Private Variables in Python

Last Updated : 16 Jul, 2024

Prerequisite: Underscore in Python
In Python, there is no existence of “Private” instance variables that cannot be accessed except inside an object. However, a convention is being followed by most Python code and coders i.e., a name prefixed with an underscore, For e.g. _geek should be treated as a non-public part of the API or any Python code, whether it is a function, a method, or a data member. While going through this we would also try to understand the concept of various forms of trailing underscores, for e.g., for _ in range(10), __init__(self).

Mangling and how it works

In Python, there is something called name mangling, which means that there is limited support for a valid use-case for class-private members basically to avoid name clashes of names with names defined by subclasses. Any identifier of the form __geek (at least two leading underscores or at most one trailing underscore) is replaced with _classname__geek, where classname is the current class name with a leading underscore(s) stripped. As long as it occurs within the definition of the class, this mangling is done. This is helpful for letting subclasses override methods without breaking intraclass method calls.
Let’s look at this example and try to find out how this underscore works:

Python

# Python code to illustrate how mangling works class Map:     def __init__(self, iterate):         self.list = []         self.__geek(iterate)     def geek(self, iterate):         for item in iterate:             self.list.append(item)      # private copy of original geek() method     __geek = geek     class MapSubclass(Map):          # provides new signature for geek() but     # does not break __init__()     def geek(self, key, value):                 for i in zip(keys, value):             self.list.append(i)

The mangling rules are designed mostly to avoid accidents but it is still possible to access or modify a variable that is considered private. This can even be useful in special circumstances, such as in the debugger.

_Single Leading Underscores

So basically one underline at the beginning of a method, function, or data member means you shouldn’t access this method because it’s not part of the API. Let’s look at this snippet of code:

Python

# Python code to illustrate  # how single underscore works def _get_errors(self):     if self._errors is None:         self.full_clean()     return self._errors  errors = property(_get_errors)

The snippet is taken from the Django source code (django/forms/forms.py). This suggests that errors are property, and it’s also a part of the API, but the method, _get_errors, is “private”, so one shouldn’t access it.

__Double Leading Underscores

Two underlines, in the beginning, cause a lot of confusion. This is about syntax rather than a convention. double underscore will mangle the attribute names of a class to avoid conflicts of attribute names between classes. For example:

Python

# Python code to illustrate how double  # underscore at the beginning works class Geek:     def _single_method(self):         pass     def __double_method(self): # for mangling         pass class Pyth(Geek):     def __double_method(self): # for mangling         pass

__Double leading and Double trailing underscores__

There’s another case of double leading and trailing underscores. We follow this while using special variables or methods (called “magic method”) such as__len__, __init__. These methods provide special syntactic features to the names. For example, __file__ indicates the location of the Python file, __eq__ is executed when a == b expression is executed.

Example:

Python

# Python code to illustrate double leading and  # double trailing underscore works class Geek:      # '__init__' for initializing, this is a     # special method       def __init__(self, ab):          self.ab = ab      # custom special method. try not to use it     def __custom__(self):          pass