make tenant field optional

doc/03_configuration.md

@@ -62,7 +62,7 @@ ingest:
 
 ## Event schema definition
 
-In this section field types and names should be defined for metadata, impression and interaction events. Metarank supports
+In this section field types and names should be defined for metadata, ranking and interaction events. Metarank supports
 the following types of fields:
 1. string: a regular UTF-8 string
 2. number: a double-precision floating-point format

doc/features/counters.md

@@ -67,7 +67,7 @@ You can configure it in the following way:
     // name of the feature extractor used as a dividend
     top: click
     // name of the feature extractor that is used as a divider
-    bottom: impression
+    bottom: ranking
     scope: item
     bucket: 24h
     periods: [7,30]

doc/tutorial_ranklens.md

@@ -13,7 +13,7 @@ This tutorial reproduces the system running on [demo.metarank.ai](https://demo.m
 
 For the data input, we are using a repackaged copy of the ranklens dataset [available here](https://github.com/metarank/metarank/tree/master/src/test/resources/ranklens/events). 
 The only difference with the original dataset is that we have converted it to a metarank-compatible data model with 
-metadata/interaction/impression [event format](./xx_event_schema.md).
+metadata/interaction/ranking [event format](./xx_event_schema.md).
 
 ### Configuration
 
@@ -83,7 +83,7 @@ features:
   - name: ctr
     type: rate
     top: click
-    bottom: impression
+    bottom: examine
     scope: item
     bucket: 24h
     periods: [7,30]

doc/xx_api_schema.md

@@ -46,7 +46,7 @@ Ranking endpoint does the real work of personalizing items that are passed to it
 }
 ```
 
-- `id`: a request identifier later used to join impression and interaction events. This will be the same value that you will pass to feedback endpoint for impression and ranking events.
+- `id`: a request identifier later used to join ranking and interaction events. This will be the same value that you will pass to feedback endpoint for impression and ranking events.
 - `user`: unique visitor identifier.
 - `session`: session identifier, a single visitor may have multiple sessions.
 - `fields`: an optional array of extra fields that you can use in your model, for more information refer to [Supported events](xx_event_schema.md).

doc/xx_event_schema.md

@@ -84,7 +84,7 @@ This information is used by personalization algorithms to understand which items
 }
 ```
 
