forked from feast-dev/feast
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathdata_source.py
More file actions
524 lines (427 loc) · 17 KB
/
data_source.py
File metadata and controls
524 lines (427 loc) · 17 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
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
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
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
214
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
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
# Copyright 2020 The Feast Authors
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# https://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import enum
from abc import ABC, abstractmethod
from typing import Any, Callable, Dict, Iterable, Optional, Tuple
from feast import type_map
from feast.data_format import StreamFormat
from feast.protos.feast.core.DataSource_pb2 import DataSource as DataSourceProto
from feast.repo_config import RepoConfig, get_data_source_class_from_type
from feast.value_type import ValueType
class SourceType(enum.Enum):
"""
DataSource value type. Used to define source types in DataSource.
"""
UNKNOWN = 0
BATCH_FILE = 1
BATCH_BIGQUERY = 2
STREAM_KAFKA = 3
STREAM_KINESIS = 4
class KafkaOptions:
"""
DataSource Kafka options used to source features from Kafka messages
"""
def __init__(
self, bootstrap_servers: str, message_format: StreamFormat, topic: str,
):
self.bootstrap_servers = bootstrap_servers
self.message_format = message_format
self.topic = topic
@classmethod
def from_proto(cls, kafka_options_proto: DataSourceProto.KafkaOptions):
"""
Creates a KafkaOptions from a protobuf representation of a kafka option
Args:
kafka_options_proto: A protobuf representation of a DataSource
Returns:
Returns a BigQueryOptions object based on the kafka_options protobuf
"""
kafka_options = cls(
bootstrap_servers=kafka_options_proto.bootstrap_servers,
message_format=StreamFormat.from_proto(kafka_options_proto.message_format),
topic=kafka_options_proto.topic,
)
return kafka_options
def to_proto(self) -> DataSourceProto.KafkaOptions:
"""
Converts an KafkaOptionsProto object to its protobuf representation.
Returns:
KafkaOptionsProto protobuf
"""
kafka_options_proto = DataSourceProto.KafkaOptions(
bootstrap_servers=self.bootstrap_servers,
message_format=self.message_format.to_proto(),
topic=self.topic,
)
return kafka_options_proto
class KinesisOptions:
"""
DataSource Kinesis options used to source features from Kinesis records
"""
def __init__(
self, record_format: StreamFormat, region: str, stream_name: str,
):
self.record_format = record_format
self.region = region
self.stream_name = stream_name
@classmethod
def from_proto(cls, kinesis_options_proto: DataSourceProto.KinesisOptions):
"""
Creates a KinesisOptions from a protobuf representation of a kinesis option
Args:
kinesis_options_proto: A protobuf representation of a DataSource
Returns:
Returns a KinesisOptions object based on the kinesis_options protobuf
"""
kinesis_options = cls(
record_format=StreamFormat.from_proto(kinesis_options_proto.record_format),
region=kinesis_options_proto.region,
stream_name=kinesis_options_proto.stream_name,
)
return kinesis_options
def to_proto(self) -> DataSourceProto.KinesisOptions:
"""
Converts an KinesisOptionsProto object to its protobuf representation.
Returns:
KinesisOptionsProto protobuf
"""
kinesis_options_proto = DataSourceProto.KinesisOptions(
record_format=self.record_format.to_proto(),
region=self.region,
stream_name=self.stream_name,
)
return kinesis_options_proto
class DataSource(ABC):
"""
DataSource that can be used to source features.
Args:
name: Name of data source, which should be unique within a project
event_timestamp_column (optional): Event timestamp column used for point in time
joins of feature values.
created_timestamp_column (optional): Timestamp column indicating when the row
was created, used for deduplicating rows.
field_mapping (optional): A dictionary mapping of column names in this data
source to feature names in a feature table or view. Only used for feature
columns, not entity or timestamp columns.
date_partition_column (optional): Timestamp column used for partitioning.
"""
name: str
event_timestamp_column: str
created_timestamp_column: str
field_mapping: Dict[str, str]
date_partition_column: str
def __init__(
self,
name: str,
event_timestamp_column: Optional[str] = None,
created_timestamp_column: Optional[str] = None,
field_mapping: Optional[Dict[str, str]] = None,
date_partition_column: Optional[str] = None,
):
"""Creates a DataSource object."""
self.name = name
self.event_timestamp_column = (
event_timestamp_column if event_timestamp_column else ""
)
self.created_timestamp_column = (
created_timestamp_column if created_timestamp_column else ""
)
self.field_mapping = field_mapping if field_mapping else {}
self.date_partition_column = (
date_partition_column if date_partition_column else ""
)
def __hash__(self):
return hash((id(self), self.name))
def __eq__(self, other):
if not isinstance(other, DataSource):
raise TypeError("Comparisons should only involve DataSource class objects.")
if (
self.name != other.name
or self.event_timestamp_column != other.event_timestamp_column
or self.created_timestamp_column != other.created_timestamp_column
or self.field_mapping != other.field_mapping
or self.date_partition_column != other.date_partition_column
):
return False
return True
@staticmethod
@abstractmethod
def from_proto(data_source: DataSourceProto) -> Any:
"""
Converts data source config in protobuf spec to a DataSource class object.
