I have a problem that I haven't been able to find a solution to, and I wonder if anyone could give some advice.
I have a mocked datacontext/objectset, done through interfaces and t4 templates, with some ninject magic, with the intent of having in memory datasets for unit testing.
However, what should you do with the foreign key values/navigation properties?
Lets say I have hotels and customers, ctx.Hotels has some values, but Customer.Hotels does not. The get is something like this if it is a one-to-one relationship:
return ((IEntityWithRelationships)this).RelationshipManager.GetRelatedReference<Hotel>("HotelModel.FK_Customers_Hotels", "Hotel").Value;
and one-to-many:
return ((IEntityWithRelationships)this).RelationshipManager.GetRelatedCollection<BookingRow>("HotelModel.FK_BookingRows_Customers", "BookingRow");
My skill level just isn't enough to even understand what is going on here.
[edit:]
Great Master Julie Lerman confirms that this is a dead end. You can't properly mock entityobjects, you need POCOs for that.
Mocking ObjectContext when you are using EntityObject based entities is mostly impossible because for example RelationshipManager is a real class which cannot be replaced with your mock. Also your entities are heavily dependent on non mockable EF code.
Note: "Mostly" because you can mock it but you need special framework intercepting calls to real objects and forwarding them to your methods instead. That is possible only with TypeMock Isolator or MS Moles.
Btw. mocking EF code is something you don't want to do - go through this answer and linked answers. Some of them targets newer EF API but the problems are still same.
Related
I'm trying to implement fully valid persistence ignorance with little effort. I have many questions though:
The simplest option
It's really straightforward - is it okay to have Entities annotated with Spring Data annotations just like in SOA (but make them really do the logic)? What are the consequences other than having to use persistance annotation in the Entities, which doesn't really follow PI principle? I mean is it really the case with Spring Data - it provides nice repositories which do what repositories in DDD should do. The problem is with Entities themself then...
The harder option
In order to make an Entity unaware of where the data it operates on came from it is natural to inject that data as an interface through constructor. Another advantage is that we always could perform lazy loading - which we have by default in Neo4j graph database for instance. The drawback is that Aggregates (which compose of Entities) will be totally aware of all data even if they don't use them - possibly it could led to debugging difficulties as data is totally exposed (DAO's would be hierarchical just like Aggregates). This would also force us to use some adapters for the repositories as they doesn't store real Entities anymore... And any translation is ugly... Another thing is that we cannot instantiate an Entity without such DAO - though there could be in-memory implementations in domain... again, more layers. Some say that injecting DAOs does break PI too.
The hardest option
The Entity could be wrapped around a lazy-loader which decides where data should come from. It could be both in-memory and in-database, and it could handle any operations which need transactions and so on. Complex layer though, but might be generic to some extent perhaps...? Have a read about it here
Do you know any other solution? Or maybe I'm missing something in mentioned ones. Please share your thoughts!
I achieve persistence ignorance (almost) for free, as a side effect of proper domain modeling.
In particular:
if you correctly define each context's boundary, you will obtain small entities without any need for lazy loading (that, actually becomes an antipattern/code smell in a DDD project)
if you can't simply use SQL into your repository, map a set of DTO to your db schema, and use them into factories to initialize entity classes.
In DDD projects, persistence ignorance is relevant for the domain model itself, not for repositories, factories and other applicative code. Indeed you are very unlikely to change the ORM and/or the DB in the future.
The only (but very strong) rational behind persistence ignorance of the domain model is separation of concerns: in the domain model you should express business invariants only! Persistence is an infrastructural concern!
For example without persistence ignorance (and with lazy loading) the domain model should handle possible exceptions from the db, it's complexity grows and business rules are buried under technological details.
Personally I find it near impossible to achieve a clean domain model when trying to use the same entities as the ORM.
My solution is to model my domain entities as I see fit and ensure that any ORM entities don't leak outside of the repositories. This means that my repositories accept and return domain entities.
