Introduction to Functional Programming – Exercises Week 3: "Lists" | TDA555 / DIT440, LP1 2015 |

Home | Schedule | Labs | Exercises | Exam | About | FAQ | Fire | Forum | TimeEdit | Links |

Introduction to Functional Programming – Exercises Week 3: "Lists" | TDA555 / DIT440, LP1 2015 |

Home | Schedule | Labs | Exercises | Exam | About | FAQ | Fire | Forum | TimeEdit | Links |

Here are some exercises designed to help you practice programming with lists
and list comprehensions.
*nub* is defined in the standard library module
*Data.List*: you must write ## 1. Permutations

## 2 (*). Sorting

## 3. Avoiding Duplicates

## 4. Pascal's Triangle

Pascal's triangle is a triangle of numbers
## 5. Erastosthenes' sieve

Eratosthenes' sieve is an ancient method for finding prime numbers. Start by
writing out all the numbers from 2 to (say) 100. The first number (2) is
prime. Now cross out all multiples of 2. The first remaining number (3) is
also prime. Cross out all multiples of 3. The first remaining number (5) is
also prime... and so on. When no numbers remain, you have found all the prime
numbers in the range you started with.
*list* as its argument, so you must see to it that the
list gets smaller in each recursive call. Take an empty argument list as your
base case.
## 6. Number Games

In these examples we'll investigate the properties of prime numbers in the
range 2 to 100. Define functions
## 7 (*). Occurrences in Lists

Define the following functions, and state their (polymorphic) types:
*bag* to convert a list into a bag. For example,
## 8. Elements and Positions

Elements which occur in lists do so at a particular position. For example, 'l'
occurs in "hello" at positions 3 and 4. Define functions
## 9 (*). List Comprehensions

Experiment with the function

You may need the following useful standard functions:

- or :: [Bool] -> Bool, returns True if any element of its argument list is True.
- and :: [Bool] -> Bool, returns True if every element of its argument list is True.
- nub :: Eq a => [a] -> [a], which removes duplicate elements from a list.

import Data.Listat the beginning of any Haskell program that uses it.

If you do not have time to do all these exercises, don't worry.
The exercises are intended to provide enough work to keep the *most*
experienced students busy. If you do all exercises marked with an (*) you have
probably understood this week's material.

Good luck!

A *permutation* of a list is another list with the same elements, but in
a possibly different order. For example, [1,2,1] is a permutation of [2,1,1],
but not of [1,2,2]. Write a function

isPermutation :: Eq a => [a] -> [a] -> Bool

that returns True if its arguments are permutations of each other.

Express suitable properties of the "reverse" function in the context of permutations.

Just as keeping a room tidy can make it easier to find things, so keeping
information organised inside a computer can make tasks easier to accomplish. For
that reason, lists are often kept *sorted*, with the least element first,
and the greatest element last. In this exercise, we develop functions to take an
unsorted list and convert it into a sorted one.

First of all, we will need to be able to check that lists are sorted. Define a function

sorted :: Ord a => [a] -> Bool

which returns True if its argument is a sorted list. (The extra type restriction "Ord a =>" is there because it only works for types that have an ordering, i.e. we can use the operations <, >, >=, etc.)

For example,

Main> sorted [1,2,3,4,5,6,7]

True

Main> sorted [1,2,3,2,4]

False

The sorting method we will use is called *insertion sort* -- imagine
sorting a collection of magazines by working through an unsorted pile and
building a sorted pile, taking each magazine in turn from the unsorted pile, and
inserting it into the right place in the sorted one. Clearly, when there are no
magazines left in the unsorted pile, then we will have a pile containing all the
magazines in the correct order. Each pile will be represented in our program by
a list.

