LINQ

LINQ: Improving our ability to search and manipulate collections of data

LINQ stands for Language INtegrated Query. It is a set of methods our various collections (e.g. List) will acquire once we add LINQ to our codebase. We will be adding the using System.Linq namespace to our code to give our List some new capabilities.

Detour through expressions

Before we can talk about LINQ we must learn a new language feature, the expression. So far we have seen methods such as this:

int MultiplyBy2(int value)
{
    return value * 2;
}

This method accepts an integer and returns an integer with twice the value. However, there is another way to express this idea.

Func<int, int> MultiplyBy2 = value => value * 2;

Let's break this down and see how it works in a similar way. First the data type for MultiplyBy2 is Func<int, int> -- The first int in the <> is the type of argument the function takes in. The last int represents the kind of return value the function produces. We are then assigning this variable the following: value => value * 2. This small bit of code is called an expression

The (value) indicates the name of that int argument we listed in the Func, and the arrow (=>) separates the list of arguments from the part afterward which is the code that should run if this code is called. In this case the value * 2.

So this function named MultiplyBy2 will take a value which is an int and return an int which is doubled.

We can write expressions to do all kinds of things. For instance we could write an expression to see if a particular Employee has a given name:

Func<Employee, string, bool> EmployeeHasName = (employee, name) => employee.Name == name;

In this case we are telling our function that it first takes an Employee object, secondarily it takes a string and finally it returns a bool. The expression defines the Employee (first) variable to be named employee and the string (second) variable to be called name. Finally the expression itself returns a bool since that is what we would get if we evaluated employee.Name == name as the == will always give us either true or false.

We could use this function as:

if (EmployeeHasName(employee, "Bob"))
{
    Console.WriteLine("Yup, that is Bob!");
}

More interestingly the expression is stored in a variable which means it can be passed to other methods.

Using LINQ and expressions.

Let's return to our expression example of using the MultiplyBy2. Suppose we had a list such as:

var scores = new List<int> { 42, 100, 98, 15 };

And suppose our task was to make a new variable equal to a list with all those values doubled? Well good for us that we have an expression that can do just that!

We would write code such as:

using System;
using System.Linq;
using System.Collections.Generic;
namespace linq
{
    class Program
    {
        static void Main(string[] args)
        {
            // Here is our original array
            var scores = new List<int> { 42, 100, 98, 15 };
            // Here is our handy Double-er
            Func<int, int> MultiplyBy2 = value => value * 2;
            // Make a new list to store the results
            var newScores = new List<int>();
            // Go through each score in the scores list
            foreach(var score in scores) {
              // Make a doubling of that score
              var doubled = MultiplyBy2(score);
              // Add it to our new list
              newScores.Add(doubled);
            }
            // Print out the scores comma separated
            Console.WriteLine(String.Join(',', newScores));
        }
    }
}

Well this is certainly nice, but let's get our new friend LINQ involved.

One of the methods that LINQ will give to our list is called Select. What Select does is go through each entry in our list, and using an expression converts each element to a new value based on what that expression does. Every new value is then added to a new List and returned. Whoa! That is exactly what our code above is doing! Let's simplify this code by using our new Select capability.

using System;
using System.Linq;
using System.Collections.Generic;
namespace linq
{
    class Program
    {
        static void Main(string[] args)
        {
            // Here is our original array
            var scores = new List<int> { 42, 100, 98, 15 };
            // Here is our handy Double-er
            Func<int, int> MultiplyBy2 = score => score * 2;
            // Make a new list by going through the `scores`
            // list, and for each item, call the `MultiplyBy2`
            // expression on that item and using the new
            // value to put into `newScores`
            var newScores = scores.Select(MultiplyBy2);
            // Print out the scores comma separated
            Console.WriteLine(String.Join(',', newScores));
        }
    }
}

Well that is much nicer. All that work of creating a new empty list, doing the foreach, calling the MultiplyBy2 expression, putting the new value into the list, etc. is all neatly captured by our friend Select. But it can be even better!

