I am using standard configuration pattern for ASP.NET Core applications and I can not bind configuration to my class as it has construtor with parameters.
In appsettings.json I included desired config:
"MyServiceConfig": {
"Identity": {
"Version": "1.0",
"ComplicatedUri": {
"Scheme": "http",
"Authority": "localhost",
"Path": "SuperService"
}
}
},
My config class and it's dependencies look like that:
public class MyServiceConfig
{
public MyIdentity Identity { get; set; }
}
public class MyIdentity
{
public string IdentityName { get; set; }
public string Version { get; set; }
public MyComplicatedUri ComplicatedProperty { get; set; }
public MyIdentity(string version, MyComplicatedUri complicatedProperty)
{
Version = version;
ComplicatedProperty = complicatedProperty;
IdentityName = complicatedProperty.Path;
}
}
public class MyComplicatedUri
{
public string Scheme { get; set; }
public string Authority { get; set; }
public string Path { get; set; }
}
I have already tried code like that:
private MyServiceConfig GetMyConfig(IConfiguration configuration)
{
var config = new MyServiceConfig();
configuration.GetSection("MyServiceConfig").Bind(config);
return config;
}
It throws exception:
'Cannot create instance of type 'MyIdentity' because it is missing
a public parameterless constructor.'
That behaviour can make sense in some cases but in that particular one not so much. Mappings could be straightforward - by property names which have public setters or by constructor parameter names.
Another idea would be adding converter in AddJsonOptions in Startup class for my types - IConfiguration.Bind could infer how to construct it but I also tried that with no success.
Have you encoutered similar problems and found some reasonable solution to that?
Edit: Adding parameterless constructor will work of course, but sometimes I need to deal with some classes from external packages I'd like to use as parts of my config class so let's assume we can not modify them. I'd like to avoid adding new types for mapping only as well. Ideally I'd like to force ASP.NET Core engine to use existing constructor with parameters and by parameter name map with json properties - which currently is not working.
You should just add a default constructor in MyIdentity class.
.bind() binds the configuration into the object using the default constructor.
So, add the required default constructor in your MyIdentity class and it will be fine.
public MyIdentity(){}
Also, you can use Options.
In ConfigureServices, add the following:
services.AddOptions();
services.ConfigureOptions<MyServiceConfig>();
and then use dependency injection to initialize it.
In addition, use your own JsonConverter
Related
So I have a simple configuration class PubsubSettings:
public class PubSubSettings
{
public string ProjectId { get; set; }
public string TopicId { get; set; }
public int PartnerId { get; set; }
public string SubscriptionId { get; set; }
}
I have previously only had one of these configured in my appsettings.json but now I want to be able to handle an arbitrary number of them.
I have another class, PubSub, that I usually inject an IOptions<PubSubSettings> into. And this, in turn, gets injected into my Worker class.
services.Configure<PubSubSettings>(configuration.GetSection(nameof(PubSubSettings)));
...
services.AddHostedService<Worker>();
So, what I want to do now, is add a new Worker as a hosted service for each entry in my AppSettings PubSubSettings section and inject the relevant IOptions<PubSubSettings> into each of these (along with the standard ILogger).
So in essence, I'd like this config block:
"PubsubSettings": [
{
"ProjectId": "project1",
"TopicId": "topic",
"PartnerId": 1,
"SubscriptionId": "sub1"
},
{
"ProjectId": "project2",
"TopicId": "topic2",
"PartnerId": 2,
"SubscriptionId": "sub2"
}
]
To end up with two hosted services being created, one with the first set of options and the other with the second.
I've seen a few questions looking for similar things but nothing I could find quite lined up with this so I'm a bit stumped. Any ideas?
The solution is Dotnet 5.
So from what I've been able to find, there's no way to do this out-of-the box.
However, This can be done manually using a combination of ActivatorUtilities and Configuration.Bind().
private void CreateWorkers(IServiceCollection services, IConfigurationRoot configuration)
{
List<PubSubSettings> pubsubSettings = new();
configuration.Bind(nameof(PubSubSettings), pubsubSettings);
foreach (PubSubSettings setting in pubsubSettings)
{
services.AddSingleton<IHostedService>(s => ActivatorUtilities.CreateInstance<Worker>(s, ActivatorUtilities.CreateInstance<PubSub.PubSub>(s, setting)));
}
}
Essentially, you can use Bind to get the configuration objects from the JSON. Then you can manually construct the Worker for the call to AddHostedService using CreateInstance.
Two calls are needed in this case, one to generate the PubSub for the worker (in which we pass the setting parameter) and the other to generate the Worker itself.
ActivatorUtilities essentially injects everything you need for the object except the parameters you've provided.
We need to use .AddSingleton<IHostedService> because of the way that the framework checks for dupes with AddHostedService().
