Ok, so let me explain a little of what I am trying to do.
I have a table called WebsitePage that contains all of the pages on my site with some information about them. Then, I have a Customer table for all of my customers. Then I have another table called CustomerWebsitePage that stores customer values for some of the columns in the WebsitePage table.
So, using the entity framework, I imported these three tables. What I want to be able to do is return a strongly typed WebsitePage list that has any values from CustomerWebsitePage if there are any values for it. So, for example, say one of my customers added a CustomerWebsitePageName for one of my website pages. I want to return a list of WebsitePages that contains the CustomerWebsitePageName instead of the WebsitePage Name in that case. But the original WebsitePage Name for everything else since it wasn't overridden.
The kicker here is that my WebsitePage table has a foreign key to itself for a Parent/Child relationship. So, I also want to return the child WebsitePages at the same time. I tried using a function import to get what I wanted, but then of course I lost the ChildPages.
I have tried just about everything to get what I want using the Entity framework and LINQ. But so far almost everything I try ends up with an exception being thrown. Here are a few:
The EntityCollection has already been initialized
The entity or complex type 'MyEntityModel.WebsitePage' cannot be constructed in a LINQ to Entities query.
I have one idea of how I can get around all this, and that would be to duplicate the ParentPageID into my WebsitePage table, but this really seems to violate a lot of principles and I really just don't want to add the maintenance headache related to this.
Anyone have any ideas how to accomplish this type of thing?
A simple DB diagram. http://images.tehone.com/screenshots/2009-08-17_013009.png
The object that you need to return is a CustomerWebsitePage and not a WebsitePage. The reason is that whatever object you return must know about the customer since the property will need it to determine which field (CustomerPage or WebsitePage) to use.
Considering that, you can have a function CustomerWebsitePage.GetAllPagesForCustomer(Customer c) that would return an enumeration of pages. However, to achieve the functionality you are looking for, you must implement some read-through properties in the CustomerWebsitePage. Let's take the example of Name. Here would be how to implement it in CustomerWebsitePage:
public string Name
{
get{ if( String.IsNullOrEmpty(CustomerWebsitePageName) )
return WebsitePage.Name;
return CustomerWebsitePageName; }
}
For the Children pages, you could have a property:
public IEnumerable<CustomerWebPage> Children
{
get
{
return WebsitePage.Children.Select( it => it.Customer.CustomerID == this.CustomerID ); }
}
Note that with this setup you couldn't run EF Linq queries on these new fields (because they exist only in the objects themselves, not in the database mapping). You can pepper the code with Load() if you want to seamlessly load all the children, but it will cost you in performance. To ensure that all the children are loaded, try the following loading function:
IEnumerable<CusomterWebPage> GetAllPagesForCustomer(Customer c)
{
return Context.CustomerWebPageSet.Include("WebsitePage").Where( it => it.Customer.CustomerID == c.CustomerID );
}
Related
What's the most preferred way to work with Entity Framework and DTOs?
Let's say that after mapping I have objects like:
Author
int id
sting name
List<Book> books
Book
int id
string name
Author author
int authorID
My DTOs
AuthorDTO
int id
sting name
BookDTO
int id
string name
int authorID
Since author can have a lot of books I don't want to retrieve all of them, when for example I'm only interested in authors.
But sometimes I might want to get few authors and filtered books or all books.
I could go with multiple queries AuthorDTO GetAuthor(int id) List<BookDTO> GetBooks(int authorID). But that means several accesses to database.
The ways I see it:
If I had in AuthorDTO field List<BookDTO> books the job could be done. But sometimes I would keep this list empty, if for example I listed only authors. And that means some unconsistency, mess and a lot of details to remember.
Return Tuple<AuthorDTO, List<BookDTO>> it might be a bit confusing.
Define new DTO.
AuthorAndBooksDTO
AuthorDTO author
List<BookDTO> books
The problem with sticking to a sinlge AuthorDTO and selectively filling the List is that you are now forced to keep track of where that DTO came from. Is the list of Books not hydrated, or does this Author simply have no books? Do I have to go back to my controller and call a different method to get a different state of the same DTO? This lacks clarity from the consumer's standpoint.
