Swap Meet

Skills Assessed

• Understanding and following code specifications
• Reading tests
• Creating classes with attributes and instance methods
• Importing modules
• Working with attributes that are lists of instances
• Implementing instance methods that interact with other instances and objects
• Implementing inheritance
• Overriding methods from superclasses and Object

Goal

You want to organize a swap meet! You own a lot of things. So do your friends! It would be awesome if each person could swap one of their things with another person's things.

For this event, you want each person to register online as a vendor. Also, they should be able to add a list of things as their inventory.

You envision an app where vendors can swap items between different inventories. But what would that backend logic look like?

For this project, given some features that the vendors want, create a set of classes, following the directions below. The directions will lead you to create many class definitions, their attributes and instance methods, and some other cool features. Vendors will be able to swap items based on values like quality, category, or id!

Setup and Workflow

Refer to the viewing-party README for detailed instructions on the One-Time Project Setup, Project Development Workflow, and Details About How to Run Tests.

For this project, there are tests that you must complete for Waves 01, 03, and 06.

Integration Tests

We provided integration tests for this project. The integration tests provided in this project do not cover all the test cases verified by the unit tests. While unit tests are small, fast, and should cover most of our program's behavior down to individual functions, integration tests verify that the various pieces of a program are working together correctly. For this project, the integration tests mainly verify that the functions work together when invoked correctly. We could classify them as the subset of Integration Tests called Smoke Tests, tests that check basic functionality of the application.

The integration tests for this project are in the directory tests/integration_tests, and the tests have the decorator @pytest.mark.integration_test provided by the pytest-integration package. Marking these tests as integration tests makes them run after the unit tests. The isolated nature and specficity of unit tests make them a much better tool for debugging. Hence, we want to run the unit tests before the integration tests.

For more on different types of software testing, here is a helpful resource.

The integration tests use the package pytest-integration. To insure that pytest is using the version installed in your venv and not the globally installed pytest, deactivate and reactive your virtual environment after you've installed the requirements. This step is important to make sure the integration tests run after the unit tests.

Code Coverage

Code coverage is a term used to describe how much application code is executed when a particular test suite is run. It is a good practice to check our code coverage, to understand how much of our code is exercised by tests vs how much is still untested. A test suite with a high percentage of coverage is likely to be testing more throughly and have fewer bugs. A code coverage tool can partner with our testing suite to give us a report illustrating the coverage of our tests.

Given that Ada provided all test cases in this project, we should anticipate high code coverage.

Review the code coverage exercise on how to use pytest-cov to generate a code coverage report. We will need to change the directory where the application code is located from student to swap_meet.

pytest --cov=swap_meet --cov-report html --cov-report term

Note: Code coverage is disabled for integration tests, since unit tests should cover all the code. source

Project Directions

This project is designed such that one could puzzle together how to implement this project without many directions. Being able to use tests to drive project completion is a skill that needs to be developed; programmers often take years to develop this skill competently.

When our test failures leave us confused and stuck, let's use the detailed project requirements below.

At submission time, no matter where you are, submit the project via Learn.

Wave 1

In Wave 1 we will create the Vendor class.

• There is a module (file) named vendor.py inside of the swap_meet package (folder)

• Inside this module, there is a class named Vendor

• Each Vendor will have an attribute named inventory, which is an empty list by default

• When we instantiate an instance of Vendor, we can optionally pass in a list with the keyword argument inventory

• Every instance of Vendor has an instance method named add, which takes in one item

• This method adds the item to the inventory

• This method returns the item that was added

• Similarly, every instance of Vendor has an instance method named remove, which takes in one item

• This method removes the matching item from the inventory

• This method returns the item that was removed

• If there is no matching item in the inventory, the method should return False

Wave 2

In Wave 2 we will create the Item class and the Vendor class' get_by_id method.

• There is a module (file) named item.py inside of the swap_meet package (folder)

• Inside this module, there is a class named Item

• Each Item will have an attribute named id, which is a unique integer by default

  • There are many ways to generate numbers, but generating numbers without duplicates takes some care. Happily, Python has a package called uuid that can help!
    • If we import the uuid package in item.py, with a little research we can use one of the functions uuid provides to create large unique numbers meant to be used as identifiers
    • Specifically, you'll need to choose which of the uuid package's functions to use, so be sure to consider which function will work best for creating a unique integer
    • Note that this package's functions return UUID objects, not integers as such, but UUID objects have an attribute int which allow us to access their value as an integer

• When we initialize an instance of Item, we can optionally pass in an integer with the keyword argument id to manually set the Item's id

• Each Item will have a function named get_category, which will return a string holding the name of the class

• Instances of Vendor have an instance method named get_by_id

  • This method takes one argument: an integer, representing an Item's id
  • This method returns the item with a matching id from the inventory
  • If there is no matching item in the inventory, the method should explicitly return None

Wave 3

In Wave 3 we will write a method to stringify (convert to a string) an Item using str() and write the method swap_items.

