fluent-plugin-bigquery 1.2.0 → 2.0.0.beta
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/.travis.yml +2 -9
- data/README.md +68 -65
- data/lib/fluent/plugin/bigquery/version.rb +1 -1
- data/lib/fluent/plugin/bigquery/writer.rb +45 -39
- data/lib/fluent/plugin/out_bigquery_base.rb +211 -0
- data/lib/fluent/plugin/out_bigquery_insert.rb +131 -0
- data/lib/fluent/plugin/out_bigquery_load.rb +220 -0
- data/test/helper.rb +3 -1
- data/test/plugin/test_out_bigquery_base.rb +579 -0
- data/test/plugin/test_out_bigquery_insert.rb +420 -0
- data/test/plugin/test_out_bigquery_load.rb +310 -0
- metadata +13 -7
- data/lib/fluent/plugin/out_bigquery.rb +0 -500
- data/test/plugin/test_out_bigquery.rb +0 -1276
@@ -0,0 +1,420 @@
|
|
1
|
+
require 'helper'
|
2
|
+
|
3
|
+
class BigQueryInsertOutputTest < Test::Unit::TestCase
|
4
|
+
def setup
|
5
|
+
Fluent::Test.setup
|
6
|
+
end
|
7
|
+
|
8
|
+
CONFIG = %[
|
9
|
+
table foo
|
10
|
+
email foo@bar.example
|
11
|
+
private_key_path /path/to/key
|
12
|
+
project yourproject_id
|
13
|
+
dataset yourdataset_id
|
14
|
+
|
15
|
+
<inject>
|
16
|
+
time_format %s
|
17
|
+
time_key time
|
18
|
+
</inject>
|
19
|
+
|
20
|
+
schema [
|
21
|
+
{"name": "time", "type": "INTEGER"},
|
22
|
+
{"name": "status", "type": "INTEGER"},
|
23
|
+
{"name": "bytes", "type": "INTEGER"},
|
24
|
+
{"name": "vhost", "type": "STRING"},
|
25
|
+
{"name": "path", "type": "STRING"},
|
26
|
+
{"name": "method", "type": "STRING"},
|
27
|
+
{"name": "protocol", "type": "STRING"},
|
28
|
+
{"name": "agent", "type": "STRING"},
|
29
|
+
{"name": "referer", "type": "STRING"},
|
30
|
+
{"name": "remote", "type": "RECORD", "fields": [
|
31
|
+
{"name": "host", "type": "STRING"},
|
32
|
+
{"name": "ip", "type": "STRING"},
|
33
|
+
{"name": "user", "type": "STRING"}
|
34
|
+
]},
|
35
|
+
{"name": "requesttime", "type": "FLOAT"},
|
36
|
+
{"name": "bot_access", "type": "BOOLEAN"},
|
37
|
+
{"name": "loginsession", "type": "BOOLEAN"}
|
38
|
+
]
|
39
|
+
]
|
40
|
+
|
41
|
+
API_SCOPE = "https://www.googleapis.com/auth/bigquery"
|
42
|
+
|
43
|
+
def create_driver(conf = CONFIG)
|
44
|
+
Fluent::Test::Driver::Output.new(Fluent::Plugin::BigQueryInsertOutput).configure(conf)
|
45
|
+
end
|
46
|
+
|
47
|
+
def stub_writer(stub_auth: true)
|
48
|
+
stub.proxy(Fluent::BigQuery::Writer).new.with_any_args do |writer|
|
49
|
+
stub(writer).get_auth { nil } if stub_auth
|
50
|
+
yield writer
|
51
|
+
writer
|
52
|
+
end
|
53
|
+
end
|
54
|
+
|
55
|
+
def test__write_with_insert_id
|
56
|
+
now = Time.now.to_i
|
57
|
+
input = {
|
58
|
+
"uuid" => "9ABFF756-0267-4247-847F-0895B65F0938",
|
59
|
+
}
|
60
|
+
expected = {
|
61
|
+
insert_id: "9ABFF756-0267-4247-847F-0895B65F0938",
|
62
|
+
json: {
|
63
|
+
uuid: "9ABFF756-0267-4247-847F-0895B65F0938",
|
64
|
+
}
|
65
|
+
}
|
66
|
+
|
67
|
+
driver = create_driver(<<-CONFIG)
|
68
|
+
table foo
|
69
|
+
email foo@bar.example
|
70
|
+
private_key_path /path/to/key
|
71
|
+
project yourproject_id
|
72
|
+
dataset yourdataset_id
|
73
|
+
|
74
|
+
insert_id_field uuid
|
75
|
+
schema [{"name": "uuid", "type": "STRING"}]
|
76
|
+
CONFIG
|
77
|
+
mock(driver.instance).insert("yourproject_id", "yourdataset_id", "foo", [expected], instance_of(Fluent::BigQuery::RecordSchema), nil)