In my experience, I've leaned the way of more DTOs instead of trying to re-use a set of basic DTOs to represent multiple different sets of data. It requires a bit more "boilerplate", having to set up a whole bunch of similar DTOs and mappings between DTO and Entity, but in the end the specificity and clarity makes the codebase easier to read and manage.
I think some clarification of the issues involved will actually solve your confusion here.
First and most importantly, your entity classes are DTOs. In fact, that's all they are. They're classes that represent a table structure in your database so that data from queries Entity Framework makes can be mapped on to them. In other words, they are literally objects that transfer data. The failing of Microsoft and subsequently far too many MVC developers is to conflate them with big-M Models described by the MVC pattern.
As a result, it makes absolutely zero sense to use Entity Framework to return one or more instances of an entity and then map that to yet another DTO class before finally utilizing it in your code. All you're doing is creating a pointless level of abstraction that adds nothing to your application but yet another thing to maintain.
As far as relationships go, that's where Entity Framework's lazy/eager loading comes in. In order to take advantage of it, though, the property representing the relationship must follow a very specific convention:
public virtual ICollection<Book> Books { get; set; }
If you type it as something like List<Book>, Entity Framework will not touch the relationship at all. It will not ever load the related entities and it will not persist changes made to that property when saving the entity back to the database. The virtual keyword allows Entity Framework to dynamically subclass your entity and override the collection property to add the logic for lazy-loading. Without that, the related entities will only ever be loaded if you explicitly use Load from the EF API.
Assuming your property is defined in that way, then you gain a whole world of abilities. If you want all books belonging to the author you can just interact with author.Books directly (iterate, query, whatever). No queries are made until you do something that requires evaluation of the queryset. EF issues just-in-time queries depending on the information you're requesting from the database. If you do want to load all the related books at the same time you retrieve the author, you can just use Include with your query:
var author = db.Authors.Include(m => m.Books).SingleOrDefault(m => m.Id == id);
My first question would be to ask why you are creating DTO's in the first place? Is there a consumer on the other end that is using this data? Is it a screen? Are you building DTO's just to build DTO's?
Since you tagged the question as MVC i'm going to assume you are sending data to a view. You probably want a ViewModel. This ViewModel should contain all the data that is shown on the View that uses it. Then use entity framework to populate the view model. This may be done with a single query using projections or something complex.
So after all that blathering. I would say you want option 3.
Just like the others said, for clarity reasons, you should avoid creating "generic" DTO's for specific cases.
When you want to sometimes have authors and some of their books then model a DTO for that.
When you need only the authors then create another DTO that is more suited for that.
Or maybe you don't need DTOs, maybe a List containing their names is enough. Or maybe you could in fact use an anonymous type, like new { AuthorId = author.Id, AuthorName = author.Name }. It depends on the situation.
If you're using ASP.NET MVC the DTO you'll want is in fact a ViewModel that best represents your page.
Based on what you've described, you're view model could be something like this
public class BookViewModel{
public int Id {get;set;}
public string Name {get;set;}
}
public class AuthorViewModel{
public int Id {get;set;}
public string Name {get;set;}
public List<BookViewModel> Books {get;set;} = new List<BookViewModel>();
}
public class AuthorsViewModel
{
public List<AuthorViewModel> Authors {get;set;} = new List<AuthorViewModel>();
//add in this class other properties, like the filters used on the page...
public void Load(){
//here you can retrieve the data from your database.
//you could do like this:
//step 1: retrieve data from DB via EF
//step 2: fill in the Authors view models from the data at step 1
}
}
//and in your controller you're calling the Load method to fill you're viewmodel with data from db.
public class AuthorsController{
public ActionResult Index(){
AuthorsViewModel model = new AuthorsViewModel();
model.Load();
return View(model);
}
}
I have had to move queries off the main database in order to meet requirements for complex authorization - for example a user with a given authorization role can only view data for individuals in the same institution.