Maybe you could try creating a class only for the object and let the PubSubSettings class only for the array:
public class PubSubSettings
{
public PubSubObject[] PubSubs { get; set; }
}
public class PubSubObject
{
public string ProjectId { get; set; }
public string TopicId { get; set; }
public int PartnerId { get; set; }
public string SubscriptionId { get; set; }
}
Then in the startup class you should use Bind to get the current value of the array to create a Worker for each PubSub:
PubSubSettings settings = new PubSubSettings();
Configuration.GetSection(nameof(PubSubSettings)).Bind(settings);
...
foreach(PubSubObject item in settings.PubSubs)
{
services.AddHostedService<Worker>();
}
Then in the PubSub class you need to search the PartnerId inside the Array.
Or you could follow the approach described in the section Named options support using IConfigureNamedOptions in the Microsoft docs: Options pattern in ASP.NET Core
I'm trying to load some settings from the appsettings file and I've got a small issue with the way lists are loaded when using the options pattern. Suppose I've got the following classes (for loading the settings):
public class Application {
public string Name { get; set; } = "";
public IEnumerable<string> Roles { get; set; } = new[] {""};
public Application ToApplicationWithoutPass() =>
new Application {
Name = Name,
Username = Username,
Roles = Roles.ToList()
};
}
public class Applications {
public IEnumerable<Application> AppList { get; set; } = new List<Application>();
}
And here is what the settings that are defined on the appsetings file look like:
"Applications": {
"AppList": [
{
"Name": "SraWebuserAdmin",
"Roles": [ "SraEntitiesWriters", "SraEntitiesReaders", "SraEntitiesLoginAccess" ]
},
...
Here are the entries from the DI setup which is done on the ConfigureServices method:
services.Configure<Applications>(options => Configuration.GetSection("Applications").Bind(options));
services.AddScoped<IApplicationAccessVerifier, ApplicationAccessVerifier>();
And, finally, here's the constructor of the ApplicationAccessVerifier class:
public ApplicationAccessVerifier(IOptionsSnapshot<Applications> applicationOptions) {
_applicationOptions = applicationOptions;
}
Now, the question: if I don't initialize the AppList property,
public class Applications {
public IEnumerable<Application> AppList { get; set; }
}
then the settings are loaded correctly.
However, if I initialized it like I've shown (making sure the filed wrapper by the property is initialized with an empty list), then the settings won't be copied to the AppList.
I find this strange since simple properties (ex.: Name on the Application class) aren't affected by the same issue.
Can anyone tell me why this happens or point me to an official documentation about it?
I want a service I can inject - or in my example get with GetService - that contains settings from my appsettings.json file.
The appsettings.json fragment looks like this:
"ExternalInterfaces": [
{
"Name": "name1",
"BaseUrl": "https://www.baseurl1.svc"
},
{
"Name": "name2",
"BaseUrl": "https://www.baseurl2.svc"
}
]
To do this I have the following interfaces:
using System.Collections.Generic;
namespace Infrastructure.Settings
{
public interface IExternalInterfaceSettingsCollection
{
IReadOnlyCollection<IExternalInterfaceSettings> Settings { get; set; }
}
}
namespace Infrastructure.Settings
{
public interface IExternalInterfaceSettings
{
string Name { get; set; }
string BaseUrl { get; set; }
}
}
and the following corresponding classes:
using System.Collections.Generic;
namespace Infrastructure.Settings
{
public class ExternalInterfaceSettingsCollection : IExternalInterfaceSettingsCollection
{
public IReadOnlyCollection<IExternalInterfaceSettings> Settings { get; set; }
}
}
namespace Infrastructure.Settings
{
public class ExternalInterfaceSettings : IExternalInterfaceSettings
{
const string DefaultName = "newExternalInterface";
const string DefaultBaseUrl = "";
public string Name { get; set; } = DefaultName;
public string BaseUrl { get; set; } = DefaultBaseUrl;
}
}
And in my Startup.cs I have this (definitely gets called with no exceptions):
services.Configure<IExternalInterfaceSettingsCollection>(settings => _configuration.GetSection("ExternalInterfaces").Bind(settings));
and this is then consumed as follows:
var externalInterfaceConfiguration = app.ApplicationServices.GetService<ExternalInterfaceSettingsCollection>();
var Setting1BaseUrl = externalInterfaceConfiguration.Settings
.SingleOrDefault(s => s.Name == "name1")?.BaseUrl;
However, in the last 3 lines, externalInterfaceConfiguration is always null.
I'm clearly missing something, but I can't see what. Any clues?
You've registered IExternalInterfaceSettings, but you're attempting to retrieve ExternalInterfaceSettings. There's no such service in the collection, so the result is null (since you used GetService<T>). If you had used GetRequiredService<T> then an exception would have been thrown telling you as much.
Then, the options pattern is not meant to bind to interfaces. The whole idea is that you're binding to a POCO that represents a specific set of settings. If you want to use an interface, I suppose that's your prerogative, but it's not going to be applicable to the options configuration. In other words, you need the following instead:
services.Configure<ExternalInterfaceSettings>(Configuration.GetSection("ExternalInterfaces"));
(Note, the action overload with Bind is superfluous. You can just pass the config section directly.)
