I am trying to do a simple JSON return but I am having issues I have the following below.
public JsonResult GetEventData()
{
var data = Event.Find(x => x.ID != 0);
return Json(data);
}
I get a HTTP 500 with the exception as shown in the title of this question. I also tried
var data = Event.All().ToList()
That gave the same problem.
Is this a bug or my implementation?
It seems that there are circular references in your object hierarchy which is not supported by the JSON serializer. Do you need all the columns? You could pick up only the properties you need in the view:
return Json(new
{
PropertyINeed1 = data.PropertyINeed1,
PropertyINeed2 = data.PropertyINeed2
});
This will make your JSON object lighter and easier to understand. If you have many properties, AutoMapper could be used to automatically map between DTO objects and View objects.
I had the same problem and solved by using Newtonsoft.Json;
var list = JsonConvert.SerializeObject(model,
Formatting.None,
new JsonSerializerSettings() {
ReferenceLoopHandling = Newtonsoft.Json.ReferenceLoopHandling.Ignore
});
return Content(list, "application/json");
This actually happens because the complex objects are what makes the resulting json object fails.
And it fails because when the object is mapped it maps the children, which maps their parents, making a circular reference to occur. Json would take infinite time to serialize it, so it prevents the problem with the exception.
Entity Framework mapping also produces the same behavior, and the solution is to discard all unwanted properties.
Just expliciting the final answer, the whole code would be:
public JsonResult getJson()
{
DataContext db = new DataContext ();
return this.Json(
new {
Result = (from obj in db.Things select new {Id = obj.Id, Name = obj.Name})
}
, JsonRequestBehavior.AllowGet
);
}
It could also be the following in case you don't want the objects inside a Result property:
public JsonResult getJson()
{
DataContext db = new DataContext ();
return this.Json(
(from obj in db.Things select new {Id = obj.Id, Name = obj.Name})
, JsonRequestBehavior.AllowGet
);
}
To sum things up, there are 4 solutions to this:
Solution 1: turn off ProxyCreation for the DBContext and restore it in the end.
private DBEntities db = new DBEntities();//dbcontext
public ActionResult Index()
{
bool proxyCreation = db.Configuration.ProxyCreationEnabled;
try
{
//set ProxyCreation to false
db.Configuration.ProxyCreationEnabled = false;
var data = db.Products.ToList();
return Json(data, JsonRequestBehavior.AllowGet);
}
catch (Exception ex)
{
Response.StatusCode = (int)HttpStatusCode.BadRequest;
return Json(ex.Message);
}
finally
{
//restore ProxyCreation to its original state
db.Configuration.ProxyCreationEnabled = proxyCreation;
}
}
Solution 2: Using JsonConvert by Setting ReferenceLoopHandling to ignore on the serializer settings.
//using using Newtonsoft.Json;
private DBEntities db = new DBEntities();//dbcontext
public ActionResult Index()
{
try
{
var data = db.Products.ToList();
JsonSerializerSettings jss = new JsonSerializerSettings { ReferenceLoopHandling = ReferenceLoopHandling.Ignore };
var result = JsonConvert.SerializeObject(data, Formatting.Indented, jss);
return Json(result, JsonRequestBehavior.AllowGet);
}
catch (Exception ex)
{
Response.StatusCode = (int)HttpStatusCode.BadRequest;
return Json(ex.Message);
}
}
Following two solutions are the same, but using a model is better because it's strong typed.
Solution 3: return a Model which includes the needed properties only.
private DBEntities db = new DBEntities();//dbcontext
public class ProductModel
{
public int Product_ID { get; set;}
public string Product_Name { get; set;}
public double Product_Price { get; set;}
}
public ActionResult Index()
{
try
{
var data = db.Products.Select(p => new ProductModel
{
Product_ID = p.Product_ID,
Product_Name = p.Product_Name,
Product_Price = p.Product_Price
}).ToList();
return Json(data, JsonRequestBehavior.AllowGet);
}
catch (Exception ex)
{
Response.StatusCode = (int)HttpStatusCode.BadRequest;
return Json(ex.Message);
}
}
Solution 4: return a new dynamic object which includes the needed properties only.