• When we stringify an instance of Item using str(), it returns "An object of type Item with id <id value>.", where <id value> is the id of the Item instance that str() was called on.
  • For example, if we had an Item instance item_a = Item(id=12345), the output of str(item_a) should be "An object of type Item with id 12345.".
  • To accomplish this, you'll want to investigate what calling str() on a class instance does and how you can override such a method. This type of overriding is known as "operator overloading", put simply, it means that the same method exhibits different behavior across instances of different classes. A simple example would be something like + which for strings means "concatenate" but for numbers, means "add", or for lists, means "combine".

The remaining tests in wave 3 imply:

• Instances of Vendor have an instance method named swap_items
  • It takes 3 arguments:
  • swap_items takes 3 arguments:
    1. an instance of another Vendor (other_vendor), representing the friend that the vendor is swapping with
    2. an instance of an Item (my_item), representing the item this Vendor instance plans to give
    3. an instance of an Item (their_item), representing the item the friend Vendor plans to give
  • The method removes my_item from this Vendor's inventory, and adds it to the friend's inventory
  • The method removes their_item from the other Vendor's inventory, and adds it to this Vendor's inventory
  • The method returns True
  • If this Vendor's inventory doesn't contain my_item or the friend's inventory doesn't contain their_item, the method returns False

Wave 4

In Wave 4 we will write one method, swap_first_item.

• Instances of Vendor have an instance method named swap_first_item
  • It takes one argument: an instance of another Vendor (other_vendor), representing the friend that the vendor is swapping with
  • This method considers the first item in the instance's inventory, and the first item in the friend's inventory
  • It removes the first item from its inventory, and adds the friend's first item
  • It removes the first item from the friend's inventory, and adds the instances first item
  • It returns True
  • If either itself or the friend have an empty inventory, the method returns False

Wave 5

In Wave 5 we will create three additional modules with three additional classes.

Our new modules should be defined as follows:

• Clothing

  • Has an attribute id that is by default a unique integer
  • Has an attribute fabric that is by default the string "Unknown"
    • This attribute describes what fabric the clothing is made from; some example values might be "Striped", "Cotton", or "Floral"
    • When we instantiate an instance of Clothing, we can optionally pass in a string with the keyword argument fabric
  • Has a function get_category that returns "Clothing"
  • Has a stringify method that returns "An object of type Clothing with id <id value>. It is made from <fabric value> fabric."
    • For example, if we had a Clothing instance with an id of 123435 and a fabric attribute that holds "Wool", its stringify method should return "An object of type Clothing with id 12345. It is made from Wool fabric."

• Decor

  • Has an attribute id that is by default a unique integer
  • Holds 2 integer attributes width and length
    • Both of these values should be 0 by default
    • When we instantiate an instance of Decor, we can optionally pass in integers with the keyword arguments width and length
  • Has a function get_category that returns "Decor"
  • Has a stringify method that returns "An object of type Decor with id <id value>. It takes up a <width value> by <length value> sized space."
    • For example, if we had a Decor instance with an id of 123435, width of 3, and length of 7, its stringify method should return "An object of type Decor with id 12345. It takes up a 3 by 7 sized space."

• Electronics

  • Has an attribute id that is by default a unique integer
  • Has an attribute type that is by default the string "Unknown"
    • This attribute describes what kind of electronic device this is. Some example values might be “Kitchen Appliance”, “Game Console”, or “Health Tracker”
    • When we initialize an instance of Electronics, we can optionally pass in a string with the keyword argument type
  • Has an function get_category that returns "Electronics"
  • Has a stringify method that returns "An object of type Electronics with id <id value>. This is a <type value> device."
    • For example, if we had an Electronics instance with an id of 123435 and type attribute of "Mobile Phone", its stringify method should return "An object of type Electronics with id 12345. This is a Mobile Phone device."

• All three new classes and the Item class have an attribute called condition, which can be optionally provided in the initializer. The default value should be 0

• All three new classes and the Item class have an instance method named condition_description, which should describe the condition in words based on the value, assuming they all range from 0 to 5.

  • These can be basic descriptions (eg. 'mint', 'heavily used') but feel free to have fun with these (e.g. 'You probably want a glove for this one...").
  • The one requirement is that all the classes share the same condition_description behavior.

Using Inheritance

Now, we may notice that these three classes hold the same types of state and have the same general behavior as Item. That makes this is a great opportunity to use inheritance! If you haven't already, go back and implement the Clothing, Decor, and Electronics classes so that they inherit from the Item class. This should eliminate repetition in your code and greatly reduce the total number of lines code in your program!

Tip: Importing Item

You'll need to refer to Item in order to declare it as a parent. To reference the Item class from these modules, try this import line:

from swap_meet.item import Item

Wave 6

In Wave 6 we will write three methods, get_by_category, get_best_by_category, and swap_best_by_category.

