airbyte-cdk 6.60.6__py3-none-any.whl → 6.60.7__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- airbyte_cdk/connector_builder/connector_builder_handler.py +32 -36
- airbyte_cdk/connector_builder/main.py +3 -3
- airbyte_cdk/connector_builder/test_reader/helpers.py +24 -2
- airbyte_cdk/sources/concurrent_source/concurrent_read_processor.py +14 -21
- airbyte_cdk/sources/concurrent_source/concurrent_source.py +29 -18
- airbyte_cdk/sources/declarative/concurrent_declarative_source.py +54 -3
- airbyte_cdk/sources/declarative/parsers/model_to_component_factory.py +4 -0
- airbyte_cdk/sources/declarative/requesters/http_requester.py +7 -1
- airbyte_cdk/sources/declarative/stream_slicers/declarative_partition_generator.py +20 -4
- airbyte_cdk/sources/declarative/stream_slicers/stream_slicer_test_read_decorator.py +2 -2
- airbyte_cdk/sources/message/concurrent_repository.py +43 -0
- airbyte_cdk/sources/streams/concurrent/partition_reader.py +44 -5
- airbyte_cdk/sources/streams/concurrent/partitions/types.py +7 -1
- airbyte_cdk/sources/utils/slice_logger.py +4 -0
- {airbyte_cdk-6.60.6.dist-info → airbyte_cdk-6.60.7.dist-info}/METADATA +1 -1
- {airbyte_cdk-6.60.6.dist-info → airbyte_cdk-6.60.7.dist-info}/RECORD +20 -19
- {airbyte_cdk-6.60.6.dist-info → airbyte_cdk-6.60.7.dist-info}/LICENSE.txt +0 -0
- {airbyte_cdk-6.60.6.dist-info → airbyte_cdk-6.60.7.dist-info}/LICENSE_SHORT +0 -0
- {airbyte_cdk-6.60.6.dist-info → airbyte_cdk-6.60.7.dist-info}/WHEEL +0 -0
- {airbyte_cdk-6.60.6.dist-info → airbyte_cdk-6.60.7.dist-info}/entry_points.txt +0 -0
@@ -3,8 +3,8 @@
|
|
3
3
|
#
|
4
4
|
|
5
5
|
|
6
|
-
from dataclasses import asdict
|
7
|
-
from typing import Any,
|
6
|
+
from dataclasses import asdict
|
7
|
+
from typing import Any, Dict, List, Mapping, Optional
|
8
8
|
|
9
9
|
from airbyte_cdk.connector_builder.test_reader import TestReader
|
10
10
|
from airbyte_cdk.models import (
|
@@ -15,45 +15,32 @@ from airbyte_cdk.models import (
|
|
15
15
|
Type,
|
16
16
|
)
|
17
17
|
from airbyte_cdk.models import Type as MessageType
|
18
|
+
from airbyte_cdk.sources.declarative.concurrent_declarative_source import (
|
19
|
+
ConcurrentDeclarativeSource,
|
20
|
+
TestLimits,
|
21
|
+
)
|
18
22
|
from airbyte_cdk.sources.declarative.declarative_source import DeclarativeSource
|
19
23
|
from airbyte_cdk.sources.declarative.manifest_declarative_source import ManifestDeclarativeSource
|
20
|
-
from airbyte_cdk.sources.declarative.parsers.model_to_component_factory import (
|
21
|
-
ModelToComponentFactory,
|
22
|
-
)
|
23
24
|
from airbyte_cdk.utils.airbyte_secrets_utils import filter_secrets
|
24
25
|
from airbyte_cdk.utils.datetime_helpers import ab_datetime_now
|
25
26
|
from airbyte_cdk.utils.traced_exception import AirbyteTracedException
|
26
27
|
|
27
|
-
DEFAULT_MAXIMUM_NUMBER_OF_PAGES_PER_SLICE = 5
|
28
|
-
DEFAULT_MAXIMUM_NUMBER_OF_SLICES = 5
|
29
|
-
DEFAULT_MAXIMUM_RECORDS = 100
|
30
|
-
DEFAULT_MAXIMUM_STREAMS = 100
|
31
|
-
|
32
28
|
MAX_PAGES_PER_SLICE_KEY = "max_pages_per_slice"
|
33
29
|
MAX_SLICES_KEY = "max_slices"
|
34
30
|
MAX_RECORDS_KEY = "max_records"
|
35
31
|
MAX_STREAMS_KEY = "max_streams"
|
36
32
|
|
37
33
|
|
38
|
-
@dataclass
|
39
|
-
class TestLimits:
|
40
|
-
__test__: ClassVar[bool] = False # Tell Pytest this is not a Pytest class, despite its name
|
41
|
-
|
42
|
-
max_records: int = field(default=DEFAULT_MAXIMUM_RECORDS)
|
43
|
-
max_pages_per_slice: int = field(default=DEFAULT_MAXIMUM_NUMBER_OF_PAGES_PER_SLICE)
|
44
|
-
max_slices: int = field(default=DEFAULT_MAXIMUM_NUMBER_OF_SLICES)
|
45
|
-
max_streams: int = field(default=DEFAULT_MAXIMUM_STREAMS)
|
46
|
-
|
47
|
-
|
48
34
|
def get_limits(config: Mapping[str, Any]) -> TestLimits:
|
49
35
|
command_config = config.get("__test_read_config", {})
|
50
|
-
|
51
|
-
command_config.get(
|
36
|
+
return TestLimits(
|
37
|
+
max_records=command_config.get(MAX_RECORDS_KEY, TestLimits.DEFAULT_MAX_RECORDS),
|
38
|
+
max_pages_per_slice=command_config.get(
|
39
|
+
MAX_PAGES_PER_SLICE_KEY, TestLimits.DEFAULT_MAX_PAGES_PER_SLICE
|
40
|
+
),
|
41
|
+
max_slices=command_config.get(MAX_SLICES_KEY, TestLimits.DEFAULT_MAX_SLICES),
|
42
|
+
max_streams=command_config.get(MAX_STREAMS_KEY, TestLimits.DEFAULT_MAX_STREAMS),
|
52
43
|
)
|
53
|
-
max_slices = command_config.get(MAX_SLICES_KEY) or DEFAULT_MAXIMUM_NUMBER_OF_SLICES
|
54
|
-
max_records = command_config.get(MAX_RECORDS_KEY) or DEFAULT_MAXIMUM_RECORDS
|
55
|
-
max_streams = command_config.get(MAX_STREAMS_KEY) or DEFAULT_MAXIMUM_STREAMS
|
56
|
-
return TestLimits(max_records, max_pages_per_slice, max_slices, max_streams)
|
57
44
|
|
58
45
|
|
59
46
|
def should_migrate_manifest(config: Mapping[str, Any]) -> bool:
|
@@ -75,21 +62,30 @@ def should_normalize_manifest(config: Mapping[str, Any]) -> bool:
|
|
75
62
|
return config.get("__should_normalize", False)
|
76
63
|
|
77
64
|
|
78
|
-
def create_source(
|
65
|
+
def create_source(
|
66
|
+
config: Mapping[str, Any],
|
67
|
+
limits: TestLimits,
|
68
|
+
catalog: Optional[ConfiguredAirbyteCatalog],
|
69
|
+
state: Optional[List[AirbyteStateMessage]],
|
70
|
+
) -> ConcurrentDeclarativeSource[Optional[List[AirbyteStateMessage]]]:
|
79
71
|
manifest = config["__injected_declarative_manifest"]