I am using the Breeze .net DocCode sample for guidance, and have copied the premise for the mapping of domain models to DTOs.
get { return ForCurrentUser(Context.Orders).Select(o => new Order {
OrderID = o.OrderID,
....
OrderDetails = o.OrderDetails.Select(od => new OrderDetail
{
ProductID = od.ProductID,
UnitPrice = od.UnitPrice
...
})
The problem is that which mapped properties to .include(entity framework method)/.expand (breeze method) is now a concern of the mapping function (for example, the above code will always return the OrderDetails collection, whether I want them or not). I would like to still only eagerly load/expand properties if the javascript client generated predicate has a .expand directive for that property.
Is this at all possible, or am I stuck with manually defining different mapping functions on the server, depending on what properties I want expanded? (I am happy to use tools such as automapper if that would solve or simplify the problem)
Thank you
You will need to use the ODataQueryOptions as a parameter to your controller method. This gives you the details of the query predicates in your server method, so that you can apply them as needed rather that having them applied automatically. This will let you expand, or not, based upon the query.
See this answer and this answer to see how it works.
I am refactoring an MVC project to make it testable. Currently the Controller uses the Entity Framework's context objects directly to ask for the required data. I started abstract this and it just doesn't work. Eventually I have an IService and an IRepository abstraction, but to describe the problem let's just look at the IRepository. Many people advise an interface with functions which return some of these: IQueriable<...>, IEnumerable<...>, IList<...>, SomeEntityObject, SomeDTO. Then when one wants to test the service layer they can implement the interface with a class which doesn't go to the database to return these.
Problem: Using linq to entities I have lazy (deferred) loading in my toolset. This is actually very useful, because my controller action functions know which data they need for the view and I didn't ask for more than required. However linq to anythingelse doesn't have lazy loading. So when my IRepository functions return any of the above mentioned things I lose lazy loading. I extended my interface with functions like "GetAnything" and "GetAnythingDeep" but it's not enough: it has to be much more fine-grained. Which would result about 5-6 functions for the same type of object, depending on the properties I want to get in the result. Maybe could be a general function with some "include properties" parameter, but I don't like that too.
Eventually atm I think if I want to make it testable that will result either much less efficient or much more complicated code. Sounds not right.
Btw I was thinking about to change the data source behind the entity model to either xml or some object data soruce, and so I could keep the linq to entities. I found that it's not supported out of the box... which is also sad: this means that entity framework means database source - not a really useful abstraction.
Specific example:
Entity objects:
Article, Language, Person. Relations: Article can have 1-N languages, and one Person (publisher).
ViewModel object:
ArticleDeepViewModel: Contains all the properties of the article, including the languages and the Name of the Person (it's for view the article, so no need for the other properties of the person).
Controller action which will return this view should get the data from somewhere.
Code before modifications:
using (var context = new Entities.Articles())
{
var article = (from a in context.Articles.Include("Languages")
where a.ID == ID
select new ViewArticleViewModel()
{
ID = a.ID,
Headline = a.Headline,
Summary = a.Summary,
Body = a.Body,
CreatedBy = a.CreatedByEntity.Name,
CreatedDate = a.CreatedDate,
Languages = (from l in context.Languages select new ViewLanguagesViewModel() { ID = l.ID, Name = l.Name, Selected = a.Languages.Contains(l) })}).Single();
this.ViewData.Model = article;
}
return View();
Code after modifications could be something like:
var article = ArticleService.GetArticleDeep(ID);
var viewModel = /* mapping */
this.ViewData.Model = viewModel;
return View();
Problem is that GetArticleDeep should return an Article object with Languages included and the entire Person object included (it shouldn't know that the viewmodel needs just the Name of the Person). Also I have so far 3 different viewmodels for an article. For example if someone wants to see the list of articles, then it's unnecessary to get the languages, the body and some other properties, however it might be useful to get the Name of the publisher (which is in the deep). Before "testable" code the controller actions could just contain the linq to entities query and get whichever data they need using lazy loading, Include function, using subqueries, referencing foreign properties (Publisher.Name) ... So there is no unnecessary query to the database and no unnecessary data transferred from the database.