This means you lose "most of your ORM goodness" and end up "using your ORM for simple CRUD operations".
Both of these trade-offs are fine for me, I would rather have a clean domain model that I can use, rather than one polluted with artefacts from my DB or ORM. It also cuts down the amount of time I spend "wrestling with my ORM" to zero.
As a side-note, I find document databases a much better fit for DDD.
Once you will provide persistence mapping in you domain model:
your code depends on framework. If you decided to change this framework, you want to change persistence layer and model layer source code - more work, more changes, more merging of code etc.
your domain model jar file depends on spring/nhibernate jars etc.
your classes become larger and larger how business code and persistence related code grows
I've to admit that I dont understand harder and hardest option.
We used separated interfaces and implementations for domain entities. Provide separated mapping files using Hibernate along with repositories.
Entities are created using factory (or repository later), identifier is generated within persistence layer, entity does not need it until it's being persisted.
Lazy loading is provided by special implementation of List once:
mapping of an entity contains it
entity/aggregate is fetched from persistence layer
The only issue is related to transaction as when you use lazy-loaded collection out of transaction scope, it fails.
I would follow the simplest option unless I ran into a stone wall. There are also pitfalls such as this when you adopt pi principle.
Somtimes some compromises are acceptable.
public class Order {
private String status;//my orm does not support enum
public Status status() {
return Status.of(this.status);
}
public is(Status status) {
return status() == status;//use status() instead of getStatus() in domain model
}
}
Can someone enlighten me about the intended (best-practice) use of "DbContext" and "DbSet" where data consumers should have no dependency on the EF whatsoever (100% POCO). While all the examples I find explain how to create POCO classes for the underlying data source (corresponding to RDBMS tables usually), they still expose "DbContext" and "DbSet" to the outside world (including MSFT's own examples on how to do data binding with these classes). This seems to defeat or (seriously) water down the benefits of using POCO. IOW, clients are still bound to the EF rather than a cleaner (IMHO) approach of creating some data store project that uses EF internally, but exposes only POCO to its users. Why do all articles on the EF seem to ignore this (am I missing something). Thanks.
You would usually create a service layer (based on the Repository pattern for example) that makes use of DbContext. Its role is to mediate between the database and the application/client. It will accept POCOs from the client and return POCOs to the client. That way, the client knows about the repository, but it doesn't know anything about DbContext.
Some more reading for you:
http://blogs.microsoft.co.il/blogs/gilf/archive/2010/01/20/using-repository-pattern-with-entity-framework.aspx
http://blogs.msdn.com/b/adonet/archive/2009/06/16/using-repository-and-unit-of-work-patterns-with-entity-framework-4-0.aspx
We are in a process of designing an application with approx 100 tables and complicated business logic. Windows Forms will be used on the client side and WCF services with MSSQL on the server.
Custom DTOs are used for client-server communication, business entities are not distributed.
Which variant of Entity Framework to use (and why):
EF 4.0 EntityObjects
EF 4.0 POCO
EF 4.1 DbContext
Something else
Database-first approach is a requirement.
Also, is it worth implementing a Repository pattern? It seems a bit redundant, as there is one level of abstraction in the mapping itself and another one in the use of DTOs. I'm currently leaned towards using auto-generated extendable repositories for each entity returning IQueryable, just to have a place to put common queries, but still allowing querying entity model directly from the Service Layer.
Which variant to use? Basically once you have custom DTO the only question is do you want to have control over entities code (their base class) and make them independent on EF? Do you want to use code first? If the answers to all questions are no then you can use EntityObjects. If you want to have entities persistence ignorant or use custom base class you should go to POCO. If you want to use code first or new DbContext API you will need EF 4.1. Some related topics:
EF 4.1 Code-first vs Model/Database-first
EF POCO code only VS EF POCO with Entity Data Model (this was related to CTP)
ADO.NET DbContext Generator vs. ADO.NET POCO Entity Generator
EF Model First or Code First Approach?
