代写代考 Copyright © 2018 by . All Rights Reserved.

Copyright © 2018 by . All Rights Reserved.
Algebraic Data Types

Learning Outcomes

Copyright By PowCoder代写 加微信 powcoder

by the End of the Lecture, Students that Complete
the Recommended Exercises should be Able to:
Describe Algebraic Data Types
Use Typeclass Derivation
Use Option Types, Maybe, Just, and Nothing Define a Recursive Data Type for a Implementing a Tree
Copyright © 2018 by . All Rights Reserved.

Reading and Suggested Exercises
Copyright © 2018 by . All Rights Reserved.
Read the following Chapter(s):
Chapter 8: “Algebraic data types intro” “Type parameters” “Derived instances”

Algebraic Data Types
Lists are Special Instances of the More General Algebraic Data Type
as noted previously, there are Two Types of List: Empty Lists and Cons Pairs
Copyright © 2018 by . All Rights Reserved.

Algebraic Data Types
this entails that If there was No List Type, then a Special Instance of a Algebraic Data Type for storing a List of Integers could be defined:
data ListInts = EmptyList | (ConsPair Integer ListInts) How could the List [1, 2, 3] be Stored
Using this Custom Type? Copyright © 2018 by . All Rights Reserved.

Algebraic Data Types
remember that “:” is an Infix Binary Operator used to Associate a Head Element with a Tail List for the Recursive Definition of List
= (1 : [2, 3])
= (1: (2 : [3]))
= (1 : (2 : (3 : [])))
Copyright © 2018 by . All Rights Reserved.

Algebraic Data Types
data ListInts = EmptyList | (ConsPair Integer ListInts)
Using this Special Type, (1 : (2 : (3 : []))) could be Written as:
(ConsPair 1 (ConsPair 2 (ConsPair 3 EmptyList)))
Copyright © 2018 by . All Rights Reserved.

Algebraic Data Types
data ListInts = EmptyList | (ConsPair Integer ListInts)
How would the sum and concat Functions Implemented Without the Tail Call Optimization) be Redefined for ListInts?
Copyright © 2018 by . All Rights Reserved.

Algebraic Data Types
data ListInts = EmptyList | (ConsPair Integer ListInts)
How would you Redefine the ListInts Type to be Generic (i.e., not only for integers)?
What Changes would be Required to Allow sum and concat to work with the Generic List Type?
Copyright © 2018 by . All Rights Reserved.

data ListInts
= EmptyList | ConsPair
sumList :: ListInts -> sumList EmptyList = 0 sumList (ConsPair h t)
concatList :: ListInts concatList EmptyList x concatList (ConsPair h
Int ListInts Int
= h + sumList t
-> ListInts -> ListInts = x
ConsPair h (concatList t x)
Copyright © 2018 by . All Rights Reserved.
Algebraic Data Types

Algebraic Data Types
data List a
= EmptyList | ConsPair a (List a)
sumList :: List a -> a
sumList EmptyList = 0
sumList (ConsPair h t) = h + sumList t
concatList :: List a -> List a -> List a concatList EmptyList x = x
concatList (ConsPair h t) x =
ConsPair h (concatList t x)
Copyright © 2018 by . All Rights Reserved.

Algebraic Data Types Simple Applications for Algebraic Data Types
include Enumerated or Categorical Types data Adjective
= Excellent | Good | Satisfactory | Poor
data Letter
= A|B|C|D|F
autoComment autoComment autoComment autoComment autoComment
:: Letter -> Adjective A = Excellent
C = Satisfactory
Copyright © 2018 by . All Rights Reserved.

Algebraic Data Types
data BooleanValue = MyFalse | MyTrue
With this Algebraic Data Type for Boolean Values, How (using Pattern Matching) could you Implement the Conjunction and Disjunction Operators?
logicalAND :: BooleanValue -> BooleanValue -> BooleanValue
logicalIOR :: BooleanValue -> BooleanValue -> BooleanValue
Copyright © 2018 by . All Rights Reserved.