|
80
|
-
|
72
|
+
|
73
|
+
# We enforce a concurrency level of 1 so that the stream is processed on a single thread
|
74
|
+
# to retain ordering for the grouping of the builder message responses.
|
75
|
+
if "concurrency_level" in manifest:
|
76
|
+
manifest["concurrency_level"]["default_concurrency"] = 1
|
77
|
+
else:
|
78
|
+
manifest["concurrency_level"] = {"type": "ConcurrencyLevel", "default_concurrency": 1}
|
79
|
+
|
80
|
+
return ConcurrentDeclarativeSource(
|
81
|
+
catalog=catalog,
|
81
82
|
config=config,
|
82
|
-
|
83
|
+
state=state,
|
83
84
|
source_config=manifest,
|
85
|
+
emit_connector_builder_messages=True,
|
84
86
|
migrate_manifest=should_migrate_manifest(config),
|
85
87
|
normalize_manifest=should_normalize_manifest(config),
|
86
|
-
|
87
|
-
emit_connector_builder_messages=True,
|
88
|
-
limit_pages_fetched_per_slice=limits.max_pages_per_slice,
|
89
|
-
limit_slices_fetched=limits.max_slices,
|
90
|
-
disable_retries=True,
|
91
|
-
disable_cache=True,
|
92
|
-
),
|
88
|
+
limits=limits,
|
93
89
|
)
|
94
90
|
|
95
91
|
|
@@ -91,12 +91,12 @@ def handle_connector_builder_request(
|
|
91
91
|
def handle_request(args: List[str]) -> str:
|
92
92
|
command, config, catalog, state = get_config_and_catalog_from_args(args)
|
93
93
|
limits = get_limits(config)
|
94
|
-
source = create_source(config, limits)
|
95
|
-
return orjson.dumps(
|
94
|
+
source = create_source(config=config, limits=limits, catalog=catalog, state=state)
|
95
|
+
return orjson.dumps( # type: ignore[no-any-return] # Serializer.dump() always returns AirbyteMessage
|
96
96
|
AirbyteMessageSerializer.dump(
|
97
97
|
handle_connector_builder_request(source, command, config, catalog, state, limits)
|
98
98
|
)
|
99
|
-
).decode()
|
99
|
+
).decode()
|
100
100
|
|
101
101
|
|
102
102
|
if __name__ == "__main__":
|
@@ -5,7 +5,7 @@
|
|
5
5
|
import json
|
6
6
|
from copy import deepcopy
|
7
7
|
from json import JSONDecodeError
|
8
|
-
from typing import Any, Dict, List, Mapping, Optional
|
8
|
+
from typing import Any, Dict, List, Mapping, Optional, Union
|
9
9
|
|
10
10
|
from airbyte_cdk.connector_builder.models import (
|
11
11
|
AuxiliaryRequest,
|
@@ -17,6 +17,8 @@ from airbyte_cdk.connector_builder.models import (
|
|
17
17
|
from airbyte_cdk.models import (
|
18
18
|
AirbyteLogMessage,
|
19
19
|
AirbyteMessage,
|
20
|
+
AirbyteStateBlob,
|
21
|
+
AirbyteStateMessage,
|
20
22
|
OrchestratorType,
|
21
23
|
TraceType,
|
22
24
|
)
|
@@ -466,7 +468,7 @@ def handle_current_slice(
|
|
466
468
|
return StreamReadSlices(
|
467
469
|
pages=current_slice_pages,
|
468
470
|
slice_descriptor=current_slice_descriptor,
|
469
|
-
state=[latest_state_message] if latest_state_message else [],
|
471
|
+
state=[convert_state_blob_to_mapping(latest_state_message)] if latest_state_message else [],
|
470
472
|
auxiliary_requests=auxiliary_requests if auxiliary_requests else [],
|
471
473
|
)
|
472
474
|
|
@@ -718,3 +720,23 @@ def get_auxiliary_request_type(stream: dict, http: dict) -> str: # type: ignore
|
|
718
720
|
Determines the type of the auxiliary request based on the stream and HTTP properties.
|
719
721
|
"""
|
720
722
|
return "PARENT_STREAM" if stream.get("is_substream", False) else str(http.get("type", None))
|
723
|
+
|
724
|
+
|
725
|
+
def convert_state_blob_to_mapping(
|
726
|
+
state_message: Union[AirbyteStateMessage, Dict[str, Any]],
|
727
|
+
) -> Dict[str, Any]:
|
728
|
+
"""
|
729
|
+
The AirbyteStreamState stores state as an AirbyteStateBlob which deceivingly is not
|
730
|
+
a dictionary, but rather a list of kwargs fields. This in turn causes it to not be
|
731
|
+
properly turned into a dictionary when translating this back into response output
|
732
|
+
by the connector_builder_handler using asdict()
|
733
|
+
"""
|
734
|
+
|
735
|
+
if isinstance(state_message, AirbyteStateMessage) and state_message.stream:
|
736
|
+
state_value = state_message.stream.stream_state
|
737
|
+
if isinstance(state_value, AirbyteStateBlob):
|
738
|
+
state_value_mapping = {k: v for k, v in state_value.__dict__.items()}
|
739
|
+
state_message.stream.stream_state = state_value_mapping # type: ignore # we intentionally set this as a Dict so that StreamReadSlices is translated properly in the resulting HTTP response
|
740
|
+
return state_message # type: ignore # See above, but when this is an AirbyteStateMessage we must convert AirbyteStateBlob to a Dict
|
741
|
+
else:
|
742
|
+
return state_message # type: ignore # This is guaranteed to be a Dict since we check isinstance AirbyteStateMessage above
|
@@ -95,11 +95,14 @@ class ConcurrentReadProcessor:
|
|
95
95
|
"""
|
96
96
|
stream_name = partition.stream_name()
|
97
97
|
self._streams_to_running_partitions[stream_name].add(partition)
|
98
|
+
cursor = self._stream_name_to_instance[stream_name].cursor
|
98
99
|
if self._slice_logger.should_log_slice_message(self._logger):
|
99
100
|
self._message_repository.emit_message(
|
100
101
|
self._slice_logger.create_slice_log_message(partition.to_slice())
|
101
102
|
)
|
102
|
-
self._thread_pool_manager.submit(
|
103
|
+
self._thread_pool_manager.submit(
|
104
|
+
self._partition_reader.process_partition, partition, cursor
|
105
|
+
)
|
103
106
|
|
104
107
|
def on_partition_complete_sentinel(
|
105
108
|
self, sentinel: PartitionCompleteSentinel
|
@@ -112,26 +115,16 @@ class ConcurrentReadProcessor:
|
|
112
115
|
"""
|
113
116
|
partition = sentinel.partition
|
114
117
|
|
115
|
-
|
116
|
-
|
117
|
-
|
118
|
-
|
119
|
-
|
120
|
-
|
121
|
-
|
122
|
-
|
123
|
-
|
124
|
-
|
125
|
-
partitions_running = self._streams_to_running_partitions[partition.stream_name()]
|
126
|
-
if partition in partitions_running:
|
127
|
-
partitions_running.remove(partition)
|
128
|
-
# If all partitions were generated and this was the last one, the stream is done
|
129
|
-
if (
|
130
|
-
partition.stream_name() not in self._streams_currently_generating_partitions
|
131
|
-
and len(partitions_running) == 0
|
132
|
-
):
|
133
|
-
yield from self._on_stream_is_done(partition.stream_name())
|
134
|
-
yield from self._message_repository.consume_queue()
|
118
|
+
partitions_running = self._streams_to_running_partitions[partition.stream_name()]
|
119
|
+
if partition in partitions_running:
|
120
|
+
partitions_running.remove(partition)
|
121
|
+
# If all partitions were generated and this was the last one, the stream is done
|
122
|
+
if (
|
123
|
+
partition.stream_name() not in self._streams_currently_generating_partitions
|
124
|
+
and len(partitions_running) == 0
|
125
|
+
):
|
126
|
+
yield from self._on_stream_is_done(partition.stream_name())
|
127
|
+
yield from self._message_repository.consume_queue()
|
135
128
|
|
136
129
|
def on_record(self, record: Record) -> Iterable[AirbyteMessage]:
|
137
130
|
"""
|
@@ -4,7 +4,7 @@
|
|
4
4
|
import concurrent
|
5
5
|
import logging
|
6
6
|
from queue import Queue
|
7
|
-
from typing import Iterable, Iterator, List
|
7
|
+
from typing import Iterable, Iterator, List, Optional
|
8
8
|
|
9
9
|
from airbyte_cdk.models import AirbyteMessage
|
10
10
|
