airbyte-cdk 6.41.9.dev41013__tar.gz → 6.42.1__tar.gz

{airbyte_cdk-6.41.9.dev41013 → airbyte_cdk-6.42.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: airbyte-cdk
-Version: 6.41.9.dev41013
+Version: 6.42.1
 Summary: A framework for writing Airbyte Connectors.
 Home-page: https://airbyte.com
 License: MIT
@@ -22,7 +22,7 @@ Provides-Extra: sql
 Provides-Extra: vector-db-based
 Requires-Dist: Jinja2 (>=3.1.2,<3.2.0)
 Requires-Dist: PyYAML (>=6.0.1,<7.0.0)
-Requires-Dist: airbyte-protocol-models-dataclasses (==0.14.1337.dev1742858109)
+Requires-Dist: airbyte-protocol-models-dataclasses (>=0.14,<0.15)
 Requires-Dist: anyascii (>=0.3.2,<0.4.0)
 Requires-Dist: avro (>=1.11.2,<1.13.0) ; extra == "file-based"
 Requires-Dist: backoff

{airbyte_cdk-6.41.9.dev41013 → airbyte_cdk-6.42.1}/airbyte_cdk/connector_builder/connector_builder_handler.py

@@ -117,5 +117,30 @@ def resolve_manifest(source: ManifestDeclarativeSource) -> AirbyteMessage:
         return error.as_airbyte_message()
 
 
+def full_resolve_manifest(source: ManifestDeclarativeSource) -> AirbyteMessage:
+    try:
+        manifest = {**source.resolved_manifest}
+        streams = manifest.get("streams", [])
+        for stream in streams:
+            stream["dynamic_stream_name"] = None
+        streams.extend(source.dynamic_streams)
+        manifest["streams"] = streams
+        return AirbyteMessage(
+            type=Type.RECORD,
+            record=AirbyteRecordMessage(
+                data={"manifest": manifest},
+                emitted_at=_emitted_at(),
+                stream="full_resolve_manifest",
+            ),
+        )
+    except AirbyteTracedException as exc:
+        return exc.as_airbyte_message()
+    except Exception as exc:
+        error = AirbyteTracedException.from_exception(
+            exc, message=f"Error full resolving manifest: {str(exc)}"
+        )
+        return error.as_airbyte_message()
+
+
 def _emitted_at() -> int:
     return ab_datetime_now().to_epoch_millis()

{airbyte_cdk-6.41.9.dev41013 → airbyte_cdk-6.42.1}/airbyte_cdk/connector_builder/main.py

@@ -12,6 +12,7 @@ from airbyte_cdk.connector import BaseConnector
 from airbyte_cdk.connector_builder.connector_builder_handler import (
     TestReadLimits,
     create_source,
+    full_resolve_manifest,
     get_limits,
     read_stream,
     resolve_manifest,
@@ -81,6 +82,8 @@ def handle_connector_builder_request(
             catalog is not None
         ), "`test_read` requires a valid `ConfiguredAirbyteCatalog`, got None."
         return read_stream(source, config, catalog, state, limits)
+    elif command == "full_resolve_manifest":
+        return full_resolve_manifest(source)
     else:
         raise ValueError(f"Unrecognized command {command}.")
 

{airbyte_cdk-6.41.9.dev41013 → airbyte_cdk-6.42.1}/airbyte_cdk/models/__init__.py

@@ -19,7 +19,6 @@ from .airbyte_protocol import (
     AirbyteMessage,
     AirbyteProtocol,
     AirbyteRecordMessage,
-    AirbyteRecordMessageFileReference,
     AirbyteStateBlob,
     AirbyteStateMessage,
     AirbyteStateStats,

{airbyte_cdk-6.41.9.dev41013 → airbyte_cdk-6.42.1}/airbyte_cdk/sources/concurrent_source/concurrent_read_processor.py

@@ -150,7 +150,6 @@ class ConcurrentReadProcessor:
             stream_name=record.stream_name,
             data_or_message=record.data,
             is_file_transfer_message=record.is_file_transfer_message,
-            file_reference=record.file_reference,
         )
         stream = self._stream_name_to_instance[record.stream_name]
 

{airbyte_cdk-6.41.9.dev41013 → airbyte_cdk-6.42.1}/airbyte_cdk/sources/declarative/async_job/job.py

