Ok. So here is the deal. I have two entities - "Product" and "Parts". The product consists of parts. And parts are reusable in other products. The relation between those entities is many-to-many. And it all works great.
The problem is that I cannot add the same part to the same product twice. EF seems to force all the related entities to be unique. Consider the following code:
var product = context.Create<Product>();
var part = GetSomePart();
Console.WriteLine(product.Parts.Count); // will output 0
// Add a part
product.Parts.Add(part);
Console.WriteLine(product.Parts.Count); // will output 1
// Add the same part again
product.Parts.Add(part);
Console.WriteLine(product.Parts.Count); // will output 1!
So ok, I get the point - avoid duplicates or something. But I need this to be possible. Is there a way to do this (to tell EF to stop enforcing unique values) without creating an additional table? Or is the only way to resolve this is to manually add the intermediate table and handle the many-to-many myself?
In this case you will have to create another table called "ProductParts" which will have an identity unique key, and which can hold references to both product and part, and they can be multiple too.
In the second add statement it will not add another object because part is already in added state. So you need to create a new object with same properties add it again.
product.Parts.Add(new Part{someProperty=part.someProperty ... ect });
if you want to reduce code you can use Automapper ( http://automapper.codeplex.com/ )to copy all properties,
product.Parts.Add(Mapper.Map<Part,Part>(part));
Related
I am using Entity Framework database-first model. I have 2 tables that are referenced by a link table.
For example:
When I update my edmx file from my database this creates the expected model:
Now what I want to do is delete an entry from the Product_User table without deleting the referenced entry in either of the related tables (Product or User).
I've tried both of these statements (together and separately) but neither seems to have any effect:
user.Products.Clear();
foreach (var product in products)
{
product.User = null;
}
Is what I'm trying to do possible with the model the way I have it now? And if so what am I doing wrong?
I noticed I can do what I'm trying to do if I add the link table explicitly to the model but I'm trying to avoid that.
Any help would be much appreciated. Thanks.
Your seconde way to go (with a loop) seems more like the correct way.
Instead of = null try to use the .Remove() function. I guess the remove function only remove the link between the two entity, not the entities related.
See this answer : Removing many to many entity Framework
Problem: When adding an object "Order" to my dbcontext, all nested objects of the order gets "readded" to the database, though the nested objects is static data and only a reference shoudl be added in the database.
Example:
The database holds 0 orders, and 3 items.
I add one order with 2 items.
Now the database hold 1 order, and 5 items. The two items in the order has been "readded" to the database, even though the items had the right primary keys before db.SaveChanges().
I realize that i may be able to attach the existing items to the dbcontext before saving changes, but is that really the only way to go? Can't EF figure out that to item already exists when the primary key matches an existing item?
Does anyone know if this is different in the new version of EF CodeFirst?
No EF cannot figure if entities are existing one or new one - both Add and Attach commands are graph oriented operations. You call them on one entity in the graph and they traverse all relations (and their relations and so on) and perform the operation for them as well.
You must figure correct state of each entity in the graph for example by using:
dbContext.Orders.Add(newOrder);
foreach(var item in newOrder.Items) {
dbContext.Entry(item).State = EntityState.Unchanged;
}
dbContext.SaveChanges();
You can use the reverse operation by calling Attach(newOrder) and set the order to Added state. The main difference will come with independent associations (for example many-to-many relations). The first approach will correctly add new relation between order and each item whereas second will not unless you manually set each relation to Added state (and changing state for relations is more complex).
I want to have an ObjectQuery that returns tracked entities (not static data), but I don't want it to load all the columns, I want some columns to load as null, I don't want to use select, since this will return an IEnumerable of the values, not tracked objects.
Is there a way to do it?
If yes, how do I then complete reloading those columns on demand?
Have you tried creating a view and then mapping the view?
By creating a view you can select the columns that you really want and only those will show up on the Entity Model.
I think the only way is to create new entity type which will not contain columns you don't need. You will map this entity type to the same table. On demand (lazy) loading works only for navigation properties.
