-->
Value generation patterns
There are three value generation patterns that can be used for properties:
I have tried renaming the field in my Foo class and specifying my (now non-conventionally-named) foreign key barId in my EntityTypeConfiguration. Builder.HasOne(x = x.Bar).WithMany.HasForeignKey('barId'); But this resulted in EF still generating a BarId column, without using it as foreign key to the Bar table. Jan 24, 2012 Adding the ForeignKey attribute, we say to Code First that we want the ManagerId properties to be used as Foreign Keys to the Person table. Run the application and check your database schema. The ManagerPersonId column does not exist anymore, there is only the ManagerId column and it is declared as a foreign key to the People table. How to set ForeignKey column name in fluent API on EF Core. I'm trying to give my Foreign Key column the name I want. The key is correctly created, but I. The ForeignKey attribute is used to specify which property is the foreign key in a relationship. In the following example, the AuthorFK property in the Book entity does not follow Entity Framework Core's convention for foreign key names.Left as it is, Entity Framework Core will create an AuthorFK field and an AuthorId field which it will configure as a foreign key. I'm a DBA and the developers in my organization want to use the Entity Framework (EF) in their applications. While I understand how EF can dynamically generate the SQL for queries, we have many existing views that I would like to be able to continue to u. In EF6 there's nothing stopping the ModelDiffer, which is common to all providers, to generate two AddForeignKeyOperations, one for PostLatestPost on the Blog table and one for BlogBlogId on each Post. Generating the SQL for adding two foreign keys is a piece of cake.
- No value generation
- Value generated on add
- Value generated on add or update
No value generation
No value generation means that you will always supply a valid value to be saved to the database. This valid value must be assigned to new entities before they are added to the context.
Value generated on add
Value generated on add means that a value is generated for new entities.
Depending on the database provider being used, values may be generated client side by EF or in the database. If the value is generated by the database, then EF may assign a temporary value when you add the entity to the context. This temporary value will then be replaced by the database generated value during SaveChanges()
.
If you add an entity to the context that has a value assigned to the property, then EF will attempt to insert that value rather than generating a new one. A property is considered to have a value assigned if it is not assigned the CLR default value (null
for string
, 0
for int
, Guid.Empty
for Guid
, etc.). For more information, see Explicit values for generated properties.
Warning
How the value is generated for added entities will depend on the database provider being used. Database providers may automatically setup value generation for some property types, but others may require you to manually setup how the value is generated.
For example, when using SQL Server, values will be automatically generated for GUID
properties (using the SQL Server sequential GUID algorithm). However, if you specify that a DateTime
property is generated on add, then you must setup a way for the values to be generated. One way to do this, is to configure a default value of GETDATE()
, see Default Values.
Value generated on add or update
Value generated on add or update means that a new value is generated every time the record is saved (insert or update).
Like value generated on add
, if you specify a value for the property on a newly added instance of an entity, that value will be inserted rather than a value being generated. It is also possible to set an explicit value when updating. For more information, see Explicit values for generated properties.
Warning
How the value is generated for added and updated entities will depend on the database provider being used. Database providers may automatically setup value generation for some property types, while others will require you to manually setup how the value is generated.
For example, when using SQL Server, byte[]
properties that are set as generated on add or update and marked as concurrency tokens, will be setup with the rowversion
data type - so that values will be generated in the database. However, if you specify that a DateTime
property is generated on add or update, then you must setup a way for the values to be generated. One way to do this, is to configure a default value of GETDATE()
(see Default Values) to generate values for new rows. You could then use a database trigger to generate values during updates (such as the following example trigger).
Value generated on add
By convention, non-composite primary keys of type short, int, long, or Guid are set up to have values generated for inserted entities, if a value isn't provided by the application. Your database provider typically takes care of the necessary configuration; for example, a numeric primary key in SQL Server is automatically set up to be an IDENTITY column.
You can configure any property to have its value generated for inserted entities as follows:
Warning
This just lets EF know that values are generated for added entities, it does not guarantee that EF will setup the actual mechanism to generate values. See Value generated on add section for more details.
Default values
On relational databases, a column can be configured with a default value; if a row is inserted without a value for that column, the default value will be used.
You can configure a default value on a property:
You can also specify a SQL fragment that is used to calculate the default value:
Specifying a default value will implicitly configure the property as value generated on add.
Value generated on add or update
Warning
This just lets EF know that values are generated for added or updated entities, it does not guarantee that EF will setup the actual mechanism to generate values. See Value generated on add or update section for more details.