|
78
|
+
|
79
|
+
driver.run do
|
80
|
+
driver.feed('tag', now, input)
|
81
|
+
end
|
82
|
+
end
|
83
|
+
|
84
|
+
def test__write_with_nested_insert_id
|
85
|
+
input = {
|
86
|
+
"data" => {
|
87
|
+
"uuid" => "809F6BA7-1C16-44CD-9816-4B20E2C7AA2A",
|
88
|
+
},
|
89
|
+
}
|
90
|
+
expected = {
|
91
|
+
insert_id: "809F6BA7-1C16-44CD-9816-4B20E2C7AA2A",
|
92
|
+
json: {
|
93
|
+
data: {
|
94
|
+
uuid: "809F6BA7-1C16-44CD-9816-4B20E2C7AA2A",
|
95
|
+
}
|
96
|
+
}
|
97
|
+
}
|
98
|
+
|
99
|
+
driver = create_driver(<<-CONFIG)
|
100
|
+
table foo
|
101
|
+
email foo@bar.example
|
102
|
+
private_key_path /path/to/key
|
103
|
+
project yourproject_id
|
104
|
+
dataset yourdataset_id
|
105
|
+
|
106
|
+
insert_id_field $.data.uuid
|
107
|
+
schema [{"name": "data", "type": "RECORD", "fields": [
|
108
|
+
{"name": "uuid", "type": "STRING"}
|
109
|
+
]}]
|
110
|
+
CONFIG
|
111
|
+
|
112
|
+
mock(driver.instance).insert("yourproject_id", "yourdataset_id", "foo", [expected], instance_of(Fluent::BigQuery::RecordSchema), nil)
|
113
|
+
|
114
|
+
driver.run do
|
115
|
+
driver.feed('tag', Fluent::EventTime.now, input)
|
116
|
+
end
|
117
|
+
end
|
118
|
+
|
119
|
+
def test_write
|
120
|
+
entry = {a: "b"}
|
121
|
+
driver = create_driver
|
122
|
+
|
123
|
+
stub_writer do |writer|
|
124
|
+
mock.proxy(writer).insert_rows('yourproject_id', 'yourdataset_id', 'foo', [{json: hash_including(entry)}], template_suffix: nil)
|
125
|
+
mock(writer.client).insert_all_table_data('yourproject_id', 'yourdataset_id', 'foo', {
|
126
|
+
rows: [{json: hash_including(entry)}],
|
127
|
+
skip_invalid_rows: false,
|
128
|
+
ignore_unknown_values: false
|
129
|
+
}, {}) do
|
130
|
+
s = stub!
|
131
|
+
s.insert_errors { nil }
|
132
|
+
s
|
133
|
+
end
|
134
|
+
end
|
135
|
+
|
136
|
+
driver.run do
|
137
|
+
driver.feed("tag", Time.now.to_i, {"a" => "b"})
|
138
|
+
end
|
139
|
+
end
|
140
|
+
|
141
|
+
def test_write_with_retryable_error
|
142
|
+
data_input = [
|
143
|
+
{ "status_code" => 500 },
|
144
|
+
{ "status_code" => 502 },
|
145
|
+
{ "status_code" => 503 },
|
146
|
+
{ "status_code" => 504 },
|
147
|
+
]
|
148
|
+
|
149
|
+
data_input.each do |d|
|
150
|
+
driver = create_driver(<<-CONFIG)
|
151
|
+
table foo
|
152
|
+
email foo@bar.example
|
153
|
+
private_key_path /path/to/key
|
154
|
+
project yourproject_id
|
155
|
+
dataset yourdataset_id
|
156
|
+
|
157
|
+
<inject>
|
158
|
+
time_format %s
|
159
|
+
time_key time
|
160
|
+
</inject>
|
161
|
+
|
162
|
+
schema [
|
163
|
+
{"name": "time", "type": "INTEGER"},
|
164
|
+
{"name": "status", "type": "INTEGER"},
|
165
|
+
{"name": "bytes", "type": "INTEGER"},
|
166
|
+
{"name": "vhost", "type": "STRING"},
|
167
|
+
{"name": "path", "type": "STRING"},
|
168
|
+
{"name": "method", "type": "STRING"},
|
169
|
+
{"name": "protocol", "type": "STRING"},
|
170
|
+
{"name": "agent", "type": "STRING"},
|
171
|
+
{"name": "referer", "type": "STRING"},
|
172
|
+
{"name": "remote", "type": "RECORD", "fields": [
|
173
|
+
{"name": "host", "type": "STRING"},
|
174
|
+
{"name": "ip", "type": "STRING"},
|
175
|
+
{"name": "user", "type": "STRING"}
|
176
|
+
]},
|
177
|
+
{"name": "requesttime", "type": "FLOAT"},
