The story: in our multi-tenant app (one PostgreSql db, multiple schemas) we need to use one DbContext against multiple schemas.
What I tried: holding a cache (Dictionary, where key is schema name, value is the context for that schema). When instatiating new context for another schema I can see that dbContext schema is still set to previous schema provided. I assume the model in context is cached internally by context type, so that is the reason I see this behavior?
So above doesn't seem to work and I found that implementing IModelCacheKeyFactory should do the trick. Does anyone know what should go into Create method though? There are no samples nor documentation anywhere.
What I found:
Dynamically changing schema in Entity Framework Core but it answers for EF6, so not much help.
Here is an example.
Derived DbContext that replaces it's ModelCacheKey (and factory) with a Custom one.
class MyDbContext : DbContext
{
public MyDbContext(string schema)
{
Schema = schema;
}
public string Schema { get; }
protected override void OnConfiguring(DbContextOptionsBuilder options)
=> options
.UseSqlServer("...")
.ReplaceService<IModelCacheKeyFactory, MyModelCacheKeyFactory>();
protected override void OnModelCreating(ModelBuilder modelBuilder)
{
modelBuilder.HasDefaultSchema(Schema);
// ...
}
}
The factory that creates the Context with a specific key.
class MyModelCacheKeyFactory : IModelCacheKeyFactory
{
public object Create(DbContext context)
=> new MyModelCacheKey(context);
}
The custom ModelCacheKey per context.
class MyModelCacheKey : ModelCacheKey
{
string _schema;
public MyModelCacheKey(DbContext context)
: base(context)
{
_schema = (context as MyDbContext)?.Schema;
}
protected override bool Equals(ModelCacheKey other)
=> base.Equals(other)
&& (other as MyModelCacheKey)?._schema == _schema;
public override int GetHashCode()
{
var hashCode = base.GetHashCode() * 397;
if (_schema != null)
{
hashCode ^= _schema.GetHashCode();
}
return hashCode;
}
}
There actually have demo project in docs https://github.com/aspnet/EntityFramework.Docs/tree/master/samples/core/DynamicModel adding post for convinience !
Related
I am trying to inject a service into my action filter but I am not getting the required service injected in the constructor. Here is what I have:
public class EnsureUserLoggedIn : ActionFilterAttribute
{
private readonly ISessionService _sessionService;
public EnsureUserLoggedIn()
{
// I was unable able to remove the default ctor
// because of compilation error while using the
// attribute in my controller
}
public EnsureUserLoggedIn(ISessionService sessionService)
{
_sessionService = sessionService;
}
public override void OnActionExecuting(ActionExecutingContext context)
{
// Problem: _sessionService is null here
if (_sessionService.LoggedInUser == null)
{
context.HttpContext.Response.StatusCode = (int)HttpStatusCode.Unauthorized;
context.Result = new JsonResult("Unauthorized");
}
}
}
And I am decorating my controller like so:
[Route("api/issues"), EnsureUserLoggedIn]
public class IssueController : Controller
{
}
Startup.cs
services.AddScoped<ISessionService, SessionService>();
Using these articles as reference:
ASP.NET Core Action Filters
Action filters, service filters and type filters in ASP.NET 5 and MVC 6
Using the filter as a ServiceFilter
Because the filter will be used as a ServiceType, it needs to be registered with the framework IoC. If the action filters were used directly, this would not be required.
Startup.cs
public void ConfigureServices(IServiceCollection services) {
services.AddMvc();
services.AddScoped<ISessionService, SessionService>();
services.AddScoped<EnsureUserLoggedIn>();
...
}
Custom filters are added to the MVC controller method and the controller class using the ServiceFilter attribute like so:
[ServiceFilter(typeof(EnsureUserLoggedIn))]
[Route("api/issues")]
public class IssueController : Controller {
// GET: api/issues
[HttpGet]
[ServiceFilter(typeof(EnsureUserLoggedIn))]
public IEnumerable<string> Get(){...}
}
There were other examples of
Using the filter as a global filter
Using the filter with base controllers
Using the filter with an order
Take a look, give them a try and see if that resolves your issue.