Edit:
My previous idea doesn't work but in some special cases you can use idea from this article. Instead of modeling single entity from single table you will model multiple entities related with 1:1 relations. Entities will not overlap in properties (except the primary key) as my previous idea assumed because it doesn't work. You will than have main entity with fields you want to load immediately and related entities which will be lazy loaded when needed.
I'm building an ASP.NET MVC site using the ADO.NET Entity Framework.
I have an entity model that includes these entities, associated by foreign keys:
Report(ID, Date, Heading, Report_Type_ID, etc.)
SubReport(ID, ReportText, etc.) - one-to-one relationship with Report.
ReportSource(ID, Name, Description) - one-to-many relationship with Sub_Report.
ReportSourceType(ID, Name, Description) - one-to-many relationship with ReportSource.
Contact (ID, Name, Address, etc.) - one-to-one relationship with Report_Source.
There is a Create.aspx page for each type of SubReport. The post event method returns a new Sub_Report entity.
Before, in my post method, I followed this process:
Set the properties for a new Report entity from the page's fields.
Set the SubReport entity's specific properties from the page's fields.
Set the SubReport entity's Report to the new Report entity created in 1.
Given an ID provided by the page, look up the ReportSource and set the Sub_Report entity's ReportSource to the found entity.
SaveChanges.
This workflow succeeded just fine for a couple of weeks. Then last week something changed and it doesn't work any more. Now instead of the save operation, I get this Exception:
UpdateException: "Entities in 'DIR2_5Entities.ReportSourceSet'
participate in the 'FK_ReportSources_ReportSourceTypes' relationship.
0 related 'ReportSourceTypes' were found. 1 'Report_Source_Types' is expected."
The debug visualizer shows the following:
The SubReport's ReportSource is set and loaded, and all of its properties are correct.
The Report_Source has a valid ReportSourceType entity attached.
In SQL Profiler the prepared SQL statement looks OK. Can anybody point me to what obvious thing I'm missing?
TIA
Notes:
The Report and SubReport are always new entities in this case.
The Report entity contains properties common to many types of reports and is used for generic queries. SubReports are specific reports with extra parameters varying by type. There is actually a different entity set for each type of SubReport, but this question applies to all of them, so I use SubReport as a simplified example.
I realise I'm late to this, but I had a similar problem and I hacked through it for about 3 hours before I came up with a solution. I'd post code, but it's at home - I can do it later if someone needs it.
Here are some things to check:
Set a breakpoint on the SaveChanges() call and examine the object context in depth. You should see a list of additions and changes to the context. When I first looked, I found that it was trying to add all my related objects rather than just point to them. In your case, the context might be trying to add a new Report_Source_Type.
Related to the previous point, but if you're retrieving the report source, make sure it is being retrieved from the database by its entity key and properly attached to the context. If not, your context might believe it to be a new item and therefore its required relationships won't be set.
From memory, I retrieved my references using the context.GetObjectByKey method, and then explicitly attached those objects to the context using the context.Attach method before assigning them to the properties of my original object.
I got this error because the table didn't have a primary key, it had a FK reference, but no PK.
After adding a PK and updating the model all is well.
Check if your ReportSource was loaded with the NoTracking option or if its EntityState == 'Detached'. If so, that is your problem, it must be loaded in the context.
This tends to happen if your database tables have a 1 - 1 relationship with each other. In your example reportsourceset expects a reportsorttypes with whatever id it is referencing. I have run into this problem when my relationship is linking two primary keys from opposite tables together.
I've got the same error because of new object instance which created "behind the scene" in "Added" state. This was not obvious.
I got this error when I added the new entity to the context but forgot to add the new entity to its parent's collection in the object graph.
For example:
Pet pet = new Pet();
context.Pets.Add(pet);
// forgot this: petOwner.Pets.Add(pet);
Error 3007: Problem in Mapping Fragments starting at lines 186, 205: Non-Primary-Key column(s) [WheelID] are being mapped in both fragments to different conceptual side properties - data inconsistency is possible because the corresponding conceptual side properties can be independently modified.
