S#arp Architecture 1.9 + Fluently.Configure() NHiberbante - nhibernate

I have a need to fluently configure nhibernate in my S#arp application so that I can use a custom NHibernate.Search directory for each of my tenants in a multi-tenant app.
However I have googled for hours looking for a solution but can't seem to find anything current that works.
Thanks,
Paul

I haven't tried this myself, but AddConfiguration takes a dictionary of cfgProperties, which I guess you can pass the tenant specific hibernate.search.default.indexBase value to.
I had a look at this, adding the key as described above will cause a problem if you attempt to use CfgHelper.LoadConfiguration() since it will return null.
But you can configure NHSearch to use different directories for each factory using the factory key:
<nhs-configuration xmlns="urn:nhs-configuration-1.0">
<search-factory sessionFactoryName="YOUR_TENANT1_FACTORY_KEY">
<property name="hibernate.search.default.indexBase">~\IndexTenant1</property>
</search-factory>
<search-factory sessionFactoryName="YOUR_TENANT2_FACTORY_KEY">
<property name="hibernate.search.default.indexBase">~\Tenant2</property>
</search-factory>
</nhs-configuration>
If you are following instructions on
http://wiki.sharparchitecture.net/Default.aspx?Page=NHibSearch
You would need to change the method GetIndexDirectory to
private string GetIndexDirectory() {
INHSConfigCollection nhsConfigCollection = CfgHelper.LoadConfiguration();
string factoryKey = SessionFactoryAttribute.GetKeyFrom(this); // Change this with however you get the factory key for your tenants,
string property = nhsConfigCollection.GetConfiguration(factoryKey).Properties["hibernate.search.default.indexBase"];
var fi = new FileInfo(property);
return Path.Combine(AppDomain.CurrentDomain.BaseDirectory, fi.Name);
}

Related

In Identity for Entity Framework Core 3.x, how do I find users by Email (minding normalization) when Email uniqueness is not enforced?

userManager.FindByEmailAsync(myEmail) throws an exception if there are multiple users with the same email.
I could use:
await context.ApplicationUsers
.FirstOrDefaultAsync(x => x.NormalizedEmail == myEmail.ToUpperInvariant());
That seems to work okay. But I'm not sure if ToUpperInvariant is the right way to check, because System.Text also has Normalize(). It won't matter right now since we are using SQL Server with a case-insensitive configuration, but I don't want things to break if we ever change that.
Am I normalizing in a way that is consistent with how Entity Framework does it? I tried to find the source code, but what I found doesn't use the NormalizedEmail field, so it's likely old.
The normalization is not done by the EF Core, but the UserManager class (using ILookupNormalizer service injected via constructor or set via KeyNormalizer property).
UserManager.FindByEmailAsync method does the normalization for you before calling the store method. The problem is that EF Core store method implementation uses SingleOrDefaultAsync which throws if there are duplicate normalized emails in the database.
To fix that, you could use UserManager.NormalizeEmail method to do the normalization, and then use FirstOrDefaultAsync query as in your sample:
var normalizedEmail = userManager.NormalizeEmail(myEmail);
var firstDuplicate = await userManager.Users
.FirstOrDefaultAsync(x => x.NormalizedEmail == normalizedEmail);

The name 'NodaTimeField' does not exist in the current context error during installation of index on RavenDB

I am using NodaTime's LocalDate in RavenDB index.
Here is an example of the index:
public class TaskIndex : AbstractIndexCreationTask<ScheduleTask>
{
public TaskIndex()
{
Map = tasks => from task in tasks
select new
{
task.Name,
PlannedStartDate = task.PlannedStartDate.AsLocalDate().Resolve(),
PlannedDueDate = task.PlannedDueDate.AsLocalDate().Resolve()
};
Index(x => x.Name, FieldIndexing.Analyzed);
Store(x => x.Name, FieldStorage.Yes);
TermVector(x => x.Name, FieldTermVector.WithPositionsAndOffsets);
}
}
I installed RavenDB-NodaTime bundle as described here.
Here is a piece of code I use to install index:
var assembly = AppDomain.CurrentDomain.Load(new AssemblyName
{
Name = "cs.Scheduling"
});
var catalog = new AssemblyCatalog(assembly);
var provider = new CompositionContainer(catalog);
var commands = documentStore.DatabaseCommands.ForDatabase(dbName);
IndexCreation.CreateIndexes(provider, commands, documentStore.Conventions);
documentStore is configured with default database, but then I use it to install index to different (tenant) database name of which comes in dbName.
During the installation of the index I got an exception: The name 'NodaTimeField' does not exist in the current context.
I have one default database which is completely different from database I try to install index for. So basically the case is similar to one described here but I am using standalone version of RavenDB server.
I tried to find out how I can do suggested there but was not able to do that:
embeddableDocumentStore.ServerIfEmbedded.Options.DatabaseLandlord.SetupTenantConfiguration += configuration =>
{
configuration.Catalog.Catalogs.Add(new TypeCatalog(typeof(DeleteOnConflict)));
configuration.Catalog.Catalogs.Add(new TypeCatalog(typeof(PutOnConflict)));
};
Version of RavenDB I am using is 2.5.2956.
RavenDB.Client.NodaTime - 2.5.10.
Hope for your help. Thanks.
In my case that was a very silly mistake. When I was installing RavenDB server some time ago I installed it into non-default destination. Later some RavenDB updates were installed into default destination (i.e. \Program Files (x86)\RavenDB). And when I was installing RavenDB-NodaTime bundle I put it into incorrect destination (\Program Files (x86)\RavenDB).
After detecting this issue and configuring RavenDB server in my correct destination properly an error described in the heading has gone away.
Hope this answer can help somebody else.
P.S. Later there was a deserialization error during reading data from db (RavenDB was not aware of how to deserialize date from string in "yyyy-MM-dd" format to LocalDate object) which I fixed by calling store.ConfigureForNodaTime(DateTimeZoneProviders.Tzdb); after store.Initialize(); call as Steven suggested in his answer.
I believe the answer is that your tenant database does not have the bundle "activated" Your database document (under settings in Raven 3) should have something like
"Raven/ActiveBundles": "Encryption;Compression;NodaTime"
Also you must call
store.ConfigureForNodaTime(DateTimeZoneProviders.Tzdb);
I call this after store.Initialize(). Once you do both of these things, you may have to fix existing data by re-saving your documents (not sure if there is another way). New data will be properly stored like '2016-2-3' format which should make your index return data.