from airbyte_cdk.sources.concurrent_source.concurrent_read_processor import ConcurrentReadProcessor
|
@@ -16,7 +16,7 @@ from airbyte_cdk.sources.concurrent_source.thread_pool_manager import ThreadPool
|
|
16
16
|
from airbyte_cdk.sources.message import InMemoryMessageRepository, MessageRepository
|
17
17
|
from airbyte_cdk.sources.streams.concurrent.abstract_stream import AbstractStream
|
18
18
|
from airbyte_cdk.sources.streams.concurrent.partition_enqueuer import PartitionEnqueuer
|
19
|
-
from airbyte_cdk.sources.streams.concurrent.partition_reader import PartitionReader
|
19
|
+
from airbyte_cdk.sources.streams.concurrent.partition_reader import PartitionLogger, PartitionReader
|
20
20
|
from airbyte_cdk.sources.streams.concurrent.partitions.partition import Partition
|
21
21
|
from airbyte_cdk.sources.streams.concurrent.partitions.types import (
|
22
22
|
PartitionCompleteSentinel,
|
@@ -43,6 +43,7 @@ class ConcurrentSource:
|
|
43
43
|
logger: logging.Logger,
|
44
44
|
slice_logger: SliceLogger,
|
45
45
|
message_repository: MessageRepository,
|
46
|
+
queue: Optional[Queue[QueueItem]] = None,
|
46
47
|
timeout_seconds: int = DEFAULT_TIMEOUT_SECONDS,
|
47
48
|
) -> "ConcurrentSource":
|
48
49
|
is_single_threaded = initial_number_of_partitions_to_generate == 1 and num_workers == 1
|
@@ -59,12 +60,13 @@ class ConcurrentSource:
|
|
59
60
|
logger,
|
60
61
|
)
|
61
62
|
return ConcurrentSource(
|
62
|
-
threadpool,
|
63
|
-
logger,
|
64
|
-
slice_logger,
|
65
|
-
|
66
|
-
|
67
|
-
|
63
|
+
threadpool=threadpool,
|
64
|
+
logger=logger,
|
65
|
+
slice_logger=slice_logger,
|
66
|
+
queue=queue,
|
67
|
+
message_repository=message_repository,
|
68
|
+
initial_number_partitions_to_generate=initial_number_of_partitions_to_generate,
|
69
|
+
timeout_seconds=timeout_seconds,
|
68
70
|
)
|
69
71
|
|
70
72
|
def __init__(
|
@@ -72,6 +74,7 @@ class ConcurrentSource:
|
|
72
74
|
threadpool: ThreadPoolManager,
|
73
75
|
logger: logging.Logger,
|
74
76
|
slice_logger: SliceLogger = DebugSliceLogger(),
|
77
|
+
queue: Optional[Queue[QueueItem]] = None,
|
75
78
|
message_repository: MessageRepository = InMemoryMessageRepository(),
|
76
79
|
initial_number_partitions_to_generate: int = 1,
|
77
80
|
timeout_seconds: int = DEFAULT_TIMEOUT_SECONDS,
|
@@ -91,25 +94,28 @@ class ConcurrentSource:
|
|
91
94
|
self._initial_number_partitions_to_generate = initial_number_partitions_to_generate
|
92
95
|
self._timeout_seconds = timeout_seconds
|
93
96
|
|
97
|
+
# We set a maxsize to for the main thread to process record items when the queue size grows. This assumes that there are less
|
98
|
+
# threads generating partitions that than are max number of workers. If it weren't the case, we could have threads only generating
|
99
|
+
# partitions which would fill the queue. This number is arbitrarily set to 10_000 but will probably need to be changed given more
|
100
|
+
# information and might even need to be configurable depending on the source
|
101
|
+
self._queue = queue or Queue(maxsize=10_000)
|
102
|
+
|
94
103
|
def read(
|
95
104
|
self,
|
96
105
|
streams: List[AbstractStream],
|
97
106
|
) -> Iterator[AirbyteMessage]:
|
98
107
|
self._logger.info("Starting syncing")
|
99
|
-
|
100
|
-
# We set a maxsize to for the main thread to process record items when the queue size grows. This assumes that there are less
|
101
|
-
# threads generating partitions that than are max number of workers. If it weren't the case, we could have threads only generating
|
102
|
-
# partitions which would fill the queue. This number is arbitrarily set to 10_000 but will probably need to be changed given more
|
103
|
-
# information and might even need to be configurable depending on the source
|
104
|
-
queue: Queue[QueueItem] = Queue(maxsize=10_000)
|
105
108
|
concurrent_stream_processor = ConcurrentReadProcessor(
|
106
109
|
streams,
|
107
|
-
PartitionEnqueuer(
|
110
|
+
PartitionEnqueuer(self._queue, self._threadpool),
|
108
111
|
self._threadpool,
|
109
112
|
self._logger,
|
110
113
|
self._slice_logger,
|
111
114
|
self._message_repository,
|
112
|
-
PartitionReader(
|
115
|
+
PartitionReader(
|
116
|
+
self._queue,
|
117
|
+
PartitionLogger(self._slice_logger, self._logger, self._message_repository),
|
118
|
+
),
|
113
119
|
)
|
114
120
|
|
115
121
|
# Enqueue initial partition generation tasks
|
@@ -117,7 +123,7 @@ class ConcurrentSource:
|
|
117
123
|
|
118
124
|
# Read from the queue until all partitions were generated and read
|
119
125
|
yield from self._consume_from_queue(
|
120
|
-
|
126
|
+
self._queue,
|
121
127
|
concurrent_stream_processor,
|
122
128
|
)
|
123
129
|
self._threadpool.check_for_errors_and_shutdown()
|
@@ -141,7 +147,10 @@ class ConcurrentSource:
|
|
141
147
|
airbyte_message_or_record_or_exception,
|
142
148
|
concurrent_stream_processor,
|
143
149
|
)
|
144
|
-
|
150
|
+
# In the event that a partition raises an exception, anything remaining in
|
151
|
+
# the queue will be missed because is_done() can raise an exception and exit
|
152
|
+
# out of this loop before remaining items are consumed
|
153
|
+
if queue.empty() and concurrent_stream_processor.is_done():
|
145
154
|
# all partitions were generated and processed. we're done here
|
146
155
|
break
|
147
156
|
|
@@ -161,5 +170,7 @@ class ConcurrentSource:
|
|
161
170
|
yield from concurrent_stream_processor.on_partition_complete_sentinel(queue_item)
|
162
171
|
elif isinstance(queue_item, Record):
|
163
172
|
yield from concurrent_stream_processor.on_record(queue_item)
|
173
|
+
elif isinstance(queue_item, AirbyteMessage):
|
174
|
+
yield queue_item
|
164
175
|
else:
|
165
176
|
raise ValueError(f"Unknown queue item type: {type(queue_item)}")
|
@@ -3,7 +3,11 @@
|
|
3
3
|
#
|
4
4
|
|
5
5
|
import logging
|
6
|
-
from
|
6
|
+
from dataclasses import dataclass, field
|
7
|
+
from queue import Queue
|
8
|
+
from typing import Any, ClassVar, Generic, Iterator, List, Mapping, MutableMapping, Optional, Tuple
|
9
|
+
|
10
|
+
from airbyte_protocol_dataclasses.models import Level
|
7
11
|
|
8
12
|
from airbyte_cdk.models import (
|
9
13
|
AirbyteCatalog,
|
@@ -48,6 +52,8 @@ from airbyte_cdk.sources.declarative.stream_slicers.declarative_partition_genera
|
|
48
52
|
StreamSlicerPartitionGenerator,
|
49
53
|
)
|
50
54
|
from airbyte_cdk.sources.declarative.types import ConnectionDefinition
|
55
|
+
from airbyte_cdk.sources.message.concurrent_repository import ConcurrentMessageRepository
|
56
|
+
from airbyte_cdk.sources.message.repository import InMemoryMessageRepository, MessageRepository
|
51
57
|
from airbyte_cdk.sources.source import TState
|
52
58
|
from airbyte_cdk.sources.streams import Stream
|
53
59
|
from airbyte_cdk.sources.streams.concurrent.abstract_stream import AbstractStream
|
@@ -58,6 +64,22 @@ from airbyte_cdk.sources.streams.concurrent.availability_strategy import (
|
|
58
64
|
from airbyte_cdk.sources.streams.concurrent.cursor import ConcurrentCursor, FinalStateCursor
|
59
65
|
from airbyte_cdk.sources.streams.concurrent.default_stream import DefaultStream
|
60
66
|
