We Keep Coding

Changing column from text to integer with Entity Framework

My table has a text column and I'd like to change to an integer. The existing text column contains values that can be converted to an integer. I don't want to change the column name.
How can I safely do this using code first without losing my data?
One option would be to create a new, temporary column. Copy the values. And then delete the old column and rename the new one. But I'm just not that sure what Entity Framework will decide to do with those changes.

So, I simply changed the column definition from:
[Display(Name = "Bill of Lading")]
[Required]
[StringLength(80)]
public string BillOfLading { get; set; }
To:
[Display(Name = "Bill of Lading")]
public int BillOfLading { get; set; }
I patched up my code so it would compile with the new type, and I added a migration, which looked like this:
public partial class TransloadingDetailBillOfLadingToInt : Migration
{
protected override void Up(MigrationBuilder migrationBuilder)
{
migrationBuilder.AlterColumn<int>(
name: "BillOfLading",
table: "TransloadingDetails",
type: "int",
nullable: false,
oldClrType: typeof(string),
oldType: "nvarchar(80)",
oldMaxLength: 80);
}
protected override void Down(MigrationBuilder migrationBuilder)
{
migrationBuilder.AlterColumn<string>(
name: "BillOfLading",
table: "TransloadingDetails",
type: "nvarchar(80)",
maxLength: 80,
nullable: false,
oldClrType: typeof(int),
oldType: "int");
}
}
I was able to run update-database on a smaller, demo database we have. On my first attempt, I had a column value that could not be converted. In that instance, I got an error and no data was lost. A good sign!
After correcting the data, I ran it again and the text columns were converted to integers. All the existing data was correctly converted. And then the same thing on my main database.
So, at least in the case of converting a text column to an integer column, where the data can be converted, this works just as you'd want it to work.

Generating non-nullable rowversion on SQL Server with EF Core 3.1

We are trying to generate a non-nullable rowversion column on SQL Server with EF Core 3.1 using Fluent API:
public class Person
{
public int Id { get; set; }
public byte[] Timestamp { get; set; }
}
public class PersonEntityConfiguration : IEntityTypeConfiguration<Person>
{
public void Configure(EntityTypeBuilder<Person> builder)
{
builder.HasKey(p => p.Id);
builder.Property(p => p.Timestamp)
.IsRowVersion()
.IsRequired();
}
}
This works fine when the entire table is new:
public partial class PersonMigration : Migration
{
protected override void Up(MigrationBuilder migrationBuilder)
{
migrationBuilder.CreateTable(
name: "Persons",
columns: table => new
{
Id = table.Column<int>(nullable: false)
.Annotation("SqlServer:Identity", "1, 1"),
Timestamp = table.Column<byte[]>(rowVersion: true, nullable: false)
},
constraints: table =>
{
table.PrimaryKey("PK_Persons", x => x.Id);
});
}
}
However, we sometimes need to add the rowversion to an existing table. In that case, EF Core generates an invalid migration:
public partial class PersonTimestampMigration : Migration
{
protected override void Up(MigrationBuilder migrationBuilder)
{
migrationBuilder.AddColumn<byte[]>(
name: "Timestamp",
table: "Persons",
rowVersion: true,
nullable: false,
defaultValue: new byte[] { });
}
}
The default value generated above will cause an exception when being applied to the database:
Failed executing DbCommand (1ms) [Parameters=[], CommandType='Text', CommandTimeout='30']
ALTER TABLE [Persons] ADD [Timestamp] rowversion NOT NULL DEFAULT 0x;
Microsoft.Data.SqlClient.SqlException (0x80131904): Defaults cannot be created on columns of data type timestamp. Table 'Persons', column 'Timestamp'.
Could not create constraint or index. See previous errors.
Is this a known bug in EF Core? The issue can be fixed by manually removing the defaultValue: new byte[] { } from the migration, but is there a way of suppressing the default value from being generated using the Fluent API?

Is this a known bug in EF Core?
It is bug/defect for sure, but probably not known, since it's happening even in the latest at this time EF Core 5.0 preview. Or is known, but with low priority (for them) - you have to check EF Core Issue Tracker.
Tried adding explicitly .HasDefaultValue(null) and .HasDefaultValueSql(null) - nothing helps, so the only option is manually removing the defaultValue: new byte[] { } from the migration. The good thing is that when you do so, it works and the column is created and populated successfully even though the table has existing records (which is the reason EF Core adds such defaultValue argument for new required columns in general, but as we see shouldn't do that for ROWVERSION).