I found several places on the web describing this error, but I simply don't understand them. (confused smiley goes here)
One
Two
Three
Four
There is something pretty fundamental here, I must be missing. Can you explain it, so that I understand it? Maybe using my real life example below?
Foreign key 1:N Wheels.Id -> Slices.WheelId
I add them to entity framework, and WheelId is not visible in the Slices-entity.
Doing some workaround (deleting the relationship from the db before adding tables to EF - then re-creating it and updating EF) I managed to get the WheelId to stay in Slices, but then I get the error mentioned at the top.
Since Slices.WheelId is an FK, you cannot expose it in your client model, period. There are ways to get the value, though.
var wheelId = someSlice.Wheels.ID;
Update In EF 4 you can do this by using FK Associations instead of independent associations.
Try to remove foreign property column from Entity set using entity model design it will solve your problem
For example
We have two tables one is customer and other one is order, using entity model design we added association between customers and orders when we do this Ado.net entity framework i will add navigation properties to both below tables.
Like
Customer.Orders - Here order is list
Order.Customer
One - Many relation.
So we need to remove property from with name CustomerId[Foreign key column] from Order entity set.
For reference:
http://social.msdn.microsoft.com/forums/en-US/adodotnetentityframework/thread/2823634f-9dd1-4547-93b5-17bb8a882ac2/
I was able to overcome this problem by the following steps:
right click the designer window
Select 'update model from database'
Select Add AND make sure that the 'Include foreign key columns in the model' checkbox is selected.
click on Finish...
I had set foreign keys up in the database but framework still wasn't pulling them in correctly. So I tried to add the association myself.
However, when I did this I would get a mapping error. It took me A WHILE but I figured out. What I did was set up the association using the entity toolbox association tool and then you have to double click on the association (1 to many) line and set the primary and foreign key there. Hopefully, this to help others who might have the same problem. I couldn't find the answer anywhere.
I had this problem for quite a different reason, and the message was slightly different; it didn't say "data inconsistency is possible because the corresponding conceptual side properties can be independently modified."
I have a table involved in my model with a binary column where I store image data. I only want this data returned when I need it (performance is a feature), so I split the table using a method similar to this. Later on, I added a property to that table, then updated the model from the database. The wizard added the property to both entity types that refer to the table with the added property. I had to delete it from one of them to solve the error.
I've had this happen because Entity Framework Update wizard mismapped some keys (or did not update?). As a result, some columns were mistakenly labeled as keys, while actual key columns were treated as plain columns.
The solution was to manually open EDMX file, find the entities, and update the keys.
Couldn't get any of the answer to work with EF6. The problem seems to be the framework doesn't import the foreign keys correctly as Associations. My solution was removing foreign keys from the tables, and then manually adding the associations using Entity Framework model, using the following steps: Entity Framework - Add Navigation Property Manually
For LinQ to Entities queries in EF1, my workaround for not having access to the foreign key as a property is with the following code, which does not produce a join query to the associated table:
dbContext.Table1s.FirstOrDefault(c => (int?)c.Table2.Id == null)
i.e, the generated SQL is:
...WHERE ([Extent1].[Table2Id] IS NULL)...
Solution is to allow deleting Rule = Cascade on Sql association.
Same thing as to be done on .edmx model, adding element to
association:
<Association Name="FK_Wheels_Slices">
<End Role="Wheels" Type= "your tipe here" Multiplicity="1">
<OnDelete Action="Cascade" />
</End>
</Association>
I had a table already mapped in EF. I added two more tables which had foreign keys in the previously added table. I then got the 3007 error.
To fix the error I deleted all three tables from the EDMX file, and then re-added them all at once together (via "Update Model from Database..."), instead of in stages.
I checked my Error List window and noticed I had errors in the model. Fixed them and all is well
in my case I solved this error by tick (include foreign key columns in the model)
- update Model from database
- tick (include foreign key columns in the model)
- finish