private DBEntities db = new DBEntities();//dbcontext
public ActionResult Index()
{
try
{
var data = db.Products.Select(p => new
{
Product_ID = p.Product_ID,
Product_Name = p.Product_Name,
Product_Price = p.Product_Price
}).ToList();
return Json(data, JsonRequestBehavior.AllowGet);
}
catch (Exception ex)
{
Response.StatusCode = (int)HttpStatusCode.BadRequest;
return Json(ex.Message);
}
}
JSON, like xml and various other formats, is a tree-based serialization format. It won't love you if you have circular references in your objects, as the "tree" would be:
root B => child A => parent B => child A => parent B => ...
There are often ways of disabling navigation along a certain path; for example, with XmlSerializer you might mark the parent property as XmlIgnore. I don't know if this is possible with the json serializer in question, nor whether DatabaseColumn has suitable markers (very unlikely, as it would need to reference every serialization API)
add [JsonIgnore] to virtuals properties in your model.
Using Newtonsoft.Json: In your Global.asax Application_Start method add this line:
GlobalConfiguration.Configuration.Formatters.JsonFormatter.SerializerSettings.ReferenceLoopHandling = Newtonsoft.Json.ReferenceLoopHandling.Ignore;
Its because of the new DbContext T4 template that is used for generating the EntityFramework entities. In order to be able to perform the change tracking, this templates uses the Proxy pattern, by wrapping your nice POCOs with them. This then causes the issues when serializing with the JavaScriptSerializer.
So then the 2 solutions are:
Either you just serialize and return the properties you need on the client
You may switch off the automatic generation of proxies by setting it on the context's configuration
context.Configuration.ProxyCreationEnabled = false;
Very well explained in the below article.
http://juristr.com/blog/2011/08/javascriptserializer-circular-reference/
Provided answers are good, but I think they can be improved by adding an "architectural" perspective.
Investigation
MVC's Controller.Json function is doing the job, but it is very poor at providing a relevant error in this case. By using Newtonsoft.Json.JsonConvert.SerializeObject, the error specifies exactly what is the property that is triggering the circular reference. This is particularly useful when serializing more complex object hierarchies.
Proper architecture
One should never try to serialize data models (e.g. EF models), as ORM's navigation properties is the road to perdition when it comes to serialization. Data flow should be the following:
Database -> data models -> service models -> JSON string
Service models can be obtained from data models using auto mappers (e.g. Automapper). While this does not guarantee lack of circular references, proper design should do it: service models should contain exactly what the service consumer requires (i.e. the properties).
In those rare cases, when the client requests a hierarchy involving the same object type on different levels, the service can create a linear structure with parent->child relationship (using just identifiers, not references).
Modern applications tend to avoid loading complex data structures at once and service models should be slim. E.g.:
access an event - only header data (identifier, name, date etc.) is loaded -> service model (JSON) containing only header data
managed attendees list - access a popup and lazy load the list -> service model (JSON) containing only the list of attendees
Avoid converting the table object directly. If relations are set between other tables, it might throw this error.
Rather, you can create a model class, assign values to the class object and then serialize it.
I'm Using the fix, Because Using Knockout in MVC5 views.
On action
return Json(ModelHelper.GetJsonModel<Core_User>(viewModel));
function
public static TEntity GetJsonModel<TEntity>(TEntity Entity) where TEntity : class
{
TEntity Entity_ = Activator.CreateInstance(typeof(TEntity)) as TEntity;
foreach (var item in Entity.GetType().GetProperties())
{
if (item.PropertyType.ToString().IndexOf("Generic.ICollection") == -1 && item.PropertyType.ToString().IndexOf("SaymenCore.DAL.") == -1)
item.SetValue(Entity_, Entity.GetPropValue(item.Name));
}
return Entity_;
}
You can notice the properties that cause the circular reference. Then you can do something like:
private Object DeCircular(Object object)
{
// Set properties that cause the circular reference to null
return object
}
//first: Create a class as your view model
public class EventViewModel
{
public int Id{get;set}
public string Property1{get;set;}
public string Property2{get;set;}
}
//then from your method
[HttpGet]
public async Task<ActionResult> GetEvent()
{
var events = await db.Event.Find(x => x.ID != 0);
List<EventViewModel> model = events.Select(event => new EventViewModel(){
Id = event.Id,
Property1 = event.Property1,
Property1 = event.Property2
}).ToList();
return Json(new{ data = model }, JsonRequestBehavior.AllowGet);
}
An easier alternative to solve this problem is to return an string, and format that string to json with JavaScriptSerializer.
