ival);
if (ival < 0) {
/* negate: cant write this as abs_ival = -ival since that
invokes undefined behaviour when ival is LONG_MIN */
abs_ival = 0U-(unsigned long)ival;
sign = -1;
}
else {
abs_ival = (unsigned long)ival;
}
/* Fast path for single-digit ints */
if (!(abs_ival >> PyLong_SHIFT)) {
v = _PyLong_New(1);
if (v) {
Py_SIZE(v) = sign;
v->ob_digit[0] = Py_SAFE_DOWNCAST(
abs_ival, unsigned long, digit);
}
return (PyObject*)v;
}
Now, we're no C master-code-haxxorz but we're also not dumb, we can see that CHECK_SMALL_INT(ival); peeking at us all seductively; we can understand it has something to do with this. Let's check it out:
#define CHECK_SMALL_INT(ival) \
do if (-NSMALLNEGINTS <= ival && ival < NSMALLPOSINTS) { \
return get_small_int((sdigit)ival); \
} while(0)
So it's a macro that calls function get_small_int if the value ival satisfies the condition:
if (-NSMALLNEGINTS <= ival && ival < NSMALLPOSINTS)
So what are NSMALLNEGINTS and NSMALLPOSINTS? Macros! Here they are:
#ifndef NSMALLPOSINTS
#define NSMALLPOSINTS 257
#endif
#ifndef NSMALLNEGINTS
#define NSMALLNEGINTS 5
#endif
So our condition is if (-5 <= ival && ival < 257) call get_small_int.
Next let's look at get_small_int in all its glory (well, we'll just look at its body because that's where the interesting things are):
PyObject *v;
assert(-NSMALLNEGINTS <= ival && ival < NSMALLPOSINTS);
v = (PyObject *)&small_ints[ival + NSMALLNEGINTS];
Py_INCREF(v);
Okay, declare a PyObject, assert that the previous condition holds and execute the assignment:
v = (PyObject *)&small_ints[ival + NSMALLNEGINTS];
small_ints looks a lot like that array we've been searching for, and it is! We could've just read the damn documentation and we would've know all along!:
/* Small integers are preallocated in this array so that they
can be shared.
The integers that are preallocated are those in the range
-NSMALLNEGINTS (inclusive) to NSMALLPOSINTS (not inclusive).
*/
static PyLongObject small_ints[NSMALLNEGINTS + NSMALLPOSINTS];
So yup, this is our guy. When you want to create a new int in the range [NSMALLNEGINTS, NSMALLPOSINTS) you'll just get back a reference to an already existing object that has been preallocated.
Since the reference refers to the same object, issuing id() directly or checking for identity with is on it will return exactly the same thing.
But, when are they allocated??
During initialization in _PyLong_Init Python will gladly enter in a for loop to do this for you:
for (ival = -NSMALLNEGINTS; ival < NSMALLPOSINTS; ival++, v++) {
Check out the source to read the loop body!
I hope my explanation has made you C things clearly now (pun obviously intented).
But, 257 is 257? What's up?
This is actually easier to explain, and I have attempted to do so already; it's due to the fact that Python will execute this interactive statement as a single block:
>>> 257 is 257
During complilation of this statement, CPython will see that you have two matching literals and will use the same PyLongObject representing 257. You can see this if you do the compilation yourself and examine its contents:
>>> codeObj = compile("257 is 257", "blah!", "exec")
>>> codeObj.co_consts
(257, None)
When CPython does the operation, it's now just going to load the exact same object:
>>> import dis
>>> dis.dis(codeObj)
1 0 LOAD_CONST 0 (257) # dis
3 LOAD_CONST 0 (257) # dis again
6 COMPARE_OP 8 (is)
So is will return True.
It depends on whether you're looking to see if 2 things are equal, or the same object.
is checks to see if they are the same object, not just equal. The small ints are probably pointing to the same memory location for space efficiency
In [29]: a = 3
In [30]: b = 3
In [31]: id(a)
Out[31]: 500729144
In [32]: id(b)
Out[32]: 500729144
You should use == to compare equality of arbitrary objects. You can specify the behavior with the __eq__, and __ne__ attributes.