With that, you'll be able to request something like IOptions<ExternalInterfaceSettings>, but you still cannot get ExternalInterfaceSettings directly from the service collection. If you want that functionality, you'll need to add an additional service registration (which can utilize an interface, this time):
services.AddScoped<IExternalInterfaceSettings, ExternalInterfaceSettings>(p =>
p.GetRequiredService<IOptions<ExternalInterfaceSettings>>().Value);
I am enforcing the problem to add membership provider into config in the ASP.Net Core application. I must add it because it references to older projects which contain the legacy code and reference to an assembly of System.Web.ApplicationServices. And in this assembly, the flow goes to line, which checks if the name of the provider is not null.
if ( providerName == null || SystemWebProxy.Membership.Providers[providerName] == null )
{
throw new ArgumentException(String.Format(CultureInfo.CurrentCulture, ApplicationServicesStrings.Membership_provider_name_invalid), "providerName" );
}
But anyway, I am adding the necessary implementation in appsettings.json:
"Membership": {
"DefaultProvider": "blahblah",
"Providers": {
"Add": {
...
"Name": "blahblah"
...
}
}}
And in the startup class I am configuring the sqlMembershipProvider in the following way:
services.Configure<System.Web.Security.SqlMembershipProvider>(Configuration.GetSection("Membership"));
Could somebody tell me, why I am not getting my provider implemented into the configuration? Maybe I am going the wrong way?
You can try something like that:
Firstly, define classes to handle appsettings.json
public class MembershipSettings{
public string DefaultProvider { get; set; }
public Provider Provider { get; set; }
}
public class Provider
{
public string Name { get; set; }
public string ConnectionStringName { get; set; } //not sure if it works
public string ApplicationName { get; set; }
//etc
}
Then inherit from SqlMembershipProvider
internal class CustomeSqlMembershipProvider : SqlMembershipProvider
{
public CustomeSqlMembershipProvider(IOptions<MembershipSettings> settings)
{
InitializeProvider(settings.Value);
}
protected virtual void InitializeProvider(MembershipSettings settings)
{
base.Initialize(settings.Provider.Name, PrepareSettings(settings.Provider));
}
private NameValueCollection PrepareSettings(Provider provider) => new NameValueCollection
{
//you can use reflection to do that as well
{ nameof(provider.ApplicationName), provider.ApplicationName},
{nameof(provider.ConnectionStringName), provider.ConnectionStringName}
//...
//etc
};
}
And in the Startup.cs add following lines:
services.Configure<MembershipSettings>(Configuration.GetSection("Membership"));
services.AddScoped<SqlMembershipProvider, CustomSqlMembershipProvider>();
And last but not least, modify your appsettings.json a little (you can adjust MembershiSettings class and CustomSqlMembershiProvider to handle and initialize list of providers)
"Membership": {
"DefaultProvider": "blahblah",
"Provider": {
...
"Name": "blahblah"
...
}
}
Note that I'm not so familiar with SqlMembershipProvider so there could be some flaws. In my opinion you should write your own MembershipProvider but if you can't (because of legacy code) then you can do something as above
I'm writing session manager class in .NET MVC 4 and I got stuck at the point of creating SqlConfiguration.
Here is one of the properties from the class:
public static MsSqlConfiguration SqlConfig { get; set; }
All working perfectly excluding the fact that I can manage sessions only from MS SQLServer.
I want to do something like this:
public static MsSqlConfiguration,SQLiteConfiguration SqlConfig { get; set; }
And I know it isn't possible so I don't know what to do.
Thanks.
You can create a class with these 2 properties and Use them. I know its very basic
class MyConfig
{
public static MsSqlConfiguration SqlConfig { get; set; }
public static SQLiteConfiguration SQLiteConfig { get; set; }
}
Use them like
public static MyConfig SqlConfig { get; set; }
What about a dictionary?
public static IReadOnlyDictionary<string, IPersistenceConfigurer> DbConfigurations =
new ReadOnlyDictionary<string, IPersistenceConfigurer>(
new Dictionary<string, IPersistenceConfigurer>
{
{ "azure", MsSqlConfiguration.MsSql2008
.ConnectionString("ConnectionString")
.Dialect<MsSqlAzure2008Dialect>()
.Driver<SqlAzureClientDriver>() },
{ "mssql", MsSqlConfiguration.MsSql2008
.ConnectionString("ConnectionString")
.Dialect<MsSql2008Dialect>() },
{ "sqlite", SQLiteConfiguration.Standard
.InMemory() },
// etc..
});
IPersistenceConfigurer is an interface that any database configuration must implement.
As it's a dictionary, you can always check if a database configuration is present by calling DbConfigurations.ContainsKey("mssql").
Another option is using a generic list of IPersistenceConfigurer (aka List<IPersistenceConfigurer>) and get a configuration using LINQ OfType<T> extension method as follows:
dbConfigs.OfType<MsSqlConfiguration >().Single()
...or
dbConfigs.Single(config => config is MsSqlConfiguration)
Another option, if you're using Dependency Injection and an IoC container such as Castle Windsor is to register a factory with the container that can provide an instance of IPersistenceConfigurer to any components that require it. This way you can register different components for the IPersistenceConfigurer service depending on which environment you are running in (as I assume that you only need one particular IPersistenceConfigurer for the application at any one time).