from airbyte_cdk.sources.streams.concurrent.helpers import get_primary_key_from_stream
|
67
|
+
from airbyte_cdk.sources.streams.concurrent.partitions.types import QueueItem
|
68
|
+
|
69
|
+
|
70
|
+
@dataclass
|
71
|
+
class TestLimits:
|
72
|
+
__test__: ClassVar[bool] = False # Tell Pytest this is not a Pytest class, despite its name
|
73
|
+
|
74
|
+
DEFAULT_MAX_PAGES_PER_SLICE: ClassVar[int] = 5
|
75
|
+
DEFAULT_MAX_SLICES: ClassVar[int] = 5
|
76
|
+
DEFAULT_MAX_RECORDS: ClassVar[int] = 100
|
77
|
+
DEFAULT_MAX_STREAMS: ClassVar[int] = 100
|
78
|
+
|
79
|
+
max_records: int = field(default=DEFAULT_MAX_RECORDS)
|
80
|
+
max_pages_per_slice: int = field(default=DEFAULT_MAX_PAGES_PER_SLICE)
|
81
|
+
max_slices: int = field(default=DEFAULT_MAX_SLICES)
|
82
|
+
max_streams: int = field(default=DEFAULT_MAX_STREAMS)
|
61
83
|
|
62
84
|
|
63
85
|
class ConcurrentDeclarativeSource(ManifestDeclarativeSource, Generic[TState]):
|
@@ -73,7 +95,9 @@ class ConcurrentDeclarativeSource(ManifestDeclarativeSource, Generic[TState]):
|
|
73
95
|
source_config: ConnectionDefinition,
|
74
96
|
debug: bool = False,
|
75
97
|
emit_connector_builder_messages: bool = False,
|
76
|
-
|
98
|
+
migrate_manifest: bool = False,
|
99
|
+
normalize_manifest: bool = False,
|
100
|
+
limits: Optional[TestLimits] = None,
|
77
101
|
config_path: Optional[str] = None,
|
78
102
|
**kwargs: Any,
|
79
103
|
) -> None:
|
@@ -81,22 +105,40 @@ class ConcurrentDeclarativeSource(ManifestDeclarativeSource, Generic[TState]):
|
|
81
105
|
# no longer needs to store the original incoming state. But maybe there's an edge case?
|
82
106
|
self._connector_state_manager = ConnectorStateManager(state=state) # type: ignore # state is always in the form of List[AirbyteStateMessage]. The ConnectorStateManager should use generics, but this can be done later
|
83
107
|
|
108
|
+
# We set a maxsize to for the main thread to process record items when the queue size grows. This assumes that there are less
|
109
|
+
# threads generating partitions that than are max number of workers. If it weren't the case, we could have threads only generating
|
110
|
+
# partitions which would fill the queue. This number is arbitrarily set to 10_000 but will probably need to be changed given more
|
111
|
+
# information and might even need to be configurable depending on the source
|
112
|
+
queue: Queue[QueueItem] = Queue(maxsize=10_000)
|
113
|
+
message_repository = InMemoryMessageRepository(
|
114
|
+
Level.DEBUG if emit_connector_builder_messages else Level.INFO
|
115
|
+
)
|
116
|
+
|
84
117
|
# To reduce the complexity of the concurrent framework, we are not enabling RFR with synthetic
|
85
118
|
# cursors. We do this by no longer automatically instantiating RFR cursors when converting
|
86
119
|
# the declarative models into runtime components. Concurrent sources will continue to checkpoint
|
87
120
|
# incremental streams running in full refresh.
|
88
|
-
component_factory =
|
121
|
+
component_factory = ModelToComponentFactory(
|
89
122
|
emit_connector_builder_messages=emit_connector_builder_messages,
|
90
123
|
disable_resumable_full_refresh=True,
|
124
|
+
message_repository=ConcurrentMessageRepository(queue, message_repository),
|
91
125
|
connector_state_manager=self._connector_state_manager,
|
92
126
|
max_concurrent_async_job_count=source_config.get("max_concurrent_async_job_count"),
|
127
|
+
limit_pages_fetched_per_slice=limits.max_pages_per_slice if limits else None,
|
128
|
+
limit_slices_fetched=limits.max_slices if limits else None,
|
129
|
+
disable_retries=True if limits else False,
|
130
|
+
disable_cache=True if limits else False,
|
93
131
|
)
|
94
132
|
|
133
|
+
self._limits = limits
|
134
|
+
|
95
135
|
super().__init__(
|
96
136
|
source_config=source_config,
|
97
137
|
config=config,
|
98
138
|
debug=debug,
|
99
139
|
emit_connector_builder_messages=emit_connector_builder_messages,
|
140
|
+
migrate_manifest=migrate_manifest,
|
141
|
+
normalize_manifest=normalize_manifest,
|
100
142
|
component_factory=component_factory,
|
101
143
|
config_path=config_path,
|
102
144
|
)
|
@@ -126,6 +168,7 @@ class ConcurrentDeclarativeSource(ManifestDeclarativeSource, Generic[TState]):
|
|
126
168
|
initial_number_of_partitions_to_generate=initial_number_of_partitions_to_generate,
|
127
169
|
logger=self.logger,
|
128
170
|
slice_logger=self._slice_logger,
|
171
|
+
queue=queue,
|
129
172
|
message_repository=self.message_repository,
|
130
173
|
)
|
131
174
|
|
@@ -287,6 +330,9 @@ class ConcurrentDeclarativeSource(ManifestDeclarativeSource, Generic[TState]):
|
|
287
330
|
self.message_repository,
|
288
331
|
),
|
289
332
|
stream_slicer=declarative_stream.retriever.stream_slicer,
|
333
|
+
slice_limit=self._limits.max_slices
|
334
|
+
if self._limits
|
335
|
+
else None, # technically not needed because create_declarative_stream() -> create_simple_retriever() will apply the decorator. But for consistency and depending how we build create_default_stream, this may be needed later
|
290
336
|
)
|
291
337
|
else:
|
292
338
|
if (
|
@@ -318,6 +364,7 @@ class ConcurrentDeclarativeSource(ManifestDeclarativeSource, Generic[TState]):
|
|
318
364
|
self.message_repository,
|
319
365
|
),
|
320
366
|
stream_slicer=cursor,
|
367
|
+
slice_limit=self._limits.max_slices if self._limits else None,
|
321
368
|
)
|
322
369
|
|
323
370
|
concurrent_streams.append(
|
@@ -349,6 +396,9 @@ class ConcurrentDeclarativeSource(ManifestDeclarativeSource, Generic[TState]):
|
|
349
396
|
self.message_repository,
|
350
397
|
),
|
351
398
|
declarative_stream.retriever.stream_slicer,
|
399
|
+
slice_limit=self._limits.max_slices
|
400
|
+
if self._limits
|
401
|
+
else None, # technically not needed because create_declarative_stream() -> create_simple_retriever() will apply the decorator. But for consistency and depending how we build create_default_stream, this may be needed later
|
352
402
|
)
|
353
403
|
|
354
404
|
final_state_cursor = FinalStateCursor(
|
@@ -410,6 +460,7 @@ class ConcurrentDeclarativeSource(ManifestDeclarativeSource, Generic[TState]):
|
|
410
460
|
self.message_repository,
|
411
461
|
),
|
412
462
|
perpartition_cursor,
|
463
|
+
slice_limit=self._limits.max_slices if self._limits else None,
|
413
464
|
)
|
414
465
|
|
415
466
|
concurrent_streams.append(
|
@@ -622,6 +622,10 @@ SCHEMA_TRANSFORMER_TYPE_MAPPING = {
|
|
622
622
|
SchemaNormalizationModel.Default: TransformConfig.DefaultSchemaNormalization,
|
623
623
|
}
|
624
624
|
|
625
|
+
# Ideally this should use the value defined in ConcurrentDeclarativeSource, but
|
626
|
+
# this would be a circular import
|
627
|
+
MAX_SLICES = 5
|
628
|
+
|
625
629
|
|
626
630
|
class ModelToComponentFactory:
|
627
631
|
EPOCH_DATETIME_FORMAT = "%s"
|
@@ -168,7 +168,13 @@ class HttpRequester(Requester):
|
|
168
168
|
next_page_token=next_page_token,
|
169
169
|
)
|
170
170
|
|
171
|
-
full_url =
|
171
|
+
full_url = (
|
172
|
+
self._join_url(url_base, path)
|
173
|
+
if url_base
|
174
|
+
else self._join_url(url, path)
|
175
|
+
if path
|
176
|
+
else url
|
177
|
+
)
