We Keep Coding

How to set ExecuteSqlCommand with Table variable

I have a query I would like to run via C# application. There is no option to do this outside of the application. I have the following code:
var keyGroupsToCleanUp = new List<string>
{
"Address",
"Manufacturer",
"Product",
"Customer",
"Picture",
"Category",
"Vendor",
"SS_A_Attachment",
"SS_A_AttachmentDownload",
"SS_MAP_EntityMapping",
"SS_MAP_EntityWidgetMapping",
};
foreach (var keyGroup in keyGroupsToCleanUp)
{
_databaseFacade.ExecuteSqlCommand($#"
DELETE
FROM GenericAttribute
WHERE KeyGroup = {keyGroup} AND [Key] = 'CommonId'
AND EntityId NOT IN (SELECT Id FROM [{keyGroup}]);
");
}
I want to loop through each name in the List and run the below query for each of them. When I try to do this, I receive the following error:
System.Data.SqlClient.SqlException (0x80131904): Invalid object name '#p1'.
From what I have gathered after searching online, this is because a Table name cannot be a string. You have to declare a variable and use this variable for the table name. I learned that a Table variable has columns that need to be declared and felt a wave of dread wash over me. None of these tables have the same column structure.
Is what I am trying to do possible? If so, how can I do it?
The GenericAttributes table is one large table that consists of six columns.
When I joined the project that this is being used on it had already been used to the point where it was irreplacable. You can save additional data for a database table in here by specifying the KeyGroup as the Database table. We have a table called "Address" and we save additional data in the GenericAttributes table for the Address (It does not make sense, I know). This causes a lot of issues because a relational database is not meant for this. The query I have written above looks for rows in the GenericAttributes Table that are now detached. For example, the EntityId 0 does not exist as an Id in Address, so it would be returned here. That row must then be deleted, because it is linked to a non-existant entityId.
This is an example of a query that would achieve that:
// Address
_databaseFacade.ExecuteSqlCommand(#"
DELETE
FROM GenericAttribute
WHERE KeyGroup = 'Address' AND [Key] = 'CommonId'
AND EntityId NOT IN (SELECT Id FROM [Address]);
");
I have to do this for 11 tables, so I wanted to make it a bit easier to do. Every query is written in the same way. The only thing that changes is the KeyGroup and the table that it looks for. These will both always have the same name.
Here is an example of another call for Products. They are the same, the only difference is the KeyGroup and the Table in the NOT IN statement.
// Product
_databaseFacade.ExecuteSqlCommand(#"
DELETE
FROM GenericAttribute
WHERE KeyGroup = 'Product' AND [Key] = 'CommonId'
AND EntityId NOT IN (SELECT Id FROM Product);
");

To ensure there is no injection vulnerability, you can use dynamic SQL with QUOTENAME
_databaseFacade.ExecuteSqlRaw(#"
DECLARE #sql nvarchar(max) = N'
DELETE
FROM GenericAttribute
WHERE KeyGroup = #keyGroup AND [Key] = ''CommonId''
AND EntityId NOT IN (SELECT Id FROM ' + {0} + ');
';
EXEC sp_executesql #sql,
N'#keyGroup nvarchar(100)',
#keyGroup = {0};
", keyGroup);
Note how ExecuteSqlRaw will interpolate the string. Do not interpolate it yourself with $

At a guess, you're using Entity Framework Core. The ExecuteSqlCommand method accepts a FormattableString, and converts any placeholders into command parameters. But your placeholders appear to be column/table names, which cannot be passed as parameters.
Since there's also an overload which accepts a string, which has different behaviour, this method has been marked as obsolete, and replaced by ExecuteSqlInterpolated and ExecuteSqlRaw.
Assuming none of your values can be influenced by the user, and you're happy that you're not going to introduce a SQL Injection vulnerability, you can use ExecuteSqlRaw instead:
_databaseFacade.ExecuteSqlRaw($#"
DELETE
FROM GenericAttribute
WHERE KeyGroup = [{keyGroup}] AND [Key] = 'CommonId'
AND EntityId NOT IN (SELECT Id FROM [{keyGroup}]);
");