What should be the IService or IRepository interface provide to get the 3-4 different level of Article objects or sometimes list of these objects?
Not sure if you are planning to stick with lazy loading, but if you want a flexible way to integrate eager loading into your repository and service layers first check out this article:
http://blogs.msdn.com/b/alexj/archive/2009/07/25/tip-28-how-to-implement-include-strategies.aspx
He basically gives you a way to build a strongly-typed include strategy like this:
var strategy = new IncludeStrategy<Article>();
strategy.Include(a => a.Author);
Which can then be passed into a general method on your repository or service layers. This way you don't have to have a separate method for each circumstance (i.e. your GetArticleDeep method).
Here is an example repository method using the above include strategy:
public IQueryable<Article> Find(Expression<Func<Article, bool>> criteria, IncludeStrategy<Article> includes)
{
var query = includes.ApplyTo(context.Articles).Where(criteria);
return query;
}
I was reading Steven Sanderson's book Pro ASP.NET MVC Framework and he suggests using a repository pattern:
public interface IProductsRepository
{
IQueryable<Product> Products { get; }
void SaveProduct(Product product);
}
He accesses the products repository directly from his Controllers, but since I will have both a web page and web service, I wanted to have add a "Service Layer" that would be called by the Controllers and the web services:
public class ProductService
{
private IProductsRepository productsRepsitory;
public ProductService(IProductsRepository productsRepository)
{
this.productsRepsitory = productsRepository;
}
public Product GetProductById(int id)
{
return (from p in productsRepsitory.Products
where p.ProductID == id
select p).First();
}
// more methods
}
This seems all fine, but my problem is that I can't use his SaveProduct(Product product) because:
1) I want to only allow certain fields to be changed in the Product table
2) I want to keep an audit log of each change made to each field of the Product table, so I would have to have methods for each field that I allow to be updated.
My initial plan was to have a method in ProductService like this:
public void ChangeProductName(Product product, string newProductName);
Which then calls IProductsRepository.SaveProduct(Product)
But there are a few problems I see with this:
1) Isn't it not very "OO" to pass in the Product object like this? However, I can't see how this code could go in the Product class since it should just be a dumb data object. I could see adding validation to a partial class, but not this.
2) How do I ensure that no one changed any other fields other than Product before I persist the change?
I'm basically torn because I can't put the auditing/update code in Product and the ProductService class' update methods just seem unnatural (However, GetProductById seems perfectly natural to me).
I think I'd still have these problems even if I didn't have the auditing requirement. Either way I want to limit what fields can be changed in one class rather than duplicating the logic in both the web site and the web services.
Is my design pattern just bad in the first place or can I somehow make this work in a clean way?
Any insight would be greatly appreciated.
I split the repository into two interfaces, one for reading and one for writing.
The reading implements IDisposeable, and reuses the same data-context for its lifetime. It returns the entity objects produced by linq to SQL. For example, it might look like:
interface Reader : IDisposeable
{
IQueryable<Product> Products;
IQueryable<Order> Orders;
IQueryable<Customer> Customers;
}
The iQueryable is important so I get the delayed evaluation goodness of linq2sql. This is easy to implement with a DataContext, and easy enough to fake. Note that when I use this interface I never use the autogenerated fields for related rows (ie, no fair using order.Products directly, calls must join on the appropriate ID columns). This is a limitation I don't mind living with considering how much easier it makes faking read repository for unit tests.
The writing one uses a separate datacontext per write operation, so it does not implement IDisposeable. It does NOT take entity objects as input or out- it takes the specific fields needed for each write operation.
When I write test code, I can substitute the readable interface with a fake implementation that uses a bunch of List<>s which I populate manually. I use mocks for the write interface. This has worked like a charm so far.
Don't get in a habit of passing the entity objects around, they're bound to the datacontext's lifetime and it leads to unfortunate coupling between your repository and its clients.
