I use code first(entity framework). when i call DbContext.SaveChanges(), I get a primary key constraint because here is already such a record. I want to override the record if it already exists in the database. Which is the simplest way to do this. Checking each time I call savechanges for repeting primary keys is too hard in my project. Thanks
Unfortunately EF demands that you know if you are adding or modifing entity. So only ways are:
Keep this knowledge in your application and set proper EntityState in ObjectStateManager.
Load entity first. If exists modify its data, if doesn't create new entity and add it to context.
Related
I am beginning playing with EF 6.x and I've got an annoying issue.
I've designed a database with simple User,Role,Permissions tables each one bond to another with classic many-to-many relation.
EF wants me to create primary keys in indirect tables (UserPermissions etc.)
but also wants me to make some mapping for those Primary Keys but I have no clue why.
Is there something wrong with my design or there Is a workaround to this issue?
Maybe Can I in some way use Unique index instead of Primary Key to satisfy EF?
There you have diagram to make it more clear:
Well, I'm embarrased that I've not came up to solution earlier.
Just added multicolumn Primary Key on columns referencing direct tables via foreign key.
e.g. CONSTRAINT pk_IdRole_IdPermission_RolePermissions PRIMARY KEY (IdRole,IdPermission)
I'm using Entity Framework 5, code first approach. As there's no built in support for updating child entities in disconnected scenario, I'm building my own mechanism to do that. At some point I need to get the property of an entity with which it has a Foreign Key relationship with another (principal) entity. I've tried to get access to CSpace through
((IObjectContextAdapter)dbContext).ObjectContext.MetadataWorkspace.GetItems<MyEntity>(System.Data.Entity.Core.Metadata.Edm.DataSpace.CSpace)
but here I got a warning that said there's no implicit conversion between MyEntity and System.Data.Entity.Core.Metadata.Edm.GlobalItem.
I can't look for a property that has Foreign Key attribute because in most of my entities I use EF convention to get foreign keys automatically. So how one would go about finding which property is used for foreign key relationship.
Thanks to #octavioccl's post I was able to do what I want. So I was in the right path to look inside ObjectContext.
I'm getting the following error when I SaveChanges after Removing an entity that has related entities containing data in the context as well. (The entity I'm deleting has the unique Primary Key). I have Cascading Delete configured at the SQL Server database level for the relation between the primary key table and the foreign key table.
"The operation failed: The relationship could not be changed because one or more of the foreign-key properties is non-nullable. When a change is made to a relationship, the related foreign-key property is set to a null value. If the foreign-key does not su...
The primary entity has its related data loaded explicitly prior to me removing the primary from the context. I assumed that EF and SQL Server would take care of the cascaded delete for me. If the related entity has no data the delete (of the primary entity) works fine. If there is data in the related entity, I get the error above.
Any suggestions?
The lesson is that EF has a learning curve. Keep exploring.
Anyway, I was "deleting" entities by setting the state to "Deleted" then calling SaveChanges. This seems to work fine if the entity has no related data. However, if you have an entity that has related "child" entities, you need to call Remove on the parent entity for the "delete" to cascade through the graph. Live and learn. I'm pretty sure this is the answer. I think for the time being I'm going to stop setting State for deletes and use Remove instead.
I want to create an entity in EF without a public key. The backing table has got a non-unique clustered key, but the data in the table conceptually doesn't have a unique primary key it can use.
It looks like EF really doesn't like this. Is there any way of getting EF to accept that the table has no primary key and make it work with it anyway, with no performance hit? I don't care if the result is read-only.
As I understand it, as the Entity Framework is based on the Domain Driven Design concept of Entities, each Entity by definition must have a unique identifier. If the concept which the data in your table represents does not conceptually have a unique identifier then it is not an Entity, in the sense intended by the framework.
With this in mind I'd define a Stored Procedure, make it available through my object context, then make the objects encapsulating this data available via a class which lazy-loads the data, manually maps it into the objects you're using and presents it in a read-only manner.
You may also be able to accomplish this by exposing a view and then mapping your entity to the view.
I see some posts on programmatically mapping a table with a composite key but I haven't found any examples to do that within xml of the edmx file. How do I do this? In the edmx of my entity type I have a key section with multiple propertyref's pointing to my composite key. However, when I go to save my changes it errors and states that there is no update function.
It seems to me, if I mark one key as the primary key and it can manage that it should be able to use a composite key as well.
Please do not reply and state that I shouldn't be using composite keys because I don't care and have no ability to change it. I didn't develop this database as it is a third-party ERP application that I'm interfacing with.
From what I can tell I may have to create a stored procedure just to save it? Isn't there a better way?
Thank You.