Hope this helps.
Global filters
You need to implement IFilterFactory:
public class AuthorizationFilterFactory : IFilterFactory
{
public bool IsReusable => false;
public IFilterMetadata CreateInstance(IServiceProvider serviceProvider)
{
// manually find and inject necessary dependencies.
var context = (IMyContext)serviceProvider.GetService(typeof(IMyContext));
return new AuthorizationFilter(context);
}
}
In Startup class instead of registering an actual filter you register your filter factory:
services.AddMvc(options =>
{
options.Filters.Add(new AuthorizationFilterFactory());
});
One more way for resolving this problem. You can get your service via Context as in the following code:
public override void OnActionExecuting(ActionExecutingContext context)
{
_sessionService = context.HttpContext.RequestServices.GetService<ISessionService>();
if (_sessionService.LoggedInUser == null)
{
context.HttpContext.Response.StatusCode = (int)HttpStatusCode.Unauthorized;
context.Result = new JsonResult("Unauthorized");
}
}
Please note that you have to register this service in Startup.cs
services.AddTransient<ISessionService, SessionService>();
Example
private ILoginService _loginService;
public override void OnActionExecuting(ActionExecutingContext context)
{
_loginService = (ILoginService)context.HttpContext.RequestServices.GetService(typeof(ILoginService));
}
Hope it helps.
After reading this article ASP.NET Core - Real-World ASP.NET Core MVC Filters (Aug 2016) I implemented it like this:
In Starup.cs / ConfigureServices:
services.AddScoped<MyService>();
In MyFilterAttribute.cs:
public class MyFilterAttribute : TypeFilterAttribute
{
public MyFilterAttribute() : base(typeof (MyFilterAttributeImpl))
{
}
private class MyFilterAttributeImpl : IActionFilter
{
private readonly MyService _sv;
public MyFilterAttributeImpl(MyService sv)
{
_sv = sv;
}
public void OnActionExecuting(ActionExecutingContext context)
{
_sv.MyServiceMethod1();
}
public void OnActionExecuted(ActionExecutedContext context)
{
_sv.MyServiceMethod2();
}
}
}
In MyFooController.cs :
[MyFilter]
public IActionResult MyAction()
{
}
Edit: Passing arguments like [MyFilter("Something")] can be done using the Arguments property of the TypeFilterAttribute class: How do I add a parameter to an action filter in asp.net? (rboe's code also shows how to inject things (the same way))
While the question implicitly refers to "filters via attributes", it is still worth highlighting that adding filters "globally by type" supports DI out-of-the-box:
[For global filters added by type] any constructor dependencies will be populated by dependency injection (DI). Adding a filter by type is equivalent to filters.Add(new TypeFilterAttribute(typeof(MyFilter))).
https://learn.microsoft.com/en-us/aspnet/core/mvc/controllers/filters?view=aspnetcore-2.2#dependency-injection
With regards to attribute-based filters:
Filters that are implemented as attributes and added directly to controller classes or action methods cannot have constructor dependencies provided by dependency injection (DI). This is because attributes must have their constructor parameters supplied where they're applied. This is a limitation of how attributes work.
https://learn.microsoft.com/en-us/aspnet/core/mvc/controllers/filters?view=aspnetcore-2.2#dependency-injection
However, as mentioned in the previous answers to the OP, there are ways of indirection that can be used to achieve DI. For the sake of completeness, here are the links to the official docs:
ServiceFilterAttribute
TypeFilterAttribute
IFilterFactory implemented on your attribute
How can I set default scheme for Openiddict-core tables?
Unfortunately EF Core does't have (not that I know) a method that would accept only scheme and (EntityTypeBuilder.ToTable) requires also table name, beside scheme.