Try following:
foreach (var keyGroup in keyGroupsToCleanUp)
{
var sql = #"DELETE FROM GenericAttribute
WHERE KeyGroup = #Group
AND [Key] = 'CommonId'
AND EntityId NOT IN (SELECT Id FROM #Group)"; // Or [#Group], depends on schema
_databaseFacade.ExecuteSqlCommand(
sql,
new SqlParameter("#Group", keyGroup));
This code assumes, that ExecuteSqlCommand in your facade follows standard Microsoft pattern (same overrides as Microsoft's ones).

Default values doesn't set in SQL Server table when creating by Entity Framework mode

I have lots of tables that contain default values, such as CreatedDateTime (getutcdate()). But right now, the value 0001-01-01 00:00:00.0000000 gets stored instead.
https://stackoverflow.com/a/35093135/7731479 --> that is not effective, I have to do it for each table manually for every database model update (edmx). How can I update all StoreGeneratedPattern to Computed automatically? Or why it does not takes computed automatically?
https://stackoverflow.com/a/43400053/7731479 --> ado.net generates all properties and I can't generate again CreatedDateTime.
Are there any automatic solution?
I am using Entity Framework and ado.net.
Person person = new Person()
{
Id = id,
Name = name,
};
AddToPerson(person);
SaveChanges();
I want to use above. I don't want use the following and assign CreatedDeteTime again because it is assigned in MSSQL with default value getutcdate().
Person person = new Person()
{
Id = id,
Name = name,
CreatedDeteTime = DateTime.UtcNow;
};
AddToPerson(person);
SaveChanges();

The configured default constraint of the SQL Server table will only be applied if you have a SQL INSERT statement that omits the column in question.
So if you insert
INSERT INTO dbo.Person(Id, Name) VALUES (42, "John Doe")
--> then your CreatedDateTime will automatically be set to the GETUTCDATE() value.
Unfortunately, if you have mapped this column in your EF model class, then this is not what happens. If you create an instance of Person in your C# code, and the CreatedDateTime column is in fact part of the model class, then EF will use something like this to insert the new person:
INSERT INTO dbo.Person(Id, Name, CreatedDateTime) VALUES (42, "John Doe", NULL)
and since now NULL is in fact provided for the CreatedDateTime column, that's the value that will be stored - or maybe it's an empty string - no matter what, the column is specified in the INSERT statement and thus the configured default constraint is not applied.
So if you want to let SQL Server kick in with the defaults, you need to make sure not to provide the column(s) in question in the INSERT statement at all. You can do this by:
having a separate model class just for inserts, which does not include those columns in question - e.g. have a NewPerson entity, that also maps to the Person table, but only consists of Name and ID for instance. Since those properties aren't there, EF cannot and will not generate an INSERT statement with them - so then the SQL Server default constraints will kick in
map the INSERT method to a SQL Server stored procedure and handle the inserting inside that procedure, by explicitly not specifying those columns you want to have take on default values

May be I'm wrong, but I have a question.
If you need to save a default date in your DB Table, why you're trying to save another date from programm level? I mean, it's easy to create a procedure and on the procedure level save the date. Something like (select getdate()...).

