Home > Articles > Programming > Windows Programming

  • Print
  • + Share This
This chapter is from the book

21.8 Retrieving Data from Multiple Tables with LINQ

In the two previous examples, we used data bindings to display data extracted using LINQ to SQL. In this section, we concentrate on LINQ to SQL features that simplify querying and combining data from multiple tables. You’ve already seen the SQL INNER JOIN operator in Section 21.4.4—LINQ to SQL provides similar capabilities and allows more complex operations as well. Figure 21.27 uses LINQ to SQL to combine and organize data from multiple tables.

Fig. 21.27 Using LINQ to perform a join and aggregate data across tables.

 1  // Fig. 21.27: JoiningTest.cs
 2  // Using LINQ to perform a join and aggregate data across tables.
 3  using System;
 4  using System.Linq;
 5
 6  namespace JoiningWithLINQ
 7  {
 8     public class JoiningTest
 9     {
10        public static void Main( string[] args )
11        {
12           // create database connection
13           BooksDataContext database = new BooksDataContext();
14
15           // get authors and ISBNs of each book they co-authored
16           var authorsAndISBNs =
17              from author in database.Authors
18              join book in database.AuthorISBNs
19                 on author.AuthorID equals book.AuthorID
20              orderby author.LastName, author.FirstName
21              select new { author.FirstName, author.LastName, book.ISBN };
22
23           Console.WriteLine( "Authors and ISBNs:" ); // display header
24
25           // display authors and ISBNs in tabular format
26           foreach ( var element in authorsAndISBNs )
27           {
28              Console.WriteLine( "\t{0,-10} {1,-10} {2,-10}",
29                 element.FirstName, element.LastName, element.ISBN );
30           } // end foreach
31
32           // get authors and titles of each book they co-authored
33           var authorsAndTitles =
34              from title in database.Titles                             
35              from book in title.AuthorISBNs                            
36              let author = book.Author                                  
37              orderby author.LastName, author.FirstName, title.BookTitle
38              select new { author.FirstName, author.LastName,           
39                 title.BookTitle };                                     
40
41           Console.WriteLine( "\nAuthors and titles:" ); // header
42
43           // display authors and titles in tabular format
44           foreach ( var element in authorsAndTitles )
45           {
46              Console.WriteLine( "\t{0,-10} {1,-10} {2}",
47                 element.FirstName, element.LastName, element.BookTitle );
48           } // end foreach
49
50           // get authors and titles of each book
51           // they co-authored; group by author
52           var titlesByAuthor =
53              from author in database.Authors
54              orderby author.LastName, author.FirstName
55              let name = author.FirstName + " " + author.LastName
56              let titles =
57                 from book in author.AuthorISBNs
58                 orderby book.Title.BookTitle
59                 select book.Title.BookTitle
60              select new { Name = name, Titles = titles };
61
62           Console.WriteLine( "\nTitles grouped by author:" ); // header
63
64           // display titles written by each author, grouped by author
65           foreach ( var author in titlesByAuthor )
66           {
67              // display author's name
68              Console.WriteLine( "\t" + author.Name + ":" );
69
70              // display titles written by that author
71              foreach ( var title in author.Titles )
72              {
73                 Console.WriteLine( "\t\t" + title );
74              } // end inner foreach
75           } // end outer foreach
76        } // end Main
77     } // end class JoiningTest
78  } // end namespace JoiningWithLINQ
Authors and ISBNs:
        Greg       Ayer        0136053033
        Harvey     Deitel      0131752421
        Harvey     Deitel      0132222205
        Harvey     Deitel      0132404168
        Harvey     Deitel      0136053033
        Harvey     Deitel      013605305X
        Harvey     Deitel      013605322X
        Harvey     Deitel      0136151574
        Harvey     Deitel      0136152503
        Paul       Deitel      0131752421
        Paul       Deitel      0132222205
        Paul       Deitel      0132404168
        Paul       Deitel      0136053033
        Paul       Deitel      013605305X
        Paul       Deitel      013605322X
        Paul       Deitel      0136151574
        Paul       Deitel      0136152503
        Dan        Quirk       0136151574