There are more things to consider when designing service layer. You should be aware of complications you will have to deal with when using EF in WCF. Your service will provide data to WinForms application and it will work with them in "detached mode". Once user will do all changes he wants to do he will post data back to the service. But here comes the problem - you must tell EF what has changed. If you for example allow user to change order with all its order items (change quantity in items, add new items, delete some items) you must say EF exactly what has changed, what was added and what was deleted. That is easy when you work with single entity but once you allow user to change object graph (especially many-to-many relations) then it is quite tough. The most common solution is loading the whole graph and merge the state from incoming DTOs to loaded and attached graph. Other solution is using Self tracking entities instead of EntityObjects/POCOs + DTOs.
When discussing repositories I would refer you to this answer which refers many other answers discussing repositories, their possible redundancy and possible mistakes when using them just to make your code testable. Generally each layer should be added only if there is real need for the layer - due to better separation of concerns.
The main advantage of POCOs is that those classes can be your DTOs, so if you've already got custom DTOs that you're using, POCO seems a bit redundant. However, there are some other advantages which may or may not have value to you, since you didn't mention unit testing as a requirement. If you plan to write unit tests, then POCO is still the way to go. You probably won't notice much difference between 4.0 POCO and 4.1 since you won't be using the code-first feature (disclaimer: I've only used 4.0 POCO, so I'm not intimately familiar with any minor differences between the two, but they seem to be more or less the same--basically I was already using POCO in 4.0 and haven't seen anything that's made me want to update everything to use 4.1).
Also, depending on whether you plan to unit-test this layer, there's still value in implementing the repository/unit of work patterns when using Entity Framework. It serves to abstract away the data access logic (the context), not the entities themselves, and allows you to do things like mocking your context in unit tests. What I do is copy the T4 template for my context and use it to create the interface, then edit the T4 template for the context and have it implement that interface and use IObjectSet<T> instead of ObjectSet<T>. So instead of:
public class MyEntitiesContext
{
public ObjectSet<MyClass> MyEntities
...
}
I end up with:
public interface IMyEntitiesContext
{
public IObjectSet<MyClass> MyEntities;
}
and
public class MyEntitiesContext : IMyEntitiesContext
{
public IObjectSet<MyClass> MyEntities
...
}
So I guess it really comes down to whether or not you plan to write unit tests for this layer. If you won't be doing anything that would require mocking out your context for testing, then the easiest thing to use would probably be 4.0 EntityObjects, since you aren't planning to pass your entities between layers and it would require the least effort to implement. If you plan to use mocking, then you'll probably want to use POCO and implement repository/unit of work.
Following the way Rob does it, I have the classes that are generated by the Linq to SQL wizard, and then a copy of those classes that are POCOs. In my repositories I return these POCOs rather than the Linq to SQL models:
return from c in DataContext.Customer
where c.ID == id
select new MyPocoModels.Customer { ID = c.ID, Name = c.Name }
I understand that the benefit of this is that the POCO models can be instantiated easier so this will make my code more testable.
I'm now moving from Linq to SQL over to Entity Framework and I'm about half way through an EF book. It seems there's a lot of goodness I'm going to lose out on by returning POCOs from my repositories rather than the EF entities.
I still haven't really embraced unit testing, so I feel like I'm wasting a lot of time creating these extra POCOs and writing the code to populate them, when all I appear to be gaining is testable code, yet I'm also gonna lose out on a lot of the benefits of the EF by not being able to track my objects.
Does anyone have any advice for a relative newb to all this ORM/Repository stuff?
Anthony
Another reason people don't like the auto-generated objects (in LINQ to SQL for example) is because of their built-in "magic".
Usually the magic is invisible and you never notice it, but when you try to do things like serialize one of those objects and then deserialize it (for example when using web services) its internal connection to the data source is broken and special hacks need to be employed to "put the magic back in".