@@ -34,6 +34,12 @@ class AsyncJob:
 
     def status(self) -> AsyncJobStatus:
         if self._timer.has_timed_out():
+            # TODO: we should account the fact that,
+            # certain APIs could send the `Timeout` status,
+            # thus we should not return `Timeout` in that case,
+            # but act based on the scenario.
+
+            # the default behavior is to return `Timeout` status and retry.
             return AsyncJobStatus.TIMED_OUT
         return self._status
 

{airbyte_cdk-6.41.9.dev41013 → airbyte_cdk-6.42.1}/airbyte_cdk/sources/declarative/async_job/job_orchestrator.py

@@ -44,16 +44,21 @@ class AsyncPartition:
     This bucket of api_jobs is a bit useless for this iteration but should become interesting when we will be able to split jobs
     """
 
-    _MAX_NUMBER_OF_ATTEMPTS = 3
+    _DEFAULT_MAX_JOB_RETRY = 3
 
-    def __init__(self, jobs: List[AsyncJob], stream_slice: StreamSlice) -> None:
+    def __init__(
+        self, jobs: List[AsyncJob], stream_slice: StreamSlice, job_max_retry: Optional[int] = None
+    ) -> None:
         self._attempts_per_job = {job: 1 for job in jobs}
         self._stream_slice = stream_slice
+        self._job_max_retry = (
+            job_max_retry if job_max_retry is not None else self._DEFAULT_MAX_JOB_RETRY
+        )
 
     def has_reached_max_attempt(self) -> bool:
         return any(
             map(
-                lambda attempt_count: attempt_count >= self._MAX_NUMBER_OF_ATTEMPTS,
+                lambda attempt_count: attempt_count >= self._job_max_retry,
                 self._attempts_per_job.values(),
             )
         )
@@ -62,7 +67,7 @@ class AsyncPartition:
         current_attempt_count = self._attempts_per_job.pop(job_to_replace, None)
         if current_attempt_count is None:
             raise ValueError("Could not find job to replace")
-        elif current_attempt_count >= self._MAX_NUMBER_OF_ATTEMPTS:
+        elif current_attempt_count >= self._job_max_retry:
             raise ValueError(f"Max attempt reached for job in partition {self._stream_slice}")
 
         new_attempt_count = current_attempt_count + 1
@@ -155,6 +160,7 @@ class AsyncJobOrchestrator:
         message_repository: MessageRepository,
         exceptions_to_break_on: Iterable[Type[Exception]] = tuple(),
         has_bulk_parent: bool = False,
+        job_max_retry: Optional[int] = None,
     ) -> None:
         """
         If the stream slices provided as a parameters relies on a async job streams that relies on the same JobTracker, `has_bulk_parent`
@@ -175,11 +181,12 @@ class AsyncJobOrchestrator:
         self._message_repository = message_repository
         self._exceptions_to_break_on: Tuple[Type[Exception], ...] = tuple(exceptions_to_break_on)
         self._has_bulk_parent = has_bulk_parent
+        self._job_max_retry = job_max_retry
 
         self._non_breaking_exceptions: List[Exception] = []
 
     def _replace_failed_jobs(self, partition: AsyncPartition) -> None:
-        failed_status_jobs = (AsyncJobStatus.FAILED,)
+        failed_status_jobs = (AsyncJobStatus.FAILED, AsyncJobStatus.TIMED_OUT)
         jobs_to_replace = [job for job in partition.jobs if job.status() in failed_status_jobs]
         for job in jobs_to_replace:
             new_job = self._start_job(job.job_parameters(), job.api_job_id())
@@ -214,7 +221,7 @@ class AsyncJobOrchestrator:
             for _slice in self._slice_iterator:
                 at_least_one_slice_consumed_from_slice_iterator_during_current_iteration = True
                 job = self._start_job(_slice)
-                self._running_partitions.append(AsyncPartition([job], _slice))
+                self._running_partitions.append(AsyncPartition([job], _slice, self._job_max_retry))
                 if self._has_bulk_parent and self._slice_iterator.has_next():
                     break
         except ConcurrentJobLimitReached:
@@ -363,7 +370,7 @@ class AsyncJobOrchestrator:
                     self._reallocate_partition(current_running_partitions, partition)
 
