---
## Learning Objectives
*After this lesson, you will be able to…*
- Define a class.
- Instantiate an object from a class.
- Create classes with default instance variables.
---
## Blueprints
All cars have things that make them a `Car`. Although the details might be different, every type of car has the same basics — it's off the same blueprint, with the same properties and methods.
- Property: A shape (could be hatchback or sedan).
- Property: A color (could be red, black, blue, or silver).
- Property: Seats (could be between 2 and 5).
- Method: Can drive.
- Method: Can park.
---
## Introduction: Objects and Classes
These properties and behaviors can be thought of as variables and functions.
`Car` blueprint:
- Properties (variables): `shape`, `color`, `seats`
- Methods (functions): `drive()` and `park()`
An actual car might have:
```python
# **Properties - Variables**:
- shape = "hatchback"
- color = "red", "black", "blue", or "silver"
- seats = 2
# **Methods - Functions:**
- drive()
- park()
- reverse()
```
**Discussion:** What might a blueprint for a chair look like?
---
## Discussion: Python Classes
In Python, the concept of blueprints and objects is common. A **class** is the blueprint for an **object**. Everything you declare in Python is an object, and it has a class.
Consider the `List` class — every list you make has the same basic concept.
Variables:
- Elements: What's in the list! E.g., `my_list = [element1, element2, element3]`.
Methods that all lists have:
- `my_list.pop()`, `my_list.append()`, `my.list.insert(index)`
What behaviors and properties do you think are in the `Dictionary` class? The `Set` class?
---
## Discussion: A `Dog` Class
We can make a class for anything! Let's create a `Dog` class.
The objects might be `greyhound`, `goldenRetriever`, `corgi`, etc.
Think about the `Dog` blueprint. What variables might our class have? What functions?
**Pro tip:** When functions are in a class, they are called "methods." A method is a function _within the context_ of a class object.
**Pro tip:** While objects are named in snake_case, classes are conventionally named in TitleCase.
---
## We Do: Defining Classes
Follow along! Let's create a new file, `Dog.py`.
Class definitions are similar to function definitions, but instead of `def`, we use `class`.
Let's declare a class for `Dog`:
```python
class Dog:
# We'll define the class here.
# Our dog will have two variables: name and age.
```
**Pro tip:** Files are usually named for their class, so the `Dog` class is in `Dog.py`.
---
## We Do: Adding Docstrings, Part 1
Using our `Dog` class from the previous exercise, let's add a _docstring_.
A [docstring](https://www.python.org/dev/peps/pep-0257/#what-is-a-docstring) is a comment that is placed at the top of a class (or function). When python precompiles the class or function (which happens in the background), it parses this comment and makes it available to the user as a helpfile when the class is accessed when typing the class or function, followed by a `?` question mark. In a Jupyter Notebook, this can also be accessed by placing your cursor at the end of a class or function and pressing `SHIFT + TAB`. Try it out! Enter `list?` in a python interpreter or Jupyter cell and run it. What is returned?
```python
In [1]: list?
Init signature: list(self, /, *args, **kwargs)
Docstring:
list() -> new empty list
list(iterable) -> new list initialized from iterable's items
Type: type
```
Here, we can see under the `Docstring` section are the 'instructions for use', and the `Init signature` is a list of arguments that can be passed to the function, which is automatically recognized and generated by the python compiler.
---
## We Do: Adding Docstrings, Part 2
Using our `Dog` class from the previous exercise, let's add a _docstring_.
The docstring for a function or method should summarize its behavior and document its arguments, return value(s), side effects, exceptions raised, and restrictions on when it can be called. Not all of these will be applicable for every function or method, so the programmer should take care to document the parts relevant. Optional arguments should be indicated. It should be documented whether keyword arguments are part of the interface.
The first line is a simple description of the class or function, followed by a blank line, followed by keyword arguments, if any.
Let's write a docstring for `Dog`:
```python
class Dog:
"""Creates Dog class, possible child class of Animal
Parameters
----------
name : str, default blank
Desired name of Dog
age : int, default 0
Age of Dog in years
"""
```
---
## We Do: The `__init__` Method
What first? Every class starts with an `__init__` method. It's:
* Where we define the class' variables.
* Short for "initialize."
* "Every time you make an object from this class, do what's in here."
Let's add this:
```python
class Dog:
"""Creates Dog class, possible child class of Animal
Parameters
----------
name : str, default blank
Desired name of Dog
age : int, default 0
Age of Dog in years
"""
def __init__(self, name="", age=0):
# Note the optional parameters and defaults.
self.name = name # All dogs have a name.
self.age = age # All dogs have an age.
