chore: move tests from akka-projection-grpc into own submodule (#796)

* chore: move tests from akka-projection-grpc into own submodule

.github/workflows/integration-tests-grpc.yml

@@ -49,7 +49,7 @@ jobs:
           jvm: ${{ matrix.jvmName }}
 
       - name: Run all integration tests with default Scala and Java ${{ matrix.jdkVersion }}
-        run: sbt "akka-projection-grpc/It/test" ${{ matrix.extraOpts }}
+        run: sbt "akka-projection-grpc-tests/It/test" ${{ matrix.extraOpts }}
         env: # Disable Ryuk resource reaper since we always spin up fresh VMs
           TESTCONTAINERS_RYUK_DISABLED: true
 

build.sbt

@@ -99,12 +99,21 @@ lazy val `durable-state` =
 
 lazy val grpc =
   Project(id = "akka-projection-grpc", base = file("akka-projection-grpc"))
-    .configs(IntegrationTest)
-    .settings(headerSettings(IntegrationTest))
-    .settings(Defaults.itSettings)
     .settings(Dependencies.grpc)
     .dependsOn(core)
     .dependsOn(eventsourced)
+    .enablePlugins(AkkaGrpcPlugin)
+    .settings(akkaGrpcCodeGeneratorSettings += "server_power_apis", IntegrationTest / fork := true)
+
+lazy val grpcTests =
+  Project(id = "akka-projection-grpc-tests", base = file("akka-projection-grpc-tests"))
+    .configs(IntegrationTest)
+    .settings(headerSettings(IntegrationTest))
+    .disablePlugins(MimaPlugin)
+    .settings(Defaults.itSettings)
+    .settings(Dependencies.grpcTest)
+    .settings(publish / skip := true)
+    .dependsOn(grpc)
     .dependsOn(testkit % Test)
     .enablePlugins(AkkaGrpcPlugin)
     .settings(akkaGrpcCodeGeneratorSettings += "server_power_apis", IntegrationTest / fork := true)

docs/src/main/paradox/grpc-replicated-event-sourcing-transport.md

@@ -1,11 +1,11 @@
 # Akka Replicated Event Sourcing over gRPC
 
 Akka Replicated Event Sourcing extends Akka Persistence allowing multiple replicas of the same entity, all accepting
-writes, for example in different data centers or cloud provider regions. This makes it possible to implement patterns 
-such as active-active and hot standby. 
+writes, for example in different data centers or cloud provider regions. This makes it possible to implement patterns
+such as active-active and hot standby.
 
 Originally, Akka Replicated Event Sourcing has required cross-replica access to the underlying replica database, which
-can be hard to open up for security and infrastructure reasons. It was also easiest to use in an 
+can be hard to open up for security and infrastructure reasons. It was also easiest to use in an
 [Akka Multi DC Cluster](https://doc.akka.io/docs/akka/current/typed/cluster-dc.html) setup
 where a single cluster spans multiple datacenters or regions, another thing that can be complicated to allow.
 
@@ -103,18 +103,18 @@ Java
 
 ### Settings
 
-The @apidoc[akka.projection.grpc.replication.*.ReplicationSettings] @scala[`apply`]@java[`create`] factory methods can 
+The @apidoc[akka.projection.grpc.replication.*.ReplicationSettings] @scala[`apply`]@java[`create`] factory methods can
 accept an entity name, a @apidoc[ReplicationProjectionProvider] and an actor system. The configuration of that system
 is expected to have a top level entry with the entity name containing this structure:
 
 Scala
-:  @@snip [config](/akka-projection-grpc/src/test/scala/akka/projection/grpc/replication/ReplicationSettingsSpec.scala) { #config }
+:  @@snip [config](/akka-projection-grpc-tests/src/test/scala/akka/projection/grpc/replication/ReplicationSettingsSpec.scala) { #config }
 
 Java
-:  @@snip [config](/akka-projection-grpc/src/test/scala/akka/projection/grpc/replication/ReplicationSettingsSpec.scala) { #config }
+:  @@snip [config](/akka-projection-grpc-tests/src/test/scala/akka/projection/grpc/replication/ReplicationSettingsSpec.scala) { #config }
 
-The entries in the block refer to the local replica while `replicas` is a list of all replicas, including the node itself, 
-with details about how to reach the replicas across the network. 
+The entries in the block refer to the local replica while `replicas` is a list of all replicas, including the node itself,
+with details about how to reach the replicas across the network.
 
 The `grpc.client` section for each of the replicas is used for setting up the Akka gRPC client and supports the same discovery, TLS
 and other connection options as when using Akka gRPC directly. For more details see @extref:[Akka gRPC configuration](akka-grpc:client/configuration.html#by-configuration).
@@ -150,10 +150,10 @@ be passed at once to `EventProducer.grpcServiceHandler` to create a single produ
 streams.
 
 Scala
-:  @@snip [ProducerApiSample.scala](/akka-projection-grpc/src/test/scala/akka/projection/grpc/replication/scaladsl/ProducerApiSample.scala) { #multi-service }
+:  @@snip [ProducerApiSample.scala](/akka-projection-grpc-tests/src/test/scala/akka/projection/grpc/replication/scaladsl/ProducerApiSample.scala) { #multi-service }
 
 Java
-:  @@snip [ReplicationCompileTest.java](/akka-projection-grpc/src/test/java/akka/projection/grpc/replication/javdsl/ReplicationCompileTest.java) { #multi-service }
+:  @@snip [ReplicationCompileTest.java](/akka-projection-grpc-tests/src/test/java/akka/projection/grpc/replication/javdsl/ReplicationCompileTest.java) { #multi-service }
 
 
 The Akka HTTP server must be running with HTTP/2, this is done through config:
@@ -167,13 +167,13 @@ Java
 ### Serialization of events
 
 The events are serialized for being passed over the wire using the same Akka serializer as configured for serializing
-the events for storage. 
+the events for storage.
 
 Note that having separate replicas increases the risk that two different serialized formats and versions of the serializer
 are running at the same time, so extra care must be taken when changing the events and their serialization and deploying
 new versions of the application to the replicas.
 
-For some scenarios it may be necessary to do a two-step deploy of format changes to not lose data, first deploy support 
+For some scenarios it may be necessary to do a two-step deploy of format changes to not lose data, first deploy support
 for a new serialization format so that all replicas can deserialize it, then a second deploy where the new field is actually
 populated with data.
 

project/Dependencies.scala

@@ -211,7 +211,9 @@ object Dependencies {
         Compile.akkaPersistenceTyped,
         Compile.akkaPersistenceQuery,
         // Only needed for Replicated Event Sourcing over gRPC
-        Compile.akkaClusterShardingTyped % Optional,
+        Compile.akkaClusterShardingTyped % Optional)
+
+  val grpcTest = deps ++= Seq(
         Test.akkaProjectionR2dbc,
         Test.postgresDriver,
         Test.akkaShardingTyped,