I have found two solutions:
1- This solution solve for all entities that has same property such as CreatedDateTime
public partial class MyEntities : ObjectContext
{
public override int SaveChanges(SaveOptions options)
{
this.DetectChanges();
foreach (var insert in this.ObjectStateManager.GetObjectStateEntries(System.Data.EntityState.Added))
{
if (insert.Entity.GetType().GetProperty("CreatedDateTime") != null && insert.Entity.GetType().GetProperty("CreatedDateTime").GetType().Name == "DateTime" && (DateTime)(insert.Entity.GetType().GetProperty("CreatedDateTime").GetValue(insert.Entity)) == DateTime.Parse("0001-01-01 00:00:00.0000000"))
insert.Entity.GetType().GetProperty("CreatedDateTime").SetValue(insert.Entity, DateTime.UtcNow, null);
}
return base.SaveChanges(options);
}
}
referance: https://stackoverflow.com/a/5965743/7731479
2-
public partial class Person
{
public Person()
{
this.CreatedDateTime = DateTime.UtcNow;
}
}
referance : DB default value ignored when creating Entity Framework model

How to achieve partial update in Entity Framework 5/6?

I am working on Entity framework with database first approach and I came across below issue.
I have a Customer table with columns col1, col2, col3 ,....,col8. I have created an entity for this table and this table has around 100 records already. Out of above 8 columns, col4 is marked as Non-null.
Class Customer
{
member col1;
member col2;
member col3;
member col4;
.
.
member col8;
}
class Main
{
//main logic to read data from database using EF
Customer obj = object of Customerwith values assigned to col1,col2 and col3 members
obj.col2=some changed value.
DBContext.SaveChanges(); //<- throws an error stating it is expecting value of col4.
}
In my application, I am trying to read the one of the record using the stored procedure using EF and stored procedure only returns col1,col2 and col3.
I am trying to save the modified value of col2 and trying to save back to database using DBContext. But it thows an error stating value of required field col4 is not provided.
FYI: I have gone through couple of forums and question and option to go with disabled verfication on SaveChanges is not feasible for me.
Is there any other way through which I can achieve partial update?

I guess EntityFramework.Utilities satisfies your conditions.
This code:
using (var db = new YourDbContext())
{
db.AttachAndModify(new BlogPost { ID = postId }).Set(x => x.Reads, 10);
db.SaveChanges();
}
will generate single SQL command:
exec sp_executesql N'UPDATE [dbo].[BlogPosts]
SET [Reads] = #0
WHERE ([ID] = #1)
',N'#0 int,#1 int',#0=10,#1=1

disabled verfication on SaveChanges is not feasible for me
Sure it is. You even have to disable validation on Save. But then you can't mark the whole entity as modified, which I think you did. You must mark individual properties as modified:
var mySmallCustomer = someService.GetCustomer(); // from sproc
mySmallCustomer.col2 = "updated";
var myLargeCustomer = new Customer();
context.Customers.Attach(myLargeCustomer);
Entry(myLargeCustomer).CurrentValues.SetValues(mySmallCustomer);
// Here it comes:
Entry(myLargeCustomer).Property(c => c.col2).IsModified = true;
context.Configuration.ValidateOnSaveEnabled = false;
context.SaveChanges();
So you see it's enough to get the "small" customer. From this object you create a stub entity (myLargeCustomer) that is used for updating the one property.

EF 5 Conditional Mapping

I'm using EF 5 Database first approach in my MVC application. all of my tables uses a Field called Deleted which is a boolean field to mark a record is deleted.
I'm trying to get rid of the requirement of having to check Deleted == false every time I query my database. The very straightforward way of doing this is to use a conditional mapping in the edmx file where EF always return data that are not deleted. That's all good.
But the problem of doing this condition mapping is that, when I want to allow the user to delete some record for e.g Address from their address book I don't have access to Delete field from EF as I used it in the conditional mapping and therefore I have to look for another option to allow user to delete a record.
The way I thought is to create a stored proc that handle the delete query and call it when I want to delete the record.
Is there a better way of doing this? Is it possible to make the Delete field accessible even it is used in the conditional mapping?