```
*Note: `self` means "each individual object made from this class." Not every "dog" has the same name!*
*Note: in the effort of brevity, we wil be omitting the docstring from future examples!*
---
## We Do: Adding a `bark_hello()` Method
All dogs have the behavior `bark`, so let's add that. This is a regular function (method), just inside the class!
```python
class Dog:
def __init__(self, name="", age=0):
# Note the defaults.
self.name = name # All dogs have a name.
self.age = age # All dogs have an age.
# All dogs have a bark function.
"""Prints a message stating name and age to stdout"""
def bark_hello(self):
print("Woof! I am called", self.name, "; I am", self.age, "human-years old.")
```
We're done defining the class!
*Note: We have also added a docstring to our new `.bark_hello()` method. This will be omitted for future examples for brevity.*
---
## Aside: Instantiating Objects From Classes
Now we have a `Dog` template!
Each `dog` object we make from this template:
* Has a name.
* Has an age.
* Can bark.
---
## We Do: How Do We Make a `Dog` Object?
We call our class name like we call a function — passing in arguments, which go to the `init`.
Add this under your class (non-indented!):
```python
# Declare the objects.
gracie = Dog("Gracie", 8)
spitz = Dog("Spitz", 5)
buck = Dog("Buck", 3)
# Test them out!
gracie.bark_hello()
print("This dog's name is", gracie.name)
print("This dog's age is", gracie.age)
spitz.bark_hello()
buck.bark_hello()
```
Try it! Run `Dog.py` like a normal Python file: `python Dog.py`.
---
## We Do: Adding Print
`__init__` is just a method. It creates variables, but we can also add a `print` statement! This will run when we create the object.
```python
class Dog:
def __init__(self, name="", age=0):
self.name = name
self.age = age
print(name, "created.") # Run when init is finished.
def bark_hello(self):
print("Woof! I am called", self.name, "; I am", self.age, "human-years old.")
fox = Dog("Fox") # Note that "Fox created." prints — and we're using the default age.
fox.bark_hello()
```
Try it!
---
## Quick Review: Classes
A class is a blueprint for an object. Some classes are built into Python, like `List`. We can always make a `list` object.
We can make a class for anything!
```python
# Created like a function; TitleCase
class Dog:
# __init__: A method (function) that happens just once, when the object is created.
def __init__(self, name="", age=0): # What's passed in to the class is used here.
# Set variables for each.
self.name = name
self.age = age
print(name, "created.") # This will run when the __init__ method is called.
# Classes can have as many methods (functions) as you'd like.
def bark_hello(self):
print("Woof! I am called", self.name, "; I am", self.age, "human-years old.")
fox = Dog("Fox") # Creating the object calls __init__. Objects are snake_case.
print("This dog's name is", fox.name) # The object now has those variables!
fox.bark_hello() # The object now has those methods and variables!
```
---
## Discussion: What About Tea?
Let's make a `TeaCup` class.
* What variables would a cup of tea have?
* What methods?
---
## A Potential `TeaCup` Class
We could say:
Variables:
* A total `capacity`.
* A current `amount`.
Methods:
* `fill()` our cup.
* `empty()` our cup.
* `drink()` some tea from our cup.
---
## Example: A `TeaCup` Class
Here's what a `TeaCup` class definition might look like in Python:
```python
class TeaCup:
def __init__(self, capacity):
# Python executes when a new cup of tea is created.
self.capacity = capacity # Total ounces the cup holds.
self.amount = 0 # Current ounces in the cup. All cups start empty!
def fill(self):
self.amount = self.capacity
def empty(self):
self.amount = 0
def drink(self, amount_drank):
self.amount -= amount_drank
# If it's empty, it stays empty!
if (self.amount == 0):
self.amount = 0
steves_cup = TeaCup(12) # Maybe a fancy tea latte.
yis_cup = TeaCup(16) # It's a rough morning!
brandis_cup = TeaCup(2) # Just a quick sip.
```
---
## Quick Knowledge Check:
```python
class TeaCup:
def __init__(self, capacity = 8):
self.capacity = capacity
self.amount = 0
```
When will the capacity be `8`?
---
## Variables for All Class Objects
Next up: new types of class variables!
Let's revisit our `Dog` class:
```python
class Dog:
def __init__(self, name="", age=0):
self.name = name
self.age = age
print(name, "created.")
def bark_hello(self):
print("Woof! I am called", self.name, "; I am", self.age, "human-years old")
```
What if there are variables that we want across all dogs?
For example, can we count how many `dog` objects we make and track it in the class?
---
## I Do: Class vs. Instance Members
We already have **instance variables**, which are specific to each `dog` object (each has its own name!).
A **class variable** is specific to the class, regardless of the object. It's created **above** `__init__`.
