airbyte-cdk 6.20.2.dev0__py3-none-any.whl → 6.21.0__py3-none-any.whl

airbyte_cdk/sources/declarative/parsers/model_to_component_factory.py

@@ -54,7 +54,7 @@ from airbyte_cdk.sources.declarative.auth.token_provider import (
     SessionTokenProvider,
     TokenProvider,
 )
-from airbyte_cdk.sources.declarative.checks import CheckStream
+from airbyte_cdk.sources.declarative.checks import CheckDynamicStream, CheckStream
 from airbyte_cdk.sources.declarative.concurrency_level import ConcurrencyLevel
 from airbyte_cdk.sources.declarative.datetime import MinMaxDatetime
 from airbyte_cdk.sources.declarative.declarative_stream import DeclarativeStream
@@ -66,12 +66,15 @@ from airbyte_cdk.sources.declarative.decoders import (
     JsonlDecoder,
     PaginationDecoderDecorator,
     XmlDecoder,
+    ZipfileDecoder,
 )
 from airbyte_cdk.sources.declarative.decoders.composite_raw_decoder import (
     CompositeRawDecoder,
     CsvParser,
     GzipParser,
     JsonLineParser,
+    JsonParser,
+    Parser,
 )
 from airbyte_cdk.sources.declarative.extractors import (
     DpathExtractor,
@@ -84,8 +87,6 @@ from airbyte_cdk.sources.declarative.extractors.record_filter import (
 )
 from airbyte_cdk.sources.declarative.incremental import (
     ChildPartitionResumableFullRefreshCursor,
-    ConcurrentCursorFactory,
-    ConcurrentPerPartitionCursor,
     CursorFactory,
     DatetimeBasedCursor,
     DeclarativeCursor,
@@ -123,6 +124,9 @@ from airbyte_cdk.sources.declarative.models.declarative_component_schema import
 from airbyte_cdk.sources.declarative.models.declarative_component_schema import (
     BearerAuthenticator as BearerAuthenticatorModel,
 )
+from airbyte_cdk.sources.declarative.models.declarative_component_schema import (
+    CheckDynamicStream as CheckDynamicStreamModel,
+)
 from airbyte_cdk.sources.declarative.models.declarative_component_schema import (
     CheckStream as CheckStreamModel,
 )
@@ -249,6 +253,9 @@ from airbyte_cdk.sources.declarative.models.declarative_component_schema import
 from airbyte_cdk.sources.declarative.models.declarative_component_schema import (
     JsonLineParser as JsonLineParserModel,
 )
+from airbyte_cdk.sources.declarative.models.declarative_component_schema import (
+    JsonParser as JsonParserModel,
+)
 from airbyte_cdk.sources.declarative.models.declarative_component_schema import (
     JwtAuthenticator as JwtAuthenticatorModel,
 )
@@ -350,6 +357,9 @@ from airbyte_cdk.sources.declarative.models.declarative_component_schema import
 from airbyte_cdk.sources.declarative.models.declarative_component_schema import (
     XmlDecoder as XmlDecoderModel,
 )
+from airbyte_cdk.sources.declarative.models.declarative_component_schema import (
+    ZipfileDecoder as ZipfileDecoderModel,
+)
 from airbyte_cdk.sources.declarative.partition_routers import (
     CartesianProductStreamSlicer,
     ListPartitionRouter,
@@ -440,7 +450,6 @@ from airbyte_cdk.sources.message import (
     InMemoryMessageRepository,
     LogAppenderMessageRepositoryDecorator,
     MessageRepository,
-    NoopMessageRepository,
 )
 from airbyte_cdk.sources.streams.concurrent.cursor import ConcurrentCursor, CursorField
 from airbyte_cdk.sources.streams.concurrent.state_converters.datetime_stream_state_converter import (
@@ -491,6 +500,7 @@ class ModelToComponentFactory:
             BasicHttpAuthenticatorModel: self.create_basic_http_authenticator,
             BearerAuthenticatorModel: self.create_bearer_authenticator,
             CheckStreamModel: self.create_check_stream,
+            CheckDynamicStreamModel: self.create_check_dynamic_stream,
             CompositeErrorHandlerModel: self.create_composite_error_handler,
             CompositeRawDecoderModel: self.create_composite_raw_decoder,
             ConcurrencyLevelModel: self.create_concurrency_level,
@@ -525,6 +535,7 @@ class ModelToComponentFactory:
             JsonDecoderModel: self.create_json_decoder,
             JsonlDecoderModel: self.create_jsonl_decoder,
             JsonLineParserModel: self.create_json_line_parser,
+            JsonParserModel: self.create_json_parser,
             GzipJsonDecoderModel: self.create_gzipjson_decoder,
             GzipParserModel: self.create_gzip_parser,
             KeysToLowerModel: self.create_keys_to_lower_transformation,
@@ -564,6 +575,7 @@ class ModelToComponentFactory:
             ConfigComponentsResolverModel: self.create_config_components_resolver,
             StreamConfigModel: self.create_stream_config,
             ComponentMappingDefinitionModel: self.create_components_mapping_definition,
+            ZipfileDecoderModel: self.create_zipfile_decoder,
         }
 
