See attached files

1 answer below »
See attached files


Microsoft Word - PA1.docx Programming Assignment 1 ECS 032B FQ 2022 Due: October 16 th, 2022 Setup: Python3.6+ to develop and run your code. Provided files: PA1 / __init__.py pythonreview.py arrays.py linkedlist.py adts.py utils.py This file is very important. Make sure to read and understand the code inside. Hint: You may not need to override some of the provided functions. test.py Run this file to test your code. tests/ test_pythonreview.py test_arrays.py test_linkedlist.py test_adts.py Submission instructions: Submit all provided files through Gradescope as a zip file. Do not remove/rename files nor change any naming conventions. You should receive your grade for correctness automatically. Rubric: Correctness = 18 points TAs will double-check your assignment to ensure that you are adhering the specifications. They may overrule your score if you break the listed restrictions. Documentation = 2 points Provide comments in code and good documentation on Gradescope. A TA will grade this based on correctness. Extra Credit +0.2 points for every day you turn in all parts (Python files + Documentation) early. You must submit all parts; the latest day will be counted. (Maximum: 2 points) Section 1: Python Review (pythonreview.py) No additional modules to be imported for this section. You may use itertools (see: https://docs.python.org/3/library/itertools.html#itertools.combinations) You may use any built-in Python functionalities. def findWithSum(arr:list, val:int, n=2:int) -> dict: Given a list of integers arr and integer val, find n integers in arr that add up to the val. Return a dictionary with keys as indices of the n elements and the values as the value of the elements. Assume that each input would have exactly one solution. If no solutions exist, then return an empty dictionary. For Section 2-4: No additional modules may be imported. Do not modify the arguments or the naming of the already defined functions. You may not use built-in Python data structures. Cannot use any of the following: https://docs.python.org/3.8/tutorial/datastructures.html Section 2: Dynamic Array (arrays.py) In this section, you will implement a dynamic array using the StaticArray class This class extends the given StaticArray class (a child class of Collections). Enforce shifting property: There should not be empty slots (i.e. None) between occupied slots. This property is not enforced in StaticArray, hence some modifications must be made. Assume that None cannot be an element in Collections Follow this resizing rule: Double the size if more than 80% is occupied. Only double upon adding new items. Halve the size if less than 20% is occupied. Only halve upon removing items. class DynamicArray(StaticArray): def __init__(self, size:int, isSet = False:bool): Constructor for collection with size empty slots. To determine whether this collection will enforce the set property, use isSet argument. def __getitem__(self, index:int) -> Any: Return the value at the index of the collection. def __setitem__(self, index:int, value:Any) -> None: Set the index of the collection to the value. def __delitem__(self, index:int) -> None: Remove the element at index of the collection. def append(self, value:Any) -> None: Add value to collection at the next open slot. def extend(self, arr:Collections) -> None: Append the values of arr to the collection. def remove(self, value:Any) -> None: Remove the first instance of the value from collection. def argwhere(self, value:Any) -> StaticArray: Find the value in your collection. Returns a StaticArray with elements containing the indices where the value exists. def __eq__(self, arr:Collections) -> bool: Check whether arr is equivalent to this collection Two collections are equal only if they share all the same values, in the correct order. Their sizes do not matter. def __len__(self) -> int: Return the number of total slots. def get_size(self) -> int: Return the number of elements in this collection. def __iter__(self) -> Any: Yield the next value (Used for iterable functions) def reallocate(self, size:int) -> None: Reallocate your collection to a new collection with size. This is where you transfer over your existing content. This function should call resize. def resize(self, size:int) -> None: Do not modify this function. Resizes your data to size. This function is destructive! Remember to back-up your previous data somewhere else. Section 3: Linked Lists (linkedlist.py) In this section, you will be implementing a linked list using the Node class. You will mostly follow the same instructions as Section 2, but with a linked list implementation. As this class extends Collections, you will have to implement all those functions. Whenever you return a value from the linked list, return it as a Node. Do not return the data of the node. The purpose of returning the node instead of the data itself is mostly just for testing and making sure that you are adhering to the requirements. In practice, you should just return the data. When you store values in the linked list, you will accept as the original data type and internally wrap it and store it as Node class LinkedList(Collections): def __init__(self, isSet = False:bool, isDoubly = False:bool, isCircular = False:bool): Constructor for linked list. To determine whether this collection will enforce the set property, use isSet argument. To determine whether this linked list is a doubly or not, use the isDoubly argument. To determine whether this linked list is a circular or not, use the isCircular argument. Section 4: Stacks and Queue ADTs (adts.py) In this section, you will be implementing stacks and queues. Use your DynamicCollection and LinkedList implementations. class StackOrQueue (): def __init__(self, isQueue = False:bool, isLinked = False:bool,): This function will initialize this collection. Implement this collection is a singly linked list if isLinked = True. Otherwise, use a dynamic array. To determine whether this collection is a queue or stack, use the isQueue argument. def peek(self) -> Any: Return the first value in the collection but do not remove the value. def push(self, value: Any) -> None: Add the value into the collection. Treat this as enqueue if isQueue is used. def pop(self) -> Any: Remove and return the first value from the collection. If collection is empty, return None. Treat this as dequeue if isQueue is used.
Answered 2 days AfterOct 05, 2022

Answer To: See attached files

Ria answered on Oct 07 2022
57 Votes
SOLUTION.PDF

Answer To This Question Is Available To Download

Related Questions & Answers

More Questions »

Submit New Assignment

Copy and Paste Your Assignment Here