|
178
|
+
{"name": "bot_access", "type": "BOOLEAN"},
|
179
|
+
{"name": "loginsession", "type": "BOOLEAN"}
|
180
|
+
]
|
181
|
+
<secondary>
|
182
|
+
type file
|
183
|
+
path error
|
184
|
+
utc
|
185
|
+
</secondary>
|
186
|
+
CONFIG
|
187
|
+
|
188
|
+
entry = {a: "b"}
|
189
|
+
stub_writer do |writer|
|
190
|
+
mock(writer.client).insert_all_table_data('yourproject_id', 'yourdataset_id', 'foo', {
|
191
|
+
rows: [{json: hash_including(entry)}],
|
192
|
+
skip_invalid_rows: false,
|
193
|
+
ignore_unknown_values: false
|
194
|
+
}, {}) do
|
195
|
+
ex = Google::Apis::ServerError.new("error", status_code: d["status_code"])
|
196
|
+
raise ex
|
197
|
+
end
|
198
|
+
end
|
199
|
+
|
200
|
+
assert_raise(Fluent::BigQuery::RetryableError) do
|
201
|
+
driver.run do
|
202
|
+
driver.feed("tag", Time.now.to_i, {"a" => "b"})
|
203
|
+
end
|
204
|
+
end
|
205
|
+
end
|
206
|
+
end
|
207
|
+
|
208
|
+
def test_write_with_not_retryable_error
|
209
|
+
driver = create_driver(<<-CONFIG)
|
210
|
+
table foo
|
211
|
+
email foo@bar.example
|
212
|
+
private_key_path /path/to/key
|
213
|
+
project yourproject_id
|
214
|
+
dataset yourdataset_id
|
215
|
+
|
216
|
+
<inject>
|
217
|
+
time_format %s
|
218
|
+
time_key time
|
219
|
+
</inject>
|
220
|
+
|
221
|
+
schema [
|
222
|
+
{"name": "time", "type": "INTEGER"},
|
223
|
+
{"name": "status", "type": "INTEGER"},
|
224
|
+
{"name": "bytes", "type": "INTEGER"},
|
225
|
+
{"name": "vhost", "type": "STRING"},
|
226
|
+
{"name": "path", "type": "STRING"},
|
227
|
+
{"name": "method", "type": "STRING"},
|
228
|
+
{"name": "protocol", "type": "STRING"},
|
229
|
+
{"name": "agent", "type": "STRING"},
|
230
|
+
{"name": "referer", "type": "STRING"},
|
231
|
+
{"name": "remote", "type": "RECORD", "fields": [
|
232
|
+
{"name": "host", "type": "STRING"},
|
233
|
+
{"name": "ip", "type": "STRING"},
|
234
|
+
{"name": "user", "type": "STRING"}
|
235
|
+
]},
|
236
|
+
{"name": "requesttime", "type": "FLOAT"},
|
237
|
+
{"name": "bot_access", "type": "BOOLEAN"},
|
238
|
+
{"name": "loginsession", "type": "BOOLEAN"}
|
239
|
+
]
|
240
|
+
<secondary>
|
241
|
+
type file
|
242
|
+
path error
|
243
|
+
utc
|
244
|
+
</secondary>
|
245
|
+
CONFIG
|
246
|
+
|
247
|
+
entry = {a: "b"}
|
248
|
+
stub_writer do |writer|
|
249
|
+
mock(writer.client).insert_all_table_data('yourproject_id', 'yourdataset_id', 'foo', {
|
250
|
+
rows: [{json: hash_including(entry)}],
|
251
|
+
skip_invalid_rows: false,
|
252
|
+
ignore_unknown_values: false
|
253
|
+
}, {}) do
|
254
|
+
ex = Google::Apis::ServerError.new("error", status_code: 501)
|
255
|
+
def ex.reason
|
256
|
+
"invalid"
|
257
|
+
end
|
258
|
+
raise ex
|
259
|
+
end
|
260
|
+
end
|
261
|
+
|
262
|
+
driver.instance_start
|
263
|
+
tag, time, record = "tag", Time.now.to_i, {"a" => "b"}
|
264
|
+
metadata = driver.instance.metadata_for_test(tag, time, record)
|
265
|
+
chunk = driver.instance.buffer.generate_chunk(metadata).tap do |c|
|
266
|
+
c.append([driver.instance.format(tag, time, record)])
|
267
|
+
end
|
268
|
+
assert_raise Fluent::BigQuery::UnRetryableError do
|
269
|
+
driver.instance.write(chunk)
|
270
|
+
end
|
271
|
+
assert_in_delta driver.instance.retry.secondary_transition_at , Time.now, 0.1
