I am working on a project with the following many-to-many model:
Book: can have multiple tags (book.tags)
Tag: can include a lot of books (tag.books)
i found that when a book is already in database and i just simply want to add a tag to that book by doing [bookMO addTagsObject:tag], i will get a fault in book.tags. using the instrument, i found the core data is trying to do "[NSObject(NSKeyValueObserver(Notification) willChangeValueForKey:withSetMutation:usingObjects]".
i also checked the actual sql getting executed and i found:
annotation: to-many relationship fault "tags" for objectID 0x20140b00 fulfilled from database.
and then a sql query that returns all the tags that contain this book. i found internally core data use a join table of Book and Tag. the primary key of that join table is simply a combination of book_id and tag_id. That joint table is not indexed. And to get all the tags that contain a book, it seems to be (i am not sure here) traversing all the rows in that joint table and therefore this operation is very expensive.
i think what is happening in my app now is, every time i want to add a tag to a book, i will have to do that linear scan in the joint table. The overall complexity is O(N^2) because each operation will do a linear scan in the join table. And I am having 10k books now and the performance is not so good..
is there any way i can avoid that tags fault triggered by kvo? or is there any way i can implement my own join table that can return the result in O(1) by indexing?
-Erben
sorry for the delayed update.
Core data is just not good for many-to-many relationships as it doesn't create the index that I want.
I ended up creating a RleationManagedObject. which has a property "book" pointing to a BookManagedObject and a property "tag" pointing to a TagManagedObject. both property are indexed (very important, default core data is not indexing both properties).
Every time I need to add "tag_1" on "book_1", I search all the RelationManagedObject with "book_1" and "tag_1". if not found, I create this relationManagedObject. I can also create relationManagedObject with <"book_1", "tag2">, or <"book_2", "tag_1">, etc. Basically I am managing the relationship by myself.
so the gotcha is core data many-to-many relationship doesn't index properly.
Also, xcode can print out the sql that it executed. The first time you opened the app it will create those tables that you defined in core data. so you can see how CoreData Framework create the default "Relationship" table without indexing properly. and thats how I found the issue.
Related
I'm using .Net 4.5, entity framework 5, database first. I have a junction (many-to-many) table in my database. For this example lets say the table is "StudentsCourses":
Students
-PkStudentId
-Name
Courses
-PkCourseId
-CourseNumber
StudentsCourses
-FkStudentId
-FkCourseId
This works just fine right now. The 'generate model from database' creates a Student entity with a navigation property to the Course entity. But here is where the trouble is:
I need to add another column to the StudentsCourses table. Lets just call this column "CourseYear". So our junction table would now look like this:
StudentsCourses
-FkStudentId
-FkCourseId
-CourseYear
So, I've added this column to the database and ran "Update Model from Database" on the edmx. I would expect to see an entity created for StudentCourses, with a navigation property to both Students and Courses. But no such entity is created. I still see the same two tables (Students & Courses) with the same navigation property as before.
I've done a lot of reading and researching, but haven't really come across a clear-cut answer. There is a wealth of information on code-first which I can't apply to my scenario. Is there a way to get what I'm after? Is it as simple as adding a PkId to the StudentCourses table? SQL Replication is preventing me from doing this. I would think the composite should suffice, but maybe EF needs a PK to do it's magic? I read a little bit about manually setting relationships, but could not find anything speaking to my particular situation. It could be that I am just missing a simple step in the process of updating the edmx from database. I've done this plenty of times when: adding new tables, adding columns, deleting columns, etc. I'm following the same steps as I always do, but maybe I need to do something different in this case?
Thanks ahead of time for any help. It is greatly appreciated. Please let me know if any more information would help.
From what I've gathered it appears as though EF will not generate a model for a table that doesn't have a Primary Key.
I'm a bit late for this, but you have the answer in this thread Updating Entity Framework Model after adding a field to a previous look up only table
As they say here, you have to delete the relationship between Students and Courses in the designer. Then update your model from the database, and make sure StudentsCourses table is checked in the Tables branch of the Add tab.