public string GetEntityInJson()
{
JavaScriptSerializer j = new JavaScriptSerializer();
var entityList = dataContext.Entitites.Select(x => new { ID = x.ID, AnotherAttribute = x.AnotherAttribute });
return j.Serialize(entityList );
}
It is important the "Select" part, which choose the properties you want in your view. Some object have a reference for the parent. If you do not choose the attributes, the circular reference may appear, if you just take the tables as a whole.
Do not do this:
public string GetEntityInJson()
{
JavaScriptSerializer j = new JavaScriptSerializer();
var entityList = dataContext.Entitites.toList();
return j.Serialize(entityList );
}
Do this instead if you don't want the whole table:
public string GetEntityInJson()
{
JavaScriptSerializer j = new JavaScriptSerializer();
var entityList = dataContext.Entitites.Select(x => new { ID = x.ID, AnotherAttribute = x.AnotherAttribute });
return j.Serialize(entityList );
}
This helps render a view with less data, just with the attributes you need, and makes your web run faster.
So I'm trying to use Autofac Automocking in ASP.NET MVC 5, but for some reason I can't get it to work.
Here's the test so far:
using (var mock = AutoMock.GetLoose())
{
const string mainUserID = "MainUserID";
const string otherUserID = "OtherUserID";
ApplicationUser user = new ApplicationUser()
{
Id = mainUserID,
UserName = "TestUser"
};
var dataProvider = mock.Mock<IDataProtectionProvider>();
dataProvider.DefaultValue = DefaultValue.Mock;
var userManagerMock = mock.Mock<ApplicationUserManager>();
}
The test fails when mocking the ApplicationUserManager. The error is this:
Result StackTrace:
at Autofac.Extras.Moq.AutoMock.Mock[T](Parameter[] parameters)
at AwenterWeb_NUnit.AccountControllerTest.<Deactivate_User>d__0.MoveNext() in C:\Users\Fabis\Documents\Docs\KvalifikÄcijas darbs 2015\AwenterWeb\AwenterWeb-NUnit\AccountControllerTest.cs:line 51
at NUnit.Framework.AsyncInvocationRegion.AsyncTaskInvocationRegion.WaitForPendingOperationsToComplete(Object invocationResult)
at NUnit.Core.NUnitAsyncTestMethod.RunTestMethod()
Result Message: System.InvalidCastException : Unable to cast object of type 'AwenterWeb.ApplicationUserManager' to type 'Moq.IMocked`1[AwenterWeb.ApplicationUserManager]'.
The same thing happens when trying to automock the ApplicationDbContext and it has a very simple constructor, so there shouldn't even be any issues with it.
I'm new to Mocking - what should I do in this scenario?
Edit: Also kind of an unrelated question, maybe you guys know - I've noticed that when creating a Moq for a DbSet using a list created previously in the test, I have to do this:
var dbSetMock = new Mock<IDbSet<DbEntity>>();
dbSetMock.Setup(m => m.Provider).Returns(data.Provider);
dbSetMock.Setup(m => m.Expression).Returns(data.Expression);
dbSetMock.Setup(m => m.ElementType).Returns(data.ElementType);
dbSetMock.Setup(m => m.GetEnumerator()).Returns(data.GetEnumerator());
It seems really unintuitive. Is there a way to just tell the mock to take the list? So something like:
dbSetMock.Setup(m => m).Returns(data);
Or any other way to create a DbSet Moq from an existing list quickly without having to write those 4 extra lines?