|
272
|
+
driver.instance_shutdown
|
273
|
+
end
|
274
|
+
|
275
|
+
def test_write_with_row_based_table_id_formatting
|
276
|
+
entry = [
|
277
|
+
{json: {a: "b", created_at: Time.local(2014,8,20,9,0,0).strftime("%Y_%m_%d")}},
|
278
|
+
]
|
279
|
+
driver = create_driver(<<-CONFIG)
|
280
|
+
<buffer created_at>
|
281
|
+
</buffer>
|
282
|
+
table foo_${created_at}
|
283
|
+
email foo@bar.example
|
284
|
+
private_key_path /path/to/key
|
285
|
+
project yourproject_id
|
286
|
+
dataset yourdataset_id
|
287
|
+
|
288
|
+
schema [
|
289
|
+
{"name": "time", "type": "INTEGER"}
|
290
|
+
]
|
291
|
+
CONFIG
|
292
|
+
|
293
|
+
stub_writer do |writer|
|
294
|
+
mock(writer.client).insert_all_table_data('yourproject_id', 'yourdataset_id', 'foo_2014_08_20', {
|
295
|
+
rows: [entry[0]],
|
296
|
+
skip_invalid_rows: false,
|
297
|
+
ignore_unknown_values: false
|
298
|
+
}, {}) { stub!.insert_errors { nil } }
|
299
|
+
end
|
300
|
+
|
301
|
+
driver.run do
|
302
|
+
driver.feed("tag", Time.now.to_i, {"a" => "b", "created_at" => Time.local(2014,8,20,9,0,0).strftime("%Y_%m_%d")})
|
303
|
+
end
|
304
|
+
end
|
305
|
+
|
306
|
+
def test_auto_create_table_by_bigquery_api
|
307
|
+
now = Time.at(Time.now.to_i)
|
308
|
+
message = {
|
309
|
+
"time" => now.to_i,
|
310
|
+
"request" => {
|
311
|
+
"vhost" => "bar",
|
312
|
+
"path" => "/path/to/baz",
|
313
|
+
"method" => "GET",
|
314
|
+
"protocol" => "HTTP/1.0",
|
315
|
+
"agent" => "libwww",
|
316
|
+
"referer" => "http://referer.example",
|
317
|
+
"time" => (now - 1).to_f,
|
318
|
+
"bot_access" => true,
|
319
|
+
"loginsession" => false,
|
320
|
+
},
|
321
|
+
"remote" => {
|
322
|
+
"host" => "remote.example",
|
323
|
+
"ip" => "192.168.1.1",
|
324
|
+
"user" => "nagachika",
|
325
|
+
},
|
326
|
+
"response" => {
|
327
|
+
"status" => 200,
|
328
|
+
"bytes" => 72,
|
329
|
+
},
|
330
|
+
}
|
331
|
+
|
332
|
+
driver = create_driver(<<-CONFIG)
|
333
|
+
table foo
|
334
|
+
email foo@bar.example
|
335
|
+
private_key_path /path/to/key
|
336
|
+
project yourproject_id
|
337
|
+
dataset yourdataset_id
|
338
|
+
|
339
|
+
<inject>
|
340
|
+
time_format %s
|
341
|
+
time_key time
|
342
|
+
</inject>
|
343
|
+
|
344
|
+
auto_create_table true
|
345
|
+
schema_path #{File.join(File.dirname(__FILE__), "testdata", "apache.schema")}
|
346
|
+
CONFIG
|
347
|
+
|
348
|
+
stub_writer do |writer|
|
349
|
+
mock(writer).insert_rows('yourproject_id', 'yourdataset_id', 'foo', [{json: Fluent::BigQuery::Helper.deep_symbolize_keys(message)}], template_suffix: nil) do
|
350
|
+
raise Fluent::BigQuery::RetryableError.new(nil, Google::Apis::ServerError.new("Not found: Table yourproject_id:yourdataset_id.foo", status_code: 404, body: "Not found: Table yourproject_id:yourdataset_id.foo"))
|
351
|
+
end
|
352
|
+
mock(writer).create_table('yourproject_id', 'yourdataset_id', 'foo', driver.instance.instance_variable_get(:@table_schema))
|
353
|
+
end
|
354
|
+
|
355
|
+
assert_raise(RuntimeError) do
|
356
|
+
driver.run do
|
357
|
+
driver.feed("tag", Fluent::EventTime.from_time(now), message)
|
358
|
+
end
|
359
|
+
end
|
360
|
+
end
|
361
|
+
|
362
|
+
def test_auto_create_partitioned_table_by_bigquery_api
|