The first step is to define a function which inserts an element into a sorted list, in the correct position so that the result is still sorted. We call it

insert' :: Ord a => a -> [a] -> [a]

(yes, insert is a standard function too) and for example,

Main> insert' 2 [1,3,4]

[1,2,3,4]

Define insert and test it, by QuickChecking the following property:

prop_insert :: Integer -> [Integer] -> Property

prop_insert x xs = sorted xs ==> sorted (insert' x xs)

Now use insert' to define

isort :: Ord a => [a] -> [a]

which sorts any list into order, using the insertion sort method. For example,

Main> isort "hello"

"ehllo"

Main> isort ["hello","clouds","hello","sky"]

["clouds","hello","hello","sky"]

Write and test a QuickCheck property of isort that only holds if isort is a
correct sorting function. (In particular, make sure that your property *fails*
for this incorrect definition of isort: isort xs = []).

In many situations, lists should not contain *duplicate* elements. For
example, a pack of cards should not contain the same card twice. Define a
function

duplicates :: Eq a => [a] -> Bool

which returns True if its argument contains duplicate elements.

Main> duplicates [1,2,3,4,5]

False

Main> duplicates [1,2,3,2]

True

*Hint:* the standard function elem, which tests whether an element
occurs in a list, is helpful here.

One way to *ensure* a list contains no duplicates is to start with a
list that might contain duplicate elements, and remove them. Define a function

removeDuplicates :: Eq a => [a] -> [a]

which returns a list containing the same elements as its argument, but without duplicates. Test it using the following property:

prop_duplicatesRemoved :: [Integer] -> Bool

prop_duplicatesRemoved xs = not (duplicates (removeDuplicates
xs))

Does this property guarantee that removeDuplicates behaves correctly? If not, what is missing?

(removeDuplicates is actually a standard function, called nub).

1 1 1 1 2 1 1 3 3 1 1 4 6 4 1 1 5 10 10 5 1 .............computed as follows:

- The first row just contains a 1.
- The following rows are computed by adding together adjacent numbers in the row above, and adding a 1 at the beginning and at the end.

Define a function

pascal :: Int -> [Int]so that pascal n computes the nth row of Pascal's triangle.

Define a function

crossOut :: Int -> [Int] -> [Int]so that crossOut m ns removes all multiples of m from ns. Try to not implement crossOut recursively, but use a list comprehension instead!

Now define a (recursive!) function

sieve :: [Int] -> [Int]which applies Eratosthenes' sieve to the list of numbers it is given, and returns a list of all the prime numbers that it found. This is a recursive function with a

Use sieve to construct the list of primes from 2 to 100.

- To test whether n is a prime (in the range 2 to 100).
- To test whether n is a sum of two primes (in the range 2 to 100).

- occursIn x xs, which returns True if x is an element of xs.
- allOccurIn xs ys, which returns True if all of the elements of xs are also elements of ys.
- sameElements xs ys, which returns True if xs and ys have exactly the same elements.
- numOccurrences x xs, which returns the number of times x occurs in xs.

In some ways, lists are like sets: both are collections of elements. But the order of elements in a list matters, while it does not matter in a set, and the number of occurrences in a list matters, while it does not matter in a set.

The concept of a *bag* is something between a list and a set: the number
of occurrences matters, but the order of elements does not. One way to
represent a bag is a list of pairs of values and the number of times the value
occurs: for example

[("a",1), ("b",2)]Define a function

bag "hello"should be

[('h',1),('e',1),('l',2),('o',1)]

- positions xs, which converts a list into a list of pairs of elements and their positions. Hint: Make use of the standard function zip.
- firstPosition x xs, which returns the first position at which x occurs in xs.
- remove1 x xs, which removes the first occurrence of x from xs. For example, remove1 'l' "hello" = "helo"
- remove n x xs, which removes the first n occurrences of x from xs.

pairs :: [a] -> [b] -> [(a,b)] pairs xs ys = [(x,y) | x<-xs, y<-ys]and see what it does.

A *Pythagorean triad* is a triple of integers (a,b,c) such that

a^2 + b^2 == c^2Find all Pythagorean triads with a<=b<=c<=100.