I have downloaded a sample code from github and run AtLeastOnceDelivery.sln
Every new run it is sending messages with it. And if I change the message namespace it shows an error started with
Error loading snapshot [SnapshotMetadata<pid: delivery, seqNr: 0, timestamp: 2018/09/24>], remaining attempts: [0]
If I could clear the persistence hopefully it will accept then changed namespace and restart messaging id.
By default, all snapshots are stored as files directly in ./snapshots directory of the application, while events are stored in the memory. Because of that you should consider using a one of the akka.persistence plugins for the production purposes.
Your problem happens because you're using akka.net default serializers (dedicated for networking) which are not very version tolerant - so changing any fields, their types, class names or namespaces makes previous version of the class non-deserializable - and in future will be subject to change. This is also why it's strongly discouraged to use default serializers for persistence.
How to make a custom Akka.NET Serializer
While there are plans to improve serializers API, at the current moment (Akka.NET v1.3.9), to make your own serializer you need to simply inherit from Akka.Serialization.Serializer class:
public sealed class MySerializer : Serializer
{
public MySerializer(ExtendedActorSystem system) : base(system) { }
public override int Identifier => /* globaly unique serializer id */;
public override bool IncludeManifest => true;
public override byte[] ToBinary(object obj)
{
// serialize object
}
public override object FromBinary(byte[] bytes, Type type)
{
// deserialize object
}
}
Keep in mind that Identifier property must be unique in cluster scope - usually values below 100 are used by akka.net internal serializers, therefore it's better to use higher values.
How to bind serializer to be used for a given type
By convention Akka.NET uses empty interfaces to mark message types that are supposed to be serialized. Then you can setup your HOCON configuration to use a specific serializer for a given interface:
akka.actor {
serializers {
my-serializer = ""MyNamespace.MySerializer, MyAssembly""
}
serialization-bindings {
""MyNamespace.MyInterface, MyAssembly"" = my-serializer
}
}
Where MyInterface is interface assigned to a message type you want to serialize/deserialize with MySerializer.
Related
I am trying to implement signal r client and server with json serialization.
Currently I am targeting .NET 5 and use Microsoft json serializer implementation.
My messages are represented by complex objects and there is an JsonConverter used for reading and writing.
What I see is that on the client the On event is never raised except if handler parameter specified as an object class.
connection.On("EntityEventAsync", (object obj) =>
{
//obj will be json object here
});
On client side I can clearly see that the messages are received as JsonConverter is called and reads the messages as it should BUT the On event is never raised.
Typed client hub code
public interface IEventsClient
{
Task EntityEventAsync(DetailedMessage message);
}
[Authorize(AuthenticationSchemes = "Basic,Bearer")]
public class EventHub : Hub<Clients.IEventsClient>
{
#region CONSTRUCTOR
public EventHub()
{
}
#endregion
}
What I can be missing here?
If some one else struggles with same problem then it be might the problem with the JsonConverter implementation as it in my case.
Its possible to enable Signal R client logging as mentioned here https://learn.microsoft.com/en-us/aspnet/core/signalr/diagnostics?view=aspnetcore-5.0 and that should make it easier to figure out the problem.
I am working on Akka.net persistence and using MongoDb as the persistent store. One of the properties on the events that we persist is of custom struct type "Rational". We have configured a custom serializer for this Rational type that serializes rational type value into a decimal value. However, we don't see that custom serializer getting invoked at all. The MongoDb shows newly inserted document with rational value of type object instead of decimal.
Below is akka.hocon configuration;
akka {
actor {
serializers {
my-rational = "RationalTypePersistence.RationalSerializer, RationalTypePersistence"
}
serialization-bindings {
"RationalTypePersistence.Rational, RationalTypePersistence" = "my-rational"
}
}
}
In debugging session also, the breakpoints set in the custom serializer's "ToBinary" and "fromBinary" methods don't get hit. The breakpoint in the constructor get hits multiple times though.
The custom serializer is extending Akka.Serialization.Serializer, and has overriden Identifier, FromBinary, ToBinary properties/methods.
Are we missing any configuration ?
I am looking for a way to dynamically select the correct dependency during runtime using google guice.
My usecase is a kotlin application which can work with either sqlite or h2 databases depending on the configuration file provided.
The file is read when the application is executed and if the database is not found, the correct one is created and migrated into.
