Class: Fluent::BigQuery::Writer

Inherits:
Object
  • Object
show all
Defined in:
lib/fluent/plugin/bigquery/writer.rb

Defined Under Namespace

Classes: Error, RetryableError, UnRetryableError

Constant Summary collapse

RETRYABLE_ERROR_REASON = 
%w(backendError internalError rateLimitExceeded tableUnavailable).freeze

Instance Method Summary collapse

Constructor Details

#initialize(log, auth_method, auth_options = {}) ⇒ Writer

Returns a new instance of Writer.



43
44
45
46
47
48
49
50
 
# File 'lib/fluent/plugin/bigquery/writer.rb', line 43

def initialize(log, auth_method, auth_options = {})
  @auth_method = auth_method
  @scope = "https://www.googleapis.com/auth/bigquery"
  @auth_options = auth_options
  @log = log

  @cached_client_expiration = Time.now + 1800
end

Instance Method Details

#client ⇒ Object 



52
53
54
55
56
57
58
59
60
61
 
# File 'lib/fluent/plugin/bigquery/writer.rb', line 52

def client
  return @client if @client && @cached_client_expiration > Time.now

  client = Google::Apis::BigqueryV2::BigqueryService.new.tap do |cl|
    cl.authorization = get_auth
  end

  @cached_client_expiration = Time.now + 1800
  @client = client
end

#create_load_job(project, dataset, table_id, upload_source, job_id, fields, ignore_unknown_values: false, max_bad_records: 0, timeout_sec: nil, open_timeout_sec: 60, auto_create_table: nil, time_partitioning_type: nil, time_partitioning_expiration: nil) ⇒ Object 



149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
 
# File 'lib/fluent/plugin/bigquery/writer.rb', line 149

def create_load_job(project, dataset, table_id, upload_source, job_id, fields, ignore_unknown_values: false, max_bad_records: 0, timeout_sec: nil, open_timeout_sec: 60, auto_create_table: nil, time_partitioning_type: nil, time_partitioning_expiration: nil)
  configuration = {
    configuration: {
      load: {
        destination_table: {
          project_id: project,
          dataset_id: dataset,
          table_id: table_id,
        },
        schema: {
          fields: fields.to_a,
        },
        write_disposition: "WRITE_APPEND",
        source_format: "NEWLINE_DELIMITED_JSON",
        ignore_unknown_values: ignore_unknown_values,
        max_bad_records: max_bad_records,
      }
    }
  }
  configuration[:configuration][:load].merge!(create_disposition: "CREATE_NEVER") if time_partitioning_type
  configuration.merge!({job_reference: {project_id: project, job_id: job_id}}) if job_id

  # If target table is already exist, omit schema configuration.
  # Because schema changing is easier.
  begin
    if client.get_table(project, dataset, table_id)
      configuration[:configuration][:load].delete(:schema)
    end
  rescue Google::Apis::ServerError, Google::Apis::ClientError, Google::Apis::AuthorizationError
    raise UnRetryableError.new("Schema is empty") if fields.empty?
  end

  res = client.insert_job(
    project,
    configuration,
    {
      upload_source: upload_source,
      content_type: "application/octet-stream",
      options: {
        timeout_sec: timeout_sec,
        open_timeout_sec: open_timeout_sec,
      }
    }
  )
  wait_load_job(project, dataset, res.job_reference.job_id, table_id)
rescue Google::Apis::ServerError, Google::Apis::ClientError, Google::Apis::AuthorizationError => e
  @client = nil

  reason = e.respond_to?(:reason) ? e.reason : nil
  log.error "job.load API", project_id: project, dataset: dataset, table: table_id, code: e.status_code, message: e.message, reason: reason

  if auto_create_table && e.status_code == 404 && /Not Found: Table/i =~ e.message
    # Table Not Found: Auto Create Table
    create_table(project, dataset, table_id, fields, time_partitioning_type: time_partitioning_type, time_partitioning_expiration: time_partitioning_expiration)
    raise "table created. send rows next time."
  end

  return wait_load_job(project, dataset, job_id, table_id) if job_id && e.status_code == 409 && e.message =~ /Job/ # duplicate load job

  if RETRYABLE_ERROR_REASON.include?(reason) || e.is_a?(Google::Apis::ServerError)
    raise RetryableError.new(nil, e)
  else
    raise UnRetryableError.new(nil, e)
  end
end

#create_table(project, dataset, table_id, record_schema, time_partitioning_type: nil, time_partitioning_expiration: nil) ⇒ Object 