             # We only remove completed / timeout jobs jobs as we want failed jobs to be re-allocated in priority
-            self._remove_completed_or_timed_out_jobs(partition)
+            self._remove_completed_jobs(partition)
 
         # update the referenced list with running partitions
         self._running_partitions = current_running_partitions
@@ -378,11 +385,7 @@ class AsyncJobOrchestrator:
     def _stop_timed_out_jobs(self, partition: AsyncPartition) -> None:
         for job in partition.jobs:
             if job.status() == AsyncJobStatus.TIMED_OUT:
-                self._abort_job(job, free_job_allocation=True)
-                raise AirbyteTracedException(
-                    internal_message=f"Job {job.api_job_id()} has timed out. Try increasing the `polling job timeout`.",
-                    failure_type=FailureType.config_error,
-                )
+                self._abort_job(job, free_job_allocation=False)
 
     def _abort_job(self, job: AsyncJob, free_job_allocation: bool = True) -> None:
         try:
@@ -392,7 +395,7 @@ class AsyncJobOrchestrator:
         except Exception as exception:
             LOGGER.warning(f"Could not free budget for job {job.api_job_id()}: {exception}")
 
-    def _remove_completed_or_timed_out_jobs(self, partition: AsyncPartition) -> None:
+    def _remove_completed_jobs(self, partition: AsyncPartition) -> None:
         """
         Remove completed or timed out jobs from the partition.
 
@@ -400,7 +403,7 @@ class AsyncJobOrchestrator:
             partition (AsyncPartition): The partition to process.
         """
         for job in partition.jobs:
-            if job.status() in [AsyncJobStatus.COMPLETED, AsyncJobStatus.TIMED_OUT]:
+            if job.status() == AsyncJobStatus.COMPLETED:
                 self._job_tracker.remove_job(job.api_job_id())
 
     def _reallocate_partition(
@@ -415,10 +418,7 @@ class AsyncJobOrchestrator:
             current_running_partitions (list): The list of currently running partitions.
             partition (AsyncPartition): The partition to reallocate.
         """
-        for job in partition.jobs:
-            if job.status() != AsyncJobStatus.TIMED_OUT:
-                # allow the FAILED jobs to be re-allocated for partition
-                current_running_partitions.insert(0, partition)
+        current_running_partitions.insert(0, partition)
 
     def _process_partitions_with_errors(self, partition: AsyncPartition) -> None:
         """

{airbyte_cdk-6.41.9.dev41013 → airbyte_cdk-6.42.1}/airbyte_cdk/sources/declarative/async_job/job_tracker.py

@@ -3,9 +3,11 @@
 import logging
 import threading
 import uuid
-from typing import Set
+from dataclasses import dataclass, field
+from typing import Any, Mapping, Set, Union
 
 from airbyte_cdk.logger import lazy_log
+from airbyte_cdk.sources.declarative.interpolation import InterpolatedString
 
 LOGGER = logging.getLogger("airbyte")
 
@@ -14,15 +16,29 @@ class ConcurrentJobLimitReached(Exception):
     pass
 
 
+@dataclass
 class JobTracker:
-    def __init__(self, limit: int):
+    limit: Union[int, str]
+    config: Mapping[str, Any] = field(default_factory=dict)
+
+    def __post_init__(self) -> None:
         self._jobs: Set[str] = set()
-        if limit < 1:
+        self._lock = threading.Lock()
+        if isinstance(self.limit, str):
+            try:
+                self.limit = int(
+                    InterpolatedString(self.limit, parameters={}).eval(config=self.config)
+                )
+            except Exception as e:
+                LOGGER.warning(
+                    f"Error interpolating max job count: {self.limit}. Setting to 1. {e}"
+                )
+                self.limit = 1
+        if self.limit < 1:
             LOGGER.warning(
-                f"The `max_concurrent_async_job_count` property is less than 1: {limit}. Setting to 1. Please update the source manifest to set a valid value."
+                f"The `max_concurrent_async_job_count` property is less than 1: {self.limit}. Setting to 1. Please update the source manifest to set a valid value."
             )
-        self._limit = 1 if limit < 1 else limit
-        self._lock = threading.Lock()
+        self._limit = self.limit if self.limit >= 1 else 1
 
     def try_to_get_intent(self) -> str:
         lazy_log(

{airbyte_cdk-6.41.9.dev41013 → airbyte_cdk-6.42.1}/airbyte_cdk/sources/declarative/concurrent_declarative_source.py