S#arp Architecture - Rhino Security (unmapped class: Rhino.Security.IUser)

I'm using S#arp Architecture 1.6 and have implemented the Rhino Security integration as per
Rhino Security - S#arp Architecture
I'm using the latest build from Rhino.Commons
My Application_EndRequest method contains
ISession session = NHibernateSession.Current;
My ComponentRegister.cs contains
container.Kernel.Register(
Component.For<IAuthorizationService>()
.ImplementedBy<AuthorizationService>()
.LifeStyle.Is(LifestyleType.Transient),
Component.For<IAuthorizationRepository>()
.ImplementedBy<AuthorizationRepository>()
.LifeStyle.Is(LifestyleType.Transient),
Component.For<IPermissionsBuilderService>()
.ImplementedBy<PermissionsBuilderService>()
.LifeStyle.Is(LifestyleType.Transient),
Component.For<IPermissionsService>()
.ImplementedBy<PermissionsService>()
.LifeStyle.Is(LifestyleType.Transient),
Component.For<IUnitOfWorkFactory>()
.ImplementedBy<NHibernateUnitOfWorkFactory>()
.LifeStyle.Is(LifestyleType.Singleton),
Component.For<Rhino.Commons.IRepository<User>>()
.ImplementedBy<NHRepository<User>>()
.LifeStyle.Is(LifestyleType.Transient)
);
container.AddFacility<FactorySupportFacility>()
.Register(Component.For<ISession>()
.UsingFactoryMethod(() => NHibernateSession.Current)
.LifeStyle.Is(LifestyleType.Transient));
I have also added RhinoSecurityPersistenceConfigurer() as per instructions.
The error I'm recieving on calling
UnitOfWork.Start()
is
An association from the table Permissions refers to an unmapped class: Rhino.Security.IUser
Does anyone know what the cause of this error may be?
Has anyone successfully integrated Rhino.Security with S#arp Architecture?
Any help would be great.
Thanks
Rich
-- Additional Details --
Thanks for all the replies so far.
I've still not been able to resolve this, so thought I'd add more details.
In my Global.asax.cs I have
private void InitializeNHibernateSession()
{
NHibernateSession.Init(
webSessionStorage,
new string[] { Server.MapPath("~/bin/SwitchSnapshot.Data.dll") },
new AutoPersistenceModelGenerator().Generate(),
Server.MapPath("~/NHibernate.config"),
null, null, new RhinoSecurityPersistenceConfigurer());
}
RhinoSecurityPersistenceConfigurer :
public Configuration ConfigureProperties(Configuration nhibernateConfig)
{
Security.Configure<User>(nhibernateConfig, SecurityTableStructure.Prefix);
return nhibernateConfig;
}
I have an AuthorizationAttribute which calls
using (UnitOfWork.Start())
The error is occuring in NHibernateUnitOfWorkFactory.cs as
sessionFactory = cfg.BuildSessionFactory();
You need an NHibernate mapping for your User class (i.e. the class that implements the IUser interface). You also need a table in the database with the correct fields for your User class.
You have to let RS do some configuration work before the SessionFactory is created. Look at the second issue here http://groups.google.com/group/sharp-architecture/browse_frm/thread/4093c52596f54d23/194f19cd08c8fdd7?q=#194f19cd08c8fdd7. It should get you in the right direction.
Thanks to all who helped out.
In the end it was my own fault.
All I needed to do was to follow the S#arp Architecture Instructions a little better.
From an old version of S#arp I had 2 config files hibernate.cfg.xml and NHibernate.config. I thought I still needed both, but all I needed was hibernate.cfg.xml for S#arp version 1.6 and mapped User.cs using Fluent NHibernate.
Other changes I did was in ComponentRegister.cs
container.Kernel.Register(
Component.For<IAuthorizationService>()
.ImplementedBy<AuthorizationService>()
.LifeStyle.Is(LifestyleType.Transient),
Component.For<IAuthorizationRepository>()
.ImplementedBy<AuthorizationRepository>()
.LifeStyle.Is(LifestyleType.Transient),
Component.For<IPermissionsBuilderService>()
.ImplementedBy<PermissionsBuilderService>()
.LifeStyle.Is(LifestyleType.Transient),
Component.For<IPermissionsService>()
.ImplementedBy<PermissionsService>()
.LifeStyle.Is(LifestyleType.Transient),
Component.For<IUnitOfWorkFactory>()
.ImplementedBy<NHibernateUnitOfWorkFactory>()
.LifeStyle.Is(LifestyleType.Singleton),
Component.For<Rhino.Commons.IRepository<User>>()
.ImplementedBy<NHRepository<User>>()
.LifeStyle.Is(LifestyleType.Transient)
);
container.Kernel.AddFacility<FactorySupportFacility>()
.Register(Component.For<ISession>()
.UsingFactoryMethod(() => NHibernateSession.Current)
.LifeStyle.Is(LifestyleType.Transient)
);
Then use the following in my code.
var authorizationService = IoC.Resolve<IAuthorizationService>();
using (UnitOfWork.Start())
{
}