363
|
+
now = Time.now
|
364
|
+
message = {
|
365
|
+
json: {
|
366
|
+
time: now.to_i,
|
367
|
+
request: {
|
368
|
+
vhost: "bar",
|
369
|
+
path: "/path/to/baz",
|
370
|
+
method: "GET",
|
371
|
+
protocol: "HTTP/1.0",
|
372
|
+
agent: "libwww",
|
373
|
+
referer: "http://referer.example",
|
374
|
+
time: (now - 1).to_f,
|
375
|
+
bot_access: true,
|
376
|
+
loginsession: false,
|
377
|
+
},
|
378
|
+
remote: {
|
379
|
+
host: "remote.example",
|
380
|
+
ip: "192.168.1.1",
|
381
|
+
user: "nagachika",
|
382
|
+
},
|
383
|
+
response: {
|
384
|
+
status: 200,
|
385
|
+
bytes: 72,
|
386
|
+
},
|
387
|
+
}
|
388
|
+
}
|
389
|
+
|
390
|
+
driver = create_driver(<<-CONFIG)
|
391
|
+
table foo
|
392
|
+
email foo@bar.example
|
393
|
+
private_key_path /path/to/key
|
394
|
+
project yourproject_id
|
395
|
+
dataset yourdataset_id
|
396
|
+
|
397
|
+
time_format %s
|
398
|
+
time_field time
|
399
|
+
|
400
|
+
auto_create_table true
|
401
|
+
schema_path #{File.join(File.dirname(__FILE__), "testdata", "apache.schema")}
|
402
|
+
|
403
|
+
time_partitioning_type day
|
404
|
+
time_partitioning_expiration 1h
|
405
|
+
CONFIG
|
406
|
+
|
407
|
+
stub_writer do |writer|
|
408
|
+
mock(writer).insert_rows('yourproject_id', 'yourdataset_id', 'foo', [message], template_suffix: nil) do
|
409
|
+
raise Fluent::BigQuery::RetryableError.new(nil, Google::Apis::ServerError.new("Not found: Table yourproject_id:yourdataset_id.foo", status_code: 404, body: "Not found: Table yourproject_id:yourdataset_id.foo"))
|
410
|
+
end
|
411
|
+
mock(writer).create_table('yourproject_id', 'yourdataset_id', 'foo', driver.instance.instance_variable_get(:@table_schema))
|
412
|
+
end
|
413
|
+
|
414
|
+
assert_raise(RuntimeError) do
|
415
|
+
driver.run do
|
416
|
+
driver.feed("tag", Fluent::EventTime.now, message[:json])
|
417
|
+
end
|
418
|
+
end
|
419
|
+
end
|
420
|
+
end
|
@@ -0,0 +1,310 @@
|
|
1
|
+
require 'helper'
|
2
|
+
|
3
|
+
class BigQueryLoadOutputTest < Test::Unit::TestCase
|
4
|
+
def setup
|
5
|
+
Fluent::Test.setup
|
6
|
+
end
|
7
|
+
|
8
|
+
SCHEMA_PATH = File.join(File.dirname(__FILE__), "testdata", "sudo.schema")
|
9
|
+
CONFIG = %[
|
10
|
+
table foo
|
11
|
+
email foo@bar.example
|
12
|
+
private_key_path /path/to/key
|
13
|
+
project yourproject_id
|
14
|
+
dataset yourdataset_id
|
15
|
+
|
16
|
+
<buffer>
|
17
|
+
@type memory
|
18
|
+
</buffer>
|
19
|
+
|
20
|
+
<inject>
|
21
|
+
time_format %s
|
22
|
+
time_key time
|
23
|
+
</inject>
|
24
|
+
|
25
|
+
schema_path #{SCHEMA_PATH}
|
26
|
+
wait_job_interval 0.1
|
27
|
+
]
|
28
|
+
|
29
|
+
API_SCOPE = "https://www.googleapis.com/auth/bigquery"
|
30
|
+
|
31
|
+
def create_driver(conf = CONFIG)
|
32
|
+
Fluent::Test::Driver::Output.new(Fluent::Plugin::BigQueryLoadOutput).configure(conf)
|
33
|
+
end
|
34
|
+
|
35
|
+
def stub_writer(stub_auth: true)
|
36
|
+
stub.proxy(Fluent::BigQuery::Writer).new.with_any_args do |writer|
|
37
|
+
stub(writer).get_auth { nil } if stub_auth
|
38
|
+
yield writer
|
39
|
+
writer
|
40
|
+
end
|
41
|
+
end
|
42
|
+
|
43
|
+
def test_write
|
44
|
+
schema_fields = Fluent::BigQuery::Helper.deep_symbolize_keys(MultiJson.load(File.read(SCHEMA_PATH)))
|
45
|
+
|
46
|
+
response_stub = stub!