If you look at ligne 73 of MoqRegistrationHandler.cs you can see that only interface is moqable using Autofac.Extras.Moq
var typedService = service as TypedService;
if (typedService == null ||
!typedService.ServiceType.IsInterface ||
typedService.ServiceType.IsGenericType && typedService.ServiceType.GetGenericTypeDefinition() == typeof(IEnumerable<>) ||
typedService.ServiceType.IsArray ||
typeof(IStartable).IsAssignableFrom(typedService.ServiceType))
return Enumerable.Empty<IComponentRegistration>();
var rb = RegistrationBuilder.ForDelegate((c, p) => CreateMock(c, typedService))
.As(service)
.InstancePerLifetimeScope();
You can change the code but it may be quite difficult to make it works with non parameter less dependency.
Can your dependencies be changed to use an interface instead of a concrete class ? if it is not possible and/or if it doesn't make sense, you can use the MockRepository to create your non parameter-less component and then inject it on the AutoMock class.
class Program
{
static void Main(string[] args)
{
using (var mock = AutoMock.GetLoose())
{
/// configure your non interface component with constructor parameters
/// if foo need more complex parameters you can get them
/// using mock.Mock<T>().Object
var fooMock = mock.MockRepository.Create<Foo>((String)null);
fooMock.SetupGet(f => f.Value).Returns("test");
// insert your instance into the container
mock.Provide<Foo>(fooMock.Object);
var bar = mock.Create<Bar>();
Console.WriteLine(bar.GetValue());
}
}
}
public class Foo
{
public Foo(String value)
{
this._value = value;
}
private readonly String _value;
public virtual String Value
{
get
{
return this._value;
}
}
}
public interface IBar
{
String GetValue();
}
public class Bar : IBar
{
public Bar(Foo foo)
{
this._foo = foo;
}
private readonly Foo _foo;
public String GetValue()
{
return this._foo.Value;
}
}
It is not a perfect solution but without big refactoring of the Autofac.Extras.Moq project I can't see any simpler way to do it.
I am in the early stages of developing a new module.
Much of it is laid out in terms of the models etc. Also have the migrations all set up and my database now has the tables for my module.
I am encountering the following error when calling ContentManager.New<myPart> and would like some help please.
Error is this:
An unhandled exception has occurred and the request was terminated. Please refresh the page. If the error persists, go back
Specified cast is not valid.
System.InvalidCastException: Specified cast is not valid.
at Orchard.ContentManagement.ContentCreateExtensions.New[T]
(IContentManager manager, String contentType)
The chunk of code that fires the exception is this:
public static T New<T>(this IContentManager manager, string contentType) where T : class, IContent {
var contentItem = manager.New(contentType);
if (contentItem == null)
return null;
var part = contentItem.Get<T>();
if (part == null)
throw new InvalidCastException();
return part;
}
Here are the various parts to my module that are related to the operation i am struggling with:
ContentPart
public class GoogleMapsSettingsPart : ContentPart<GoogleMapsSettingsPartRecord>
{
public string ApiKey {
get { return Record.ApiKey; }
set { Record.ApiKey = value; }
}
}
ContentPartRecord
public class GoogleMapsSettingsPartRecord : ContentPartRecord
{
public virtual string ApiKey { get; set; }
}
Handler
public GoogleMapsSettingsPartHandler(IRepository<GoogleMapsSettingsPartRecord> repository)
{
Filters.Add(StorageFilter.For(repository));
}
Migration for this table
// Settings Table
SchemaBuilder.CreateTable("GoogleMapsSettingsPartRecord", table => table
.ContentPartRecord()
.Column("ApiKey", DbType.String, c => c.WithLength(60))
);
Some of the code from the controller for this model etc
public AdminController(IContentManager contentManager, IShapeFactory shapeFactory, IServiceLocatorService serviceLocatorService, INotifier notifier)
{
_contentManager = contentManager;
_serviceLocatorService = serviceLocatorService;
_notifier = notifier;
Shape = shapeFactory;
T = NullLocalizer.Instance;
}
/// <summary>
/// Display Settings
/// </summary>
/// <returns></returns>
public ActionResult Settings()
{
var settings = _serviceLocatorService.GoogleMapsSettings;
var editor = CreateSettingsEditor(settings);
var model = _services.ContentManager.BuildEditor(settings);
return View((object)model);
}
Finally - the Services where my call throws this exception
private GoogleMapsSettingsPart _settings;
public GoogleMapsSettingsPart GoogleMapsSettings
{
get {
if (_settings == null)
{
_settings = _contentManager.Query<GoogleMapsSettingsPart, GoogleMapsSettingsPartRecord>().List().FirstOrDefault();
if (_settings == null)
{
_settings = _contentManager.New<GoogleMapsSettingsPart>("GoogleMapsSettings");
}
}
return _settings;
}
}
The actual line where the exception happens is _settings = _contentManager.New<GoogleMapsSettingsPart>("GoogleMapsSettings");
I have tried all sorts of stuff in place of "GoogleMapsSettings" though nothing is working.
