Kafka 生产者模块（二）：KafkaProducer 的 send 接口发送消息

        send接口会把消息经过处理后，放在一个缓存中，由后台sender线程从缓存中取出，然后发送到服务端，这一篇介绍放入缓存的send接口。

了解了 KafkaProducer 的字段定义和对象的构造过程之后，下面正式开始对消息收集的过程进行分析，相关实现位于 KafkaProducer#send 方法中：

    public Future<RecordMetadata> send(ProducerRecord<K, V> record, Callback callback) {
        // intercept the record, which can be potentially modified; this method does not throw exceptions
        // 遍历注册的拦截器对待发送的消息执行拦截修改
        ProducerRecord<K, V> interceptedRecord = this.interceptors.onSend(record);
        // 调用 doSend 方法开始发送消息
        return doSend(interceptedRecord, callback);
    }

方法 ProducerInterceptor#doSend 用于收集消息的过程

    private Future<RecordMetadata> doSend(ProducerRecord<K, V> record, Callback callback) {
        TopicPartition tp = null;
        try {
            throwIfProducerClosed();
            // first make sure the metadata for the topic is available
            ClusterAndWaitTime clusterAndWaitTime;
            try {
                // 1. 获取 kafka 集群元数据信息，如果当前请求的是新 topic，或者指定的分区超过已知的分区范围，则会触发更新集群元数据信息
                clusterAndWaitTime = waitOnMetadata(record.topic(), record.partition(), maxBlockTimeMs);
            } catch (KafkaException e) {
                if (metadata.isClosed())
                    throw new KafkaException("Producer closed while send in progress", e);
                throw e;
            }
            long remainingWaitMs = Math.max(0, maxBlockTimeMs - clusterAndWaitTime.waitedOnMetadataMs);
            Cluster cluster = clusterAndWaitTime.cluster;
            // 2 基于注册的序列化器对 key 执行序列化
            byte[] serializedKey;
            try {
                serializedKey = keySerializer.serialize(record.topic(), record.headers(), record.key());
            } catch (ClassCastException cce) {
                throw new SerializationException("Can't convert key of class " + record.key().getClass().getName() +
                        " to class " + producerConfig.getClass(ProducerConfig.KEY_SERIALIZER_CLASS_CONFIG).getName() +
                        " specified in key.serializer", cce);
            }
            // 3. 基于注册的序列化器对 value 执行序列化
            byte[] serializedValue;
            try {
                serializedValue = valueSerializer.serialize(record.topic(), record.headers(), record.value());
            } catch (ClassCastException cce) {
                throw new SerializationException("Can't convert value of class " + record.value().getClass().getName() +
                        " to class " + producerConfig.getClass(ProducerConfig.VALUE_SERIALIZER_CLASS_CONFIG).getName() +
                        " specified in value.serializer", cce);
            }
            // 4. 为当前消息选择合适的分区，如果未明确指定的话，则基于注册的分区器为当前消息计算分区
            int partition = partition(record, serializedKey, serializedValue, cluster);
            // 消息投递的目标 topic 分区
            tp = new TopicPartition(record.topic(), partition);

            setReadOnly(record.headers());
            /* 5. 将消息追加到消息收集器（RecordAccumulator）中 */
            Header[] headers = record.headers().toArray();
            // 计算当前消息大小，并校验消息是否过大
            int serializedSize = AbstractRecords.estimateSizeInBytesUpperBound(apiVersions.maxUsableProduceMagic(),
                    compressionType, serializedKey, serializedValue, headers);
            ensureValidRecordSize(serializedSize);
            // 如果未明确为当前消息指定时间戳，则设置为当前时间戳
            long timestamp = recor