Since MultiplyBy2 is simply score => score * 2 we can put that code directly into Select() and our code becomes:

using System;
using System.Linq;
using System.Collections.Generic;
namespace linq
{
    class Program
    {
        static void Main(string[] args)
        {
            // Here is our original array
            var scores = new List<int> { 42, 100, 98, 15 };
            // Make a new list by going through the `scores`
            // list, and for each item, call the expression
            // on that item and using the new value to put
            // into `newScores`
            var newScores = scores.Select(score => score * 2);
            // Print out the scores comma separated
            Console.WriteLine(String.Join(',', newScores));
        }
    }
}

How nice and neat! What is nice is that Select is a generic method that we can use to do all kinds of processing. Maybe we need another variable that stores all the scores if we had just done a little better and scored one more point each.

var slightlyBetterScores = scores.Select(score => score + 1);

The power of LINQ methods that take expressions

We are about to see many different LINQ methods that each work by starting with a collection and applying an expression to its elements in different ways. Which method we will reach for when writing code depends on the behavior we are looking for. We must simply find the appropriate method and supply it an expression that does the work we want to do.

Later on we are going to see how LINQ works equally well on a List stored in memory and rows of a database. Thus learning how to effectively use LINQ is a skill we will re-use quite often in our C# programming.

Example

Let's see some examples of using LINQ on a specific list of data. In this case we are going to use a List of objects, specifically Movie

The definition of the Movie class is:

public class Movie
{
    public int Id { get; set; }
    public string Name { get; set; }
    public string Tagline { get; set; }
    public DateTime ReleasedDate { get; set; }
    public int Screenings { get; set; }
    public double PricePerTicket { get; set; }
    public double TotalRevenue { get; set; }
    public double Cost { get; set; }
    public double Budget { get; set; }
}

and our list is:

var movies = new List<Movie>()
{
    new Movie()
    {
        Id = 1,
        Name = "Dorm Daze (National Lampoon Presents Dorm Daze)",
        Tagline = "Multi-tiered modular standardization",
        ReleasedDate = DateTime.Parse("3/27/2019"),
        Screenings = 186,
        PricePerTicket = 11,
        TotalRevenue = 13361359,
        Cost = 18274829,
        Budget = 8210089
    },
    new Movie()
    {
        Id = 2,
        Name = "Born Yesterday",
        Tagline = "Managed empowering open system",
        ReleasedDate = DateTime.Parse("2/12/2014"),
        Screenings = 184,
        PricePerTicket = 11,
        TotalRevenue = 6563796,
        Cost = 9021912,
        Budget = 11364786
    },
    new Movie()
    {
        Id = 3,
        Name = "Darjeeling Limited, The",
        Tagline = "Quality-focused actuating initiative",
        ReleasedDate = DateTime.Parse("8/21/2013"),
        Screenings = 177,
        PricePerTicket = 10,
        TotalRevenue = 17851792,
        Cost = 5441889,
        Budget = 12144397
    },
    new Movie()
    {
        Id = 4,
        Name = "Offside",
        Tagline = "Enhanced homogeneous migration",
        ReleasedDate = DateTime.Parse("4/18/2019"),
        Screenings = 169,
        PricePerTicket = 11,
        TotalRevenue = 1445952,
        Cost = 4008467,
        Budget = 7417825
    },
    new Movie()
    {
        Id = 5,
        Name = "Superman vs. The Elite",
        Tagline = "Stand-alone systematic model",
        ReleasedDate = DateTime.Parse("12/7/2016"),
        Screenings = 124,
        PricePerTicket = 19,
        TotalRevenue = 13737676,
        Cost = 18893333,
        Budget = 6585110
    },
    new Movie()
    {
        Id = 6,
        Name = "Body Snatchers",
        Tagline = "Diverse holistic data-warehouse",
        ReleasedDate = DateTime.Parse("1/12/2007"),
        Screenings = 170,
        PricePerTicket = 10,
        TotalRevenue = 10540575,
        Cost = 12946787,
        Budget = 9237906
    },
    new Movie()
    {
        Id = 7,
        Name = "Death and Cremation",
        Tagline = "Ergonomic local knowledge base",
        ReleasedDate = DateTime.Parse("4/1/2013"),
        Screenings = 138,
        PricePerTicket = 10,
        TotalRevenue = 12361644,
        Cost = 7326663,
        Budget = 16829534
    },
    new Movie()
    {
        Id = 8,
        Name = "Other End of the Line, The",
        Tagline = "Up-sized demand-driven policy",
        ReleasedDate = DateTime.Parse("11/15/2016"),
        Screenings = 169,
        PricePerTicket = 12,
        TotalRevenue = 6371172,
        Cost = 17279838,
        Budget = 14274676
    },
    new Movie()
    {
        Id = 9,
        Name = "Our Mother's House",
        Tagline = "Enhanced methodical algorithm",
        ReleasedDate = DateTime.Parse("7/20/2018"),
        Screenings = 188,
        PricePerTicket = 17,
        TotalRevenue = 3544170,
        Cost = 7953388,
        Budget = 19636220
    },
    new Movie()
    {
        Id = 10,
        Name = "Everything I Can See From Here",
        Tagline = "Synchronized 24/7 utilization",
        ReleasedDate = DateTime.Parse("7/26/2012"),
        Screenings = 84,
        PricePerTicket = 4,
        TotalRevenue = 14520267,
        Cost = 2766779,
        Budget = 2478292
    },
    new Movie()
    {
        Id = 11,
        Name = "My Rainy Days",
        Tagline = "Cloned static array",
        ReleasedDate = DateTime.Parse("8/4/2015"),
        Screenings = 104,
        PricePerTicket = 15,
        TotalRevenue = 6860536,
        Cost = 6622076,
        Budget = 1091525
    },
    new Movie()
    {
        Id = 12,
        Name = "Five Graves to Cairo",
        Tagline = "Ergonomic heuristic capacity",
        ReleasedDate = DateTime.Parse("10/25/2013"),
        Screenings = 65,
        PricePerTicket = 17,
        TotalRevenue = 13595001,
        Cost = 3736299,
        Budget = 724740
    },
    new Movie()
    {
        Id = 13,
        Name = "Hunted, The",
        Tagline = "Multi-channelled object-oriented groupware",
        ReleasedDate = DateTime.Parse("2/4/2014"),
        Screenings = 185,
        PricePerTicket = 7,
        TotalRevenue = 13273082,
        Cost = 14879296,
        Budget = 7461416
    },
    new Movie()
    {
        Id = 14,
        Name = "Charlie Chan's Courage",
        Tagline = "Implemented interactive installation",
        ReleasedDate = DateTime.Parse("5/25/2006"),
        Screenings = 50,
        PricePerTicket = 10,
        TotalRevenue = 15695655,
        Cost = 11372062,
        Budget = 9089553
    },
    new Movie()
    {
        Id = 15,
        Name = "When Will I Be Loved",
        Tagline = "Networked uniform toolset",
        ReleasedDate = DateTime.Parse("8/25/2015"),
        Screenings = 165,
        PricePerTicket = 21,
        TotalRevenue = 10095292,
        Cost = 16020659,
        Budget = 15707348
    },
    new Movie()
    {
        Id = 16,
        Name = "Viva Las Vegas",
        Tagline = "Digitized dedicated capability",
        ReleasedDate = DateTime.Parse("7/4/2015"),
        Screenings = 85,
        PricePerTicket = 16,
        TotalRevenue = 16406383,
        Cost = 9854228,
        Budget = 16042287
    },
    new Movie()
    {
        Id = 17,
        Name = "Topaze",
        Tagline = "Advanced high-level benchmark",
        ReleasedDate = DateTime.Parse("12/1/2010"),
        Screenings = 60,
        PricePerTicket = 4,
        TotalRevenue = 13809680,
        Cost = 12667720,
        Budget = 14805773
    },
    new Movie()
    {
        Id = 18,
        Name = "The Clinic",
        Tagline = "Polarized regional solution",
        ReleasedDate = DateTime.Parse("4/20/2013"),
        Screenings = 128,
        PricePerTicket = 8,
        TotalRevenue = 17416537,
        Cost = 3435812,
        Budget = 8818065
    },
    new Movie()
    {
        Id = 19,
        Name = "The Land Before Time X: The Great Longneck Migration",
        Tagline = "Adaptive dedicated workforce",
        ReleasedDate = DateTime.Parse("10/10/2008"),
        Screenings = 170,
        PricePerTicket = 21,
        TotalRevenue = 5720197,
        Cost = 10514309,
        Budget = 3781872
    },
    new Movie()
    {
        Id = 20,
        Name = "Tarzan",
        Tagline = "Polarized intangible productivity",
        ReleasedDate = DateTime.Parse("12/31/2006"),
        Screenings = 105,
        PricePerTicket = 19,
        TotalRevenue = 6338974,
        Cost = 18402771,
        Budget = 844331
    },
    new Movie()
    {
        Id = 21,
        Name = "Jaws",
        Tagline = "When a killer shark unleashes chaos on a beach community, it's up to a local sheriff, a marine biologist, and an old seafarer to hunt the beast down. ",
        ReleasedDate = DateTime.Parse("1/1/1975"),
        Screenings = 105,
        PricePerTicket = 7,
        TotalRevenue = 6338974,
        Cost = 18402771,
        Budget = 844331
    },
};

