Use specialized sorted collections

Grade School
Grade School in C#
using System.Collections.Generic;
using System.Linq;

public class GradeSchool
{
    private readonly SortedDictionary<int, SortedSet<string>> _roster = new();

    public bool Add(string student, int grade)
    {
        if (_roster.Values.Any(students => students.Contains(student)))
            return false;

        if (!_roster.TryAdd(grade, new SortedSet<string> { student }))
            _roster[grade].Add(student);

        return true;
    }

    public IEnumerable<string> Roster() => _roster.Values.SelectMany(students => students);

    public IEnumerable<string> Grade(int grade) =>
        _roster.TryGetValue(grade, out var students) ? students : Enumerable.Empty<string>();
}

Choice of data structures

A key component of this exercise is on to return the data in a specific order. Whilst LINQ has an OrderBy() method that works on any collection type, there are also specialized data structures that are specifically designed to have sorting built-in.

So instead of using an array of list, we'll be using the SortedDictionary<K, T> and SortedSet<T> data structures. These data structures will always return data in sorted order (either some default or custom ordering), so we don't have to worry about that.

We can store the students in a grade (which is a collection of strings) in a SortedSet<string>, as we cannot have duplicate names in a grade.

As an example, here's how iterating a SortedSet<string> looks like:

var names = new SortedSet<string>();
names.Add("john");
names.Add("paul");
names.Add("george");
names.Add("ringo");

foreach (var name in names)
    Console.WriteLine(name);

// Outputs: george, john, paul, ringo

For the mapping between a grade (which is an integer) and its students, we'll use a SortedDictionary<int, SortedSet<string>>. We'll refer to this grade to student mapping as our roster, and define a private readonly field for it:

private readonly SortedDictionary<int, SortedSet<string>> _roster = new();

Method: Add()

The first step is to prevent a student from being added twice (regardless whether the grade matches). We can do that by check if any grade contains that student, and if so, return false:

if (_roster.Values.Any(students => students.Contains(student)))
    return false;

Having verified that the student does not exist, to add the student to the right grade, we first need to check if there already are students in the specified grade. If so, we'll add the student to the existing students set, otherwise we'll create a new set with the student:

if (_roster.ContainsKey(grade))
    _roster[grade].Add(student);
else
    _roster[grade] = new SortedSet<string> { student };
Note

Not that we don't worry about sorting, this will all happen transparently due to our choice of data structures.

We could also rewrite this to:

if (!_roster.ContainsKey(grade))
    _roster[grade] = new SortedSet<string>();

_roster[grade].Add(student);

To me, this reads somewhat better.

A slightly more performant approach is to use the TryAdd() method, which takes a key and the value to add when the key does not exist, whilst also returning a bool indicating if the key did not exist and the value was just added:

if (!_roster.TryAdd(grade, new SortedSet<string> { student }))
    _roster[grade].Add(student);

Finally, we'll need to return true to indicate that a new student was added:

return true;

Method: Roster()

To get all the students names in the roster, order ascendingly by grade and then by studnet name, we can just iterate over

public IEnumerable<string> Roster()
{
    var students = new List<string>();

    foreach (var (grade, studentsInGrade) in _roster)
        students.AddRange(studentsInGrade);

    return students;
}

We can shorten this by using the dictionary's Values property to only return its values: _roster.Values. This will, for our dictionary, return a collection of SortedSet<string> instances. To only return a single, flat enumerable, we can use Enumerable.SelectMany():

public IEnumerable<string> Roster()
{
    return _roster.Values.SelectMany(students => students);
}

Shortening

We can shorten our single-statement method to an expression-bodied method:

public IEnumerable<string> Roster() =>
    _roster.Values.SelectMany(students => students);

Method: Grade()

For the Grade() method, we'll have to deal with two cases:

  1. The roster has students in the specified grade
  2. The roster does not have students in the specified grade

An initial attempt to implement this logic could look like this:

public IEnumerable<string> Grade(int grade)
{
    if (_roster.ContainsKey(grade))
        return _roster[grade];

    return Enumerable.Empty<string>();
}
Note

The pattern of checking if a dictionary contains a key followed by accessing that value by key after that check is slightly wasteful, as it will require accessing the dictionary twice. For this pattern, you should instead use TryGetValue(), which returns a bool value indicating if the key exists and has an out parameter for the retrieved value (if any)

Note

We're using Enumerable.Empty<string>(), which is best practice when returning an empty enumerable (for performance reasons).

With this knowledge, let's rewrite out method a bit:

public IEnumerable<string> Grade(int grade)
{
    if (_roster.TryGetValue(grade, out var students))
        return students;

    return Enumerable.Empty<string>();
}

We've now gotten rid of the additional indexer access. Nice!

Shortening

We could rewrite our logic to a single statement using the ternary operator:

public IEnumerable<string> Grade(int grade)
{
    return _roster.TryGetValue(grade, out var students)
        ? students
        : Enumerable.Empty<string>();
}

As our method now only has a single statement, we can then rewrite it to an expression-bodied method:

public IEnumerable<string> Grade(int grade) =>
    _roster.TryGetValue(grade, out var students) ? students : Enumerable.Empty<string>();
24th Apr 2024 · Found it useful?