@@ -25,7 +25,6 @@ from airbyte_cdk.sources.declarative.incremental.per_partition_with_global impor
     PerPartitionWithGlobalCursor,
 )
 from airbyte_cdk.sources.declarative.manifest_declarative_source import ManifestDeclarativeSource
-from airbyte_cdk.sources.declarative.models import FileUploader
 from airbyte_cdk.sources.declarative.models.declarative_component_schema import (
     ConcurrencyLevel as ConcurrencyLevelModel,
 )
@@ -207,20 +206,6 @@ class ConcurrentDeclarativeSource(ManifestDeclarativeSource, Generic[TState]):
             # these legacy Python streams the way we do low-code streams to determine if they are concurrent compatible,
             # so we need to treat them as synchronous
 
-            file_uploader = None
-            if isinstance(declarative_stream, DeclarativeStream):
-                file_uploader = (
-                    self._constructor.create_component(
-                        model_type=FileUploader,
-                        component_definition=name_to_stream_mapping[declarative_stream.name][
-                            "file_uploader"
-                        ],
-                        config=config,
-                    )
-                    if "file_uploader" in name_to_stream_mapping[declarative_stream.name]
-                    else None
-                )
-
             if (
                 isinstance(declarative_stream, DeclarativeStream)
                 and name_to_stream_mapping[declarative_stream.name]["type"]
@@ -288,7 +273,6 @@ class ConcurrentDeclarativeSource(ManifestDeclarativeSource, Generic[TState]):
                                 declarative_stream.get_json_schema(),
                                 retriever,
                                 self.message_repository,
-                                file_uploader,
                             ),
                             stream_slicer=declarative_stream.retriever.stream_slicer,
                         )
@@ -319,7 +303,6 @@ class ConcurrentDeclarativeSource(ManifestDeclarativeSource, Generic[TState]):
                                 declarative_stream.get_json_schema(),
                                 retriever,
                                 self.message_repository,
-                                file_uploader,
                             ),
                             stream_slicer=cursor,
                         )
@@ -339,7 +322,6 @@ class ConcurrentDeclarativeSource(ManifestDeclarativeSource, Generic[TState]):
                             else None,
                             logger=self.logger,
                             cursor=cursor,
-                            supports_file_transfer=bool(file_uploader),
                         )
                     )
                 elif (
@@ -351,7 +333,6 @@ class ConcurrentDeclarativeSource(ManifestDeclarativeSource, Generic[TState]):
                             declarative_stream.get_json_schema(),
                             declarative_stream.retriever,
                             self.message_repository,
-                            file_uploader,
                         ),
                         declarative_stream.retriever.stream_slicer,
                     )
@@ -372,7 +353,6 @@ class ConcurrentDeclarativeSource(ManifestDeclarativeSource, Generic[TState]):
                             cursor_field=None,
                             logger=self.logger,
                             cursor=final_state_cursor,
-                            supports_file_transfer=bool(file_uploader),
                         )
                     )
                 elif (
@@ -412,7 +392,6 @@ class ConcurrentDeclarativeSource(ManifestDeclarativeSource, Generic[TState]):
                             declarative_stream.get_json_schema(),
                             retriever,
                             self.message_repository,
-                            file_uploader,
                         ),
                         perpartition_cursor,
                     )
@@ -427,7 +406,6 @@ class ConcurrentDeclarativeSource(ManifestDeclarativeSource, Generic[TState]):
                             cursor_field=perpartition_cursor.cursor_field.cursor_field_key,
                             logger=self.logger,
                             cursor=perpartition_cursor,
-                            supports_file_transfer=bool(file_uploader),
                         )
                     )
                 else:

{airbyte_cdk-6.41.9.dev41013 → airbyte_cdk-6.42.1}/airbyte_cdk/sources/declarative/declarative_component_schema.yaml