-- `id`: a request identifier later used to join impression and interaction events. Should match the value that is sent to [ranking API](xx_api_schema.md).
+- `id`: a request identifier later used to join ranking and interaction events. Should match the value that is sent to [ranking API](xx_api_schema.md).
 - `user`: unique visitor identifier.
 - `session`: session identifier, a single visitor may have multiple sessions.
 - `fields`: an optional array of extra fields that you can use in your model, as described above.
@@ -105,7 +105,7 @@ The `type` field must match the `name` provided in the [Configuration](03_config
 {
   "event": "interaction",
   "id": "0f4c0036-04fb-4409-b2c6-7163a59f6b7d",// required
-  "impression": "81f46c34-a4bb-469c-8708-f8127cd67d27", //required
+  "ranking": "81f46c34-a4bb-469c-8708-f8127cd67d27", //required
   "timestamp": "1599391467000",// required
   "user": "user1",// required
   "session": "session1",// required
@@ -119,7 +119,7 @@ The `type` field must match the `name` provided in the [Configuration](03_config
 ```
 
 - `id`: a request identifier.
-- `impression`: an identifier of the parent raning event event
+- `ranking`: an identifier of the parent raning event event
 - `user`: unique visitor identifier.
 - `session`: session identifier, a single visitor may have multiple sessions.
 - `type`: internal name of the event.

src/main/scala/ai/metarank/mode/inference/api/RankApi.scala

@@ -98,8 +98,8 @@ case class RankApi(
 }
 
 object RankApi {
-
-  implicit val requestDecoder: EntityDecoder[IO, RankingEvent]   = jsonOf
+  import ai.metarank.model.Event.EventCodecs._
+  implicit val requestDecoder: EntityDecoder[IO, RankingEvent]   = jsonOf[IO, RankingEvent]
   implicit val itemScoreEncoder: EntityEncoder[IO, RankResponse] = jsonEncoderOf
 
   object ExplainParamDecoder             extends OptionalQueryParamDecoderMatcher[Boolean]("explain")

src/main/scala/ai/metarank/model/Event.scala

@@ -20,8 +20,8 @@ object Event {
       id: EventId,
       item: ItemId,
       timestamp: Timestamp,
-      fields: List[Field],
-      tenant: String
+      fields: List[Field] = Nil,
+      tenant: String = "default"
   ) extends Event {}
 
   sealed trait FeedbackEvent extends Event {
@@ -34,9 +34,9 @@ object Event {
       timestamp: Timestamp,
       user: UserId,
       session: SessionId,
-      fields: List[Field],
+      fields: List[Field] = Nil,
       items: List[ItemRelevancy],
-      tenant: String
+      tenant: String = "default"
   ) extends FeedbackEvent
 
   case class InteractionEvent(
@@ -47,20 +47,26 @@ object Event {
       user: UserId,
       session: SessionId,
       `type`: String,
-      fields: List[Field],
-      tenant: String
+      fields: List[Field] = Nil,
+      tenant: String = "default"
   ) extends FeedbackEvent
 
   case class ItemRelevancy(id: ItemId, relevancy: Option[Double] = None)
   object ItemRelevancy {
     def apply(id: ItemId, relevancy: Double) = new ItemRelevancy(id, Some(relevancy))
   }
 
-  implicit val relevancyCodec: Codec[ItemRelevancy]      = deriveCodec
-  implicit val metadataCodec: Codec[MetadataEvent]       = deriveCodec
-  implicit val rankingCodec: Codec[RankingEvent]         = deriveCodec
-  implicit val interactionCodec: Codec[InteractionEvent] = deriveCodec
+  object EventCodecs {
+    implicit val conf                                      = Configuration.default.withDefaults
+    implicit val relevancyCodec: Codec[ItemRelevancy]      = deriveCodec
+    implicit val metadataCodec: Codec[MetadataEvent]       = deriveConfiguredCodec
+    implicit val rankingCodec: Codec[RankingEvent]         = deriveConfiguredCodec
+    implicit val interactionCodec: Codec[InteractionEvent] = deriveConfiguredCodec
+  }
 
+  import EventCodecs.metadataCodec
+  import EventCodecs.rankingCodec
+  import EventCodecs.interactionCodec
   implicit val conf = Configuration.default
     .withDiscriminator("event")
     .withKebabCaseMemberNames

src/main/scala/ai/metarank/source/FileEventSource.scala

@@ -14,12 +14,16 @@ import io.circe.parser._
 import java.io.{ByteArrayOutputStream, InputStream}
 import ai.metarank.flow.DataStreamOps._
 import ai.metarank.util.Logging
+import org.apache.flink.connector.file.src.enumerate.NonSplittingRecursiveEnumerator
 
 case class FileEventSource(path: String) extends EventSource {
   override def eventStream(env: StreamExecutionEnvironment)(implicit ti: TypeInformation[Event]): DataStream[Event] =
     env
       .fromSource(
-        source = FileSource.forRecordStreamFormat(EventStreamFormat(), new Path(path)).processStaticFileSet().build(),
+        source = FileSource
+          .forRecordStreamFormat(EventStreamFormat(), new Path(path))
+          .processStaticFileSet()
+          .build(),
         watermarkStrategy = EventWatermarkStrategy(),
         sourceName = "events-source"
       )
@@ -53,7 +57,7 @@ object FileEventSource {
       if (line != null) decode[Event](line) match {
         case Left(value) =>
           logger.error(s"cannot decode line ${line}", value)
-          throw new IllegalArgumentException("json decoding error")
+          throw new IllegalArgumentException(s"json decoding error '$value' on line '${line}'")
         case Right(value) =>
           value
       }

src/test/scala/ai/metarank/model/EventJsonTest.scala

@@ -14,7 +14,6 @@ class EventJsonTest extends AnyFlatSpec with Matchers {
                  |  "id": "81f46c34-a4bb-469c-8708-f8127cd67d27",
                  |  "item": "product1",
                  |  "timestamp": "1599391467000", 
-                 |  "tenant": "default",
                  |  "fields": [
                  |    {"name": "title", "value": "Nice jeans"},
                  |    {"name": "price", "value": 25.0},
@@ -37,6 +36,52 @@ class EventJsonTest extends AnyFlatSpec with Matchers {
       )
     )
   }
+  it should "decode metadata with tenant" in {
+    val json = """{
+                 |  "event": "metadata",
+                 |  "id": "81f46c34-a4bb-469c-8708-f8127cd67d27",
+                 |  "item": "product1",
+                 |  "timestamp": "1599391467000", 
+                 |  "fields": [
+                 |    {"name": "title", "value": "Nice jeans"},
+                 |    {"name": "price", "value": 25.0},
+                 |    {"name": "color", "value": ["blue", "black"]},
+                 |    {"name": "availability", "value": true}
+                 |  ],
+                 |  "tenant": "foo"
+                 |}""".stripMargin
+    decode[Event](json) shouldBe Right(
+      MetadataEvent(
+        id = EventId("81f46c34-a4bb-469c-8708-f8127cd67d27"),
+        item = ItemId("product1"),
+        timestamp = Timestamp(1599391467000L),
+        fields = List(
+          StringField("title", "Nice jeans"),
+          NumberField("price", 25),
+          StringListField("color", List("blue", "black")),
+          BooleanField("availability", true)
+        ),
+        tenant = "foo"
+      )
+    )
+  }
+  it should "decode metadata with empty fields" in {
+    val json = """{
+                 |  "event": "metadata",
+                 |  "id": "81f46c34-a4bb-469c-8708-f8127cd67d27",
+                 |  "item": "product1",
+                 |  "timestamp": "1599391467000"
+                 |}""".stripMargin
+    decode[Event](json) shouldBe Right(
+      MetadataEvent(
+        id = EventId("81f46c34-a4bb-469c-8708-f8127cd67d27"),
+        item = ItemId("product1"),
+        timestamp = Timestamp(1599391467000L),
+        fields = Nil,
+        tenant = "default"
+      )
+    )
+  }
 
   it should "decode ranking" in {
     val json = """{
@@ -45,7 +90,6 @@ class EventJsonTest extends AnyFlatSpec with Matchers {
                  |  "timestamp": "1599391467000",
                  |  "user": "user1",
                  |  "session": "session1",
-                 |  "tenant": "default",
                  |  "fields": [
                  |      {"name": "query", "value": "jeans"},
                  |      {"name": "source", "value": "search"}
@@ -85,7 +129,6 @@ class EventJsonTest extends AnyFlatSpec with Matchers {
                  |  "timestamp": "1599391467000",
                  |  "user": "user1",
                  |  "session": "session1",
-                 |  "tenant": "default",
                  |  "type": "purchase",
                  |  "item": "product1",
                  |  "fields": [