Computed columns
On some relational databases, a column can be configured to have its value computed in the database, typically with an expression referring to other columns:
Note
In some cases the column's value is computed every time it is fetched (sometimes called virtual columns), and in others it is computed on every update of the row and stored (sometimes called stored or persisted columns). This varies across database providers.
No value generation
Disabling value generation on a property is typically necessary if a convention configures it for value generation. For example, if you have a primary key of type int, it will be implicitly set configured as value generated on add; you can disable this via the following:
-->Code First Migrations is the recommended way to evolve your application's database schema if you are using the Code First workflow. Migrations provide a set of tools that allow:
- Create an initial database that works with your EF model
- Generating migrations to keep track of changes you make to your EF model
- Keep your database up to date with those changes
The following walkthrough will provide an overview Code First Migrations in Entity Framework. You can either complete the entire walkthrough or skip to the topic you are interested in. The following topics are covered:
Building an Initial Model & Database
Before we start using migrations we need a project and a Code First model to work with. For this walkthrough we are going to use the canonical Blog and Post model.
- Create a new MigrationsDemo Console application
- Add the latest version of the EntityFramework NuGet package to the project
- Tools –> Library Package Manager –> Package Manager Console
- Run the Install-Package EntityFramework command
- Add a Model.cs file with the code shown below. This code defines a single Blog class that makes up our domain model and a BlogContext class that is our EF Code First context
- Now that we have a model it’s time to use it to perform data access. Update the Program.cs file with the code shown below.
Run your application and you will see that a MigrationsCodeDemo.BlogContext database is created for you.
Enabling Migrations
It’s time to make some more changes to our model.
- Let’s introduce a Url property to the Blog class.
If you were to run the application again you would get an InvalidOperationException stating The model backing the 'BlogContext' context has changed since the database was created. Consider using Code First Migrations to update the database (http://go.microsoft.com/fwlink/?LinkId=238269).
As the exception suggests, it’s time to start using Code First Migrations. The first step is to enable migrations for our context.
Run the Enable-Migrations command in Package Manager Console
This command has added a Migrations folder to our project. This new folder contains two files:
The Configuration class. This class allows you to configure how Migrations behaves for your context. For this walkthrough we will just use the default configuration.Because there is just a single Code First context in your project, Enable-Migrations has automatically filled in the context type this configuration applies to.
An InitialCreate migration. This migration was generated because we already had Code First create a database for us, before we enabled migrations. The code in this scaffolded migration represents the objects that have already been created in the database. In our case that is the Blog table with a BlogId and Name columns. The filename includes a timestamp to help with ordering.If the database had not already been created this InitialCreate migration would not have been added to the project. Instead, the first time we call Add-Migration the code to create these tables would be scaffolded to a new migration.
Multiple Models Targeting the Same Database
When using versions prior to EF6, only one Code First model could be used to generate/manage the schema of a database. This is the result of a single __MigrationsHistory table per database with no way to identify which entries belong to which model.
Starting with EF6, the Configuration class includes a ContextKey property. This acts as a unique identifier for each Code First model. A corresponding column in the __MigrationsHistory table allows entries from multiple models to share the table. By default, this property is set to the fully qualified name of your context.
Generating & Running Migrations
Code First Migrations has two primary commands that you are going to become familiar with.
- Add-Migration will scaffold the next migration based on changes you have made to your model since the last migration was created
- Update-Database will apply any pending migrations to the database
We need to scaffold a migration to take care of the new Url property we have added. The Add-Migration command allows us to give these migrations a name, let’s just call ours AddBlogUrl.
- Run the Add-Migration AddBlogUrl command in Package Manager Console
- In the Migrations folder we now have a new AddBlogUrl migration. The migration filename is pre-fixed with a timestamp to help with ordering
We could now edit or add to this migration but everything looks pretty good. Let’s use Update-Database to apply this migration to the database.
- Run the Update-Database command in Package Manager Console
- Code First Migrations will compare the migrations in our Migrations folder with the ones that have been applied to the database. It will see that the AddBlogUrl migration needs to be applied, and run it.
The MigrationsDemo.BlogContext database is now updated to include the Url column in the Blogs table.
Customizing Migrations
So far we’ve generated and run a migration without making any changes. Now let’s look at editing the code that gets generated by default.
- It’s time to make some more changes to our model, let’s add a new Rating property to the Blog class
- Let's also add a new Post class
- We'll also add a Posts collection to the Blog class to form the other end of the relationship between Blog and Post
We'll use the Add-Migration command to let Code First Migrations scaffold its best guess at the migration for us. We’re going to call this migration AddPostClass.
- Run the Add-Migration AddPostClass command in Package Manager Console.
