Python - Variable Scope

---

A variable in Python is a symbols name to the object in computer's memory. Python works on the concept of namespaces to define the context for various identifiers such as functions, variables etc. A namespace is a collection of symbolic names defined in the current context.

Python provides the following types of namespaces −

• Built-in namespace contains built-in functions and built-in exceptions. They are loaded in the memory as soon as Python interpreter is loaded and remain till the interpreter is running.

• Global namespace contains any names defined in the main program. These names remain in memory till the program is running.

• Local namespace contains names defined inside a function. They are available till the function is running.

These namespaces are nested one inside the other. Following diagram shows relationship between namespaces.

The life of a certain variable is restricted to the namespace in which it is defined. As a result, it is not possible to access a variable present in the inner namespace from any outer namespace.

globals() Function

Python's standard library includes a built-in function globals(). It returns a dictionary of symbols currently available in global namespace.

Run the globals() function directly from the Python prompt.

>>> globals()
{'__name__': '__main__', '__doc__': None, '__package__': None, '__loader__': <class '_frozen_importlib.BuiltinImporter'>, '__spec__': None, '__annotations__': {}, '__builtins__': <module 'builtins' (built-in)>}

It can be seen that the builtins module which contains definitions of all built-in functions and built-in exceptions is loaded.

Save the following code that contains few variables and a function with few more variables inside it.

name = 'TutorialsPoint'
marks = 50
result = True
def myfunction():
   a = 10
   b = 20
   return a+b
   
print (globals())

Calling globals() from inside this script returns following dictionary object −

{'__name__': '__main__', '__doc__': None, '__package__': None, '__loader__': <_frozen_importlib_external.SourceFileLoader object at 0x00000263E7255250>, '__spec__': None, '__annotations__': {}, '__builtins__': <module 'builtins' (built-in)>, '__file__': 'C:\\Users\\user\\examples\\main.py', '__cached__': None, 'name': 'TutorialsPoint', 'marks': 50, 'result': True, 'myfunction': <function myfunction at 0x00000263E72004A0>}

The global namespace now contains variables in the program and their values and the function object in it (and not the variables in the function).

locals() Function

Python's standard library includes a built-in function locals(). It returns a dictionary of symbols currently available in namespace of the function.

Modify the above script to print dictionary of global and local namespaces from within the function.

name = 'TutorialsPoint'
marks = 50
result = True
def myfunction():
   a = 10
   b = 20
   c = a+b
   print ("globals():", globals())
   print ("locals():", locals())
   return c
myfunction()

The output shows that locals() returns a dictionary of variables and their values currently available in the function.

globals(): {'__name__': '__main__', '__doc__': None, '__package__': None, '__loader__': <_frozen_importlib_external.SourceFileLoader object at 0x00000169AE265250>, '__spec__': None, '__annotations__': {}, '__builtins__': <module 'builtins' (built-in)>, '__file__': 'C:\\Users\\mlath\\examples\\main.py', '__cached__': None, 'name': 'TutorialsPoint', 'marks': 50, 'result': True, 'myfunction': <function myfunction at 0x00000169AE2104A0>}
locals(): {'a': 10, 'b': 20, 'c': 30}

Since both globals() and locals functions return dictionary, you can access value of a variable from respective namespace with dictionary get() method or index operator.

print (globals()['name']) #displays TutorialsPoint
   print (locals().get('a')) #displays 10

Namespace Conflict

If a variable of same name is present in global as well as local scope, Python interpreter gives priority to the one in local namespace.

marks = 50 # this is a global variable
def myfunction():
   marks = 70 # this is a local variable
   print (marks)
   
myfunction()
print (marks) # prints global value

It will produce the following output −

70
50

If you try to manipulate value of a global variable from inside a function, Python raises UnboundLocalError.

marks = 50 # this is a global variable
def myfunction():
   marks = marks + 20
   print (marks)

myfunction()
print (marks) # prints global value

It will produce the following output −

   marks = marks + 20
           ^^^^^
UnboundLocalError: cannot access local variable 'marks' where it is not associated with a value

To modify a global variable, you can either update it with a dictionary syntax, or use the global keyword to refer it before modifying.

var1 = 50 # this is a global variable
var2 = 60 # this is a global variable
def myfunction():
   "Change values of global variables"
   globals()['var1'] = globals()['var1']+10
   global var2
   var2 = var2 + 20

myfunction()
print ("var1:",var1, "var2:",var2) #shows global variables with changed values

It will produce the following output −

var1: 60 var2: 80

Lastly, if you try to access a local variable in global scope, Python raises NameError as the variable in local scope can't be accessed outside it.

var1 = 50 # this is a global variable
var2 = 60 # this is a global variable
def myfunction(x, y):
   total = x+y
   print ("Total is a local variable: ", total)

myfunction(var1, var2)
print (total) # This gives NameError

It will produce the following output −

Total is a local variable: 110
Traceback (most recent call last):
   File "C:\Users\user\examples\main.py", line 9, in <module>
   print (total) # This gives NameError
          ^^^^^
NameError: name 'total' is not defined