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
 
# File 'lib/fluent/plugin/bigquery/writer.rb', line 63

def create_table(project, dataset, table_id, record_schema, time_partitioning_type: nil, time_partitioning_expiration: nil)
  create_table_retry_limit = 3
  create_table_retry_wait = 1
  create_table_retry_count = 0

  begin
    definition = {
      table_reference: {
        table_id: table_id,
      },
      schema: {
        fields: record_schema.to_a,
      }
    }

    if time_partitioning_type
      definition[:time_partitioning] = {
        type: time_partitioning_type.to_s.upcase,
        expiration_ms: time_partitioning_expiration ? time_partitioning_expiration * 1000 : nil
      }.compact
    end
    client.insert_table(project, dataset, definition, {})
    log.debug "create table", project_id: project, dataset: dataset, table: table_id
    @client = nil
  rescue Google::Apis::ServerError, Google::Apis::ClientError, Google::Apis::AuthorizationError => e
    @client = nil

    message = e.message
    if e.status_code == 409 && /Already Exists:/ =~ message
      log.debug "already created table", project_id: project, dataset: dataset, table: table_id
      # ignore 'Already Exists' error
      return
    end

    reason = e.respond_to?(:reason) ? e.reason : nil
    log.error "tables.insert API", project_id: project, dataset: dataset, table: table_id, code: e.status_code, message: message, reason: reason

    if RETRYABLE_ERROR_REASON.include?(reason) && create_table_retry_count < create_table_retry_limit
      sleep create_table_retry_wait
      create_table_retry_wait *= 2
      create_table_retry_count += 1
      retry
    else
      raise UnRetryableError.new("failed to create table in bigquery", e)
    end
  end
end

#fetch_schema(project, dataset, table_id) ⇒ Object 



111
112
113
114
115
116
117
118
119
120
121
122
 
# File 'lib/fluent/plugin/bigquery/writer.rb', line 111

def fetch_schema(project, dataset, table_id)
  res = client.get_table(project, dataset, table_id)
  schema = res.schema.fields.as_json
  log.debug "Load schema from BigQuery: #{project}:#{dataset}.#{table_id} #{schema}"

  schema
rescue Google::Apis::ServerError, Google::Apis::ClientError, Google::Apis::AuthorizationError => e
  @client = nil
  message = e.message
  log.error "tables.get API", project_id: project, dataset: dataset, table: table_id, code: e.status_code, message: message
  nil
end

#insert_rows(project, dataset, table_id, rows, skip_invalid_rows: false, ignore_unknown_values: false, template_suffix: nil, timeout_sec: nil, open_timeout_sec: 60) ⇒ Object 



124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
 
# File 'lib/fluent/plugin/bigquery/writer.rb', line 124

def insert_rows(project, dataset, table_id, rows, skip_invalid_rows: false, ignore_unknown_values: false, template_suffix: nil, timeout_sec: nil, open_timeout_sec: 60)
  body = {
    rows: rows,
    skip_invalid_rows: skip_invalid_rows,
    ignore_unknown_values: ignore_unknown_values,
  }
  body.merge!(template_suffix: template_suffix) if template_suffix
  res = client.insert_all_table_data(project, dataset, table_id, body, {
    options: {timeout_sec: timeout_sec, open_timeout_sec: open_timeout_sec}
  })
  log.debug "insert rows", project_id: project, dataset: dataset, table: table_id, count: rows.size
  log.warn "insert errors", insert_errors: res.insert_errors.to_s if res.insert_errors && !res.insert_errors.empty?
rescue Google::Apis::ServerError, Google::Apis::ClientError, Google::Apis::AuthorizationError => e
  @client = nil

  reason = e.respond_to?(:reason) ? e.reason : nil
  log.error "tabledata.insertAll API", project_id: project, dataset: dataset, table: table_id, code: e.status_code, message: e.message, reason: reason

  if RETRYABLE_ERROR_REASON.include?(reason)
    raise RetryableError.new(nil, e)
  else
    raise UnRetryableError.new(nil, e)
  end
end

#wait_load_job(project, dataset, job_id, table_id, retryable: true) ⇒ Object 



215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
 
# File 'lib/fluent/plugin/bigquery/writer.rb', line 215

def wait_load_job(project, dataset, job_id, table_id, retryable: true)
  wait_interval = 10
  _response = client.get_job(project, job_id)

  until _response.status.state == "DONE"
    log.debug "wait for load job finish", state: _response.status.state, job_id: _response.job_reference.job_id
    sleep wait_interval
    _response = client.get_job(project, _response.job_reference.job_id)
  end

  errors = _response.status.errors
  if errors
    errors.each do |e|
      log.error "job.insert API (rows)", job_id: job_id, project_id: project, dataset: dataset, table: table_id, message: e.message, reason: e.reason
    end
  end

  error_result = _response.status.error_result
  if error_result
    log.error "job.insert API (result)", job_id: job_id, project_id: project, dataset: dataset, table: table_id, message: error_result.message, reason: error_result.reason
    if retryable && RETRYABLE_ERROR_REASON.include?(error_result.reason)
      raise RetryableError.new("failed to load into bigquery, retry")
    else
      raise UnRetryableError.new("failed to load into bigquery, and cannot retry")
    end
  end

  log.debug "finish load job", state: _response.status.state
end