|
172
178
|
|
173
179
|
return full_url
|
174
180
|
|
@@ -1,8 +1,11 @@
|
|
1
|
-
# Copyright (c)
|
1
|
+
# Copyright (c) 2025 Airbyte, Inc., all rights reserved.
|
2
2
|
|
3
|
-
from typing import Any, Iterable, Mapping, Optional
|
3
|
+
from typing import Any, Iterable, Mapping, Optional, cast
|
4
4
|
|
5
5
|
from airbyte_cdk.sources.declarative.retrievers import Retriever
|
6
|
+
from airbyte_cdk.sources.declarative.stream_slicers.stream_slicer_test_read_decorator import (
|
7
|
+
StreamSlicerTestReadDecorator,
|
8
|
+
)
|
6
9
|
from airbyte_cdk.sources.message import MessageRepository
|
7
10
|
from airbyte_cdk.sources.streams.concurrent.partitions.partition import Partition
|
8
11
|
from airbyte_cdk.sources.streams.concurrent.partitions.partition_generator import PartitionGenerator
|
@@ -83,10 +86,23 @@ class DeclarativePartition(Partition):
|
|
83
86
|
|
84
87
|
class StreamSlicerPartitionGenerator(PartitionGenerator):
|
85
88
|
def __init__(
|
86
|
-
self,
|
89
|
+
self,
|
90
|
+
partition_factory: DeclarativePartitionFactory,
|
91
|
+
stream_slicer: StreamSlicer,
|
92
|
+
slice_limit: Optional[int] = None,
|
87
93
|
) -> None:
|
88
94
|
self._partition_factory = partition_factory
|
89
|
-
|
95
|
+
|
96
|
+
if slice_limit:
|
97
|
+
self._stream_slicer = cast(
|
98
|
+
StreamSlicer,
|
99
|
+
StreamSlicerTestReadDecorator(
|
100
|
+
wrapped_slicer=stream_slicer,
|
101
|
+
maximum_number_of_slices=slice_limit,
|
102
|
+
),
|
103
|
+
)
|
104
|
+
else:
|
105
|
+
self._stream_slicer = stream_slicer
|
90
106
|
|
91
107
|
def generate(self) -> Iterable[Partition]:
|
92
108
|
for stream_slice in self._stream_slicer.stream_slices():
|
@@ -4,10 +4,10 @@
|
|
4
4
|
|
5
5
|
from dataclasses import dataclass
|
6
6
|
from itertools import islice
|
7
|
-
from typing import Any, Iterable
|
7
|
+
from typing import Any, Iterable
|
8
8
|
|
9
9
|
from airbyte_cdk.sources.streams.concurrent.partitions.stream_slicer import StreamSlicer
|
10
|
-
from airbyte_cdk.sources.types import StreamSlice
|
10
|
+
from airbyte_cdk.sources.types import StreamSlice
|
11
11
|
|
12
12
|
|
13
13
|
@dataclass
|
@@ -0,0 +1,43 @@
|
|
1
|
+
# Copyright (c) 2025 Airbyte, Inc., all rights reserved.
|
2
|
+
|
3
|
+
from queue import Queue
|
4
|
+
from typing import Callable, Iterable
|
5
|
+
|
6
|
+
from airbyte_cdk.models import AirbyteMessage, Level
|
7
|
+
from airbyte_cdk.sources.message.repository import LogMessage, MessageRepository
|
8
|
+
from airbyte_cdk.sources.streams.concurrent.partitions.types import QueueItem
|
9
|
+
|
10
|
+
|
11
|
+
class ConcurrentMessageRepository(MessageRepository):
|
12
|
+
"""
|
13
|
+
Message repository that immediately loads messages onto the queue processed on the
|
14
|
+
main thread. This ensures that messages are processed in the correct order they are
|
15
|
+
received. The InMemoryMessageRepository implementation does not have guaranteed
|
16
|
+
ordering since whether to process the main thread vs. partitions is non-deterministic
|
17
|
+
and there can be a lag between reading the main-thread and consuming messages on the
|
18
|
+
MessageRepository.
|
19
|
+
|
20
|
+
This is particularly important for the connector builder which relies on grouping
|
21
|
+
of messages to organize request/response, pages, and partitions.
|
22
|
+
"""
|
23
|
+
|
24
|
+
def __init__(self, queue: Queue[QueueItem], message_repository: MessageRepository):
|
25
|
+
self._queue = queue
|
26
|
+
self._decorated_message_repository = message_repository
|
27
|
+
|
28
|
+
def emit_message(self, message: AirbyteMessage) -> None:
|
29
|
+
self._decorated_message_repository.emit_message(message)
|
30
|
+
for message in self._decorated_message_repository.consume_queue():
|
31
|
+
self._queue.put(message)
|
32
|
+
|
33
|
+
def log_message(self, level: Level, message_provider: Callable[[], LogMessage]) -> None:
|
34
|
+
self._decorated_message_repository.log_message(level, message_provider)
|
35
|
+
for message in self._decorated_message_repository.consume_queue():
|
36
|
+
self._queue.put(message)
|
37
|
+
|
38
|
+
def consume_queue(self) -> Iterable[AirbyteMessage]:
|
39
|
+
"""
|
40
|
+
This method shouldn't need to be called because as part of emit_message() we are already
|
41
|
+
loading messages onto the queue processed on the main thread.
|
42
|
+
"""
|
43
|
+
yield from []
|
@@ -1,14 +1,45 @@
|
|
1
|
-
#
|
2
|
-
|
3
|
-
|
1
|
+
# Copyright (c) 2025 Airbyte, Inc., all rights reserved.
|
2
|
+
|
3
|
+
import logging
|
4
4
|
from queue import Queue
|
5
|
+
from typing import Optional
|
5
6
|
|
6
7
|
from airbyte_cdk.sources.concurrent_source.stream_thread_exception import StreamThreadException
|
8
|
+
from airbyte_cdk.sources.message.repository import MessageRepository
|
9
|
+
from airbyte_cdk.sources.streams.concurrent.cursor import Cursor
|
7
10
|
from airbyte_cdk.sources.streams.concurrent.partitions.partition import Partition
|
8
11
|
from airbyte_cdk.sources.streams.concurrent.partitions.types import (
|
9
12
|
PartitionCompleteSentinel,
|
10
13
|
QueueItem,
|
11
14
|
)
|
15
|
+
from airbyte_cdk.sources.utils.slice_logger import SliceLogger
|
16
|
+
|
17
|
+
|
18
|
+
# Since moving all the connector builder workflow to the concurrent CDK which required correct ordering
|
19
|
+
# of grouping log messages onto the main write thread using the ConcurrentMessageRepository, this
|
20
|
+
# separate flow and class that was used to log slices onto this partition's message_repository
|
21
|
+
# should just be replaced by emitting messages directly onto the repository instead of an intermediary.
|
22
|
+
class PartitionLogger:
|
23
|
+
"""
|
24
|
+
Helper class that provides a mechanism for passing a log message onto the current
|
25
|
+
partitions message repository
|
26
|
+
"""
|
27
|
+
|
28
|
+
def __init__(
|
29
|
+
self,
|
30
|
+
slice_logger: SliceLogger,
|
31
|
+
logger: logging.Logger,
|
32
|
+
message_repository: MessageRepository,
|
33
|
+
):
|
34
|
+
self._slice_logger = slice_logger
|
35
|
+
self._logger = logger
|
36
|
+
self._message_repository = message_repository
|
37
|
+
|
38
|
+
def log(self, partition: Partition) -> None:
|
39
|
+
if self._slice_logger.should_log_slice_message(self._logger):
|
40
|
+
self._message_repository.emit_message(
|
41
|
+
self._slice_logger.create_slice_log_message(partition.to_slice())
|
42
|
+
)
|
12
43
|
|
13
44
|
|
14
45
|
class PartitionReader:
|
@@ -18,13 +49,16 @@ class PartitionReader:
|
|
18
49
|
|
19
50
|
_IS_SUCCESSFUL = True
|
20
51
|
|
21
|
-
def __init__(
|
52
|
+
def __init__(
|
53
|
+
self, queue: Queue[QueueItem], partition_logger: Optional[PartitionLogger] = None
|
54
|
+
) -> None:
|
22
55
|
"""
|
23
56
|
:param queue: The queue to put the records in.