I have a working solution for Soft Delete in Entity Framework Code First that may help.
The key is that you add a discriminator to every model that you want to be able to soft delete. In code first that is done like this:
modelBuilder.Entity<Foo>().Map(m => m.Requires("IsDeleted").HasValue(false));
This makes it invisible to the context and therefore you have to do the deletes using sql.
If this is the equivalent of your "conditional mapping" in Database First then one way to modify the sql is to override SaveChanges and run sql from there:
public override int SaveChanges()
{
foreach (var entry in ChangeTracker.Entries()
.Where(p => p.State == EntityState.Deleted
&& p.Entity is ModelBase))//I do have a base class for entities with a single
//"ID" property - all my entities derive from this,
//but you could use ISoftDelete here
SoftDelete(entry);
return base.SaveChanges();
}
private void SoftDelete(DbEntityEntry entry)
{
var e = entry.Entity as ModelBase;
string tableName = GetTableName(e.GetType());
Database.ExecuteSqlCommand(
String.Format("UPDATE {0} SET IsDeleted = 1 WHERE ID = #id", tableName)
, new SqlParameter("id", e.ID));
//Marking it Unchanged prevents the hard delete
//entry.State = EntityState.Unchanged;
//So does setting it to Detached:
//And that is what EF does when it deletes an item
//http://msdn.microsoft.com/en-us/data/jj592676.aspx
entry.State = EntityState.Detached;
}
Method used to Get Table Name explained here
That is the way I used to do it. Probably irrelevant to your Database First approach in EF5, but I have now moved to doing it in stored procedures. EF6 Code First generates CreateStoredProcedure calls in Migration files. I replace these with this.CreateDeleteProcedure("dbo.Foo_Delete", "[dbo].[Foos]"); - which is a call to my own extension method:
public static class MigrationExtensions
{
internal static string DeleteSqlFormat
{
//I also hard delete anything deleted more than a day ago in the same table
get { return "DELETE FROM {0} WHERE IsDeleted = 1 AND DATEADD(DAY, 1, DeletedAt) < GETUTCDATE(); UPDATE {0} SET IsDeleted = 1, DeletedAt = GETUTCDATE() WHERE ID = #ID;"; }
}
internal static void CreateDeleteProcedure(this DbMigration migration, string procName, string tableName)
{
migration.CreateStoredProcedure(
procName,
p => new
{
ID = p.Int(),
},
body:
string.Format(MigrationExtensions.DeleteSqlFormat, tableName)
);
}
}

How can I retrieve Id of inserted entity using Entity framework?

I have a problem with Entity Framework in ASP.NET. I want to get the Id value whenever I add an object to database. How can I do this?
According to Entity Framework the solution is:
using (var context = new EntityContext())
{
var customer = new Customer()
{
Name = "John"
};
context.Customers.Add(customer);
context.SaveChanges();
int id = customer.CustomerID;
}
This doesn't get the database table identity, but gets the assigned ID of the entity, if we delete a record from the table the seed identity will not match the entity ID.

It is pretty easy. If you are using DB generated Ids (like IDENTITY in MS SQL) you just need to add entity to ObjectSet and SaveChanges on related ObjectContext. Id will be automatically filled for you:
using (var context = new MyContext())
{
context.MyEntities.Add(myNewObject);
context.SaveChanges();
int id = myNewObject.Id; // Yes it's here
}
Entity framework by default follows each INSERT with SELECT SCOPE_IDENTITY() when auto-generated Ids are used.

I had been using Ladislav Mrnka's answer to successfully retrieve Ids when using the Entity Framework however I am posting here because I had been miss-using it (i.e. using it where it wasn't required) and thought I would post my findings here in-case people are looking to "solve" the problem I had.
Consider an Order object that has foreign key relationship with Customer. When I added a new customer and a new order at the same time I was doing something like this;
var customer = new Customer(); //no Id yet;
var order = new Order(); //requires Customer.Id to link it to customer;
context.Customers.Add(customer);
context.SaveChanges();//this generates the Id for customer
order.CustomerId = customer.Id;//finally I can set the Id
However in my case this was not required because I had a foreign key relationship between customer.Id and order.CustomerId
All I had to do was this;
var customer = new Customer(); //no Id yet;
var order = new Order{Customer = customer};
context.Orders.Add(order);
context.SaveChanges();//adds customer.Id to customer and the correct CustomerId to order
Now when I save the changes the id that is generated for customer is also added to order. I've no need for the additional steps
I'm aware this doesn't answer the original question but thought it might help developers who are new to EF from over-using the top-voted answer for something that may not be required.
This also means that updates complete in a single transaction, potentially avoiding orphin data (either all updates complete, or none do).