|
47
|
+
|
48
|
+
driver = create_driver
|
49
|
+
stub_writer do |writer|
|
50
|
+
mock(writer).fetch_load_job(is_a(Fluent::BigQuery::Writer::JobReference)) { response_stub }
|
51
|
+
mock(writer).commit_load_job(is_a(String), response_stub)
|
52
|
+
|
53
|
+
mock(writer.client).get_table('yourproject_id', 'yourdataset_id', 'foo') { nil }
|
54
|
+
|
55
|
+
mock(writer.client).insert_job('yourproject_id', {
|
56
|
+
configuration: {
|
57
|
+
load: {
|
58
|
+
destination_table: {
|
59
|
+
project_id: 'yourproject_id',
|
60
|
+
dataset_id: 'yourdataset_id',
|
61
|
+
table_id: 'foo',
|
62
|
+
},
|
63
|
+
schema: {
|
64
|
+
fields: schema_fields,
|
65
|
+
},
|
66
|
+
write_disposition: "WRITE_APPEND",
|
67
|
+
source_format: "NEWLINE_DELIMITED_JSON",
|
68
|
+
ignore_unknown_values: false,
|
69
|
+
max_bad_records: 0,
|
70
|
+
}
|
71
|
+
}
|
72
|
+
}, {upload_source: duck_type(:write, :sync, :rewind), content_type: "application/octet-stream"}) do
|
73
|
+
stub!.job_reference.stub!.job_id { "dummy_job_id" }
|
74
|
+
end
|
75
|
+
end
|
76
|
+
|
77
|
+
driver.run do
|
78
|
+
driver.feed("tag", Time.now.to_i, {"a" => "b"})
|
79
|
+
end
|
80
|
+
end
|
81
|
+
|
82
|
+
def test_write_with_prevent_duplicate_load
|
83
|
+
driver = create_driver(<<-CONFIG)
|
84
|
+
table foo
|
85
|
+
email foo@bar.example
|
86
|
+
private_key_path /path/to/key
|
87
|
+
project yourproject_id
|
88
|
+
dataset yourdataset_id
|
89
|
+
|
90
|
+
<buffer>
|
91
|
+
@type memory
|
92
|
+
</buffer>
|
93
|
+
|
94
|
+
<inject>
|
95
|
+
time_format %s
|
96
|
+
time_key time
|
97
|
+
</inject>
|
98
|
+
|
99
|
+
schema_path #{SCHEMA_PATH}
|
100
|
+
prevent_duplicate_load true
|
101
|
+
CONFIG
|
102
|
+
schema_fields = Fluent::BigQuery::Helper.deep_symbolize_keys(MultiJson.load(File.read(SCHEMA_PATH)))
|
103
|
+
|
104
|
+
response_stub = stub!