Algebraic Data Types
data BooleanValue = MyFalse | MyTrue
logicalAND MyFalse _ = MyFalse logicalAND _ MyFalse = MyFalse logicalAND _ _ = MyTrue
logicalIOR MyTrue MyTrue = MyTrue logicalIOR MyTrue _ = MyTrue logicalIOR _ MyTrue = MyTrue logicalIOR _ _ = MyFalse
Can the Same Functionality be Accomplished using Fewer Patterns?
Copyright © 2018 by . All Rights Reserved.

Algebraic Data Types
What was the Nature of the Problem Encountered during Testing?
Copyright © 2018 by . All Rights Reserved.

Typeclass Derivation
showBooleanValue :: BooleanValue -> String showBooleanValue MyTrue = “true” showBooleanValue MyFalse = “false”
fortunately there is a Simpler Approach for Ensuring that a Result can be “Shown” … using Derivation
Copyright © 2018 by . All Rights Reserved.

Typeclass Derivation
data BooleanValue = MyFalse | MyTrue deriving (Show) logicalAND MyTrue MyTrue = MyTrue
logicalAND _ _ = MyFalse
logicalIOR MyFalse MyFalse = MyFalse logicalIOR _ _ = MyTrue
now What Happens during logicalAND MyFalse MyTrue ?
Copyright © 2018 by . All Rights Reserved.

Typeclass Derivation
the deriving Keyword allows an Algebraic Data Type to become an Instances of a Typeclass
this Grants “Special” Functionality Automatically Typeclasses like Show, Eq, Ord are for Types that can
Naturally be Shown, Equated, and Ordered, respectively
Copyright © 2018 by . All Rights Reserved.

Algebraic Data Types
Each of the List Types has its own Constructor
the Empty List is Associated with Constructor “[]” and the Cons Pair Variant (that Combines a Head and a Tail) is Associated with the Constructor “:”
How many Values are Required by Each of these Constructors?
Copyright © 2018 by . All Rights Reserved.

Algebraic Data Types suppose you were providing Functionality for
Computing Properties of Circles and Rectangles?
conceptually Circles and Rectangles would both be Instances of a Shape Class, so What Algebraic Data Type could you Define?
What are the Constructors and What does Each Constructor Require?
Copyright © 2018 by . All Rights Reserved.

Algebraic Data Types
a Function for Creating a Rectangle would Require a Width and a Height, while a Function for Creating a Circle would only Require a Radius
data Shape =
Rectangle Float Float | Circle Float
Copyright © 2018 by . All Rights Reserved.

Algebraic Data Types
data Shape =
Rectangle Float Float | Circle Float
What is the Type Definition for a Function that Computes the Area of a Circle (but Not a Rectangle) ?
How would you Implement this Function? Copyright © 2018 by . All Rights Reserved.

Algebraic Data Types
data Shape =
Rectangle Float Float | Circle Float
area :: Shape -> Float area (Circle r) = pi * r * r
Why would area :: Shape -> Float be Incorrect? Why would area (Shape r) = pi * r * r be Incorrect?
How would you Add the Functionality for Rectangular Areas and Circular and Rectangular Perimeters?
Copyright © 2018 by . All Rights Reserved.

Algebraic Data Types
recalling that the :t and :type Commands in GHCi will Provide the Type Definition, What is the Response from GHCi to Each of the Following?
:t area :t perimeter :t Circle :t Rectangle
Copyright © 2018 by . All Rights Reserved.

Option Types
when Multiplying Integers (i.e., the Argument
Type is Integer), what is the Type of the Product? when Multiplying Floats (i.e., the Argument
Type is Float), what is the Type of the Product? when Dividing Integers (i.e., the Argument
Type is Integer), what is the Type of the Quotient? when Dividing Floats (i.e., the Argument
Type is Float), what is the Type of the Quotient? Copyright © 2018 by . All Rights Reserved.