Just starting out with Entity Framework (Code First) and I have to say I am having a lot of problems with it when loading SQL data that is fairly complex. For example, let's say I have the following tables which stores which animals belongs to which regions in the world and the animal are also categorized.
Table: Region
Id: integer
Name string
Table AnimalCategory
Id integer
Name: string
RegionId: integer -- Refers back Region
Table Animal
Id integer
AnimalCategoryId integer -- Refers back AnimalCategory
Let's say I want to create a query with Entity Framework that would load all Animals for a specific region. The easiest thing to do is to create 3 Entities Region, AnimalCategory, and Animal and use LINQ to load the data.
But let's say I am not interested in loading any AnimalCategory information and define an Entity class just to represent AnimalCategory so that I can do the JOIN. How can I do this with Entity Framework? Even with many of its Mapping functions I still don't think this is possible.
In non Entity Framework solutions this is easy to accomplish by using INNER JOINs in SPs or inline SQL. So what are my options in Entity Framework? Shall I pollute my data model with these useless tables just so I can do a JOIN?
It's a matter of choice I guess. EF choose to support many-to-many associations with transparent junction tables, i.e. where junction tables only have two foreign keys to the associated entities. They simply didn't choose to support this far less common "skipping one-to-many-to-many" scenario in a similar manner.
And I can imagine why.
To start with, in a many-to-many association, the junction table is nothing but that: a junction, an association. However, in a chain of one-to-many (or many-to-one) associations it would be exceptional for any of the involved tables to be just an association. In your example...
Animal → AnimalCategory → Region
...AnimalCategory would only have a primary key (Id) and a foreign key (RegionId). That would be useless though: Animal might just as well have a RegionId itself. There's no reason to support a data model that doesn't make sense.
What you're after though, is a model in which the table in the middle does carry information (AnimalCategory.Name), but where you'd like to map it as a transparent junction table, because a particular class model doesn't need this information.
Your focus seems to be on reading data. But EF has to support all CRUD actions. The problem here would be: how to deal with inserts? Suppose Name is a required field. There would be no way to supply its value.
Another problem would be that a statement like...
region.Animals.Add(animal);
...could mean two things:
add an Animal and a new AnimalCategory, the latter referring to the Region.
Add an Animal referring to an existing AnimalCategory - without being able to choose which one.
EF wouldn't want to choose for some default behavior. You'd have to make the choice yourself, so you can't do without access to AnimalCategory.
I'm really new to Entity Framework (currently using EF5) and vs2012 and am having difficulty trying to figure something out.
I have an .edmx that was generated from my database. It has two tables in it: Item and 3rdPartyItem. In short, the Item table is the main table for all the company items while the 3rdPartyItem table is a table that is used to hold additional attributes for items. This 3rdPartyItem table was created by an outside company and is used with their software for the company, so I can't mess with either table. What I'm trying to do is show a list of Items in a grid but I need to show a combination of all the fields for both tables. In vs2012, if I create a relationship and make it 'zero-to-one' (because for each record in the Item table, there doesn't necessarily have to be one in the 3rdPartyItem table), vs complains about not being mapped correctly. When I set the mapping, it then complains that there's multiple relationships. I did some research and found that you can't create a relationship with 2 primary keys, so I was thinking that was the problem. But how can I change the .edmx so that in code, I can access Items and 3rdPartyItem like so:
var items = dbContext.Items;
items.3rdPartyItem.SomeField <--field from 3rdPartyItem table.
not too sure if it's even possible, but it would be very, very helpful if so. Any ideas?
What you're looking for is table-per-type (TPT) mapping inheritance. You can view an MSDN walkthrough here (although you'd want your base type to be instantiable):
http://msdn.microsoft.com/en-us/data/jj618293.aspx
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).
We have a customer very large table with over 500 columns (i know someone does that!)
Many of these columns are in fact foreign keys to other tables.
We also have the requirement to eager load some of the related tables.
Is there any way in Linq to SQL or Dynamic Linq to specify what columns to be retrieved from the database?