There's nothing OpenIddict-specific to handle that, but it's easily achievable using the regular EF hooks. Here's an example:
public class ApplicationDbContext : IdentityDbContext<ApplicationUser>
{
public ApplicationDbContext(DbContextOptions options)
: base(options) { }
protected override void OnModelCreating(ModelBuilder builder)
{
base.OnModelCreating(builder);
// Customize the ASP.NET Identity model and override the defaults if needed.
// For example, you can rename the ASP.NET Identity table names and more.
// Add your customizations after calling base.OnModelCreating(builder);
foreach (var entity in builder.Model.GetEntityTypes())
{
for (var type = entity; type != null; type = type.BaseType)
{
if (type.ClrType.Assembly == typeof(OpenIddictApplication).Assembly)
{
entity.SqlServer().Schema = "security";
break;
}
}
}
}
}
My session uses an NHInterceptor to add INotifyPropertyChanged support to models.
// I use the session generated here to fetch Data
public class SessionServiceImpl : ISessionService
{
[Inject]
public ISessionFactory SessionFactory { get; set; }
[Inject]
public NhChangeNotificationInterceptorImpl ChangeNotificationInterceptor { get; set; }
public ISession GetSession() // reduced code here
{
return SessionFactory.OpenSession(ChangeNotificationInterceptor);
}
}
// This is the interceptor implementation
public class NhChangeNotificationInterceptorImpl : EmptyInterceptor, IInterceptor
{
[Inject]
public ISessionFactory SessionFactory { get; set; }
[Inject]
public ViewModelProxyFactory ProxyFactory { get; set; }
public override object Instantiate(string entityTypeName, EntityMode entityMode, object id)
{
Type type = Type.GetType(entityTypeName);
if (type == null) { /* Throw Exception*/ }
bool isViewModel = false;
while (type != typeof(object))
{
Type tempType = type.BaseType;
if (tempType == typeof(ViewModelBase))
{
isViewModel = true;
break;
}
}
if (entityMode == EntityMode.Poco && isViewModel)
{
var instance = ProxyFactory.CreateProxy(type);
SessionFactory.GetClassMetadata(entityTypeName).SetIdentifier(instance, id, entityMode);
return instance;
}
return base.Instantiate(entityTypeName, entityMode, id);
}
}
The ProxyFactory uses Castle to create proxies that add change notification functionality. That means all my objects come from the DB as Castle Proxies, which are AFAIK transparent.
Whenever I pass one of those NH Generated MVVM-proxies into Session.Save(), all's fine.
Now, as data driven Applications go, I also need to create new instances and save them. I can create instances of the model type and save them via the session all right. Creating a MVVM proxy instance (using Ninject to ensure that the same SessionFactory and ProxyFactory instances are used all over) and throwing this into Session.Save() results in the following:
"NHibernate.MappingException".
Message=No persister for: Castle.Proxies.FieldDescriptionProxy
Source=NHibernate
StackTrace:
at NHibernate.Impl.SessionFactoryImpl.GetEntityPersister(String entityName)
at NHibernate.Impl.SessionImpl.GetEntityPersister(String entityName, Object obj)
at NHibernate.Event.Default.AbstractSaveEventListener.SaveWithGeneratedId(Object entity, String entityName, Object anything, IEventSource source, Boolean requiresImmediateIdAccess)
at NHibernate.Event.Default.DefaultSaveOrUpdateEventListener.SaveWithGeneratedOrRequestedId(SaveOrUpdateEvent event)
at NHibernate.Event.Default.DefaultSaveEventListener.SaveWithGeneratedOrRequestedId(SaveOrUpdateEvent event)
at NHibernate.Event.Default.DefaultSaveOrUpdateEventListener.EntityIsTransient(SaveOrUpdateEvent event)
at NHibernate.Event.Default.DefaultSaveEventListener.PerformSaveOrUpdate(SaveOrUpdateEvent event)
at NHibernate.Event.Default.DefaultSaveOrUpdateEventListener.OnSaveOrUpdate(SaveOrUpdateEvent event)
at NHibernate.Impl.SessionImpl.FireSave(SaveOrUpdateEvent event)
at NHibernate.Impl.SessionImpl.Save(Object obj)
at Interpretation.UI.Service.Impl.Dao.FieldDao.SaveFields(IList`1 fields, ISession session) in C:\...\FieldDao.cs:Zeile 51.