|
24
57
|
"""
|
25
58
|
self._queue = queue
|
59
|
+
self._partition_logger = partition_logger
|
26
60
|
|
27
|
-
def process_partition(self, partition: Partition) -> None:
|
61
|
+
def process_partition(self, partition: Partition, cursor: Cursor) -> None:
|
28
62
|
"""
|
29
63
|
Process a partition and put the records in the output queue.
|
30
64
|
When all the partitions are added to the queue, a sentinel is added to the queue to indicate that all the partitions have been generated.
|
@@ -37,8 +71,13 @@ class PartitionReader:
|
|
37
71
|
:return: None
|
38
72
|
"""
|
39
73
|
try:
|
74
|
+
if self._partition_logger:
|
75
|
+
self._partition_logger.log(partition)
|
76
|
+
|
40
77
|
for record in partition.read():
|
41
78
|
self._queue.put(record)
|
79
|
+
cursor.observe(record)
|
80
|
+
cursor.close_partition(partition)
|
42
81
|
self._queue.put(PartitionCompleteSentinel(partition, self._IS_SUCCESSFUL))
|
43
82
|
except Exception as e:
|
44
83
|
self._queue.put(StreamThreadException(e, partition.stream_name()))
|
@@ -4,6 +4,7 @@
|
|
4
4
|
|
5
5
|
from typing import Any, Union
|
6
6
|
|
7
|
+
from airbyte_cdk.models import AirbyteMessage
|
7
8
|
from airbyte_cdk.sources.concurrent_source.partition_generation_completed_sentinel import (
|
8
9
|
PartitionGenerationCompletedSentinel,
|
9
10
|
)
|
@@ -34,5 +35,10 @@ class PartitionCompleteSentinel:
|
|
34
35
|
Typedef representing the items that can be added to the ThreadBasedConcurrentStream
|
35
36
|
"""
|
36
37
|
QueueItem = Union[
|
37
|
-
Record,
|
38
|
+
Record,
|
39
|
+
Partition,
|
40
|
+
PartitionCompleteSentinel,
|
41
|
+
PartitionGenerationCompletedSentinel,
|
42
|
+
Exception,
|
43
|
+
AirbyteMessage,
|
38
44
|
]
|
@@ -11,6 +11,10 @@ from airbyte_cdk.models import AirbyteLogMessage, AirbyteMessage, Level
|
|
11
11
|
from airbyte_cdk.models import Type as MessageType
|
12
12
|
|
13
13
|
|
14
|
+
# Once everything runs on the concurrent CDK and we've cleaned up the legacy flows, we should try to remove
|
15
|
+
# this class and write messages directly to the message_repository instead of through the logger because for
|
16
|
+
# cases like the connector builder where ordering of messages is important, using the logger can cause
|
17
|
+
# messages to be grouped out of order. Alas work for a different day.
|
14
18
|
class SliceLogger(ABC):
|
15
19
|
"""
|
16
20
|
SliceLogger is an interface that allows us to log slices of data in a uniform way.
|
@@ -15,11 +15,11 @@ airbyte_cdk/config_observation.py,sha256=7SSPxtN0nXPkm4euGNcTTr1iLbwUL01jy-24V1H
|
|
15
15
|
airbyte_cdk/connector.py,sha256=N6TUlrZOMjLAI85JrNAKkfyTqnO5xfBCw4oEfgjJd9o,4254
|
16
16
|
airbyte_cdk/connector_builder/README.md,sha256=Hw3wvVewuHG9-QgsAq1jDiKuLlStDxKBz52ftyNRnBw,1665
|
17
17
|
airbyte_cdk/connector_builder/__init__.py,sha256=4Hw-PX1-VgESLF16cDdvuYCzGJtHntThLF4qIiULWeo,61
|
18
|
-
airbyte_cdk/connector_builder/connector_builder_handler.py,sha256=
|
19
|
-
airbyte_cdk/connector_builder/main.py,sha256=
|
18
|
+
airbyte_cdk/connector_builder/connector_builder_handler.py,sha256=jRtSfj3aca006-01Hax-THJpuoysd8QR6JPGnr8q1Xg,6371
|
19
|
+
airbyte_cdk/connector_builder/main.py,sha256=F9bmdz252pvGXAdDgPwIOPw3fl5fwTU41uG49BQyItI,3883
|
20
20
|
airbyte_cdk/connector_builder/models.py,sha256=9pIZ98LW_d6fRS39VdnUOf3cxGt4TkC5MJ0_OrzcCRk,1578
|
21
21
|
airbyte_cdk/connector_builder/test_reader/__init__.py,sha256=iTwBMoI9vaJotEgpqZbFjlxRcbxXYypSVJ9YxeHk7wc,120
|
22
|
-
airbyte_cdk/connector_builder/test_reader/helpers.py,sha256=
|
22
|
+
airbyte_cdk/connector_builder/test_reader/helpers.py,sha256=5GSrK9EVBDm5dwtudVbA-73EHh53-niRA-oj8eQVFHI,29236
|
23
23
|
airbyte_cdk/connector_builder/test_reader/message_grouper.py,sha256=LDNl-xFQwA4RsUpn7684KbWaVH-SWWBIwhHvIgduLTE,7090
|
24
24
|
airbyte_cdk/connector_builder/test_reader/reader.py,sha256=3jLy3tUUHkG1rmGWrZuo4SmPYNVD9oiAqy8mdaUwzvo,21301
|
25
25
|
airbyte_cdk/connector_builder/test_reader/types.py,sha256=hPZG3jO03kBaPyW94NI3JHRS1jxXGSNBcN1HFzOxo5Y,2528
|
@@ -57,8 +57,8 @@ airbyte_cdk/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
|
57
57
|
airbyte_cdk/sources/__init__.py,sha256=45J83QsFH3Wky3sVapZWg4C58R_i1thm61M06t2c1AQ,1156
|
58
58
|
airbyte_cdk/sources/abstract_source.py,sha256=50vxEBRByiNhT4WJkiFvgM-C6PWqKSJgvuNC_aeg2cw,15547
|
59
59
|
airbyte_cdk/sources/concurrent_source/__init__.py,sha256=3D_RJsxQfiLboSCDdNei1Iv-msRp3DXsas6E9kl7dXc,386
|
60
|
-
airbyte_cdk/sources/concurrent_source/concurrent_read_processor.py,sha256=
|
61
|
-
airbyte_cdk/sources/concurrent_source/concurrent_source.py,sha256=
|
60
|
+
airbyte_cdk/sources/concurrent_source/concurrent_read_processor.py,sha256=iEm3st8ds4n6_Bn8J2hLK3bQuE5J6ihUitIexX4ADGQ,12240
|
61
|
+
airbyte_cdk/sources/concurrent_source/concurrent_source.py,sha256=Npc18v1ERfQyTjBwIDErTtoBtY11GMml5o_jU3ppRA0,8457
|
62
62
|
airbyte_cdk/sources/concurrent_source/concurrent_source_adapter.py,sha256=f9PIRPWn2tXu0-bxVeYHL2vYdqCzZ_kgpHg5_Ep-cfQ,6103
|
63
63
|