My database structure contains the Database (Interface), H2Database: Database, SQLiteDatabase: Database and the module binding class which looks like this:
class DatabaseModule: KotlinModule() {
override fun configure() {
bind<Database>().annotatedWith<configuration.H2>().to<H2Database>()
bind<Database>().annotatedWith<configuration.SQLite>().to<SQLiteDatabase>()
}
}
So far, with SQlite alone, I would simply request the dependency using:
#Inject
#SQLite
private lateinit var database: Database
How would I make this selection during runtime?
Without knowing too much about the specific of your code, I'll offer three general approaches.
(Also, I have never used Kotlin. I hope Java samples are enough for you to figure things out.)
First Approach
It sounds like you need some non-trivial logic to determine which Database implementation is the right one to use. This is a classic case for a ProviderBinding. Instead binding Database to a specific implementation, you bind Database to a class that is responsible providing instances (a Provider). For example, you might have this class:
public class MyDatabaseProvider.class implements Provider<Database> {
#Inject
public MyDatabaseProvider.class(Provider<SQLiteDatabase> sqliteProvider, Provider<H2Database> h2Provider) {
this.sqliteProvider = sqliteProvider;
this.h2Provider = h2Provider;
}
public Database get() {
// Logic to determine database type goes here
if (isUsingSqlite) {
return sqliteProvider.get();
} else if (isUsingH2) {
return h2Provider.get();
} else {
throw new ProvisionException("Could not determine correct database implementation.");
}
}
}
(Side note: This sample code gets you a new instance every time. It is fairly straightforward to make this also return a singleton instance.)
Then, to use it, you have two options. In your module, you would bind Database not to a specific implementation, but to your DatabaseProvider. Like this:
protected void configure() {
bind(Database.class).toProvider(MyDatabaseProvider.class);
}
The advantage of this approach is that you don't need to know the correct database implementation until Guice tries to construct an object that requires Database as one of its constructor args.
Second Approach
You could create a DatabaseRoutingProxy class which implements Database and then delegates to the correct database implementation. (I've used this pattern professionally. I don't think there's an "official" name for this design pattern, but you can find a discussion here.) This approach is based on lazy loading with Provider using the Providers that Guice automatically creates(1) for every bound type.
public class DatabaseRoutingProxy implements Database {
private Provider<SqliteDatabse> sqliteDatabaseProvider;
private Provider<H2Database> h2DatabaseProvider;
#Inject
public DatabaseRoutingProxy(Provider<SqliteDatabse> sqliteDatabaseProvider, Provider<H2Database> h2DatabaseProvider) {
this.sqliteDatabaseProvider = sqliteDatabaseProvider;
this.h2DatabaseProvider = h2DatabaseProvider;
}
// Not an overriden method
private Database getDatabase() {
boolean isSqlite = // ... decision logic, or maintain a decision state somewhere
// If these providers don't return singletons, then you should probably write some code
// to call the provider once and save the result for future use.
if (isSqlite) {
return sqliteDatabaseProvider.get();
} else {
return h2DatabaseProvider.get();
}
}
#Override
public QueryResult queryDatabase(QueryInput queryInput) {
return getDatabase().queryDatabase(queryInput);
}
// Implement rest of methods here, delegating as above
}
And in your Guice module:
protected void configure() {
bind(Database.class).to(DatabaseRoutingProxy.class);
// Bind these just so that Guice knows about them. (This might not actually be necessary.)
bind(SqliteDatabase.class);
bind(H2Database.class);
}
The advantage of this approach is that you don't need to be able to know which database implementation to use until you actually make a database call.
Both of these approaches have been assuming that you cannot instantiate an instance of H2Database or SqliteDatabase unless the backing database file actually exists. If it's possible to instantiate the object without the backing database file, then your code becomes much simpler. (Just have a router/proxy/delegator/whatever that takes the actual Database instances as the constructor args.)
Third Approach
This approach is completely different then the last two. It seems to me like your code is actually dealing with two questions:
Does a database actually exist? (If not, then make one.)
Which database exists? (And get the correct class to interact with it.)