With POCOs, you don't have to worry about those sorts of things and can get a better separation between your data and service layers. The downside of course is that you have to write lots of boring POCO -> magic object and magic object -> POCO conversion code. But in the end I think it's usually worth it, especially for large or complex projects.
The main reason is that a lot of people like to develop their model with a specific mindset: like DDD for instance. They might want to use a specific pattern (like Spec or State) for things like statuses (instead of enums) - or you might want to use a Factory for instantiation.
OO breaks when you try to use Tables as Objects when things get more complex. Simple sites work OK - but when you get to big big things, it gets ugly.
So - as always - it depends what you think your project will turn into.
My experience is that when you start writting some complex queries .Include method is worthless and you will find yourself either:
a) Writting a lot of queries to get the data you want or
b) abusing of annonymous types to load the data in a single query and then writting a lot of code just to pass that data to your entities.
POCOs are the way to go, IMHO.
I'm starting a new project and have decided to try to incorporate DDD patterns and also include Linq to Entities. When I look at the EF's ObjectContext it seems to be performing the functions of both Repository and Unit of Work patterns:
Repository in the sense that the underlying data level interface is abstracted from the entity representation and I can request and save data through the ObjectContext.
Unit Of Work in the sense that I can write all my inserts/updates to the objectContext and execute them all in one shot when I do a SaveChanges().
It seems redundant to put another layer of these patterns on top of the EF ObjectContext? It also seems that the Model classes can be incorporated directly on top of the EF generated entities using 'partial class'.
I'm new at DDD so please let me know if I'm missing something here.
I don't think that the Entity Framework is a good implementation of Repository, because:
The object context is insufficiently abstract to do good unit testing of things which reference it, since it is bound to the DB access. Having an IRepository reference instead works much better for creating unit tests.
When a client has access to the ObjectContext, the client can do pretty much anything it cares to. The only real control you have over this at all is to make certain types or properties private. It is hard to implement good data security this way.
On a non-trivial model, the ObjectContext is insufficiently abstract. You may, for example, have both tables and stored procedures mapped to the same entity type. You don't really want the client to have to distinguish between the two mappings.
On a related note, it is difficult to write comprehensive and well-enforce business rules and entity code. Indeed, whether or not it this is even a good idea is debatable.
On the other hand, once you have an ObjectContext, implementing the Repository pattern is trivial. Indeed, for cases that are not particularly complex, the Repository is something of a wrapper around the ObjectContext and the Entity types.
I would say that you should look at the ObjectContext as your UnitOfWork, and not as a repository.
An ObjectContext cannot be a repository -imho- since it is 'to generic'.
You should create your own Repositories, which have specialized methods (like GetCustomersWithGoldStatus for instance) next to the regular CRUD methods.
So, what I would do, is create repositories (one for each aggregate-root), and let those repositories use the ObjectContext.
I like to have a repository layer for the following reasons:
EF gotcha's
When you look at some of the current tutorials on EF (Code First version), it is apparent there's a number of gotcha's to be handled, particularly around object graphs (entities containing entities) and disconnected scenarios. I think a repository layer is great for wrapping these up in one place.
A clear picture of data access mechanisms
A repository gives a specific picture as to how the BL is accessing and updating the data store. It exposes methods that have a clear single purpose, and can be tested independently of the BL. Standard example from the textbooks, Find() to find a single entity. A more application specific example, Clear() to clear down a db table.
A place for optimizations
Inevitably you come up against performance hits when using vanilla EF. I use the repository to hide the optimization mechanisms from the BL.
Examples,
GetKeys() to project cached keys from the tables (for Insert/Update decisions). The reading of key only is faster and uses less memory than reading the full entity.
Bulk load via SqlBulkCopy. EF will insert by individual SQL statements. If you want a single statement to insert multiple rows, SqlBulkCopy is a good mechanism. The repository encapsulates this and provides metadata for SqlBulkCopy. As well as the Insert method, you need a StartBatch() and EndBatch() method, which is also an argument for a UnitOfWork layer.