InnerException:
Any ideas what goes wrong here (or what I might have forgotten)?
EDIT : Now got it working, but why do I have to add recognition logic (see code below) to the interceptor for instances created outside, while instances created inside can be persisted as the are?
public override string GetEntityName(object entity)
{
Type type = entity.GetType();
if (type.FullName.StartsWith("Castle.Proxies") &&
type.FullName.EndsWith("Proxy"))
{
return type.BaseType.FullName;
}
return base.GetEntityName(entity);
}
Implementing the GetEntityName method did the trick.
public override string GetEntityName(object entity)
{
Type type = entity.GetType();
if (type.FullName.StartsWith("Castle.Proxies") &&
type.FullName.EndsWith("Proxy"))
{
return type.BaseType.FullName;
}
return base.GetEntityName(entity);
}
article and sample code(Intercepting Entity Creation) demonstrating the use of dynamic proxies to implement (WPF) change notification. As seen there you'll have to use the same proxy generator in NHibernate and outside to implement the recognition of the proxies for NHibernate (see class DataBindingIntercepter method GetEntityName) .
why do I have to add recognition logic (see code below) to the interceptor for instances created outside, while instances created inside can be persisted as the are?
Methods on ISession only add instances to dictionary to search for dirty ones on Flush. Since all instances loaded are automaticly part of this dictionary the session immediatly "knows" the entity and does nothing in SaveOrUpdate. For other instances coming from outside it first has to get the appropriate mapping (using the entitiyname which defaults to the class fullname) to know which properties form the primary key.
My aim is to implement a quite generic search mechanism. Here's the general idea:
you can search based on any property of the entity you're searching for (for example- by Employee's salary, or by Department name etc.).
Each property you can search by is represented by a class, which inherits from EntityProperty:
public abstract class EntityProperty<T>
where T:Entity
{
public enum Operator
{
In,
NotIn,
}
/// <summary>
/// Name of the property
/// </summary>
public abstract string Name { get; }
//Add a search term to the given query, using the given values
public abstract IQueryable<T> AddSearchTerm(IQueryable<T> query, IEnumerable<object> values);
public abstract IQueryable<T> AddSortingTerm(IQueryable<T> query);
protected Operator _operator = Operator.In;
protected bool _sortAscending = false;
public EntityProperty(Operator op)
{
_operator = op;
}
//use this c'tor if you're using the property for sorting only
public EntityProperty(bool sortAscending)
{
_sortAscending = sortAscending;
}
}
all of the properties you're searching / sorting by are stored in a simple collection class:
public class SearchParametersCollection<T>
where T: Entity
{
public IDictionary<EntityProperty<T>,IEnumerable<object>> SearchProperties { get; private set; }
public IList<EntityProperty<T>> SortProperties { get; private set; }
public SearchParametersCollection()
{
SearchProperties = new Dictionary<EntityProperty<T>, IEnumerable<object>>();
SortProperties = new List<EntityProperty<T>>();
}
public void AddSearchProperty(EntityProperty<T> property, IEnumerable<object> values)
{
SearchProperties.Add(property, values);
}
public void AddSortProperty(EntityProperty<T> property)
{
if (SortProperties.Contains(property))
{
throw new ArgumentException(string.Format("property {0} already exists in sorting order", property.Name));
}
SortProperties.Add(property);
}
}
now, all the repository class has to do is:
protected IEnumerable<T> Search<T>(SearchParametersCollection<T> parameters)
where T : Entity
{
IQueryable<T> query = this.Session.Linq<T>();
foreach (var searchParam in parameters.SearchProperties)
{
query = searchParam.Key.AddSearchTerm(query, searchParam.Value);
}
//add order
foreach (var sortParam in parameters.SortProperties)
{
query = sortParam.AddSortingTerm(query);
}
return query.AsEnumerable();
}
for example, here's a class which implements searching a user by their full name:
public class UserFullName : EntityProperty<User>
{
public override string Name
{
get { return "Full Name"; }
}
public override IQueryable<User> AddSearchTerm(IQueryable<User> query, IEnumerable<object> values)
{
switch (_operator)
{
case Operator.In:
//btw- this doesn't work with nHibernate... :(
return query.Where(u => (values.Cast<string>().Count(v => u.FullName.Contains(v)) > 0));
case Operator.NotIn:
return query.Where(u => (values.Cast<string>().Count(v => u.FullName.Contains(v)) == 0));
default:
throw new InvalidOperationException("Unrecognized operator " + _operator.ToString());
}
}
public override IQueryable<User> AddSortingTerm(IQueryable<User> query)
{
return (_sortAscending) ? query.OrderBy(u => u.FullName) : query.OrderByDescending(u => u.FullName);
}
public UserFullName(bool sortAscending)
: base(sortAscending)
{
}
public UserFullName(Operator op)
: base(op)
{
}
}
my questions are:
1. firstly- am I even on the right track? I don't know of any well-known method for achieving what I want, but I may be wrong...
2. it seems to me that the Properties classes should be in the domain layer and not in the DAL, since I'd like the controller layers to be able to use them. However, that prevents me from using any nHibernate-specific implementation of the search (i.e any other interface but Linq). Can anybody think of a solution that would enable me to utilize the full power of nH while keeping these classes visible to upper layers? I've thought about moving them to the 'Common' project, but 'Common' has no knowledge of the Model entities, and I'd like to keep it that way.
3. as you can see by my comment for the AddSearchTerm method- I haven't really been able to implement 'in' operator using nH (I'm using nH 2.1.2 with Linq provider). any sugggestions in that respect would be appriciated. (see also my question from yesterday).
thanks!
If you need good API to query NHIbernate objects then you should use ICriteria (for NH 2.x) or QueryOver (for NH 3.x).
You over complicating DAL with these searches. Ayende has a nice post about why you should not do it
I ended up using query objects, which greatly simplified things.
I think this falls under the concept of contextual binding, but the Ninject documentation, while very thorough, does not have any examples close enough to my current situation for me to really be certain. I'm still pretty confused.
I basically have classes that represent parameter structures for queries. For instance..
class CurrentUser {
string Email { get; set; }
}
And then an interface that represents its database retrieval (in the data layer)
class CurrentUserQuery : IQueryFor<CurrentUser> {
public CurrentUserQuery(ISession session) {
this.session = session;
}
public Member ExecuteQuery(CurrentUser parameters) {
var member = session.Query<Member>().Where(n => n.Email == CurrentUser.Email);
// validation logic
return member;
}
}
Now then, what I want to do is to establish a simple class that can take a given object and from it get the IQueryFor<T> class, construct it from my Ninject.IKernel (constructor parameter), and perform the ExecuteQuery method on it, passing through the given object.
The only way I have been able to do this was to basically do the following...
Bind<IQueryFor<CurrentUser>>().To<CurrentUserQuery>();
This solves the problem for that one query. But I anticipate there will be a great number of queries... so this method will become not only tedious, but also very prone to redundancy.
I was wondering if there is an inherit way in Ninject to incorporate this kind of behavior.
:-
In the end, my (ideal) way of using this would be ...
class HomeController : Controller {
public HomeController(ITransit transit) {
// injection of the transit service
}
public ActionResult CurrentMember() {
var member = transit.Send(new CurrentUser{ Email = User.Identity.Name });
}
}
Obviously that's not going to work right, since the Send method has no way of knowing the return type.
I've been dissecting Rhino Service Bus extensively and project Alexandria to try and make my light, light, lightweight implementation.