You need to reload the entity after saving changes. Because it has been altered by a database trigger which cannot be tracked by EF. SO we need to reload the entity again from the DB,
db.Entry(MyNewObject).GetDatabaseValues();
Then
int id = myNewObject.Id;
Look at #jayantha answer in below question:
How can I get Id of the inserted entity in Entity framework when using defaultValue?
Looking #christian answer in below question may help too:
Entity Framework Refresh context?

You have to set the property of StoreGeneratedPattern to identity and then try your own code.
Or else you can also use this.
using (var context = new MyContext())
{
context.MyEntities.AddObject(myNewObject);
context.SaveChanges();
int id = myNewObject.Id; // Your Identity column ID
}

The object you're saving should have a correct Id after propagating changes into database.

I come across a situation where i need to insert the data in the database & simultaneously require the primary id using entity framework.
Solution :
long id;
IGenericQueryRepository<myentityclass, Entityname> InfoBase = null;
try
{
InfoBase = new GenericQueryRepository<myentityclass, Entityname>();
InfoBase.Add(generalinfo);
InfoBase.Context.SaveChanges();
id = entityclassobj.ID;
return id;
}

Repository.addorupdate(entity, entity.id);
Repository.savechanges();
Var id = entity.id;
This will work.

You can get ID only after saving, instead you can create a new Guid and assign before saving.

All answers are very well suited for their own scenarios, what i did different is that i assigned the int PK directly from object (TEntity) that Add() returned to an int variable like this;
using (Entities entities = new Entities())
{
int employeeId = entities.Employee.Add(new Employee
{
EmployeeName = employeeComplexModel.EmployeeName,
EmployeeCreatedDate = DateTime.Now,
EmployeeUpdatedDate = DateTime.Now,
EmployeeStatus = true
}).EmployeeId;
//...use id for other work
}
so instead of creating an entire new object, you just take what you want :)
EDIT For Mr. #GertArnold :

There are two strategies:
Use Database-generated ID (int or GUID)
Cons:
You should perform SaveChanges() to get the ID for just saved entities.
Pros:
Can use int identity.
Use client generated ID - GUID only.
Pros:
Minification of SaveChanges operations.
Able to insert a big graph of new objects per one operation.
Cons:
Allowed only for GUID

When you use EF 6.x code first
[DatabaseGenerated(DatabaseGeneratedOption.Identity)]
public Guid Id { get; set; }
and initialize a database table, it will put a
(newsequentialid())
inside the table properties under the header Default Value or Binding, allowing the ID to be populated as it is inserted.
The problem is if you create a table and add the
[DatabaseGenerated(DatabaseGeneratedOption.Identity)]
part later, future update-databases won't add back the (newsequentialid())
To fix the proper way is to wipe migration, delete database and re-migrate... or you can just add (newsequentialid()) into the table designer.

I am using MySQL DB & I have an AUTO_INCREMENT field Id .
I was facing the same issue with EF.
I tried below lines, but it was always returning 0.
await _dbContext.Order_Master.AddAsync(placeOrderModel.orderMaster);
await _dbContext.SaveChangesAsync();
int _orderID = (int)placeOrderModel.orderMaster.Id;
But I realized my mistake and corrected it.
The Mistake I was doing: I was passing 0 in my orderMaster model for Id field
Solution worked: Once I removed the Id field from my orderMaster model, It started working.
I know it was very silly mistake, but just putting here if anyone is missing this.