Class: AthenaUDF::BaseUDF

Inherits:
Object
  • Object
show all
Includes:
Utils
Defined in:
lib/athena-udf/base_udf.rb

Instance Attribute Summary collapse

Class Method Summary collapse

Instance Method Summary collapse

Methods included from Utils

#get_record_batch_bytes, #get_record_batch_index, #get_schema_bytes, #read_record_batches, #read_schema

Constructor Details

#initialize(event:, context:) ⇒ BaseUDF

rubocop:disable Lint/UnusedMethodArgument



33
34
35
36
 
# File 'lib/athena-udf/base_udf.rb', line 33

def initialize(event:, context:) # rubocop:disable Lint/UnusedMethodArgument
  @logger = Logger.new($stdout)
  @logger.level = Logger.const_get(ENV.fetch('LOG_LEVEL', 'WARN').upcase)
end

Instance Attribute Details

#logger ⇒ Object (readonly)

Returns the value of attribute logger.



14
15
16
 
# File 'lib/athena-udf/base_udf.rb', line 14

def logger
  @logger
end

Class Method Details

.capabilities ⇒ Object

About capabilities: github.com/awslabs/aws-athena-query-federation/blob/f52d929a109099a1e7180fa242e26331137ed84c/athena-federation-sdk/src/main/java/com/amazonaws/athena/connector/lambda/handlers/FederationCapabilities.java#L29-L32



29
30
31
 
# File 'lib/athena-udf/base_udf.rb', line 29

def self.capabilities
  1
end

.lambda_handler(event:, context:) ⇒ Object 



16
17
18
19
20
21
22
23
24
25
26
 
# File 'lib/athena-udf/base_udf.rb', line 16

def self.lambda_handler(event:, context:)
  instance = new(event:, context:)
  incoming_type = event['@type']
  if incoming_type == 'PingRequest'
    return instance.handle_ping(event)
  elsif incoming_type == 'UserDefinedFunctionRequest'
    return instance.handle_udf_request(event)
  end

  raise "Unknown event type #{incoming_type} from Athena"
end

Instance Method Details

#handle_athena_record(input_schema, output_schema, records) ⇒ Object

Raises:

  • (NotImplementedError) 


79
80
81
 
# File 'lib/athena-udf/base_udf.rb', line 79

def handle_athena_record(input_schema, output_schema, records)
  raise NotImplementedError
end

#handle_ping(event) ⇒ Object 



38
39
40
41
42
43
44
45
46
 
# File 'lib/athena-udf/base_udf.rb', line 38

def handle_ping(event)
  {
    '@type' => 'PingResponse',
    'catalogName' => 'event',
    'queryId' => event['queryId'],
    'sourceType' => 'athena_udf',
    'capabilities' => self.class.capabilities,
  }
end

#handle_udf_request(event) ⇒ Object 



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
 
# File 'lib/athena-udf/base_udf.rb', line 48

def handle_udf_request(event)
  # Cannot find a way to write Arrow::RecordBatch to a buffer directly in Ruby.

  output_schema = read_schema(Base64.decode64(event['outputSchema']['schema']))
  output_builder = Arrow::RecordBatchBuilder.new(output_schema)

  input_schema_data = Base64.decode64(event['inputRecords']['schema'])
  input_records_data = Base64.decode64(event['inputRecords']['records'])
  read_record_batches(input_schema_data, input_records_data) do |input_schema, record_batch|
    logger.info("Processing #{record_batch.size} records")
    output_builder.append_records(
      record_batch.each_record.map do |record|
        handle_athena_record(input_schema, output_schema, record)
      end,
    )
  end

  output_record_batch = output_builder.flush
  output_records_bytes = get_record_batch_bytes(output_schema, output_record_batch)

  {
    '@type' => 'UserDefinedFunctionResponse',
    'methodName' => event['methodName'],
    'records' => {
      'aId' => SecureRandom.uuid,
      'schema' => event['outputSchema']['schema'],
      'records' => Base64.strict_encode64(output_records_bytes),
    },
  }
end