@@ -47,7 +47,12 @@ properties:
   max_concurrent_async_job_count:
     title: Maximum Concurrent Asynchronous Jobs
     description: Maximum number of concurrent asynchronous jobs to run. This property is only relevant for sources/streams that support asynchronous job execution through the AsyncRetriever (e.g. a report-based stream that initiates a job, polls the job status, and then fetches the job results). This is often set by the API's maximum number of concurrent jobs on the account level. Refer to the API's documentation for this information.
-    type: integer
+    type:
+      - integer
+      - string
+    examples:
+      - 3
+      - "{{ config['max_concurrent_async_job_count'] }}"
   metadata:
     type: object
     description: For internal Airbyte use only - DO NOT modify manually. Used by consumers of declarative manifests for storing related metadata.
@@ -1422,33 +1427,6 @@ definitions:
             - "$ref": "#/definitions/LegacyToPerPartitionStateMigration"
             - "$ref": "#/definitions/CustomStateMigration"
         default: []
-      file_uploader:
-        title: File Uploader
-        description: (experimental) Describes how to fetch a file
-        type: object
-        required:
-          - type
-          - requester
-          - download_target_extractor
-        properties:
-          type:
-            type: string
-            enum: [ FileUploader ]
-          requester:
-            description: Requester component that describes how to prepare HTTP requests to send to the source API.
-            anyOf:
-              - "$ref": "#/definitions/CustomRequester"
-              - "$ref": "#/definitions/HttpRequester"
-          download_target_extractor:
-            description: Responsible for fetching the url where the file is located. This is applied on each records and not on the HTTP response
-            anyOf:
-              - "$ref": "#/definitions/CustomRecordExtractor"
-              - "$ref": "#/definitions/DpathExtractor"
-          file_extractor:
-            description: Responsible for fetching the content of the file. If not defined, the assumption is that the whole response body is the file content
-            anyOf:
-              - "$ref": "#/definitions/CustomRecordExtractor"
-              - "$ref": "#/definitions/DpathExtractor"
       $parameters:
         type: object
         additional_properties: true
@@ -2219,7 +2197,8 @@ definitions:
         type: object
         additionalProperties: true
   JsonDecoder:
-    title: Json Decoder
+    title: JSON
+    description: Select 'JSON' if the response is formatted as a JSON object.
     type: object
     required:
       - type
@@ -2228,8 +2207,8 @@ definitions:
         type: string
         enum: [JsonDecoder]
   JsonlDecoder:
-    title: JSONL Decoder
-    description: Use this if the response consists of JSON objects separated by new lines (`\n`) in JSONL format.
+    title: JSON Lines
+    description: Select 'JSON Lines' if the response consists of JSON objects separated by new lines ('\n') in JSONL format.
     type: object
     required:
       - type
@@ -2354,8 +2333,8 @@ definitions:
         type: object
         additionalProperties: true
   IterableDecoder:
-    title: Iterable Decoder
-    description: Use this if the response consists of strings separated by new lines (`\n`). The Decoder will wrap each row into a JSON object with the `record` key.
+    title: Iterable
+    description: Select 'Iterable' if the response consists of strings separated by new lines (`\n`). The string will then be wrapped into a JSON object with the `record` key.
     type: object
     required:
       - type
@@ -2364,8 +2343,8 @@ definitions:
         type: string
         enum: [IterableDecoder]
   XmlDecoder:
-    title: XML Decoder
-    description: Use this if the response is XML.
+    title: XML
+    description: Select 'XML' if the response consists of XML-formatted data.
     type: object
     required:
       - type
@@ -2396,8 +2375,8 @@ definitions:
         type: object
         additionalProperties: true
   ZipfileDecoder:
-    title: Zipfile Decoder
-    description: Decoder for response data that is returned as zipfile(s).
+    title: ZIP File
+    description: Select 'ZIP file' for response data that is returned as a zipfile. Requires specifying an inner data type/decoder to parse the unzipped data.
     type: object
     additionalProperties: true
     required:
@@ -2921,7 +2900,7 @@ definitions:
         title: Lazy Read Pointer
         description: If set, this will enable lazy reading, using the initial read of parent records to extract child records.
         type: array
-        default: [ ]
+        default: []
         items:
           - type: string
         interpolation_context:
@@ -3226,7 +3205,7 @@ definitions:
     properties:
       type:
         type: string
-        enum: [ StateDelegatingStream ]
+        enum: [StateDelegatingStream]
       name:
         title: Name
         description: The stream name.
@@ -3281,12 +3260,14 @@ definitions:
           - "$ref": "#/definitions/CustomPartitionRouter"
           - "$ref": "#/definitions/ListPartitionRouter"
           - "$ref": "#/definitions/SubstreamPartitionRouter"
+          - "$ref": "#/definitions/GroupingPartitionRouter"
           - type: array
             items:
               anyOf:
                 - "$ref": "#/definitions/CustomPartitionRouter"
                 - "$ref": "#/definitions/ListPartitionRouter"
                 - "$ref": "#/definitions/SubstreamPartitionRouter"
+                - "$ref": "#/definitions/GroupingPartitionRouter"
       decoder:
         title: Decoder
         description: Component decoding the response so records can be extracted.
@@ -3303,6 +3284,8 @@ definitions:
         type: object
         additionalProperties: true
   GzipDecoder:
+    title: gzip
+    description: Select 'gzip' for response data that is compressed with gzip. Requires specifying an inner data type/decoder to parse the decompressed data.
     type: object
     required:
       - type
@@ -3318,6 +3301,8 @@ definitions:
           - "$ref": "#/definitions/JsonDecoder"
           - "$ref": "#/definitions/JsonlDecoder"
   CsvDecoder:
+    title: CSV
+    description: "Select 'CSV' for response data that is formatted as CSV (comma-separated values). Can specify an encoding (default: 'utf-8') and a delimiter (default: ',')."
     type: object
     required:
       - type
@@ -3448,12 +3433,14 @@ definitions:
           - "$ref": "#/definitions/CustomPartitionRouter"
           - "$ref": "#/definitions/ListPartitionRouter"
           - "$ref": "#/definitions/SubstreamPartitionRouter"
+          - "$ref": "#/definitions/GroupingPartitionRouter"
           - type: array
             items:
               anyOf:
                 - "$ref": "#/definitions/CustomPartitionRouter"
                 - "$ref": "#/definitions/ListPartitionRouter"
                 - "$ref": "#/definitions/SubstreamPartitionRouter"
+                - "$ref": "#/definitions/GroupingPartitionRouter"
       decoder:
         title: Decoder
         description: Component decoding the response so records can be extracted.
@@ -3570,6 +3557,44 @@ definitions:
       $parameters:
         type: object
         additionalProperties: true
+  GroupingPartitionRouter:
+    title: Grouping Partition Router
+    description: >
+      A decorator on top of a partition router that groups partitions into batches of a specified size.
+      This is useful for APIs that support filtering by multiple partition keys in a single request.
+      Note that per-partition incremental syncs may not work as expected because the grouping
+      of partitions might change between syncs, potentially leading to inconsistent state tracking.
+    type: object
+    required:
+      - type
+      - group_size
+      - underlying_partition_router
+    properties:
+      type:
+        type: string
+        enum: [GroupingPartitionRouter]
+      group_size:
+        title: Group Size
+        description: The number of partitions to include in each group. This determines how many partition values are batched together in a single slice.
+        type: integer
+        examples:
+          - 10
+          - 50
+      underlying_partition_router:
+        title: Underlying Partition Router
+        description: The partition router whose output will be grouped. This can be any valid partition router component.
+        anyOf:
+          - "$ref": "#/definitions/CustomPartitionRouter"
+          - "$ref": "#/definitions/ListPartitionRouter"
+          - "$ref": "#/definitions/SubstreamPartitionRouter"
+      deduplicate:
+        title: Deduplicate Partitions
+        description: If true, ensures that partitions are unique within each group by removing duplicates based on the partition key.
+        type: boolean
+        default: true
+      $parameters:
+        type: object
+        additionalProperties: true
   WaitUntilTimeFromHeader:
     title: Wait Until Time Defined In Response Header
     description: Extract time at which we can retry the request from response header and wait for the difference between now and that time.
@@ -3741,6 +3766,13 @@ definitions:
       type:
         type: string
         enum: [DynamicDeclarativeStream]
+      name:
+        title: Name
+        description: The dynamic stream name.
+        type: string
+        default: ""
+        example:
+          - "Tables"
       stream_template:
         title: Stream Template
         description: Reference to the stream template.

