Merge branch 'stored-state' into stored-offsets

[demos/kafka/training] / src / main / java / de / juplo / kafka / EndlessConsumer.java

diff --git a/src/main/java/de/juplo/kafka/EndlessConsumer.java b/src/main/java/de/juplo/kafka/EndlessConsumer.java
index da2f8f0..2563204 100644 (file)
--- a/src/main/java/de/juplo/kafka/EndlessConsumer.java
+++ b/src/main/java/de/juplo/kafka/EndlessConsumer.java
@@ -1,16 +1,17 @@
 package de.juplo.kafka;
 
 import lombok.extern.slf4j.Slf4j;
+import org.apache.kafka.clients.consumer.ConsumerRebalanceListener;
 import org.apache.kafka.clients.consumer.ConsumerRecord;
 import org.apache.kafka.clients.consumer.ConsumerRecords;
 import org.apache.kafka.clients.consumer.KafkaConsumer;
+import org.apache.kafka.common.TopicPartition;
 import org.apache.kafka.common.errors.WakeupException;
 import org.apache.kafka.common.serialization.StringDeserializer;
 
 import javax.annotation.PreDestroy;
 import java.time.Duration;
-import java.util.Arrays;
-import java.util.Properties;
+import java.util.*;
 import java.util.concurrent.ExecutionException;
 import java.util.concurrent.ExecutorService;
 import java.util.concurrent.Future;
@@ -21,28 +22,37 @@ import java.util.concurrent.atomic.AtomicBoolean;
 public class EndlessConsumer implements Runnable
 {
   private final ExecutorService executor;
+  private final PartitionStatisticsRepository repository;
   private final String bootstrapServer;
   private final String groupId;
   private final String id;
   private final String topic;
+  private final String autoOffsetReset;
 
   private AtomicBoolean running = new AtomicBoolean();
   private long consumed = 0;
   private KafkaConsumer<String, String> consumer = null;
   private Future<?> future = null;
 
+  private final Map<Integer, Map<String, Integer>> seen = new HashMap<>();
+
+
   public EndlessConsumer(
       ExecutorService executor,
+      PartitionStatisticsRepository repository,
       String bootstrapServer,
       String groupId,
       String clientId,
-      String topic)
+      String topic,
+      String autoOffsetReset)
   {
     this.executor = executor;
+    this.repository = repository;
     this.bootstrapServer = bootstrapServer;
     this.groupId = groupId;
     this.id = clientId;
     this.topic = topic;
+    this.autoOffsetReset = autoOffsetReset;
   }
 
   @Override
@@ -54,14 +64,52 @@ public class EndlessConsumer implements Runnable
       props.put("bootstrap.servers", bootstrapServer);
       props.put("group.id", groupId);
       props.put("client.id", id);
-      props.put("auto.offset.reset", "earliest");
+      props.put("enable.auto.commit", false);
+      props.put("auto.offset.reset", autoOffsetReset);
+      props.put("metadata.max.age.ms", "1000");
       props.put("key.deserializer", StringDeserializer.class.getName());
       props.put("value.deserializer", StringDeserializer.class.getName());
 
       this.consumer = new KafkaConsumer<>(props);
 
       log.info("{} - Subscribing to topic {}", id, topic);
-      consumer.subscribe(Arrays.asList(topic));
+      consumer.subscribe(Arrays.asList(topic), new ConsumerRebalanceListener()
+      {
+        @Override
+        public void onPartitionsRevoked(Collection<TopicPartition> partitions)
+        {
+          partitions.forEach(tp ->
+          {
+            log.info("{} - removing partition: {}", id, tp);
+            Map<String, Integer> removed = seen.remove(tp.partition());
+            for (String key : removed.keySet())
+            {
+              log.info(
+                  "{} - Seen {} messages for partition={}|key={}",
+                  id,
+                  removed.get(key),
+                  tp.partition(),
+                  key);
+            }
+            repository.save(new StatisticsDocument(tp.partition(), removed, consumer.position(tp)));
+          });
+        }
+
+        @Override
+        public void onPartitionsAssigned(Collection<TopicPartition> partitions)
+        {
+          partitions.forEach(tp ->
+          {
+            log.info("{} - adding partition: {}", id, tp);
+            StatisticsDocument document =
+                repository
+                    .findById(Integer.toString(tp.partition()))
+                    .orElse(new StatisticsDocument(tp.partition()));
+            consumer.seek(tp, document.offset);
+            seen.put(tp.partition(), document.statistics);
+          });
+        }
+      });
 
       while (true)
       {
@@ -82,7 +130,24 @@ public class EndlessConsumer implements Runnable
               record.key(),
               record.value()
           );
+
+          Integer partition = record.partition();
+          String key = record.key() == null ? "NULL" : record.key();
+          Map<String, Integer> byKey = seen.get(partition);
+
+          if (!byKey.containsKey(key))
+            byKey.put(key, 0);
+
+          int seenByKey = byKey.get(key);
+          seenByKey++;
+          byKey.put(key, seenByKey);
         }
+
+        seen.forEach((partiton, statistics) -> repository.save(
+            new StatisticsDocument(
+                partiton,
+                statistics,
+                consumer.position(new TopicPartition(topic, partiton)))));
       }
     }
     catch(WakeupException e)
@@ -102,6 +167,10 @@ public class EndlessConsumer implements Runnable
     }
   }
 
+  public Map<Integer, Map<String, Integer>> getSeen()
+  {
+    return seen;
+  }
 
   public synchronized void start()
   {