How many movies are there?

To determine how many movies there are, we can ask the collection for its count:

Console.WriteLine($"There are {movies.Count()} total movies");

Turn the list of movies into a list of their names

var movieNames = movies.Select(movie => movie.Name);

We can also use the index of the item in the list as well.

var movieNames = movies.Select((movie, index) => $"The movie named {movie.Name} is at position {index}");

How many movies had more than 100 or more screenings?

First we use the Where LINQ statement to generate a new List<Movie> containing just the movies that have a Screenings property of 100 or more. Then we use Count on that List<Movie> to show the total.

var popularMovies = movies.Where(movie => movie.Screenings >= 100);
Console.WriteLine($"There are {popularMovies.Count()} popular movies");

Combine where and select to get the names of the movies with more than 100 screenings

var popularMovies = movies.Where(movie => movie.Screenings >= 100);
var popularMovieNames = popularMovies.Select(movie => movie.Name);
var popularMoviesNamesInOneLine = movies.Where(movie => movie.Screenings >= 100).Select(movie => movie.Name);

How many movies had less than 100 screenings

This is similar to the above, but with a different expression passed to Where

var unpopularMovies = movies.Where(movie => movie.Screenings < 100);
Console.WriteLine($"There are {unpopularMovies.Count()} unpopular movies");

Using Aggregate to total the revenues

We can use the Aggregate method to start currentTotalRevenue at 0.0, and each time through the loop, update currentTotalRevenue to be the existing value plus the current movie's TotalRevenue. This will give us a grand total.

var totalRevenue = movies.Aggregate(0.0, (currentTotalRevenue, movie) => currentTotalRevenue + movie.TotalRevenue);

Using Aggregate to total the gross revenue

This is similar to the above but with a different expression for totaling

var totalGross = movies.Aggregate(0.0, (currentGross, movie) => currentGross + movie.TotalRevenue - movie.Cost);
Console.WriteLine($"The total gross is ${totalGross}");

Using Sum to compute the total revenue

In this example we first take all the movies and use Select to generate a new list of all the revenues. Then we use Sum to add up the values. This is conceptually simpler than Aggregate

var allRevenues = movies.Select(movie => movie.TotalRevenue);
var totalRevenue = allRevenues.Sum();

Determine if there are any old movies

We can use the Where clause to find all the old movies

var theOldMovies = movies.Where(movie => movie.ReleasedDate.Year < 1965);
if (theOldMovies.Count() > 0)
{
    // Do something
}

Determine if there are ANY old movies

// Just one true or false if they are ALLLLL old movies.
var areAllOldMovies = movies.All(movie => movie.ReleasedDate.Year < 1965);
if (areAllOldMovies)
{
    Console.WriteLine("Yes, all movies before 1965");
}
else
{
    Console.WriteLine("No, there is at least one movie after 1965");
}

Counting a condition

We could count the number of movies that cost more than \$10 in two different ways:

var numberOfMoviesCostingMoreThanTenBucks = movies.Count(movie => movie.PricePerTicket > 10);
Console.WriteLine($"There are {numberOfMoviesCostingMoreThanTenBucks} that cost more than $10");

// This is kinda the same.
var anotherWayToCountMoviesCostingMoreThan10 = movies.Where(movie => movie.PricePerTicket > 10).Count();
Console.WriteLine($"There are {anotherWayToCountMoviesCostingMoreThan10} that cost more than $10");

References

See this quick reference guide for an overview of the frequently used enumeration methods with examples.

Here is a list of all the LINQ methods.