Authors and titles:
        Greg       Ayer        Simply Visual Basic 2008
        Harvey     Deitel      C How to Program
        Harvey     Deitel      C++ How to Program
        Harvey     Deitel      Internet & World Wide Web How to Program
        Harvey     Deitel      Java How to Program
        Harvey     Deitel      Simply Visual Basic 2008
        Harvey     Deitel      Visual Basic 2008 How to Program
        Harvey     Deitel      Visual C# 2008 How to Program
        Harvey     Deitel      Visual C++ 2008 How to Program
        Paul       Deitel      C How to Program
        Paul       Deitel      C++ How to Program
        Paul       Deitel      Internet & World Wide Web How to Program
        Paul       Deitel      Java How to Program
        Paul       Deitel      Simply Visual Basic 2008
        Paul       Deitel      Visual Basic 2008 How to Program
        Paul       Deitel      Visual C# 2008 How to Program
        Paul       Deitel      Visual C++ 2008 How to Program
        Dan        Quirk       Visual C++ 2008 How to Program

Titles grouped by author:
        Greg Ayer:
                Simply Visual Basic 2008
        Harvey Deitel:
                C How to Program
                C++ How to Program
                Internet & World Wide Web How to Program
                Java How to Program
                Simply Visual Basic 2008
                Visual Basic 2008 How to Program
                Visual C# 2008 How to Program
                Visual C++ 2008 How to Program
        Paul Deitel:
                C How to Program
                C++ How to Program
                Internet & World Wide Web How to Program
                Java How to Program
                Simply Visual Basic 2008
                Visual Basic 2008 How to Program
                Visual C# 2008 How to Program
                Visual C++ 2008 How to Program
        Dan Quirk:
                Visual C++ 2008 How to Program

The code combines data from the three tables in the Books database and displays the relationships between the book titles and authors in three different ways. The LINQ to SQL classes used in this example were created using the steps described in Section 21.6.1. As in previous examples, the BooksDataContext object (declared in line 13) is needed to be able to query the database.

The first query in the example (lines 17–21) returns results identical to those in Fig. 21.19. It uses LINQ’s join clause, which functions like SQL’s INNER JOIN operator—the generated SQL is nearly identical to the SQL given earlier in Section 21.4.4. As in the SQL example, only rows with the same AuthorID are joined together. Like the from clause, the join clause introduces a range variable—unlike the from clause, it specifies a criterion for joining. The join clause uses equals instead of the == comparison operator because the join criterion is not an arbitrary Boolean expression—you may only join based on equality. Like nested repetition statements, join clauses cause multiple range variables to be in scope—other clauses can access both range variables to combine data from multiple tables (lines 20–21).

The second query (lines 34–39) gives similar output, but it does not use the join query operator. Operations that would require a join in SQL often do not need one in LINQ to SQL, because it automatically creates properties based on foreign-key relationships. These properties enable you to easily access related rows in other tables. Line 35 uses the generated AuthorISBNs property of the Title class to query only the rows in the AuthorISBN table that link to that row of the Titles table. It does this by using multiple from clauses in the same query. In this example the inner from clause iterates over data related to the outer range variable, but the sequences iterated over may be completely unrelated. As with a join clause, both range variables may be used in later clauses. The author variable created in the let clause (line 36) refers to book.Author, demonstrating the automatically generated link between the AuthorISBN and Authors tables based on the foreign-key relationship between them.

Lines 53–60 contain the final query in the example. Instead of returning a flat result set, with data laid out in relational-style rows and columns, the results from this query are hierarchical. Each element in the results contains the name of an Author and a list of Titles that the author wrote. The LINQ query does this by using a nested query in the second let clause (lines 56–59). The outer query iterates over the authors in the database. The inner query (lines 57–59) takes a specific author and retrieves all titles that the author worked on. It does this by navigating the properties created by the foreign-key relationships in the database. The book range variable represents each pair of AuthorID and ISBN in the AuthorISBN table belonging to the author range variable of the outer query. It accesses the Title property of book to retrieve the row in the Titles table with that ISBN and then uses the BookTitle property to include the title of the book in the results. This list of titles is placed into the Titles property of the anonymous type created in the select clause, which also has a Name property that contains the author’s full name. These results are then displayed using nested foreach statements (lines 65–75).

Relational databases cannot return this kind of hierarchical result set, so, unlike the previous two queries, it would be impossible to write a query like this in SQL. Before LINQ, you’d have had to retrieve the results in a flat table like the other two queries, then transform them into the desired format. LINQ does this work for you, allowing you to ignore the relational storage model and concentrate on the object structure that fits your application.

  • + Share This
  • 🔖 Save To Your Account