|
105
|
+
stub_writer do |writer|
|
106
|
+
mock(writer).fetch_load_job(is_a(Fluent::BigQuery::Writer::JobReference)) { response_stub }
|
107
|
+
mock(writer).commit_load_job(is_a(String), response_stub)
|
108
|
+
|
109
|
+
mock(writer.client).get_table('yourproject_id', 'yourdataset_id', 'foo') { nil }
|
110
|
+
|
111
|
+
mock(writer.client).insert_job('yourproject_id', {
|
112
|
+
configuration: {
|
113
|
+
load: {
|
114
|
+
destination_table: {
|
115
|
+
project_id: 'yourproject_id',
|
116
|
+
dataset_id: 'yourdataset_id',
|
117
|
+
table_id: 'foo',
|
118
|
+
},
|
119
|
+
schema: {
|
120
|
+
fields: schema_fields,
|
121
|
+
},
|
122
|
+
write_disposition: "WRITE_APPEND",
|
123
|
+
source_format: "NEWLINE_DELIMITED_JSON",
|
124
|
+
ignore_unknown_values: false,
|
125
|
+
max_bad_records: 0,
|
126
|
+
},
|
127
|
+
},
|
128
|
+
job_reference: {project_id: 'yourproject_id', job_id: satisfy { |x| x =~ /fluentd_job_.*/}} ,
|
129
|
+
}, {upload_source: duck_type(:write, :sync, :rewind), content_type: "application/octet-stream"}) do
|
130
|
+
stub!.job_reference.stub!.job_id { "dummy_job_id" }
|
131
|
+
end
|
132
|
+
end
|
133
|
+
|
134
|
+
driver.run do
|
135
|
+
driver.feed("tag", Time.now.to_i, {"a" => "b"})
|
136
|
+
end
|
137
|
+
end
|
138
|
+
|
139
|
+
def test_write_with_retryable_error
|
140
|
+
driver = create_driver
|
141
|
+
schema_fields = Fluent::BigQuery::Helper.deep_symbolize_keys(MultiJson.load(File.read(SCHEMA_PATH)))
|
142
|
+
|
143
|
+
driver.instance_start
|
144
|
+
tag, time, record = "tag", Time.now.to_i, {"a" => "b"}
|
145
|
+
metadata = driver.instance.metadata_for_test(tag, time, record)
|
146
|
+
chunk = driver.instance.buffer.generate_chunk(metadata).tap do |c|
|
147
|
+
c.append([driver.instance.format(tag, time, record)])
|
148
|
+
end
|
149
|
+
|
150
|
+
stub_writer do |writer|
|
151
|
+
mock(writer.client).get_table('yourproject_id', 'yourdataset_id', 'foo') { nil }
|
152
|
+
|
153
|
+
mock(writer.client).insert_job('yourproject_id', {
|
154
|
+
configuration: {
|
155
|
+
load: {
|
156
|
+
destination_table: {
|
157
|
+
project_id: 'yourproject_id',
|
158
|
+
dataset_id: 'yourdataset_id',
|
159
|
+
table_id: 'foo',
|
160
|
+
},
|
161
|
+
schema: {
|
162
|
+
fields: schema_fields,
|
163
|
+
},
|
164
|
+
write_disposition: "WRITE_APPEND",
|
165
|
+
source_format: "NEWLINE_DELIMITED_JSON",
|
166
|
+
ignore_unknown_values: false,
|
167
|
+
max_bad_records: 0,
|
168
|
+
}
|
169
|
+
}
|
170
|
+
}, {upload_source: duck_type(:write, :sync, :rewind), content_type: "application/octet-stream"}) do
|
171
|
+
stub!.job_reference.stub!.job_id { "dummy_job_id" }
|
172
|
+
end
|
173
|
+
|
174
|
+
mock(writer.client).get_job('yourproject_id', 'dummy_job_id') do
|
175
|
+
stub! do |s|
|
176
|
+
s.id { 'dummy_job_id' }
|
177
|
+
s.configuration.stub! do |_s|
|
178
|
+
_s.load.stub! do |__s|
|
179
|
+
__s.destination_table.stub! do |___s|
|
180
|
+
___s.project_id { 'yourproject_id' }
|
181
|
+
___s.dataset_id { 'yourdataset_id' }
|
182
|
+
___s.table_id { 'foo' }
|
183
|
+
end
|
184
|
+
end
|
185
|
+
end
|
186
|
+
s.status.stub! do |_s|
|
187
|
+
_s.state { 'DONE' }
|
188
|
+
_s.errors { [] }
|
189
|
+
_s.error_result.stub! do |__s|
|
190
|
+
__s.message { 'error' }
|
191
|
+
__s.reason { 'backendError' }
|
192
|
+
end
|
193
|
+
end
|
194
|
+
end
|
195
|
+
end
|
196
|
+
end
|
197
|
+
|
198
|
+
assert_raise Fluent::BigQuery::RetryableError do
