Class: Kafka::FetchOperation

Inherits:
Object
  • Object
show all
Defined in:
lib/kafka/fetch_operation.rb

Overview

Fetches messages from one or more partitions.

operation = Kafka::FetchOperation.new(
  cluster: cluster,
  logger: logger,
  min_bytes: 1,
  max_wait_time: 10,
)

# These calls will schedule fetches from the specified topics/partitions.
operation.fetch_from_partition("greetings", 42, offset: :latest, max_bytes: 100000)
operation.fetch_from_partition("goodbyes", 13, offset: :latest, max_bytes: 100000)

operation.execute

Instance Method Summary collapse

Constructor Details

#initialize(cluster:, logger:, min_bytes: 1, max_bytes: 10485760, max_wait_time: 5) ⇒ FetchOperation 



23
24
25
26
27
28
29
30
 
# File 'lib/kafka/fetch_operation.rb', line 23

def initialize(cluster:, logger:, min_bytes: 1, max_bytes: 10485760, max_wait_time: 5)
  @cluster = cluster
  @logger = logger
  @min_bytes = min_bytes
  @max_bytes = max_bytes
  @max_wait_time = max_wait_time
  @topics = {}
end

Instance Method Details

#execute ⇒ Object 



46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
 
# File 'lib/kafka/fetch_operation.rb', line 46

def execute
  @cluster.add_target_topics(@topics.keys)
  @cluster.refresh_metadata_if_necessary!

  topics_by_broker = {}

  if @topics.none? {|topic, partitions| partitions.any? }
    raise NoPartitionsToFetchFrom
  end

  @topics.each do |topic, partitions|
    partitions.each do |partition, options|
      broker = @cluster.get_leader(topic, partition)

      topics_by_broker[broker] ||= {}
      topics_by_broker[broker][topic] ||= {}
      topics_by_broker[broker][topic][partition] = options
    end
  end

  topics_by_broker.flat_map {|broker, topics|
    resolve_offsets(broker, topics)

    options = {
      max_wait_time: @max_wait_time * 1000, # Kafka expects ms, not secs
      min_bytes: @min_bytes,
      max_bytes: @max_bytes,
      topics: topics,
    }

    response = broker.fetch_messages(**options)

    response.topics.flat_map {|fetched_topic|
      fetched_topic.partitions.map {|fetched_partition|
        begin
          Protocol.handle_error(fetched_partition.error_code)
        rescue Kafka::OffsetOutOfRange => e
          e.topic = fetched_topic.name
          e.partition = fetched_partition.partition
          e.offset = topics.fetch(e.topic).fetch(e.partition).fetch(:fetch_offset)

          raise e
        rescue Kafka::Error => e
          topic = fetched_topic.name
          partition = fetched_partition.partition
          @logger.error "Failed to fetch from #{topic}/#{partition}: #{e.message}"
          raise e
        end

        messages = fetched_partition.messages.map {|message|
          FetchedMessage.new(
            message: message,
            topic: fetched_topic.name,
            partition: fetched_partition.partition,
          )
        }

        FetchedBatch.new(
          topic: fetched_topic.name,
          partition: fetched_partition.partition,
          highwater_mark_offset: fetched_partition.highwater_mark_offset,
          messages: messages,
        )
      }
    }
  }
rescue Kafka::ConnectionError, Kafka::LeaderNotAvailable, Kafka::NotLeaderForPartition
  @cluster.mark_as_stale!

  raise
end

#fetch_from_partition(topic, partition, offset: :latest, max_bytes: 1048576) ⇒ Object 



32
33
34
35
36
37
38
39
40
41
42
43
44
 
# File 'lib/kafka/fetch_operation.rb', line 32

def fetch_from_partition(topic, partition, offset: :latest, max_bytes: 1048576)
  if offset == :earliest
    offset = -2
  elsif offset == :latest
    offset = -1
  end

  @topics[topic] ||= {}
  @topics[topic][partition] = {
    fetch_offset: offset,
    max_bytes: max_bytes,
  }
end