Args:
data_source: A protobuf representation of a DataSource.
Returns:
A DataSource class object.
Raises:
ValueError: The type of DataSource could not be identified.
"""
if data_source.data_source_class_type:
cls = get_data_source_class_from_type(data_source.data_source_class_type)
return cls.from_proto(data_source)
if data_source.request_data_options and data_source.request_data_options.schema:
data_source_obj = RequestDataSource.from_proto(data_source)
elif data_source.file_options.file_format and data_source.file_options.file_url:
from feast.infra.offline_stores.file_source import FileSource
data_source_obj = FileSource.from_proto(data_source)
elif (
data_source.bigquery_options.table_ref or data_source.bigquery_options.query
):
from feast.infra.offline_stores.bigquery_source import BigQuerySource
data_source_obj = BigQuerySource.from_proto(data_source)
elif data_source.redshift_options.table or data_source.redshift_options.query:
from feast.infra.offline_stores.redshift_source import RedshiftSource
data_source_obj = RedshiftSource.from_proto(data_source)
elif data_source.snowflake_options.table or data_source.snowflake_options.query:
from feast.infra.offline_stores.snowflake_source import SnowflakeSource
data_source_obj = SnowflakeSource.from_proto(data_source)
elif (
data_source.kafka_options.bootstrap_servers
and data_source.kafka_options.topic
and data_source.kafka_options.message_format
):
data_source_obj = KafkaSource.from_proto(data_source)
elif (
data_source.kinesis_options.record_format
and data_source.kinesis_options.region
and data_source.kinesis_options.stream_name
):
data_source_obj = KinesisSource.from_proto(data_source)
else:
raise ValueError("Could not identify the source type being added.")
return data_source_obj
@abstractmethod
def to_proto(self) -> DataSourceProto:
"""
Converts a DataSourceProto object to its protobuf representation.
"""
raise NotImplementedError
def validate(self, config: RepoConfig):
"""
Validates the underlying data source.
Args:
config: Configuration object used to configure a feature store.
"""
raise NotImplementedError
@staticmethod
@abstractmethod
def source_datatype_to_feast_value_type() -> Callable[[str], ValueType]:
"""
Returns the callable method that returns Feast type given the raw column type.
"""
raise NotImplementedError
def get_table_column_names_and_types(
self, config: RepoConfig
) -> Iterable[Tuple[str, str]]:
"""
Returns the list of column names and raw column types.
Args:
config: Configuration object used to configure a feature store.
"""
raise NotImplementedError
def get_table_query_string(self) -> str:
"""
Returns a string that can directly be used to reference this table in SQL.
"""
raise NotImplementedError
class KafkaSource(DataSource):
def validate(self, config: RepoConfig):
pass
def get_table_column_names_and_types(
self, config: RepoConfig
) -> Iterable[Tuple[str, str]]:
pass
def __init__(
self,
name: str,
event_timestamp_column: str,
bootstrap_servers: str,
message_format: StreamFormat,
topic: str,
created_timestamp_column: Optional[str] = "",
field_mapping: Optional[Dict[str, str]] = None,
date_partition_column: Optional[str] = "",
):
super().__init__(
name,
event_timestamp_column,
created_timestamp_column,
field_mapping,
date_partition_column,
)
self.kafka_options = KafkaOptions(
bootstrap_servers=bootstrap_servers,
message_format=message_format,
topic=topic,
)
def __eq__(self, other):
if not isinstance(other, KafkaSource):
raise TypeError(
"Comparisons should only involve KafkaSource class objects."