If you can solve question 1 before even creating the guice injector that needs to know the answer to question 2, then you don't need to do anything complicated. You can just have a database module like this:
public class MyDatabaseModule extends AbstractModule {
public enum DatabaseType {
SQLITE,
H2
}
private DatabaseType databaseType;
public MyDatabaseModule(DatabaseType databaseType) {
this.databaseType = databaseType;
}
protected void configure() {
if (SQLITE.equals(databaseType)) {
bind(Database.class).to(SqliteDatabase.class);
} else if (H2.equals(databaseType)) {
bind(Database.class).to(H2Database.class);
}
}
}
Since you've separated out questions 1 & 2, when you create the injector that will use the MyDatabaseModule, you can pass in the appropriate value for the constructor argument.
Notes
The Injector documentation states that there will exist a Provider<T> for every binding T. I have successfully created bindings without creating the corresponding provider, therefore Guice must be automatically creating a Provider for configured bindings. (Edit: I found more documentation that states this more clearly.)
public class Model {
}
public class SuperclassDTO {
private boolean funny = true;
public boolean isFunny() {
return funny;
}
public boolean setFunny(boolean f) {
this.funny = f;
}
}
public class SubclassDTO extends SuperclassDTO {
}
new SubclassDTO().isFunny() //returns true
SubclassDTO dto = binder.bindFromBusinessObject(SubclassDTO.class, new Model());
dto.isFunny(); //returns false!!!!
Isn't this weird? Model class does not have a "funny" field but somehow dto is bind with a wrong value. First I thought jDTO required "getFunny" convention, so it couldn't read the value and just set it "false" but changing the getter name to "getFunny" does not resolve the issue, plus I'm not allowed to modify SuperclassDTO. How can I bind the correct value?
Jdto version 1.4 by the way...
The behavior you're experiencing is a "side effect" of the convention over configuration approach. All the fields on the DTO are configured unless you mark them as transient, either by using the #DTOTransient annotation or the transient configuration on the XML file. If a configured field does not have a corresponding field on the source bean, it will be set with default values and that is the reason why you're experiencing this behavior.
You have some options to overcome this issue:
Add the #DTOTransient annotation to the DTO.
Since you're not able to modify the DTO, you could configure it through XML.
Use Binding lifecycle to Restore the value. By adding code on the subclass.
You might as well submit a bug report on the jDTO issue tracker on github.
Is there any switch that instructs svcutil to generate DataContract properties with their names as defined in code? For example when I create a proxy which uses the following DataContract:
[DataContract(Namespace = "http://schemas.mynamespace.com/2012/08")]
public class MyDataContract
{
[DataMember(IsRequired = true, Order = 0)]
private int _id;
public int Id
{
get { return _id; }
set { _id = value; }
}
}
I get this DataContract on the proxy generated class:
public partial class MyDataContract : object
{
private int _idField;
[System.Runtime.Serialization.DataMemberAttribute(IsRequired=true)]
public int _id
{
get
{
return this._idField;
}
set
{
this._idField = value;
}
}
}
The order property of the DataMemberAttribute is always ommited as well for the first 3 properties and a MessageContract ommits an IDisposable implementation.
Well, I cannot comment on the omitted order, but I may be able to help on the rest:
One usually specifies the DataMember attribute on the property, not on the field. The data-contract itself does not distinguish between a property and a field, but it knows the name, type, if it is mandatory, etc. etc.
Added: What Chris said: With [DataMember(Name="whateveryouwant")] you'll be able to set a name different from the field/property name. I do not like such usage, though, but it is helpful when refactoring code, but still keeping the API compatible.
Only other DataContract (and some intrinsically supported) types are serialized to/from messages. IDisposable seems not to be one.
Serializing the inherited IDisposable of a MessageContract would not make any sense. A message-contract is the .NET representation of a SOAP message. It literally has nothing else to do but to provide a 1:1 mapping between what is in the SOAP message XML, and the accessible .NET types.
A message is part of a ServiceContract, in that it specifies which kind of message must be sent to a certain operation to be a valid invocation, and another (response-)message contract specifies how the data, that the operation returns, will be structured. Nothing else; it is a data-aggregate.
If you want to capture the result of a service-operation on the client, and for convenience automatically send a message back upon going out of scope (or for instance unregistering from a service), you will have to implement this on the client-side. Be aware, however, that the service must not require this to happen (due to lost connections, crashes, etc.).
Use the name property on DataMember attribute
Such as:
[DataMember(Name="myname")]