airbyte_cdk/sources/concurrent_source/partition_generation_completed_sentinel.py,sha256=z1t-rAZBsqVidv2fpUlPHE9JgyXsITuGk4AMu96mXSQ,696
|
64
64
|
airbyte_cdk/sources/concurrent_source/stream_thread_exception.py,sha256=-q6mG2145HKQ28rZGD1bUmjPlIZ1S7-Yhewl8Ntu6xI,764
|
@@ -86,7 +86,7 @@ airbyte_cdk/sources/declarative/checks/check_stream.py,sha256=QeExVmpSYjr_CnghHu
|
|
86
86
|
airbyte_cdk/sources/declarative/checks/connection_checker.py,sha256=MBRJo6WJlZQHpIfOGaNOkkHUmgUl_4wDM6VPo41z5Ss,1383
|
87
87
|
airbyte_cdk/sources/declarative/concurrency_level/__init__.py,sha256=5XUqrmlstYlMM0j6crktlKQwALek0uiz2D3WdM46MyA,191
|
88
88
|
airbyte_cdk/sources/declarative/concurrency_level/concurrency_level.py,sha256=YIwCTCpOr_QSNW4ltQK0yUGWInI8PKNY216HOOegYLk,2101
|
89
|
-
airbyte_cdk/sources/declarative/concurrent_declarative_source.py,sha256=
|
89
|
+
airbyte_cdk/sources/declarative/concurrent_declarative_source.py,sha256=ntVyaOmYt3b7rFUjmloDwE2Vwr9kAWMr1HryZicEnwY,30101
|
90
90
|
airbyte_cdk/sources/declarative/datetime/__init__.py,sha256=4Hw-PX1-VgESLF16cDdvuYCzGJtHntThLF4qIiULWeo,61
|
91
91
|
airbyte_cdk/sources/declarative/datetime/datetime_parser.py,sha256=_zGNGq31RNy_0QBLt_EcTvgPyhj7urPdx6oA3M5-r3o,3150
|
92
92
|
airbyte_cdk/sources/declarative/datetime/min_max_datetime.py,sha256=0BHBtDNQZfvwM45-tY5pNlTcKAFSGGNxemoi0Jic-0E,5785
|
@@ -141,7 +141,7 @@ airbyte_cdk/sources/declarative/parsers/custom_exceptions.py,sha256=wnRUP0Xeru9R
|
|
141
141
|
airbyte_cdk/sources/declarative/parsers/manifest_component_transformer.py,sha256=2UdpCz3yi7ISZTyqkQXSSy3dMxeyOWqV7OlAS5b9GVg,11568
|
142
142
|
airbyte_cdk/sources/declarative/parsers/manifest_normalizer.py,sha256=EtKjS9c94yNp3AwQC8KUCQaAYW5T3zvFYxoWYjc_buI,19729
|
143
143
|
airbyte_cdk/sources/declarative/parsers/manifest_reference_resolver.py,sha256=pJmg78vqE5VfUrF_KJnWjucQ4k9IWFULeAxHCowrHXE,6806
|
144
|
-
airbyte_cdk/sources/declarative/parsers/model_to_component_factory.py,sha256=
|
144
|
+
airbyte_cdk/sources/declarative/parsers/model_to_component_factory.py,sha256=ONwdHGA5j4g-Iv2LRzcpXXpHf1Zm9ofgLcXYDyDadYk,181582
|
145
145
|
airbyte_cdk/sources/declarative/partition_routers/__init__.py,sha256=TBC9AkGaUqHm2IKHMPN6punBIcY5tWGULowcLoAVkfw,1109
|
146
146
|
airbyte_cdk/sources/declarative/partition_routers/async_job_partition_router.py,sha256=VelO7zKqKtzMJ35jyFeg0ypJLQC0plqqIBNXoBW1G2E,3001
|
147
147
|
airbyte_cdk/sources/declarative/partition_routers/cartesian_product_stream_slicer.py,sha256=c5cuVFM6NFkuQqG8Z5IwkBuwDrvXZN1CunUOM_L0ezg,6892
|
@@ -166,7 +166,7 @@ airbyte_cdk/sources/declarative/requesters/error_handlers/default_http_response_
|
|
166
166
|
airbyte_cdk/sources/declarative/requesters/error_handlers/error_handler.py,sha256=Tan66odx8VHzfdyyXMQkXz2pJYksllGqvxmpoajgcK4,669
|
167
167
|
airbyte_cdk/sources/declarative/requesters/error_handlers/http_response_filter.py,sha256=E-fQbt4ShfxZVoqfnmOx69C6FUPWZz8BIqI3DN9Kcjs,7935
|
168
168
|
airbyte_cdk/sources/declarative/requesters/http_job_repository.py,sha256=pVzIDdfGs1eAZo9F6zeFYKlEmEqanhNvZLKFCHkdmNo,14348
|
169
|
-
airbyte_cdk/sources/declarative/requesters/http_requester.py,sha256=
|
169
|
+
airbyte_cdk/sources/declarative/requesters/http_requester.py,sha256=cktdjnOu-o98smdCdrWC361iWNamPTZ-csT32OFh00c,18920
|
170
170
|
airbyte_cdk/sources/declarative/requesters/paginators/__init__.py,sha256=uArbKs9JKNCt7t9tZoeWwjDpyI1HoPp29FNW0JzvaEM,644
|
171
171
|
airbyte_cdk/sources/declarative/requesters/paginators/default_paginator.py,sha256=SB-Af3CRb4mJwhm4EKNxzl_PK2w5QS4tqrSNNMO2IV4,12760
|
172
172
|
airbyte_cdk/sources/declarative/requesters/paginators/no_pagination.py,sha256=b1-zKxYOUMHn7ahdWpzKEzfG4A7s_WQWy-vzRqZWzME,2152
|
@@ -220,9 +220,9 @@ airbyte_cdk/sources/declarative/schema/schema_loader.py,sha256=kjt8v0N5wWKA5zyLn
|
|
220
220
|
airbyte_cdk/sources/declarative/spec/__init__.py,sha256=9FYO-fVOclrwjAW4qwRTbZRVopTc9rOaauAJfThdNCQ,177
|
221
221
|
airbyte_cdk/sources/declarative/spec/spec.py,sha256=SwL_pfXZgcLYLJY-MAeFMHug9oYh2tOWjgG0C3DoLOY,3602
|
222
222
|
airbyte_cdk/sources/declarative/stream_slicers/__init__.py,sha256=UX-cP_C-9FIFFPL9z8nuxu_rglssRsMOqQmQHN8FLB8,341
|
223
|
-
airbyte_cdk/sources/declarative/stream_slicers/declarative_partition_generator.py,sha256=
|
223
|
+
airbyte_cdk/sources/declarative/stream_slicers/declarative_partition_generator.py,sha256=GTG1nqXVxbT7wnU-Q9hftlcWAqj-y4zR2KCnLWNIcv4,4084
|
224
224
|
airbyte_cdk/sources/declarative/stream_slicers/stream_slicer.py,sha256=SOkIPBi2Wu7yxIvA15yFzUAB95a3IzA8LPq5DEqHQQc,725
|
225
|
-
airbyte_cdk/sources/declarative/stream_slicers/stream_slicer_test_read_decorator.py,sha256=
|
225
|
+
airbyte_cdk/sources/declarative/stream_slicers/stream_slicer_test_read_decorator.py,sha256=4vit5ADyhoZnd1psRVeM5jdySYzhjwspLVXxh8vt1M8,944
|
226
226
|
airbyte_cdk/sources/declarative/transformations/__init__.py,sha256=CPJ8TlMpiUmvG3624VYu_NfTzxwKcfBjM2Q2wJ7fkSA,919
|
227
227
|
airbyte_cdk/sources/declarative/transformations/add_fields.py,sha256=Eg1jQtRObgzxbtySTQs5uEZIjEklsoHFxYSPf78x6Ng,5420
|
228
228
|