As you can check in source file intobject.c, Python caches small integers for efficiency. Every time you create a reference to a small integer, you are referring the cached small integer, not a new object. 257 is not an small integer, so it is calculated as a different object.
It is better to use == for that purpose.
I think your hypotheses is correct. Experiment with id (identity of object):
In [1]: id(255)
Out[1]: 146349024
In [2]: id(255)
Out[2]: 146349024
In [3]: id(257)
Out[3]: 146802752
In [4]: id(257)
Out[4]: 148993740
In [5]: a=255
In [6]: b=255
In [7]: c=257
In [8]: d=257
In [9]: id(a), id(b), id(c), id(d)
Out[9]: (146349024, 146349024, 146783024, 146804020)
It appears that numbers <= 255 are treated as literals and anything above is treated differently!
There's another issue that isn't pointed out in any of the existing answers. Python is allowed to merge any two immutable values, and pre-created small int values are not the only way this can happen. A Python implementation is never guaranteed to do this, but they all do it for more than just small ints.
For one thing, there are some other pre-created values, such as the empty tuple, str, and bytes, and some short strings (in CPython 3.6, it's the 256 single-character Latin-1 strings). For example:
>>> a = ()
>>> b = ()
>>> a is b
True
But also, even non-pre-created values can be identical. Consider these examples:
>>> c = 257
>>> d = 257
>>> c is d
False
>>> e, f = 258, 258
>>> e is f
True
And this isn't limited to int values:
>>> g, h = 42.23e100, 42.23e100
>>> g is h
True
Obviously, CPython doesn't come with a pre-created float value for 42.23e100. So, what's going on here?
The CPython compiler will merge constant values of some known-immutable types like int, float, str, bytes, in the same compilation unit. For a module, the whole module is a compilation unit, but at the interactive interpreter, each statement is a separate compilation unit. Since c and d are defined in separate statements, their values aren't merged. Since e and f are defined in the same statement, their values are merged.
You can see what's going on by disassembling the bytecode. Try defining a function that does e, f = 128, 128 and then calling dis.dis on it, and you'll see that there's a single constant value (128, 128)
>>> def f(): i, j = 258, 258
>>> dis.dis(f)
1 0 LOAD_CONST 2 ((128, 128))
2 UNPACK_SEQUENCE 2
4 STORE_FAST 0 (i)
6 STORE_FAST 1 (j)
8 LOAD_CONST 0 (None)
10 RETURN_VALUE
>>> f.__code__.co_consts
(None, 128, (128, 128))
>>> id(f.__code__.co_consts[1], f.__code__.co_consts[2][0], f.__code__.co_consts[2][1])
4305296480, 4305296480, 4305296480
You may notice that the compiler has stored 128 as a constant even though it's not actually used by the bytecode, which gives you an idea of how little optimization CPython's compiler does. Which means that (non-empty) tuples actually don't end up merged:
>>> k, l = (1, 2), (1, 2)
>>> k is l
False
Put that in a function, dis it, and look at the co_consts—there's a 1 and a 2, two (1, 2) tuples that share the same 1 and 2 but are not identical, and a ((1, 2), (1, 2)) tuple that has the two distinct equal tuples.
There's one more optimization that CPython does: string interning. Unlike compiler constant folding, this isn't restricted to source code literals:
>>> m = 'abc'
>>> n = 'abc'
>>> m is n
True
On the other hand, it is limited to the str type, and to strings of internal storage kind "ascii compact", "compact", or "legacy ready", and in many cases only "ascii compact" will get interned.
At any rate, the rules for what values must be, might be, or cannot be distinct vary from implementation to implementation, and between versions of the same implementation, and maybe even between runs of the same code on the same copy of the same implementation.