To address your need for the auditing/logging of changes, just today I put the finishing touches on a system I'll suggest for your consideration. The idea is to serialize (easily done if you are using LTS entity objects and through the magic of the DataContractSerializer) the "before" and "after" state of your object, then save these to a logging table.
My logging table has columns for the date, username, a foreign key to the affected entity, and title/quick summary of the action, such as "Product was updated". There is also a single column for storing the change itself, which is a general-purpose field for storing a mini-XML representation of the "before and after" state. For example, here's what I'm logging:
<ProductUpdated>
<Deleted><Product ... /></Deleted>
<Inserted><Product ... /></Inserted>
</ProductUpdated>
Here is the general purpose "serializer" I used:
public string SerializeObject(object obj)
{
// See http://msdn.microsoft.com/en-us/library/bb546184.aspx :
Type t = obj.GetType();
DataContractSerializer dcs = new DataContractSerializer(t);
StringBuilder sb = new StringBuilder();
XmlWriterSettings settings = new XmlWriterSettings();
settings.OmitXmlDeclaration = true;
XmlWriter writer = XmlWriter.Create(sb, settings);
dcs.WriteObject(writer, obj);
writer.Close();
string xml = sb.ToString();
return xml;
}
Then, when updating (can also be used for logging inserts/deletes), grab the state before you do your model-binding, then again afterwards. Shove into an XML wrapper and log it! (or I suppose you could use two columns in your logging table for these, although my XML approach allows me to attach any other information that might be helpful).
Furthermore, if you want to only allow certain fields to be updated, you'll be able to do this with either a "whitelist/blacklist" in your controller's action method, or you could create a "ViewModel" to hand in to your controller, which could have the restrictions placed upon it that you desire. You could also look into the many partial methods and hooks that your LTS entity classes should have on them, which would allow you to detect changes to fields that you don't want.
Good luck! -Mike
Update:
For kicks, here is how I deserialize an entity (as I mentioned in my comment), for viewing its state at some later point in history: (After I've extracted it from the log entry's wrapper)
public Account DeserializeAccount(string xmlString)
{
MemoryStream s = new MemoryStream(Encoding.Unicode.GetBytes(xmlString));
DataContractSerializer dcs = new DataContractSerializer(typeof(Product));
Product product = (Product)dcs.ReadObject(s);
return product;
}
I would also recommend reading Chapter 13, "LINQ in every layer" in the book "LINQ in Action". It pretty much addresses exactly what I've been struggling with -- how to work LINQ into a 3-tier design. I'm leaning towards not using LINQ at all now after reading that chapter.
I am pretty new to ASP.NET MVC and I am trying to get my head around some of the design concepts at the moment. One thing I am currently stuck on is how (best) to handle a situation such as that described below.
Suppose I have a page that needs to render a number of "sections". For example, on the left hand side there is a list that is data driven and then the selected item in the list displays a further list in another section on the page. For better understanding lets suggest the left hand list is a list of movie categories and the other list displays a list of movies that are contained within that category, along with the various movie details.
Now, I have some form of ORM such as Entity Framework, LINQ to SQL or whatever that maps the tblCategory and tblMovie database tables into Category and Movie entities respectively. These entities live in the MyMVCApp.Data.Entities namespace. I then use a repository pattern located in the MyMVCApp.Data namespace to encapsulate the queries (via LINQ) against these entities to return our model entities.
This is where my first question lies. Should the repository be returning view model entities or domain entities which are then extended to produce view model entities? In my naive mind I see the entities returned from the ORM as simply containers for the data with domain entities containing the business logic. So surely there should be an abstration here?
Anyway, back to my scenario. Now lets assume I wish to present the two sections described at the beginning of this. I am assuming that what should be happening here is my domain entity model would contain a list of categories and their associated movies. So something like a List each of which contains a List.
This model would have been populated somewhere. This is my second question. Say my assumption above is correct and it is simply data entities are returned from the ORM. Now I have a namespace/project called MyMVCApp.Core.Model (or the like) with some domain entities such as the Movie and Category entities mentioned in the previous paragraph. Do these entities have methods on them to retrieve the data from the ORM and populate themselves? Or does the repository retrieve these populated entity models? And another question on this part, if I have a Movie and Customer entity in my ORM is it acceptable to be having domain entities named the same?
