I'm trying to set Id manually for some entries.
I have a Customer class with an Identity field
modelBuilder.Entity<Customer>().Property(f => f.Id).HasDatabaseGeneratedOption(System.ComponentModel.DataAnnotations.Schema.DatabaseGeneratedOption.Identity);
and in the insert part I have
using (var transaction = ctx.Database.BeginTransaction())
{
ctx.Database.ExecuteSqlCommand("SET IDENTITY_INSERT [dbo].[Customers] ON");
ctx.Customers.AddRange(customersList);
ctx.SaveChanges();
ctx.Database.ExecuteSqlCommand("SET IDENTITY_INSERT [dbo].[Customers] OFF");
transaction.Commit();
}
but I'm still getting the exception
Explicit value must be specified for identity column in table
'Customers' either when IDENTITY_INSERT is set to ON or when a
replication user is inserting into a NOT FOR REPLICATION identity
column.
In the StateEntries of my exception, I can see that all the fields that are stored in another table like "Contacts" seems to be the cause
customer.Contacts = new List<Contact> {new Contact {Name = "Test", … }}
How can I make it work?
That error message is misleading. You are succeeding in turning IDENTITY_INSERT on, but EF is still generating an INSERT statement under the assumption that the key value will be generated by the IDENTITY column.
So you additionally must generate an INSERT statement containing the key value. You can, of course, use .ExecuteSqlCommand() to perform the INSERT as well. Or you can get EF to do it, but you must use a model where that Entity does not have database-generated keys. OnModelCreating configures the model only once per ModelType, so you can't just put a switch in there.
Instead you can create a subtype of your DbContext for seeding, overriding the entity configuration of the base DbContext. Like this:
class Db_Seed: Db
{
public Db_Seed(string constr) : base(constr)
{
Database.SetInitializer<Db_Seed>(null); //no initializer
}
protected override void OnModelCreating(DbModelBuilder modelBuilder)
{
base.OnModelCreating(modelBuilder);
modelBuilder.Entity<Customer>().Property(c => c.Id).HasDatabaseGeneratedOption(DatabaseGeneratedOption.None);
}
}
You still must set IDENTITY_INSERT on (since the table does have an IDENTITY), but with this context EF will not generate an INSERT statement that assumes the server will generate the key.
Related
I found related question but my issue seems to be different.
Running the following code:
var dbitem = context.MyDatabaseItems.Single(p => p.Id == someId);
context.Update(dbitem);
context.SaveChanges();
Results in "Cannot update identity column 'Id'". Table behind is a bit special. "Id" is NOT the primary key for different reasons. Primary key consists of combination of other fields. No matter what I do: detaching, reattaching etc etc the existing item I am unable to save the entity even if I do not change it (see the code).
However this Id is unique and auto generated.
The builder is the following:
builder.Property(p => p.Id)
.ValueGeneratedOnAdd();
builder.HasKey(p => new { p.BusinessDay, p.ClientId, p.Version });
BusinessDay is dateTime, CLientId and Version are integers.
What is going on here?
There are two metadata properties which control the update behavior called BeforeSaveBehavior and AfterSaveBehavior.
For auto generated keys the later is assumed to be Ignore, i.e. never update. For non key auto generated properties it must be configured explicitly (note that there is no fluent API for that so far, so you have to use the metadata API directly), e.g.
// First define the new key
builder.HasKey(p => new { p.BusinessDay, p.ClientId, p.Version });
// Then configure the auto generated column
// This (especially the `SetAfterUpdateBehavior` call) must be after
// unassociating the property as a PK, otherwise you'll get an exception
builder.Property(p => p.Id)
.ValueGeneratedOnAdd()
.Metadata.SetAfterSaveBehavior(PropertySaveBehavior.Ignore); // <--
This does not change the database schema (model), hence no migration is needed. Just the EF Core update entity behavior.
My code:
Models.Resource r = new Models.Resource();
r.Name = txtName.Text;
r.ResourceType = resTypes.Find(rt => rt.Name == "Content");
r.ResourceContents.Add(_resourceContent.Find(rc => rc.ID == _resourceContentID));
ctx.Resource.Add(r);
ctx.SaveChanges();
ctx.SaveChanges() causes the error:
Cannot insert explicit value for identity column in table 'Resources' when IDENTITY_INSERT is set to OFF.
Looking at what's being sent to SQL:
ADO.NET:Execute NonQuery "INSERT [dbo].[Resources]([ID], [Name], [Description], [IsOnFile],
[ContentOwnerAlias], [ContentOwnerGroup], [ResourceTypes_ID])
VALUES (#0, #1, #2, #3, #4, #5, NULL)"
My POCO Resource has ID as a Key:
public partial class Resource
{
public Resource()
{
}
[Key]
public int ID { get; set; }
And my Map code:
public class ResourceMap : EntityTypeConfiguration<Resource>
{
public ResourceMap()
{
// Primary Key
this.HasKey(t => t.ID);
How do I tell Entity to not send the Key ID field to the database?