It can be worth learning the rules for one specific Python for the fun of it. But it's not worth relying on them in your code. The only safe rule is:
Do not write code that assumes two equal but separately-created immutable values are identical (don't use x is y, use x == y)
Do not write code that assumes two equal but separately-created immutable values are distinct (don't use x is not y, use x != y)
Or, in other words, only use is to test for the documented singletons (like None) or that are only created in one place in the code (like the _sentinel = object() idiom).
For immutable value objects, like ints, strings or datetimes, object identity is not especially useful. It's better to think about equality. Identity is essentially an implementation detail for value objects - since they're immutable, there's no effective difference between having multiple refs to the same object or multiple objects.
is is the identity equality operator (functioning like id(a) == id(b)); it's just that two equal numbers aren't necessarily the same object. For performance reasons some small integers happen to be memoized so they will tend to be the same (this can be done since they are immutable).
PHP's === operator, on the other hand, is described as checking equality and type: x == y and type(x) == type(y) as per Paulo Freitas' comment. This will suffice for common numbers, but differ from is for classes that define __eq__ in an absurd manner:
class Unequal:
def __eq__(self, other):
return False
PHP apparently allows the same thing for "built-in" classes (which I take to mean implemented at C level, not in PHP). A slightly less absurd use might be a timer object, which has a different value every time it's used as a number. Quite why you'd want to emulate Visual Basic's Now instead of showing that it is an evaluation with time.time() I don't know.
Greg Hewgill (OP) made one clarifying comment "My goal is to compare object identity, rather than equality of value. Except for numbers, where I want to treat object identity the same as equality of value."
This would have yet another answer, as we have to categorize things as numbers or not, to select whether we compare with == or is. CPython defines the number protocol, including PyNumber_Check, but this is not accessible from Python itself.
We could try to use isinstance with all the number types we know of, but this would inevitably be incomplete. The types module contains a StringTypes list but no NumberTypes. Since Python 2.6, the built in number classes have a base class numbers.Number, but it has the same problem:
import numpy, numbers
assert not issubclass(numpy.int16,numbers.Number)
assert issubclass(int,numbers.Number)
By the way, NumPy will produce separate instances of low numbers.
I don't actually know an answer to this variant of the question. I suppose one could theoretically use ctypes to call PyNumber_Check, but even that function has been debated, and it's certainly not portable. We'll just have to be less particular about what we test for now.
In the end, this issue stems from Python not originally having a type tree with predicates like Scheme's number?, or Haskell's type class Num. is checks object identity, not value equality. PHP has a colorful history as well, where === apparently behaves as is only on objects in PHP5, but not PHP4. Such are the growing pains of moving across languages (including versions of one).
It also happens with strings:
>>> s = b = 'somestr'
>>> s == b, s is b, id(s), id(b)
(True, True, 4555519392, 4555519392)
Now everything seems fine.
>>> s = 'somestr'
>>> b = 'somestr'
>>> s == b, s is b, id(s), id(b)
(True, True, 4555519392, 4555519392)
That's expected too.
>>> s1 = b1 = 'somestrdaasd ad ad asd as dasddsg,dlfg ,;dflg, dfg a'
>>> s1 == b1, s1 is b1, id(s1), id(b1)
(True, True, 4555308080, 4555308080)
>>> s1 = 'somestrdaasd ad ad asd as dasddsg,dlfg ,;dflg, dfg a'
>>> b1 = 'somestrdaasd ad ad asd as dasddsg,dlfg ,;dflg, dfg a'
>>> s1 == b1, s1 is b1, id(s1), id(b1)
(True, False, 4555308176, 4555308272)
Now that's unexpected.
What’s New In Python 3.8: Changes in Python behavior:
The compiler now produces a SyntaxWarning when identity checks (is and
is not) are used with certain types of literals (e.g. strings, ints).
These can often work by accident in CPython, but are not guaranteed by
the language spec. The warning advises users to use equality tests (==
and !=) instead.