```python
class Dog:
### Here, we define class variables. ###
# These are the same for ALL dogs.
total_dogs = 0
def __init__(self, name="", age=0):
### These are instance variables. ###
self.name = name
self.age = age
print(name, "created.")
def bark_hello(self):
print("Woof! I am called", self.name, "; I am", self.age, "human-years old")
print("There are", Dog.total_dogs, "dogs in this room!") # There's no "self" — we call the Dog class name!
molly = Dog("Molly", 8)
molly.bark_hello()
sheera = Dog("Sheera", 5)
sheera.bark_hello()
```
---
## I Do: Tallying Dogs
We can increment the `class` variable any time.
```python
class Dog:
total_dogs = 0
def __init__(self, name="", age=0):
self.name = name
self.age = age
Dog.total_dogs += 1 # We can increment this here!
print(name, "created:")
def bark_hello(self):
print("Woof! I am called", self.name, "; I am", self.age, "human-years old.")
print("There are", Dog.total_dogs, "dogs in this room!")
molly = Dog("Molly", 8)
molly.bark_hello()
sheera = Dog("Sheera", 5)
sheera.bark_hello()
```
---
## Partner Exercise: Create a Music Genre Class
Pair up! Create a new file, `Band.py`.
- Define a class, `Band`, with these instance variables: `"genre"`, `"band_name"`, and `"albums_released"` (defaulting to `0`).
- Give `Band` a method called `print_stats()`, which prints a string like `"The rock band Queen has 15 albums."`
- Create a class variable, `number_of_bands`, that tracks the number of bands created.
Test your code with calls like:
```python
my_band = ("Queen", 15, "rock")
```
**Bonus:** If the genre provided isn't `"pop"`, `"classical"`, or `"rock"`, print out `"This isn't a genre I know."`
---
## Partner Exercise: Create a `BankAccount` Class
Switch drivers! Create a new class (and file), `Bank.py`.
Bank accounts should:
* Be created with the `accountType` property (either `"savings"` or `"checking"`).
* Keep track of its current `balance`, which always starts at `0`.
* Have access to `deposit()` and `withdraw()` methods, which take in an integer and update `balance` accordingly.
* Have a class-level variable tracking the total amount of money in all accounts, adding or subtracting whenever `balance` changes.
**Bonus:** Start each account with an additional `overdraftFees` property that begins at `0`. If a call to `withdraw()` ends with the `balance` below `0`, then `overdraftFees` should be incremented by `20`.
---
## Knowledge Check: Select the Best Answer
Consider the following class definition for `Cat`:
```python
class Cat:
def __init__(self, name='Lucky'):
self.name = name
self.fur = short
```
How would you instantiate a `Cat` object with the `name` attribute `'Furball'`?
1. `mycat = Cat(name='Furball')`
2. `furball = Cat`
3. `mycat = Cat(self, name='Furball')`
4. `mycat = Cat.init(name='Furball')`
---
## Knowledge Check: Select All That Apply.
Which of the following statements are true about the ```self``` argument in class definitions?
- The user does not need to supply `self` when using instance methods.
- The `self` argument is a reference to the instance object.
- Any variable assigned with `self` (e.g., `self.var`) will be shared across instances of the class.
- With an instance object, `obj`, entering `obj.self.var` will provide the value for `var` for that instance.
---
## Knowledge Check: Select the Best Answer
Consider the following code:
```python
class Shape(object):
possible = ['triangle','square','circle','pentagon','polygon','rectangle']
def __init__(self, label='triangle'):
self.label = label
def is_possible(self):
if self.label in self.possible:
print('This is possible')
else:
print('This is impossible')
square = Shape(label='square')
wormhole = Shape(label='wormhole')
square.possible.append('wormhole')
```
If you were to enter `wormhole.is_possible()`, would the outcome be `"This is possible"` or `"This is impossible"`?
---
## Summary: Discussion
Let's chat! Can anyone explain:
- What a class is?
- What `__init__` does?
- What an object is?
- The point of `self`?
- The two types of variables?
---
## Summary and Q&A
Class:
* A pre-defined structure that contains attributes and behaviors grouped together.
* The blueprint.
* Defined via a method call.
* Contains an `__init__` method that takes in parameters to be assigned to an object.
* E.g., the `Dog` class; the `List` class.
Object:
* An instance of a class structure.
* The items built from the blueprint.
* E.g., the `gracie` object; the `my_list` object.
---
## Summary: Types of Variables in a Class
Instance variables:
- Contain data types declared in the class but defined in each object.
- Each `dog` has its own name and age.
- Each `my_list` has its own elements.
Class variables:
- Contain data and methods that span across all objects.
- How many `dog` objects are there in total?
- The `self` keyword lets us distinguish between variables that exist at the class level versus in each object.