airbyte_cdk/sources/declarative/transformations/config_transformations/__init__.py,sha256=GaU3ezFa5opeDgdlNohX6TXsWJlOD2jOfJXQWeQCh7E,263
|
@@ -300,6 +300,7 @@ airbyte_cdk/sources/file_based/types.py,sha256=INxG7OPnkdUP69oYNKMAbwhvV1AGvLRHs
|
|
300
300
|
airbyte_cdk/sources/http_config.py,sha256=OBZeuyFilm6NlDlBhFQvHhTWabEvZww6OHDIlZujIS0,730
|
301
301
|
airbyte_cdk/sources/http_logger.py,sha256=H93kPAujHhPmXNX0JSFG3D-SL6yEFA5PtKot9Hu3TYA,1690
|
302
302
|
airbyte_cdk/sources/message/__init__.py,sha256=y98fzHsQBwXwp2zEa4K5mxGFqjnx9lDn9O0pTk-VS4U,395
|
303
|
+
airbyte_cdk/sources/message/concurrent_repository.py,sha256=axSiL2Rm1z6ENtx_ZZ7B_UfLNesGX-lBMQLJvIlmCN4,1999
|
303
304
|
airbyte_cdk/sources/message/repository.py,sha256=SG7avgti_-dj8FcRHTTrhgLLGJbElv14_zIB0SH8AIc,4763
|
304
305
|
airbyte_cdk/sources/source.py,sha256=KIBBH5VLEb8BZ8B9aROlfaI6OLoJqKDPMJ10jkAR7nk,3611
|
305
306
|
airbyte_cdk/sources/specs/transfer_modes.py,sha256=sfSVO0yT6SaGKN5_TP0Nl_ftG0yPhecaBv0WkhAEXA8,932
|
@@ -325,12 +326,12 @@ airbyte_cdk/sources/streams/concurrent/default_stream.py,sha256=3SBjFa1z955pSE_2
|
|
325
326
|
airbyte_cdk/sources/streams/concurrent/exceptions.py,sha256=JOZ446MCLpmF26r9KfS6OO_6rGjcjgJNZdcw6jccjEI,468
|
326
327
|
airbyte_cdk/sources/streams/concurrent/helpers.py,sha256=S6AW8TgIASCZ2UuUcQLE8OzgYUHWt2-KPOvNPwnQf-Q,1596
|
327
328
|
airbyte_cdk/sources/streams/concurrent/partition_enqueuer.py,sha256=2t64b_z9cEPmlHZnjSiMTO8PEtEdiAJDG0JcYOtUqAE,3363
|
328
|
-
airbyte_cdk/sources/streams/concurrent/partition_reader.py,sha256=
|
329
|
+
airbyte_cdk/sources/streams/concurrent/partition_reader.py,sha256=0QKCJYBuCqiuHp0GqcOTvzsJS1Sp7qTNfjXESvs71Vk,3417
|
329
330
|
airbyte_cdk/sources/streams/concurrent/partitions/__init__.py,sha256=4Hw-PX1-VgESLF16cDdvuYCzGJtHntThLF4qIiULWeo,61
|
330
331
|
airbyte_cdk/sources/streams/concurrent/partitions/partition.py,sha256=CmaRcKn8y118No3qvbRV9DBeAUKv17lrVgloR4Y9TwU,1490
|
331
332
|
airbyte_cdk/sources/streams/concurrent/partitions/partition_generator.py,sha256=_ymkkBr71_qt1fW0_MUqw96OfNBkeJngXQ09yolEDHw,441
|
332
333
|
airbyte_cdk/sources/streams/concurrent/partitions/stream_slicer.py,sha256=zQPikLIt0yhP9EwZaPglRTIqFCauo4pSsJk_7kYq9Aw,1406
|
333
|
-
airbyte_cdk/sources/streams/concurrent/partitions/types.py,sha256=
|
334
|
+
airbyte_cdk/sources/streams/concurrent/partitions/types.py,sha256=EdPHmMcyZhRYtO2cniIQAQpzPfGCBpzmAJ3NTVS4qbo,1249
|
334
335
|
airbyte_cdk/sources/streams/concurrent/state_converters/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
335
336
|
airbyte_cdk/sources/streams/concurrent/state_converters/abstract_stream_state_converter.py,sha256=CCxCbgvUugxiWpHX8-dkkJHWKDjL5iwiIbOUj8KIJ9c,7079
|
336
337
|
airbyte_cdk/sources/streams/concurrent/state_converters/datetime_stream_state_converter.py,sha256=x8MLm1pTMfLNHvMF3P1ixYkYt_xjpbaIwnvhY_ofdBo,8076
|
@@ -364,7 +365,7 @@ airbyte_cdk/sources/utils/casing.py,sha256=QC-gV1O4e8DR4-bhdXieUPKm_JamzslVyfABL
|
|
364
365
|
airbyte_cdk/sources/utils/files_directory.py,sha256=z8Dmr-wkL1sAqdwCST4MBUFAyMHPD2cJIzVdAuCynp8,391
|
365
366
|
airbyte_cdk/sources/utils/record_helper.py,sha256=7wL-pDYrBpcmZHa8ORtiSOqBZJEZI5hdl2dA1RYiatk,2029
|
366
367
|
airbyte_cdk/sources/utils/schema_helpers.py,sha256=bR3I70-e11S6B8r6VK-pthQXtcYrXojgXFvuK7lRrpg,8545
|
367
|
-
airbyte_cdk/sources/utils/slice_logger.py,sha256=
|
368
|
+
airbyte_cdk/sources/utils/slice_logger.py,sha256=M1TvcYGMftXR841XdJmeEpKpQqrdOD5X-qsspfAMKMs,2168
|
368
369
|
airbyte_cdk/sources/utils/transform.py,sha256=0LOvIJg1vmg_70AiAVe-YHMr-LHrqEuxg9cm1BnYPDM,11725
|
369
370
|
airbyte_cdk/sources/utils/types.py,sha256=41ZQR681t5TUnOScij58d088sb99klH_ZENFcaYro_g,175
|
370
371
|
airbyte_cdk/sql/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
@@ -424,9 +425,9 @@ airbyte_cdk/utils/slice_hasher.py,sha256=EDxgROHDbfG-QKQb59m7h_7crN1tRiawdf5uU7G
|
|
424
425
|
airbyte_cdk/utils/spec_schema_transformations.py,sha256=-5HTuNsnDBAhj-oLeQXwpTGA0HdcjFOf2zTEMUTTg_Y,816
|
425
426
|
airbyte_cdk/utils/stream_status_utils.py,sha256=ZmBoiy5HVbUEHAMrUONxZvxnvfV9CesmQJLDTAIWnWw,1171
|
426
427
|
airbyte_cdk/utils/traced_exception.py,sha256=C8uIBuCL_E4WnBAOPSxBicD06JAldoN9fGsQDp463OY,6292
|
427
|
-
airbyte_cdk-6.60.
|
428
|
-
airbyte_cdk-6.60.
|
429
|
-
airbyte_cdk-6.60.
|
430
|
-
airbyte_cdk-6.60.
|
431
|
-
airbyte_cdk-6.60.
|
432
|
-
airbyte_cdk-6.60.
|
428
|
+
airbyte_cdk-6.60.7.dist-info/LICENSE.txt,sha256=Wfe61S4BaGPj404v8lrAbvhjYR68SHlkzeYrg3_bbuM,1051
|
429
|
+
airbyte_cdk-6.60.7.dist-info/LICENSE_SHORT,sha256=aqF6D1NcESmpn-cqsxBtszTEnHKnlsp8L4x9wAh3Nxg,55
|
430
|
+
airbyte_cdk-6.60.7.dist-info/METADATA,sha256=8UFDCTLKAEkz9_HGFQhGgKVW3I0w-oPneO4SHQIHU24,6477
|
431
|
+
airbyte_cdk-6.60.7.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
|
432
|
+
airbyte_cdk-6.60.7.dist-info/entry_points.txt,sha256=AKWbEkHfpzzk9nF9tqBUaw1MbvTM4mGtEzmZQm0ZWvM,139
|
433
|
+
airbyte_cdk-6.60.7.dist-info/RECORD,,
|
File without changes
|
File without changes
|
File without changes
|
File without changes
|