• Vendor objects have an instance method named get_by_category

  • This method takes one argument: a string, representing a category
  • This method returns a list of objects in the inventory with that category
  • If there are no items in the inventory that match the category argument, the method returns an empty list

• Vendors have a method named get_best_by_category, which will get the item with the best condition in a certain category

  • It takes one argument: a string that represents a category
  • This method looks through the instance's inventory for the item with the highest condition and matching category
    • It returns this item
    • If there are no items in the inventory that match the category, it returns None
    • It returns a single item even if there are duplicates (two or more of the same item with the same condition)

The remaining tests in wave 6 imply:

• Vendors have a method named swap_best_by_category, which will swap the best item of certain categories with another Vendor
  • It takes in three arguments
    • other_vendor, which represents another Vendor instance to trade with
    • my_priority, which represents a category that the Vendor wants to receive
    • their_priority, which represents a category that other_vendor wants to receive
  • The best item in my inventory that matches their_priority category is swapped with the best item in other_vendor's inventory that matches my_priority
    • It returns True
    • If the Vendor has no item that matches their_priority category, swapping does not happen, and it returns False
    • If other_vendor has no item that matches my_priority category, swapping does not happen, and it returns False

DRYing up the code

To further reduce the amount of repeated code in your project, consider how swap_best_by_category and swap_first_item might be able to make use of swap_items. Is there a way that these methods could incorporate a call to swap_items into the body of these methods?

Try it out and see if the tests still pass! If you can't get them to pass with this refactor, you can always return to the most recent working commit before you submit the project!

Optional Enhancements

Should a project be completed before submission, and there is a desire for optional enhancements, consider this idea:

• Items have age

  • Add an age attribute to all Items
  • Implement a Vendor method named swap_by_newest, using any logic that seems appropriate
  • Write unit tests for swap_by_newest

• Take a look for error handling opportunities

  • What issues could arise if we pass a string (or any object other than an integer) for the id of an Item? How could we prevent that?
  • What other opportunities for error handling do you see?

• What is our test suite missing?

  • Identify gaps or edge cases it'd be helpful to cover
  • Write tests for the cases you identify

*********************************************************************

*********************************************************************

Wave 1

In Wave 1 we will create the Vendor class.

class Vendor: def init(self, inventory=none): if none: self.inventory = [] esle inventory = inventory

def add(self, item) inventory.append return added_item

def remove(self, item): if item in inventory inventory.remove return removed_item if no item return false

Wave 2

class Item: def init (self, id): we can optionally pass in an integer with the keyword argument id to manually set the Item's id self.id = int (unique)

def get_category(self): return class name in a string

import the uuid package in item.py to create large unique numbers UUID objects have an attribute int which allow us to access their value as an integer

class Vendor: def init(self)

def get_by_id(self, item_id) #takes one argument: an integer, representing an Item's id -> ween vendor and item for this method? return item_in_inventor_with_item_id or None

Wave 3

class Item def str(self, item): returns "An object of type Item with id <id value>." For example, if we had an Item instance item_a = Item(id=12345)

class Vendor

def swap_items(self, other_vendor, my_item, their_item): -> my_item plans to give remove my_item from self.vendor append to other_vendor remove their_item from other_vendor append self.vendor[] return True

if not my_item in vendor or not their_item in other_vendor: return False

Wave 4

In Wave 4 we will write one method, swap_first_item.

class Vendor: def swap_first_item(self, other_vendor): remove.self.inventory[0] append.other_vendor_inventory[0] remove.other_vendor_inventory_item[0] append.self_vendor.inventory return True if []empty == False

Wave 5

from swap_meet.item import Item class Clothing(self, fabric): self.fabric = "Unknown" id get id from Item class

def get_category(self, fabric, id): returns clothing returns `"An object of type Clothing with id . It is made from .

from swap_meet.item import Item class Decor(self, width, length, id): get id from Item class self.width = 0 self.lenght = 0

def get_category(self, width, length, id) returns `"An object of type Decor with id . It takes up a by sized space."

from swap_meet.item import Item class Electronics(self, type): get id from Item class self.type = "Unnown"

def get_category(self, type): get id from Item class returns "An object of type Electronics with id <id value>. This is a <type value> device."

Wave 6

class Vendor: def get_by_category(self, "string_category"): return category_list if no item in inventory_list == "string_category": return []

def get_best_by_category(self, "string_category"): if category is highest_condition: return category_item else: if inventory doesn't match category_item return None if duplicates return set(category_item)

def swap_best_by_category(self, other_vendor, my_priority, their_priority): best_item.self.inventory match best_item.their_inventory <-> swap returns True if self.vendor and other_vendor no match no swap and other way around returns False

DRYing up the code

To further reduce the amount of repeated code in your project, consider how swap_best_by_category and swap_first_item might be able to make use of swap_items. Is there a way that these methods could incorporate a call to swap_items into the body of these methods?

Try it out and see if the tests still pass! If you can't get them to pass with this refactor, you can always return to the most recent working commit before you submit the project!

Optional Enhancements

Should a project be completed before submission, and there is a desire for optional enhancements, consider this idea:

• Items have age

  • Add an age attribute to all Items
  • Implement a Vendor method named swap_by_newest, using any logic that seems appropriate
  • Write unit tests for swap_by_newest

• Take a look for error handling opportunities

  • What issues could arise if we pass a string (or any object other than an integer) for the id of an Item? How could we prevent that?
  • What other opportunities for error handling do you see?

• What is our test suite missing?

  • Identify gaps or edge cases it'd be helpful to cover
  • Write tests for the cases you identify