Update
I have been able to get a fairly desired result using .NET 4.0 dynamic objects, such as the following...
dynamic Send<T>(object message);
And then declaring my interface...
public interface IQueryFor<T,K>
{
K Execute(T message);
}
And then its use ...
public class TestCurrentMember
{
public string Email { get; set; }
}
public class TestCurrentMemberQuery : IConsumerFor<TestCurrentMember, Member>
{
private readonly ISession session;
public TestCurrentMemberQuery(ISession session) {
this.session = session;
}
public Member Execute(TestCurrentMember user)
{
// query the session for the current member
var member = session.Query<Member>()
.Where(n => n.Email == user.Email).SingleOrDefault();
return member;
}
}
And then in my Controller...
var member = Transit.Send<TestCurrentMemberQuery>(
new TestCurrentMember {
Email = User.Identity.Name
}
);
effectively using the <T> as my 'Hey, This is what implements the query parameters!'. It does work, but I feel pretty uncomfortable with it. Is this an inappropriate use of the dynamic function of .NET 4.0? Or is this more the reason why it exists in the first place?
Update (2)
For the sake of consistency and keeping this post relative to just the initial question, I'm opening up a different question for the dynamic issue.
Yes, you should be able to handle this with Ninject Conventions. I am just learning the Conventions part of Ninject, and the documentation is sparse; however, the source code for the Conventions extension is quite light and easy to read/navigate, also Remo Gloor is very helpful both here and on the mailing list.
The first thing I would try is a GenericBindingGenerator (changing the filters and scope as needed for your application):
internal class YourModule : NinjectModule
{
public override void Load()
{
Kernel.Scan(a => {
a.From(System.Reflection.Assembly.GetExecutingAssembly());
a.InTransientScope();
a.BindWith(new GenericBindingGenerator(typeof(IQueryFor<>)));
});
}
}
The heart of any BindingGenerator is this interface:
public interface IBindingGenerator
{
void Process(Type type, Func<IContext, object> scopeCallback, IKernel kernel);
}
The Default Binding Generator simply checks if the name of the class matches the name of the interface:
public void Process(Type type, Func<IContext, object> scopeCallback, IKernel kernel)
{
if (!type.IsInterface && !type.IsAbstract)
{
Type service = type.GetInterface("I" + type.Name, false);
if (service != null)
{
kernel.Bind(service).To(type).InScope(scopeCallback);
}
}
}
The GenericBindingGenerator takes a type as a constructor argument, and checks interfaces on classes scanned to see if the Generic definitions of those interfaces match the type passed into the constructor:
public GenericBindingGenerator(Type contractType)
{
if (!contractType.IsGenericType && !contractType.ContainsGenericParameters)
{
throw new ArgumentException("The contract must be an open generic type.", "contractType");
}
this._contractType = contractType;
}
public void Process(Type type, Func<IContext, object> scopeCallback, IKernel kernel)
{
Type service = this.ResolveClosingInterface(type);
if (service != null)
{
kernel.Bind(service).To(type).InScope(scopeCallback);
}
}
public Type ResolveClosingInterface(Type targetType)
{
if (!targetType.IsInterface && !targetType.IsAbstract)
{
do
{
foreach (Type type in targetType.GetInterfaces())
{
if (type.IsGenericType && (type.GetGenericTypeDefinition() == this._contractType))
{
return type;
}
}
targetType = targetType.BaseType;
}
while (targetType != TypeOfObject);
}
return null;
}
So, when the Conventions extension scans the class CurrentUserQuery it will see the interface IQueryFor<CurrentUser>. The generic definition of that interface is IQueryFor<>, so it will match and that type should get registered for that interface.
Lastly, there is a RegexBindingGenerator. It tries to match interfaces of the classes scanned to a Regex given as a constructor argument. If you want to see the details of how that operates, you should be able to peruse the source code for it now.
Also, you should be able to write any implementation of IBindingGenerator that you may need, as the contract is quite simple.