Code First Migrations did a pretty good job of scaffolding these changes, but there are some things we might want to change:
- First up, let’s add a unique index to Posts.Title column(Adding in line 22 & 29 in the code below).
- We’re also adding a non-nullable Blogs.Rating column. If there is any existing data in the table it will get assigned the CLR default of the data type for new column (Rating is integer, so that would be 0). But we want to specify a default value of 3 so that existing rows in the Blogs table will start with a decent rating.(You can see the default value specified on line 24 of the code below)
Our edited migration is ready to go, so let’s use Update-Database to bring the database up-to-date. This time let’s specify the –Verbose flag so that you can see the SQL that Code First Migrations is running.
Unique Key
- Run the Update-Database –Verbose command in Package Manager Console.
Data Motion / Custom SQL
So far we have looked at migration operations that don’t change or move any data, now let’s look at something that needs to move some data around. There is no native support for data motion yet, but we can run some arbitrary SQL commands at any point in our script.
- Let’s add a Post.Abstract property to our model. Later, we’re going to pre-populate the Abstract for existing posts using some text from the start of the Content column.
We'll use the Add-Migration command to let Code First Migrations scaffold its best guess at the migration for us.
- Run the Add-Migration AddPostAbstract command in Package Manager Console.
- The generated migration takes care of the schema changes but we also want to pre-populate the Abstract column using the first 100 characters of content for each post. We can do this by dropping down to SQL and running an UPDATE statement after the column is added.(Adding in line 12 in the code below)
Our edited migration is looking good, so let’s use Update-Database to bring the database up-to-date. We’ll specify the –Verbose flag so that we can see the SQL being run against the database.
- Run the Update-Database –Verbose command in Package Manager Console.
Ef6 Not Generating Foreign Key Name Correctly Mean
Migrate to a Specific Version (Including Downgrade)
So far we have always upgraded to the latest migration, but there may be times when you want upgrade/downgrade to a specific migration.
Let’s say we want to migrate our database to the state it was in after running our AddBlogUrl migration. We can use the –TargetMigration switch to downgrade to this migration.
- Run the Update-Database –TargetMigration: AddBlogUrl command in Package Manager Console.
This command will run the Down script for our AddBlogAbstract and AddPostClass migrations.
If you want to roll all the way back to an empty database then you can use the Update-Database –TargetMigration: $InitialDatabase command.
Getting a SQL Script
If another developer wants these changes on their machine they can just sync once we check our changes into source control. Once they have our new migrations they can just run the Update-Database command to have the changes applied locally. However if we want to push these changes out to a test server, and eventually production, we probably want a SQL script we can hand off to our DBA.
- Run the Update-Database command but this time specify the –Script flag so that changes are written to a script rather than applied. We’ll also specify a source and target migration to generate the script for. We want a script to go from an empty database ($InitialDatabase) to the latest version (migration AddPostAbstract).If you don’t specify a target migration, Migrations will use the latest migration as the target. If you don't specify a source migrations, Migrations will use the current state of the database.
- Run the Update-Database -Script -SourceMigration: $InitialDatabase -TargetMigration: AddPostAbstract command in Package Manager Console
Code First Migrations will run the migration pipeline but instead of actually applying the changes it will write them out to a .sql file for you. Once the script is generated, it is opened for you in Visual Studio, ready for you to view or save.
Generating Idempotent Scripts
Starting with EF6, if you specify –SourceMigration $InitialDatabase then the generated script will be ‘idempotent’. Idempotent scripts can upgrade a database currently at any version to the latest version (or the specified version if you use –TargetMigration). The generated script includes logic to check the __MigrationsHistory table and only apply changes that haven't been previously applied.
Automatically Upgrading on Application Startup (MigrateDatabaseToLatestVersion Initializer)
If you are deploying your application you may want it to automatically upgrade the database (by applying any pending migrations) when the application launches. You can do this by registering the MigrateDatabaseToLatestVersion database initializer. A database initializer simply contains some logic that is used to make sure the database is setup correctly. This logic is run the first time the context is used within the application process (AppDomain).
We can update the Program.cs file, as shown below, to set the MigrateDatabaseToLatestVersion initializer for BlogContext before we use the context (Line 14). Note that you also need to add a using statement for the System.Data.Entity namespace (Line 5).
When we create an instance of this initializer we need to specify the context type (BlogContext) and the migrations configuration (Configuration) - the migrations configuration is the class that got added to our Migrations folder when we enabled Migrations.
Now whenever our application runs it will first check if the database it is targeting is up-to-date, and apply any pending migrations if it is not.