Finally, I assume the controller now has this populated list of Category and Movie objects and passes it back to the view. I am guessing it is best to have each of the sections described at the beginning as partial views and passing the populated model to each? So, it may be the IndexController which retrieves the populated CategoryMovies entity, passing this to the Categories partial view and the Movies partial view. I would then need to somehow determine the selected Category (quesrystring?) and display the appropriate list of Movies in that category within the view.
OK, so if anyone has got to this point in my ramblings I will take a deep bow. I hope I have explained my mixed up thoughts and questions in sufficient detail for someone to provide some kind of enlightenment.
Thanks for listening! :-)
Since you didn't mention it, I will assume you are new to DDD concepts. DDD compliments the "M" in MVC by placing logic where it belongs. And, I think a good amount could be applied here.
In strict DDD form, I would use your Movie example as an Aggregate Root (a DDD concept). Within Movie, you would have business logic (methods) that obtain the categories and related entities directly related to Movie (i.e. Categories-this-movie-belongs-in). This assumes the "category" list you want to display is a list of categories this movie is in.
public class Movie
{
private IList<Category> _categories;
public IList<Category> FetchCategories()
{
// you can lazy-load here, use Linq-to-Sql, etc.
return _categories;
}
public void AddCategory(Category c)
{
_categories.Add(c);
}
}
Of course, you can treat Categories as a Value Object (VO, a DDD concept) here if you don't have an identity on them.
Now, what gets more interesting is if you do want to keep an identity on your Categories, treating them as Aggregate Roots with multiple entities and relations to other VOs and alike. In this case, you would want to utilize the DDD concept of Services to combine the two aggregates into your requested Model you want to pass to your Controller. This allows you to create business rules around loading the categories.
public class Movie
{...}
public class Category
{...}
public class MovieCategory : Movie
{
private IList<Category> _categories;
public IList<Category> Categories
{
get
{
return _categories;
}
internal set
{
_categories = value;
}
}
}
public class MovieCategoryService
{
public MovieCategory FetchMovieCategoryByMovieId(int id)
{
MovieCategory mc = _movieRepository.FetchMovie(id);
// do some logic here, some checks, etc
// to obtain your Categories list. Maybe querystring?
IList<Category> cats = ...;
mc.Categories = cats;
return mc;
}
}
The concept is that you now have logic in the domain layer, in this MovieCategoryService, to handle the complex creation of these relationships, or using Factories to create complex types as well.
This gives you a model you can pass around to different View and PartialViews.
The final step in your original post is how to get this into the View. I've been playing with the ViewModel approach to this issue. Creating ViewModels either dynamically, or as a class, and hanging all of the entities on this ViewModel. A recent StackOverflow question hits on this concept. ViewModel Best Practices
I've passed the Model directly into Views, and I am now trying this approach. It does clean things up, as your DOmain Model really should be disconnected from the "logic" on your page. I find myself thinking, "Dang it. I need to get this partialview populated based on this ID. Guess that means another method on the entity." Going the ViewModel approach removes the logic/thinking from the Domain Model, and puts it where it belong - in the Application/UI layer of where you are fine-tuning your view/partialview.
First of all, I think you should just get started with a simple project and try the various scenarios out you pictured in your long question :). There is nothing written in stone, you can use whatever architecture with datalayers and services or whatever, you like !
Or does the repository retrieve these populated entity models?
I would say yes. Your controller graps these from the service and gets them, populated and all and just moves them to the view in order to be displayed.
I am guessing it is best to have each of the sections described at the beginning as partial views and passing the populated model to each?
And again i think you're on the right track ;).
It is a nice idea to give a view model entity to the view instead of the whole domain entity itself thereby providing the view only the required stuffs to work on
First of all return a List of Categories from your Domain, then construct a Movie view which takes the category name as a parameter something like this
- <Host>/Movies/Comedy
- <Host>/Movies/Horror
which would in turn display the movies which belongs to that particular category