If your PK is generated by the database (like an identity) you have to configure it in your Map.
public class ResourceMap : EntityTypeConfiguration<Resource>
{
public ResourceMap()
{
// Primary Key
this.HasKey(t => t.ID);
this.Property(t => t.ID).HasDatabaseGeneratedOption(DatabaseGeneratedOption.Identity);
}
}
You do not need the HasKey(t => t.ID) Fluent API mapping or the [Key] Data Attribute because by convention EF will assume that an integer field named ID is the key and is database generated.
As an aside, I'd recommend that when you are not following conventions you should choose one method or the other - otherwise you are repeating yourself and when you want to change something you need to change it in 2 places.
I'm not sure why the field in the database isn't already database generated - maybe when you define the field via the fluent api you have to specify that too. What I do know is that in order to make EF change a key field to be database generated you will need to drop the table.
So - rollback the migration or drop the table / database, then remove the data attribute, remove the fluent mapping and recreate.
This issue is currently on a "backlog" in the entity framework. If you want to vote for it you can do that here: Migrations: does not detect changes to DatabaseGeneratedOption
Other References:
Identity problem in EF
Switching Identity On/Off With A Custom Migration Operation
In my WCF service I am using Entity Framework .NET 4.0,
in my database i have this table:
CREATE TABLE [dbo].[Tracking](
[TrackingID] [uniqueidentifier] ROWGUIDCOL NOT NULL,
...
CONSTRAINT [PK_Tracking] PRIMARY KEY CLUSTERED
(
[TrackingID] ASC
)
) ON [DATA]
ALTER TABLE [dbo].[Tracking] ADD CONSTRAINT [DF_Tracking_TrackingID] DEFAULT (newid()) FOR [TrackingID]
GO
when i insert a record Entity framewrok has prepopulated the TrackingID as "00000000-0000-0000-0000-000000000000". I have setting field property to be Computed and Identity bu no such luck any ideas: here is my code snippet:
using (var context = new DB.PTLEntities())
{
var tracking = new DB.Tracking();
context.Trackings.AddObject(tracking);
context.SaveChanges();
trackingID = tracking.TrackingID;
}
Looks like EF doesnt think this column is an identity column so its putting in default(Guid). Take a look here for some details on making a guid an identity column (it actually goes through your exact example) http://leedumond.com/blog/using-a-guid-as-an-entitykey-in-entity-framework-4/
I got same issue. Following solution worked for me.
There is a need to change the StoreGeneratedPattern value for the property in the EF designer. Set it to Identity. This will cause EF to avoid setting the value on an insert, then capture it after the insert is complete.
If your EF version is 4, then you may have trouble with this using the designer only. You may have to edit the .edmx manually and set the StoreGeneratedPattern in the storage model itself.
Example:
< Property Name="EmpId" Type="uniqueidentifier" Nullable="false" StoreGeneratedPattern="Identity" />
In my case (EF >= 4.1) I needed to add a default to the table itself:
ALTER TABLE MyTable ADD DEFAULT (newid()) FOR [Id]
Also if you're using your own context, you should tell EF that the column is the identity column:
public class ShedContext : DbContext
{
protected override void OnModelCreating(DbModelBuilder modelBuilder)
{
modelBuilder.Entity<DraftFromWebOneOff>()
.Property(d => d.Id)
.HasDatabaseGeneratedOption(DatabaseGeneratedOption.Identity);
base.OnModelCreating(modelBuilder);
}
}
I am using Entity Framework 4.3 and using Code Fist.
I have a class
public class User
{
public int UserId{get;set;}
public string UserName{get;set;}
}
How do I tell Entity Framework that UserName has to be unique when creating database table?
I would prefer to use data anotations instead of configuration file if possible.
In Entity Framework 6.1+ you can use this attribute on your model:
[Index(IsUnique=true)]
You can find it in this namespace:
using System.ComponentModel.DataAnnotations.Schema;
If your model field is a string, make sure it is not set to nvarchar(MAX) in SQL Server or you will see this error with Entity Framework Code First:
Column 'x' in table 'dbo.y' is of a type that is invalid for use as a key column in an index.
The reason is because of this:
SQL Server retains the 900-byte limit for the maximum total size of all index key columns."