I am looking for a linq statement that actually HAS this effect on the generated SQL Statement:
SELECT Id, Name FROM Book
When we run the reguar query generated by EF, SQL Server throws an error that you have reached the maximum number of columns that can be selected within a query!!!
Any help is much appreciated!
Yes exactly this is the case, the table has 500 columns and is self referencing our tool automatically eager loads the first level relations and this hits the SQL limit on number of columns that can be queried.
I was hoping that I can set to only load limited columns of the related Entities such as Id and Name (which is used in the UI to view the record to user)
I guess the other option is to control what FK columns should be eager loaded. However this still remains problem for tables that has a binary or ntext column which you may not want to load all the times.
Is there a way to hook multiple models (Entities) to the same table in Code First? We tried doing this I think the effort failed miserably.
Yes you can return only subset of columns by using projection:
var result = from x in context.LargeTable
select new { x.Id, x.Name };
The problem: projection and eager loading doesn't work together. Once you start using projections or custom joins you are changing shape of the query and you cannot use Include (EF will ignore it). The only way in such scenario is to manually include relations in the projected result set:
var result = from x in context.LargeTable
select new {
Id = x.Id,
Name = x.Name,
// You can filter or project relations as well
RelatedEnitites = x.SomeRelation.Where(...)
};
You can also project to specific type BUT that specific type must not be mapped (so you cannot for example project to LargeTable entity from my sample). Projection to the mapped entity can be done only on materialized data in Linq-to-objects.
Edit:
There is probably some misunderstanding how EF works. EF works on top of entities - entity is what you have mapped. If you map 500 columns to the entity, EF simply use that entity as you defined it. It means that querying loads entity and persisting saves entity.
Why it works this way? Entity is considered as atomic data structure and its data can be loaded and tracked only once - that is a key feature for ability to correctly persist changes back to the database. It doesn't mean that you should not load only subset of columns if you need it but you should understand that loading subset of columns doesn't define your original entity - it is considered as arbitrary view on data in your entity. This view is not tracked and cannot be persisted back to database without some additional effort (simply because EF doesn't hold any information about the origin of the projection).
EF also place some additional constraints on the ability to map the entity
Each table can be normally mapped only once. Why? Again because mapping table multiple times to different entities can break ability to correctly persist those entities - for example if any non-key column is mapped twice and you load instance of both entities mapped to the same record, which of mapped values will you use during saving changes?
There are two exceptions which allow you mapping table multiple times
Table per hierarchy inheritance - this is a mapping where table can contains records from multiple entity types defined in inheritance hierarchy. Columns mapped to the base entity in the hierarchy must be shared by all entities. Every derived entity type can have its own columns mapped to its specific properties (other entity types have these columns always empty). It is not possible to share column for derived properties among multiple entities. There must also be one additional column called discriminator telling EF which entity type is stored in the record - this columns cannot be mapped as property because it is already mapped as type discriminator.
Table splitting - this is direct solution for the single table mapping limitation. It allows you to split table into multiple entities with some constraints:
There must be one-to-one relation between entities. You have one central entity used to load the core data and all other entities are accessible through navigation properties from this entity. Eager loading, lazy loading and explicit loading works normally.
The relation is real 1-1 so both parts or relation must always exists.
Entities must not share any property except the key - this constraint will solve the initial problem because each modifiable property is mapped only once
Every entity from the split table must have a mapped key property
Insertion requires whole object graph to be populated because other entities can contain mapped required columns
Linq-to-Sql also contains ability to mark a column as lazy loaded but this feature is currently not available in EF - you can vote for that feature.
It leads to your options for optimization
Use projections to get read-only "view" for entity
You can do that in Linq query as I showed in the previous part of this answer
You can create database view and map it as a new "entity"
In EDMX you can also use Defining query or Query view to encapsulate either SQL or ESQL projection in your mapping
Use table splitting
EDMX allows you splitting table to many entities without any problem
Code first allows you splitting table as well but there are some problems when you split table to more than two entities (I think it requires each entity type to have navigation property to all other entity types from split table - that makes it really hard to use).
Create stored procedures that query the number of columns needed and then call the stored procs from code.