Add new Required Field to one of table with EF Code First Migration

I am using EF Code First Migration. I already have lots of data on production Db and I would like to intorduce a non nullable field. How it could be possible?
Currently it throws an error:
The column cannot contain null values. [ Column name = Test,Table name = 'MyTable']

The strategy I generally use for this is to first introduce the new column as optional, populate it, and then make it required. You can actually make separate migration steps by making separate migrations for each of these steps or manually change the automatically generated migration code to make the change all happen in one migration. I will be describing how to use a single migration.
If you add a new [Required] field, the autogenerated migration may look like this which will fail if dbo.MyTable has data in it:
public override void Up()
{
AddColumn("dbo.MyTable", "MyColumn", c => c.String(nullable: false));
}
public override void Down()
{
DropColumn("dbo.MyTable", "MyColumn");
}
You can edit the migration to initially add the column as optional, prepopulate it, and then mark it required. You can use Sql() to perform the prepopulation. Depending on your data, you may desire to calculate the new column’s value based on other columns or even another table and you can do that by writing the appropriate SQL.
public override void Up()
{
AddColumn("dbo.MyTable", "MyColumn", c => c.String());
Sql("UPDATE dbo.MyTable SET MyColumn = COALESCE(SomeOtherColumn, 'My Fallback Value')");
AlterColumn("dbo.MyTable", "MyColumn", c => c.String(nullable: false));
}
public override void Down()
{
DropColumn("dbo.MyTable", "MyColumn");
}
If you merely want to set the same value on all rows, you can skip the Sql() step and add the column as required with a defaultValue before removing the defaultValue. This allows you to set all rows to the same value during migration while still requiring you to manually specify the value when adding new rows. This approach supposedly doesn’t work on older versions of EF. I’m not sure which version of EF is required, but it should at least work on modern ones (≥6.2):
public override void Up()
{
AddColumn("dbo.MyTable", "MyColumn", c => c.String(defaultValue: "Some Value", nullable: false));
}
public override void Down()
{
DropColumn("dbo.MyTable", "MyColumn");
}

You can add a defaultValue in the migration file something like this:
protected override void Up(MigrationBuilder migrationBuilder)
{
migrationBuilder.AddColumn<string>(
name: "MyColumn",
table: "dbo.MyTable",
nullable: false,
defaultValue: "");
}
And if the column is bool so you just put the defaultValue false or true depend on your need.

In EFCore 3.1 you can just add the required attribute and it automatically sets a default value
Boolean defaults to false
[Required]
public Boolean myField {get;set;}
Integer defaults to 0
[Required]
public int myField {get;set;}
String defaults to ""
[Required]
public string myField {get;set;}

EF5 Code First - Changing A Column Type With Migrations

I am new to EF5 Code First and I'm tinkering with a proof-of-concept before embarking on a project at work.
I have initially created a model that looked something like
public class Person {
public int Id { get; set; }
public string FirstName { get; set;}
public string Surname {get;set;}
public string Location {get;set;}
}
And I added a few records using a little MVC application I stuck on the top.
Now I want to change the Location column to an enum, something like:
public class Person {
public int Id { get; set; }
public string FirstName { get; set;}
public string Surname {get;set;}
public Locations Location {get;set;}
}
public enum Locations {
London = 1,
Edinburgh = 2,
Cardiff = 3
}
When I add the new migration I get:
AlterColumn("dbo.People", "Location", c => c.Int(nullable: false));
but when I run update-database I get an error
Conversion failed when converting the nvarchar value 'London' to data type int.
Is there a way in the migration to truncate the table before it runs the alter statement?
I know I can open the database and manually do it, but is there a smarter way?

The smartest way is probably to not alter types. If you need to do this, I'd suggest you to do the following steps:
Add a new column with your new type
Use Sql() to take over the data from the original column using an update statement
Remove the old column
Rename the new column
This can all be done in the same migration, the correct SQL script will be created. You can skip step 2 if you want your data to be discarded. If you want to take it over, add the appropriate statement (can also contain a switch statement).
Unfortunately Code First Migrations do not provide easier ways to accomplish this.
Here is the example code:
AddColumn("dbo.People", "LocationTmp", c => c.Int(nullable: false));
Sql(#"
UPDATE dbp.People
SET LocationTmp =
CASE Location
WHEN 'London' THEN 1
WHEN 'Edinburgh' THEN 2
WHEN 'Cardiff' THEN 3
ELSE 0
END
");
DropColumn("dbo.People", "Location");
RenameColumn("dbo.People", "LocationTmp", "Location");

Based on #JustAnotherUserYouMayKnow's answer, but easier.
Try firstly execute Sql() command and then AlterColumn():
Sql(#"
UPDATE dbo.People
SET Location =
CASE Location
WHEN 'London' THEN 1
WHEN 'Edinburgh' THEN 2
WHEN 'Cardiff' THEN 3
ELSE 0
END
");
AlterColumn("dbo.People", "Location", c => c.Int(nullable: false));