I'm pretty sure at this point it's something simple, though it's avoiding me..My limited knowledge of Orchard is stumping me
Any help would be appreciated :)
The exception is thrown because your content type does not have the part you specified to get.
_contentManager.New<GoogleMapsSettingsPart>("GoogleMapsSettings");
This method call creates a new content item of type GoogleMapSettings and gets the content item as a GoogleMapsSettingsPart. However, it seems that GoogleMapSettings content type does not have a GoogleMapsSettingsPart. That's why the exception gets thrown here.
var part = contentItem.Get<T>();
if (part == null)
throw new InvalidCastException();
You must either attach the part dynamically to your content type or do it in a migration (or manually in the admin, but that's not a good idea). Your migration should look like this.
this.ContentDefinitionManager.AlterTypeDefinition("GoogleMapsSettings",
alt => alt
.WithPart("GoogleMapsSettingsPart");
Ok, so I fixed it...
My understanding of how Orchard works is still very much in the learning stages.
for this particular operation I didn't want to have a content type in the admin - though not sure why after adding the ContentType it still didn't work...
anyway, adding the lines below to my handler took care of the rest. I believe it's actually creating a temporary type so one isn't needed in the system.
public GoogleMapsSettingsPartHandler(IRepository<GoogleMapsSettingsPartRecord> repository)
{
Filters.Add(new ActivatingFilter<GoogleMapsSettingsPart>("GoogleMapsSettings"));
Filters.Add(StorageFilter.For(repository));
Filters.Add(new TemplateFilterForRecord<GoogleMapsSettingsPartRecord>("GoogleMapsSettings", "Parts/GoogleMapsSettings"));
}
I'v got the same error, but in my case it was everything ok with migration class.
The reason was unlucky merge, which deleted my driver class of my part.