there are Some Situations (i.e., Expressions) where the Result Type is Not Definite
e.g., an Operator that Produces a Number in Certain Cases but Something Else in Other Cases
Copyright © 2018 by . All Rights Reserved.
Option Types

with the Understanding that Division (applied to floating point numbers) can Result in Either a Float or a Special Value in cases where the Divisor is 0 (usually called “Undefined”)…
…How would you Define a Custom Type to Handle Possible Quotients?
Copyright © 2018 by . All Rights Reserved.
Option Types

data QuotientType =
UndefinedType | FloatType Float
floatDivision :: Float -> Float -> QuotientType floatDivision _ 0 = UndefinedType floatDivision x y = FloatType (x / y)
showQuotientType :: QuotientType -> String showQuotientType UndefinedType = “Don’t Divide by Zero!” showQuotientType (FloatType z) = show z
Copyright © 2018 by . All Rights Reserved.
Option Types

data QuotientType =
UndefinedType | FloatType Float deriving (Show)
floatDivision :: Float -> Float -> QuotientType floatDivision _ 0 = UndefinedType floatDivision x y = FloatType (x / y)
Copyright © 2018 by . All Rights Reserved.
Option Types

in fact it is Not Necessary to Develop Custom Algebraic Data Types for every Operator that exhibits this kind of Behaviour…
…for these cases, use Maybe, Just, and Nothing
Copyright © 2018 by . All Rights Reserved.
Option Types

Maybe, Just, and Nothing the Maybe Data Type is an Enumerated Type
data Maybe a = Just a | Nothing deriving (Eq, Ord)
this allows you to potentially Work With a Value that Might Not Exist or Might Not Be Available
if the Value Exists and is Available, it is “Wrapped” in the just Constructor, and Otherwise it is nothing
Copyright © 2018 by . All Rights Reserved.

Maybe, Just, and Nothing
Using Maybe, How would you Implement a Function that can Safely Return the 2nd Element of a List?
n.b., in this context, safely means only making the attempt if you know the element exists
Copyright © 2018 by . All Rights Reserved.

Copyright © 2018 by . All Rights Reserved.
Maybe, Just, and Nothing
import Data.Maybe safelyGet2nd :: [a] -> (Maybe a) safelyGet2nd [] = Nothing
safelyGet2nd (h:t)
| length t == 0 = Nothing
| otherwise = Just (head t)

Maybe, Just, and Nothing an Alternative Approach is to Use Monads
safelyGet2nd’ x = case x of
(h:(a:b)) -> Just a
_ -> Nothing
Monads are used to Represent Sequences of Computations (allowing for things like User Input that would Violate Referential Transparency)
n.b., due to the complications that can arise, monads are beyond the scope of this course
Copyright © 2018 by . All Rights Reserved.

Recursive Data Structures
Algebraic Data Types can also be Used to Define Other Data Types Recursively
What are the Possible Variants of a Binary Tree of Integers (i.e., Every Internal node contains an Integer and has No More than Two Child nodes)?
n.b., you must decide on value constructor labels and the arguments necessary for each
Copyright © 2018 by . All Rights Reserved.

Recursive Data Structures
How would you Encode this Tree? Copyright © 2018 by . All Rights Reserved.

data NTree
= NilT | Node Int
(Node 1 NilT NilT)
(Node 2 NilT NilT)
Recursive Data Structures
Copyright © 2018 by . All Rights Reserved.

Recursive Data Structures
Using a Style similar to that used for Recursive List Processing, How would Write a Function to Compute the Sum of the Elements in such a Tree?
Copyright © 2018 by . All Rights Reserved.

Recursive Data Structures
data NTree = NilT | Node Int NTree NTree
sumTree :: NTree -> Int
sumTree NilT = 0
sumTree (Node n lt rt) = n + (sumTree lt) + (sumTree rt)
How would you Trace the Following?
sumTree (Node 7 (Node 4 (Node 1 NilT NilT) NilT) (Node 2 NilT NilT))
Copyright © 2018 by . All Rights Reserved.

Recursive Data Structures
What about the Trees Used to Represent Arithmetic Expressions?
data Expr =
Lit Int | Add Expr Expr | Sub Expr Expr | Represent 𝟏−𝟐 +𝟑? Copyright © 2018 by . All Rights Reserved.

Recursive Data Structures
How would you Write Functions to Evaluate Expressions Encoded with this Structure?
n.b., to evaluate an expression tree you evaluate subtrees an perform operations
Copyright © 2018 by . All Rights Reserved.

程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com