(from: http://msdn.microsoft.com/en-us/library/ms191241.aspx )
You can solve this by setting a maximum string length on your model:
[StringLength(450)]
Your model will look like this now in EF CF 6.1+:
public class User
{
public int UserId{get;set;}
[StringLength(450)]
[Index(IsUnique=true)]
public string UserName{get;set;}
}
Update:
if you use Fluent:
public class UserMap : EntityTypeConfiguration<User>
{
public UserMap()
{
// ....
Property(x => x.Name).IsRequired().HasMaxLength(450).HasColumnAnnotation("Index", new IndexAnnotation(new[] { new IndexAttribute("Index") { IsUnique = true } }));
}
}
and use in your modelBuilder:
protected override void OnModelCreating(DbModelBuilder modelBuilder)
{
// ...
modelBuilder.Configurations.Add(new UserMap());
// ...
}
Update 2
for EntityFrameworkCore see also this topic: https://github.com/aspnet/EntityFrameworkCore/issues/1698
Update 3
for EF6.2 see: https://github.com/aspnet/EntityFramework6/issues/274
Update 4
ASP.NET Core Mvc 2.2 with EF Core:
[DatabaseGenerated(DatabaseGeneratedOption.Identity)]
public Guid Unique { get; set; }
EF doesn't support unique columns except keys. If you are using EF Migrations you can force EF to create unique index on UserName column (in migration code, not by any annotation) but the uniqueness will be enforced only in the database. If you try to save duplicate value you will have to catch exception (constraint violation) fired by the database.
In EF 6.2 using FluentAPI, you can use HasIndex()
modelBuilder.Entity<User>().HasIndex(u => u.UserName).IsUnique();
From your code it becomes apparent that you use POCO. Having another key is unnecessary: you can add an index as suggested by juFo.
If you use Fluent API instead of attributing UserName property your column annotation should look like this:
this.Property(p => p.UserName)
.HasColumnAnnotation("Index", new IndexAnnotation(new[] {
new IndexAttribute("Index") { IsUnique = true }
}
));
This will create the following SQL script:
CREATE UNIQUE NONCLUSTERED INDEX [Index] ON [dbo].[Users]
(
[UserName] ASC
)
WITH (
PAD_INDEX = OFF,
STATISTICS_NORECOMPUTE = OFF,
SORT_IN_TEMPDB = OFF,
IGNORE_DUP_KEY = OFF,
DROP_EXISTING = OFF,
ONLINE = OFF,
ALLOW_ROW_LOCKS = ON,
ALLOW_PAGE_LOCKS = ON
) ON [PRIMARY]
If you attempt to insert multiple Users having the same UserName you'll get a DbUpdateException with the following message:
Cannot insert duplicate key row in object 'dbo.Users' with unique index 'Index'.
The duplicate key value is (...).
The statement has been terminated.
Again, column annotations are not available in Entity Framework prior to version 6.1.
Note that in Entity Framework 6.1 (currently in beta) will support the IndexAttribute to annotate the index properties which will automatically result in a (unique) index in your Code First Migrations.
Solution for EF4.3
Unique UserName
Add data annotation over column as:
[Index(IsUnique = true)]
[MaxLength(255)] // for code-first implementations
public string UserName{get;set;}
Unique ID
,
I have added decoration [Key] over my column and done.
Same solution as described here: https://msdn.microsoft.com/en-gb/data/jj591583.aspx
IE:
[Key]
public int UserId{get;set;}
Alternative answers
using data annotation
[Key, DatabaseGenerated(DatabaseGeneratedOption.Identity)]
[Column("UserId")]
using mapping
mb.Entity<User>()
.HasKey(i => i.UserId);
mb.User<User>()
.Property(i => i.UserId)
.HasDatabaseGeneratedOption(DatabaseGeneratedOption.Identity)
.HasColumnName("UserId");
I am using the Entity Framework to update my database.
The Employee table has an employeeId primary key field.
When I instantiate an employee object, the employeeId defaults to zero.
I want to insert the employee object into the database, ignoring the primary key value of zero.
Yet I get this exception;
Cannot insert explicit value for identity column in table 'Employees' when IDENTITY_INSERT is set to OFF.
What should I be doing to stop this?
public void Add()
{
using (SHPContainerEntities db = new SHPContainerEntities())
{
// Set default values
this.StartDate = DateTime.Now;
// Start date now
this.SHP_UserRoleId = db.SHP_UserRoles.Where(x => x.RoleName == "Employee").Single().UserRoleId;
// Default user role is "Employee". This can be changed later.
this.DepartmentId = 1;
// This is a temporary fix.
db.AddToEmployees(this);
//db.Employees.AddObject(this);
db.SaveChanges();
}
}
I fixed the problem.
I updated the database and I forgot to update my edmx file to reflect that change.