I know this doesn't apply directly to the question but could be helpful to someone. In my problem, I accidentally made a year field a datetime and I was trying to figure out how to delete all the data and then switch the data type to an int.
When doing an add-migration, EF wanted to just update the column. I had to delete what they wanted to do and add my own code. I basically just dropped the column and added a new column. Here is what worked for me.
protected override void Up(MigrationBuilder migrationBuilder)
{
migrationBuilder.DropColumn(
name: "TestingPeriodYear",
table: "ControlActivityIssue");
migrationBuilder.AddColumn<int>(
name: "TestingPeriodYear",
table: "ControlActivityIssue",
nullable: true);
}
protected override void Down(MigrationBuilder migrationBuilder)
{
migrationBuilder.DropColumn(
name: "TestingPeriodYear",
table: "ControlActivityIssue");
migrationBuilder.AddColumn<DateTime>(
name: "TestingPeriodYear",
table: "ControlActivityIssue",
nullable: true);
}

How to rename a database column in Entity Framework 5 Code First migrations without losing data?

I got the default ASP.NET MVC 4 template successfully running with EF 5.0 Code First Migrations. However, when I update a model property name, the corresponding table column data is dropped by EF 5.0.
Is it somehow possible to rename the table column without dropping data in an automated way?

Manually edit the Up and Down methods of the migration to use the RenameColumn method to replace the AddColumn and DropColumn that it automatically generates for you.

As already said, replace the AddColumn and DropColumn that is automatically generated with RenameColumn.
Example:
namespace MyProject.Model.Migrations
{
using System;
using System.Data.Entity.Migrations;
public partial class RenameMyColumn : DbMigration
{
public override void Up()
{
// Remove the following auto-generated lines
AddColumn("dbo.MyTable", "NewColumn", c => c.String(nullable: false, maxLength: 50));
DropColumn("dbo.MyTable", "OldColumn");
// Add this line
RenameColumn("dbo.MyTable", "OldColumn", "NewColumn");
}
public override void Down()
{
// Remove the following auto-generated lines
AddColumn("dbo.MyTable", "OldColumn", c => c.String(nullable: false, maxLength: 50));
DropColumn("dbo.MyTable", "NewColumn");
// Add this line
RenameColumn("dbo.MyTable", "NewColumn", "OldColumn");
}
}
}

You can get the migration to call RenameColumn for you if you do this:
[Column("NewName")]
public string OldName { get; set; }
Here is the generated migration:
public override void Up()
{
RenameColumn(table: "Schema.MyTable", name: "OldName", newName: "NewName");
}
public override void Down()
{
RenameColumn(table: "Schema.MyTable", name: "NewName", newName: "OldName");
}
If you want your property and DB column to be the same name, you can rename the property later and remove the Column attribute.

you have 2 steps to rename column in code first migration
The first step,you add ColumnAttribute above your column which are changed, and then update-database command
[Column("Content")]
public string Description { set; get; }
The second step,
add-migration yournamechange command in order to create a partial class DbMigration.
add into up and down method here
RenameColumn("yourDatabase","name","newName");
public override void Up()
{
RenameColumn("dbo.your_database", "oldColumn",
"newColumn");
}
public override void Down()
{
RenameColumn("dbo.your_database", "newColumn",
"oldColumn");
}
Because when you connect, your database and model class will communicate via name_column at database and name_type at property method in model above.

Now, this answer is based on my knowledge of EF4.3, so I hope the migrations work roughly the same in EF5 :) After you've created a migration, you should be able to add code in the Up and Down methods, between the dropping of the old property and the creation of the new property. This code should move the property data in the correct direction. I've solved it with the SQL() method in where you can enter raw SQL to perform the data move.
In the Up method of the migration:
SQL("update [TheTable] set [NewColumn] = [OldColumn]");
and in the Down() method:
SQL("update [TheTable] set [OldColumn] = [NewColumn]");
The disadvantage of this approach is that you might couple your code with the database you're working with at the moment (since you're writing raw DB-specific SQL). There might be other methods available for data movement as well.
More info available here: MSDN

Adding to Josh Gallagher's answer:
In some places the sp_RENAME syntax is described like this:
sp_RENAME 'TableName.[OldColumnName]' , '[NewColumnName]', 'COLUMN'
However, that will actually include the brackets in the new column name.
DbMigration's RenameColumn() method will probably do the same, so avoid using brackets when specifying the new column name.
Also, the auto-generated commands in Up() & Down() include DropPrimaryKey() and AddPrimaryKey() if the column being renamed is part of the primary key. These are not needed when using RenameColumn(). The underlying sp_RENAME automatically updates the primary key if needed.

In Entity Framework Core 3.1.1, If you need to rename a column -
Add migration as below
migrationBuilder.RenameColumn(
name: "oldname",
table: "tablename",
newName: "newname",
schema: "schema");

like +Josh Gallagher said, you can use up() for doing things like:
public override void Up()
{
RenameColumn("dbo.atable","oldname","newname");
AddColumn("dbo.anothertable", "columname", c => c.String(maxLength: 250));
}
i've found this a good help in gettin into migration ;)