Just look at this code of Activating method of ContentPartDriverCoordinator class. In my case there was no partInfo for my content part and resulted part became ContentPart, so casting throws an exception
var partInfos = _drivers.SelectMany(cpp => cpp.GetPartInfo()).ToList();
foreach (var typePartDefinition in contentTypeDefinition.Parts) {
var partName = typePartDefinition.PartDefinition.Name;
var partInfo = partInfos.FirstOrDefault(pi => pi.PartName == partName);
var part = partInfo != null
? partInfo.Factory(typePartDefinition)
: new ContentPart { TypePartDefinition = typePartDefinition };
context.Builder.Weld(part);
}
I am trying to create a dataannotations attribute that controls field visiblity based on settings in a database. The attribute will be used within a system that will be used by multiple clients. Further, the visibility of the field needs to be able to change on the fly. I know I could do an if statement around each field in the view, but I am trying to avoid that and keep the visibility control within the view model as follows:
[Visible(FirstName)]
public string FirstName { get; set; }
I have tried creating this custom attribute that gets the value from a method from a resource class called ResourceType (which is generated using T4 and contains the necessary code to hit the database):
public class VisibleAttribute : Attribute, IMetadataAware
{
/// <summary>
/// Whether this field is visible
/// </summary>
public bool Hidden { get; set; }
public VisibleAttribute(string theFieldName)
{
ResourceType resources = new ResourceType();
Type _resourceType = typeof(ResourceType);
MethodInfo getHidden = _resourceType.GetMethod("IsHidden");
object[] requiredParams = new object[] { theFieldName };
Hidden = (bool)getHidden.Invoke(resources, requiredParams);
}
public void OnMetadataCreated(ModelMetadata metadata)
{
metadata.ShowForEdit = !Hidden;
metadata.HideSurroundingHtml = Hidden;
}
}
Here is an excerpt of the ResourceType class:
public class ResourceType
{
public const string Creditors_SecondaryCreditorsPayOffYesNo_Require = "Prop_Creditors_SecondaryCreditorsPayOffYesNo_Require";
public static string Prop_FieldName_Require
{
get { return GetHiddenOption(FieldName) ? "true" : "false"; }
}
internal static bool GetHiddenOption(string fieldName)
{
< < Logic here to get the option from the database > >
}
I have also tried the same attribute but with the following constructor:
public VisibleAttribute(string theFieldName)
{
ResourceType resources = new ResourceType();
Type _resourceType = typeof(ResourceType);
PropertyInfo getHidden = _resourceType.GetProperty(theFieldName);
Hidden = (bool)getHidden.GetValue
}
The problem I have with these two attempts is that, since the code is in the constructor, it only runs the first time I load the page after an IIS reset. So, any further changes I make to the visibility settings are not reflected without amother IIS reset.
I also tried creating a custom DataAnnotationsModelMetadataProvider that attempts to only load the setting once per page request:
public class EGTDataAnnotationsModelMetadataProvider : DataAnnotationsModelMetadataProvider
{
protected override ModelMetadata CreateMetadata(IEnumerable<Attribute> attributes, Type containerType,
Func<object> modelAccessor, Type modelType, string propertyName)
{
var data = base.CreateMetadata(attributes, containerType, modelAccessor, modelType, propertyName);
var visibleAttributeMetadata = attributes.SingleOrDefault(a => typeof(VisibleAttribute) == a.GetType());
if (visibleAttributeMetadata != null)
{
VisibleAttribute visibleAttribte = (VisibleAttribute)visibleAttributeMetadata;
if (!visibleAttribte.VisibleIsSet)
{
PropertyInfo getHidden = visibleAttribte.ResourceType.GetProperty("Prop_" + WebUtils.RemoveSectionNameSpace(visibleAttribte.SectionName) + "_" + visibleAttribte.FieldName + "_Hide");
visibleAttribte.IsHidden = bool.Parse(getHidden.GetValue(null, null).ToString());
data.HideSurroundingHtml = visibleAttribte.IsHidden;
data.ShowForEdit = !visibleAttribte.IsHidden;
visibleAttribte.VisibleIsSet = true;
}
else
{
data.HideSurroundingHtml = visibleAttribte.IsHidden;
data.ShowForEdit = !visibleAttribte.IsHidden;
}
}
return data;
}
}
One issue I have with the ModelMetadataProvider is that the CreateMetadata method runs many times for a single field during a single request. It is very inefficient code, and a huge decrease in performace, to call the database 5-10+ times per request to get a setting that has not changed since the beginning of the request. If I try to set a flag indicating I've already loaded the setting, I'm back to the same scenario as above where I don't see the setting change until after an IIS reset.
I'm hoping someone can give me some pointers as to what methods I can employ to see the database changes real time. Or am I trying to do the impossible? Thanks in advance.
You could combine the metadata provider approach with caching the value just that single request.