SessionFactory - one factory for multiple databases

We have a situation where we have multiple databases with identical schema, but different data in each. We're creating a single session factory to handle this.
The problem is that we don't know which database we'll connect to until runtime, when we can provide that. But on startup to get the factory build, we need to connect to a database with that schema. We currently do this by creating the schema in an known location and using that, but we'd like to remove that requirement.
I haven't been able to find a way to create the session factory without specifying a connection. We don't expect to be able to use the OpenSession method with no parameters, and that's ok.
Any ideas?
Thanks
Andy
Either implement your own IConnectionProvider or pass your own connection to ISessionFactory.OpenSession(IDbConnection) (but read the method's comments about connection tracking)
The solution we came up with was to create a class which manages this for us. The class can use some information in the method call to do some routing logic to figure out where the database is, and then call OpenSession passing the connection string.
You could also use the great NuGet package from brady gaster for this. I made my own implementation from his NHQS package and it works very well.
You can find it here:
http://www.bradygaster.com/Tags/nhqs
good luck!
Came across this and thought Id add my solution for future readers which is basically what Mauricio Scheffer has suggested which encapsulates the 'switching' of CS and provides single point of management (I like this better than having to pass into each session call, less to 'miss' and go wrong).
I obtain the connecitonstring during authentication of the client and set on the context then, using the following IConnectinProvider implementation, set that value for the CS whenever a session is opened:
/// <summary>
/// Provides ability to switch connection strings of an NHibernate Session Factory (use same factory for multiple, dynamically specified, database connections)
/// </summary>
public class DynamicDriverConnectionProvider : DriverConnectionProvider, IConnectionProvider
{
protected override string ConnectionString
{
get
{
var cxnObj = IsWebContext ?
HttpContext.Current.Items["RequestConnectionString"]:
System.Runtime.Remoting.Messaging.CallContext.GetData("RequestConnectionString");
if (cxnObj != null)
return cxnObj.ToString();
//catch on app startup when there is not request connection string yet set
return base.ConnectionString;
}
}
private static bool IsWebContext
{
get { return (HttpContext.Current != null); }
}
}
Then wire it in during NHConfig:
var configuration = Fluently.Configure()
.Database(MsSqlConfiguration.MsSql2005
.Provider<DynamicDriverConnectionProvider>() //Like so

NHibernate HybridSessionBuilder, how to switch hibernate cfg based upon url values

I am using the HybridSessionBuilder supplied by Palermo and his team .. link ..
We have our staging environments set up so that the url will be one of the following based on the environment
web-test.company.com
web-cert.company.com
web.company.com
what we normally do is take a look at the url and if it has "-test" we use the test configurations and so on (connection strings, etc).
This is the first project that uses nhibernate in this type of environment. What would be a good way to tell the Session Builder to use the correct hibernate cfg (I will build 1 for each environment).
The HybridSessionBuilder lives in an infrastructure layer and is injected into repositories via StructureMap.
Here's how I select a single configuration file using the HybridSessionBuilder:
public Configuration GetConfiguration()
{
var configuration = new Configuration();
string cfgFile = Path.GetDirectoryName(Assembly.GetAssembly(this.GetType()).CodeBase) +
"\\com.Data.nHibernate.cfg.xml";
configuration.Configure(cfgFile);
configuration.AddAssembly("com.Data");
return configuration;
}
If you want to select configuration files based on the URL I would just identify the call stack that leads to this function and pass in either an enum value or the config file's name directly.