|
199
|
+
driver.instance.write(chunk)
|
200
|
+
end
|
201
|
+
driver.instance_shutdown
|
202
|
+
end
|
203
|
+
|
204
|
+
def test_write_with_not_retryable_error
|
205
|
+
driver = create_driver(<<-CONFIG)
|
206
|
+
table foo
|
207
|
+
email foo@bar.example
|
208
|
+
private_key_path /path/to/key
|
209
|
+
project yourproject_id
|
210
|
+
dataset yourdataset_id
|
211
|
+
|
212
|
+
<buffer>
|
213
|
+
@type memory
|
214
|
+
</buffer>
|
215
|
+
|
216
|
+
<inject>
|
217
|
+
time_format %s
|
218
|
+
time_key time
|
219
|
+
</inject>
|
220
|
+
|
221
|
+
schema_path #{SCHEMA_PATH}
|
222
|
+
<secondary>
|
223
|
+
@type file
|
224
|
+
path error
|
225
|
+
utc
|
226
|
+
</secondary>
|
227
|
+
CONFIG
|
228
|
+
schema_fields = Fluent::BigQuery::Helper.deep_symbolize_keys(MultiJson.load(File.read(SCHEMA_PATH)))
|
229
|
+
|
230
|
+
driver.instance_start
|
231
|
+
tag, time, record = "tag", Time.now.to_i, {"a" => "b"}
|
232
|
+
metadata = driver.instance.metadata_for_test(tag, time, record)
|
233
|
+
chunk = driver.instance.buffer.generate_chunk(metadata).tap do |c|
|
234
|
+
c.append([driver.instance.format(tag, time, record)])
|
235
|
+
end
|
236
|
+
|
237
|
+
stub_writer do |writer|
|
238
|
+
mock(writer.client).get_table('yourproject_id', 'yourdataset_id', 'foo') { nil }
|
239
|
+
|
240
|
+
mock(writer.client).insert_job('yourproject_id', {
|
241
|
+
configuration: {
|
242
|
+
load: {
|
243
|
+
destination_table: {
|
244
|
+
project_id: 'yourproject_id',
|
245
|
+
dataset_id: 'yourdataset_id',
|
246
|
+
table_id: 'foo',
|
247
|
+
},
|
248
|
+
schema: {
|
249
|
+
fields: schema_fields,
|
250
|
+
},
|
251
|
+
write_disposition: "WRITE_APPEND",
|
252
|
+
source_format: "NEWLINE_DELIMITED_JSON",
|
253
|
+
ignore_unknown_values: false,
|
254
|
+
max_bad_records: 0,
|
255
|
+
}
|
256
|
+
}
|
257
|
+
}, {upload_source: duck_type(:write, :sync, :rewind), content_type: "application/octet-stream"}) do
|
258
|
+
stub!.job_reference.stub!.job_id { "dummy_job_id" }
|
259
|
+
end
|
260
|
+
|
261
|
+
mock(writer.client).get_job('yourproject_id', 'dummy_job_id') do
|
262
|
+
stub! do |s|
|
263
|
+
s.id { 'dummy_job_id' }
|
264
|
+
s.configuration.stub! do |_s|
|
265
|
+
_s.load.stub! do |__s|
|
266
|
+
__s.destination_table.stub! do |___s|
|
267
|
+
___s.project_id { 'yourproject_id' }
|
268
|
+
___s.dataset_id { 'yourdataset_id' }
|
269
|
+
___s.table_id { 'foo' }
|
270
|
+
end
|
271
|
+
end
|
272
|
+
end
|
273
|
+
s.status.stub! do |_s|
|
274
|
+
_s.state { 'DONE' }
|
275
|
+
_s.errors { [] }
|
276
|
+
_s.error_result.stub! do |__s|
|
277
|
+
__s.message { 'error' }
|
278
|
+
__s.reason { 'invalid' }
|
279
|
+
end
|
280
|
+
end
|
281
|
+
end
|
282
|
+
end
|
283
|
+
end
|
284
|
+
|
285
|
+
assert_raise Fluent::BigQuery::UnRetryableError do
|
286
|
+
driver.instance.write(chunk)
|
287
|
+
end
|
288
|
+
assert_in_delta driver.instance.retry.secondary_transition_at , Time.now, 0.1
|
289
|
+
driver.instance_shutdown
|
290
|
+
end
|
291
|
+
|
292
|
+
private
|
293
|
+
|
294
|
+
def create_response_stub(response)
|
295
|
+
case response
|
296
|
+
when Hash
|
297
|
+
root = stub!
|
298
|
+
response.each do |k, v|
|
299
|
+
root.__send__(k) do
|
300
|
+
create_response_stub(v)
|
301
|
+
end
|
302
|
+
end
|
303
|
+
root
|
304
|
+
when Array
|
305
|
+
response.map { |item| create_response_stub(item) }
|
306
|
+
else
|
307
|
+
response
|
308
|
+
end
|
309
|
+
end
|
310
|
+
end
|