For this you could use the Items dictionary in the current HttpContext. Be careful with this as a redirect will cause the items to be cleared:
string cacheKey = String.Format("IsVisible-{0}", propertyName)
if(!HttpContext.Current.Items.Contains(cacheKey))
HttpContext.Current.Items[cacheKey] = //get setting from db
bool isVisible = (bool)HttpContext.Current.Items[cacheKey];
You can also consider using the ASP .Net Cache in case that you prefer caching the value not just for the current request (Although you mentioned that within your metadata provider you were trying to load the setting once per request)
I have been scratching my head all morning behind this but still haven't been able to figure out what might be causing this.
I have a composite repository object that references two other repositories. I'm trying to instantiate a Model type in my LINQ query (see first code snippet).
public class SqlCommunityRepository : ICommunityRepository
{
private WebDataContext _ctx;
private IMarketRepository _marketRepository;
private IStateRepository _stateRepository;
public SqlCommunityRepository(WebDataContext ctx, IStateRepository stateRepository, IMarketRepository marketRepository)
{
_ctx = ctx;
_stateRepository = stateRepository;
_marketRepository = marketRepository;
}
public IQueryable<Model.Community> Communities
{
get
{
return (from comm in _ctx.Communities
select new Model.Community
{
CommunityId = comm.CommunityId,
CommunityName = comm.CommunityName,
City = comm.City,
PostalCode = comm.PostalCode,
Market = _marketRepository.GetMarket(comm.MarketId),
State = _stateRepository.GetState(comm.State)
}
);
}
}
}
The repository objects that I'm passing in look like this
public class SqlStateRepository : IStateRepository
{
private WebDataContext _ctx;
public SqlStateRepository(WebDataContext ctx)
{
_ctx = ctx;
}
public IQueryable<Model.State> States
{
get
{
return from state in _ctx.States
select new Model.State()
{
StateId = state.StateId,
StateName = state.StateName
};
}
}
public Model.State GetState(string stateName)
{
var s = (from state in States
where state.StateName.ToLower() == stateName
select state).FirstOrDefault();
return new Model.State()
{
StateId = s.StateId,
StateName = s.StateName
};
}
AND
public class SqlMarketRepository : IMarketRepository
{
private WebDataContext _ctx;
public SqlMarketRepository(WebDataContext ctx)
{
_ctx = ctx;
}
public IQueryable<Model.Market> Markets
{
get
{
return from market in _ctx.Markets
select new Model.Market()
{
MarketId = market.MarketId,
MarketName = market.MarketName,
StateId = market.StateId
};
}
}
public Model.Market GetMarket(int marketId)
{
return (from market in Markets
where market.MarketId == marketId
select market).FirstOrDefault();
}
}
This is how I'm wiring it all up:
WebDataContext ctx = new WebDataContext();
IMarketRepository mr = new SqlMarketRepository(ctx);
IStateRepository sr = new SqlStateRepository(ctx);
ICommunityRepository cr = new SqlCommunityRepository(ctx, sr, mr);
int commCount = cr.Communities.Count();
The last line in the above snippet is where it fails. When I debug through the instantiation (new Model.Community), it never goes into any of the other repository methods. I do not have a relationship between the underlying tables behind these three objects. Would this be the reason that LINQ to SQL is not able to build the expression tree right?
These are non-hydrated queries, not fully-hydrated collections.
The Communities query differs from the other two because it calls methods as objects are hydrated. These method calls are not translatable to SQL.
Normally this isn't a problem. For example: if you say Communities.ToList(), it will work and the methods will be called from the objects as they are hydrated.
If you modify the query such that the objects aren't hydrated, for example: when you say Communities.Count(), linq to sql attempts to send the method calls into the database and throws since it cannot. It does this even though those method calls ultimately would not affect the resulting count.
The simplest fix (if you truly expect fully hydrated collections) is to add ToList to the community query, hydrating it.
Try adding another repository method that looks like this:
public int CommunitiesCount()
{
get { return _ctx.Communities.Count(); }
}
This will allow you to return a count without exposing the entire object tree to the user, which is what I think you're trying to do anyway.
As you may have already guessed, I suspect that what you are calling the anonymous types are at fault (they're not really anonymous types; they are actual objects, which you are apparently partially populating in an effort to hide some of the fields from the end user).