)
if (
self.kafka_options.bootstrap_servers
!= other.kafka_options.bootstrap_servers
or self.kafka_options.message_format != other.kafka_options.message_format
or self.kafka_options.topic != other.kafka_options.topic
):
return False
return True
@staticmethod
def from_proto(data_source: DataSourceProto):
return KafkaSource(
name=data_source.name,
field_mapping=dict(data_source.field_mapping),
bootstrap_servers=data_source.kafka_options.bootstrap_servers,
message_format=StreamFormat.from_proto(
data_source.kafka_options.message_format
),
topic=data_source.kafka_options.topic,
event_timestamp_column=data_source.event_timestamp_column,
created_timestamp_column=data_source.created_timestamp_column,
date_partition_column=data_source.date_partition_column,
)
def to_proto(self) -> DataSourceProto:
data_source_proto = DataSourceProto(
name=self.name,
type=DataSourceProto.STREAM_KAFKA,
field_mapping=self.field_mapping,
kafka_options=self.kafka_options.to_proto(),
)
data_source_proto.event_timestamp_column = self.event_timestamp_column
data_source_proto.created_timestamp_column = self.created_timestamp_column
data_source_proto.date_partition_column = self.date_partition_column
return data_source_proto
@staticmethod
def source_datatype_to_feast_value_type() -> Callable[[str], ValueType]:
return type_map.redshift_to_feast_value_type
def get_table_query_string(self) -> str:
raise NotImplementedError
class RequestDataSource(DataSource):
"""
RequestDataSource that can be used to provide input features for on demand transforms
Args:
name: Name of the request data source
schema: Schema mapping from the input feature name to a ValueType
"""
name: str
schema: Dict[str, ValueType]
def __init__(
self, name: str, schema: Dict[str, ValueType],
):
"""Creates a RequestDataSource object."""
super().__init__(name)
self.schema = schema
def validate(self, config: RepoConfig):
pass
def get_table_column_names_and_types(
self, config: RepoConfig
) -> Iterable[Tuple[str, str]]:
pass
@staticmethod
def from_proto(data_source: DataSourceProto):
schema_pb = data_source.request_data_options.schema
schema = {}
for key in schema_pb.keys():
schema[key] = ValueType(schema_pb.get(key))
return RequestDataSource(name=data_source.name, schema=schema)
def to_proto(self) -> DataSourceProto:
schema_pb = {}
for key, value in self.schema.items():
schema_pb[key] = value.value
options = DataSourceProto.RequestDataOptions(schema=schema_pb)
data_source_proto = DataSourceProto(
name=self.name,
type=DataSourceProto.REQUEST_SOURCE,
request_data_options=options,
)
return data_source_proto
def get_table_query_string(self) -> str:
raise NotImplementedError
@staticmethod
def source_datatype_to_feast_value_type() -> Callable[[str], ValueType]:
raise NotImplementedError
class KinesisSource(DataSource):
def validate(self, config: RepoConfig):
pass
def get_table_column_names_and_types(
self, config: RepoConfig
) -> Iterable[Tuple[str, str]]:
pass
@staticmethod
def from_proto(data_source: DataSourceProto):
return KinesisSource(
name=data_source.name,
field_mapping=dict(data_source.field_mapping),
record_format=StreamFormat.from_proto(
data_source.kinesis_options.record_format
),
region=data_source.kinesis_options.region,
stream_name=data_source.kinesis_options.stream_name,
event_timestamp_column=data_source.event_timestamp_column,
created_timestamp_column=data_source.created_timestamp_column,
date_partition_column=data_source.date_partition_column,
)
@staticmethod
def source_datatype_to_feast_value_type() -> Callable[[str], ValueType]:
pass
def get_table_query_string(self) -> str:
raise NotImplementedError
def __init__(
self,
name: str,
event_timestamp_column: str,
created_timestamp_column: str,
record_format: StreamFormat,
region: str,
stream_name: str,
field_mapping: Optional[Dict[str, str]] = None,
date_partition_column: Optional[str] = "",
):
super().__init__(
name,
event_timestamp_column,
created_timestamp_column,
field_mapping,
date_partition_column,
)
self.kinesis_options = KinesisOptions(
record_format=record_format, region=region, stream_name=stream_name
)
def __eq__(self, other):
if other is None:
return False
if not isinstance(other, KinesisSource):
raise TypeError(
"Comparisons should only involve KinesisSource class objects."
)
if (
self.name != other.name
or self.kinesis_options.record_format != other.kinesis_options.record_format
or self.kinesis_options.region != other.kinesis_options.region
or self.kinesis_options.stream_name != other.kinesis_options.stream_name
):
return False
return True
def to_proto(self) -> DataSourceProto:
data_source_proto = DataSourceProto(
name=self.name,
type=DataSourceProto.STREAM_KINESIS,
field_mapping=self.field_mapping,
kinesis_options=self.kinesis_options.to_proto(),
)
data_source_proto.event_timestamp_column = self.event_timestamp_column
data_source_proto.created_timestamp_column = self.created_timestamp_column
data_source_proto.date_partition_column = self.date_partition_column
return data_source_proto