{airbyte_cdk-6.41.9.dev41013 → airbyte_cdk-6.42.1}/airbyte_cdk/sources/declarative/incremental/concurrent_partition_cursor.py

@@ -79,6 +79,7 @@ class ConcurrentPerPartitionCursor(Cursor):
         connector_state_manager: ConnectorStateManager,
         connector_state_converter: AbstractStreamStateConverter,
         cursor_field: CursorField,
+        use_global_cursor: bool = False,
     ) -> None:
         self._global_cursor: Optional[StreamState] = {}
         self._stream_name = stream_name
@@ -106,7 +107,7 @@ class ConcurrentPerPartitionCursor(Cursor):
         self._lookback_window: int = 0
         self._parent_state: Optional[StreamState] = None
         self._number_of_partitions: int = 0
-        self._use_global_cursor: bool = False
+        self._use_global_cursor: bool = use_global_cursor
         self._partition_serializer = PerPartitionKeySerializer()
         # Track the last time a state message was emitted
         self._last_emission_time: float = 0.0

{airbyte_cdk-6.41.9.dev41013 → airbyte_cdk-6.42.1}/airbyte_cdk/sources/declarative/interpolation/macros.py

@@ -156,7 +156,7 @@ def duration(datestring: str) -> Union[datetime.timedelta, isodate.Duration]:
 
 
 def format_datetime(
-    dt: Union[str, datetime.datetime], format: str, input_format: Optional[str] = None
+    dt: Union[str, datetime.datetime, int], format: str, input_format: Optional[str] = None
 ) -> str:
     """
     Converts datetime to another format
@@ -170,9 +170,13 @@ def format_datetime(
     """
     if isinstance(dt, datetime.datetime):
         return dt.strftime(format)
-    dt_datetime = (
-        datetime.datetime.strptime(dt, input_format) if input_format else str_to_datetime(dt)
-    )
+
+    if isinstance(dt, int):
+        dt_datetime = DatetimeParser().parse(dt, input_format if input_format else "%s")
+    else:
+        dt_datetime = (
+            datetime.datetime.strptime(dt, input_format) if input_format else str_to_datetime(dt)
+        )
     return DatetimeParser().format(dt=dt_datetime, format=format)
 
 

{airbyte_cdk-6.41.9.dev41013 → airbyte_cdk-6.42.1}/airbyte_cdk/sources/declarative/manifest_declarative_source.py

@@ -106,6 +106,7 @@ class ManifestDeclarativeSource(DeclarativeSource):
             AlwaysLogSliceLogger() if emit_connector_builder_messages else DebugSliceLogger()
         )
 
+        self._config = config or {}
         self._validate_source()
 
     @property
@@ -116,6 +117,12 @@ class ManifestDeclarativeSource(DeclarativeSource):
     def message_repository(self) -> MessageRepository:
         return self._message_repository
 
+    @property
+    def dynamic_streams(self) -> List[Dict[str, Any]]:
+        return self._dynamic_stream_configs(
+            manifest=self._source_config, config=self._config, with_dynamic_stream_name=True
+        )
+
     @property
     def connection_checker(self) -> ConnectionChecker:
         check = self._source_config["check"]
@@ -348,13 +355,16 @@ class ManifestDeclarativeSource(DeclarativeSource):
         return stream_configs
 
     def _dynamic_stream_configs(
-        self, manifest: Mapping[str, Any], config: Mapping[str, Any]
+        self,
+        manifest: Mapping[str, Any],
+        config: Mapping[str, Any],
+        with_dynamic_stream_name: Optional[bool] = None,
     ) -> List[Dict[str, Any]]:
         dynamic_stream_definitions: List[Dict[str, Any]] = manifest.get("dynamic_streams", [])
         dynamic_stream_configs: List[Dict[str, Any]] = []
         seen_dynamic_streams: Set[str] = set()
 
-        for dynamic_definition in dynamic_stream_definitions:
+        for dynamic_definition_index, dynamic_definition in enumerate(dynamic_stream_definitions):
             components_resolver_config = dynamic_definition["components_resolver"]
 
             if not components_resolver_config:
@@ -393,6 +403,11 @@ class ManifestDeclarativeSource(DeclarativeSource):
                 # Ensure that each stream is created with a unique name
                 name = dynamic_stream.get("name")
 
+                if with_dynamic_stream_name:
+                    dynamic_stream["dynamic_stream_name"] = dynamic_definition.get(
+                        "name", f"dynamic_stream_{dynamic_definition_index}"
+                    )
+
                 if not isinstance(name, str):
                     raise ValueError(
                         f"Expected stream name {name} to be a string, got {type(name)}."