         # Needed for the case where we need to perform a second parse on the fields of a custom component
@@ -843,6 +855,12 @@ class ModelToComponentFactory:
     def create_check_stream(model: CheckStreamModel, config: Config, **kwargs: Any) -> CheckStream:
         return CheckStream(stream_names=model.stream_names, parameters={})
 
+    @staticmethod
+    def create_check_dynamic_stream(
+        model: CheckDynamicStreamModel, config: Config, **kwargs: Any
+    ) -> CheckDynamicStream:
+        return CheckDynamicStream(stream_count=model.stream_count, parameters={})
+
     def create_composite_error_handler(
         self, model: CompositeErrorHandlerModel, config: Config, **kwargs: Any
     ) -> CompositeErrorHandler:
@@ -874,8 +892,6 @@ class ModelToComponentFactory:
         stream_namespace: Optional[str],
         config: Config,
         stream_state: MutableMapping[str, Any],
-        message_repository: Optional[MessageRepository] = None,
-        runtime_lookback_window: Optional[int] = None,
         **kwargs: Any,
     ) -> ConcurrentCursor:
         component_type = component_definition.get("type")
@@ -933,11 +949,6 @@ class ModelToComponentFactory:
             if evaluated_lookback_window:
                 lookback_window = parse_duration(evaluated_lookback_window)
 
-        if runtime_lookback_window and lookback_window:
-            lookback_window = max(lookback_window, runtime_lookback_window)
-        elif runtime_lookback_window:
-            lookback_window = runtime_lookback_window
-
         connector_state_converter: DateTimeStreamStateConverter
         connector_state_converter = CustomFormatConcurrentStreamStateConverter(
             datetime_format=datetime_format,
@@ -1016,7 +1027,7 @@ class ModelToComponentFactory:
             stream_name=stream_name,
             stream_namespace=stream_namespace,
             stream_state=stream_state,
-            message_repository=message_repository or self._message_repository,
+            message_repository=self._message_repository,
             connector_state_manager=state_manager,
             connector_state_converter=connector_state_converter,
             cursor_field=cursor_field,
@@ -1028,63 +1039,6 @@ class ModelToComponentFactory:
             cursor_granularity=cursor_granularity,
         )
 
-    def create_concurrent_cursor_from_perpartition_cursor(
-        self,
-        state_manager: ConnectorStateManager,
-        model_type: Type[BaseModel],
-        component_definition: ComponentDefinition,
-        stream_name: str,
-        stream_namespace: Optional[str],
-        config: Config,
-        stream_state: MutableMapping[str, Any],
-        partition_router: PartitionRouter,
-        **kwargs: Any,
-    ) -> ConcurrentPerPartitionCursor:
-        component_type = component_definition.get("type")
-        if component_definition.get("type") != model_type.__name__:
-            raise ValueError(
-                f"Expected manifest component of type {model_type.__name__}, but received {component_type} instead"
-            )
-
-        datetime_based_cursor_model = model_type.parse_obj(component_definition)
-
-        if not isinstance(datetime_based_cursor_model, DatetimeBasedCursorModel):
-            raise ValueError(
-                f"Expected {model_type.__name__} component, but received {datetime_based_cursor_model.__class__.__name__}"
-            )
-
-        interpolated_cursor_field = InterpolatedString.create(
-            datetime_based_cursor_model.cursor_field,
-            parameters=datetime_based_cursor_model.parameters or {},
-        )
-        cursor_field = CursorField(interpolated_cursor_field.eval(config=config))
-
-        # Create the cursor factory
-        cursor_factory = ConcurrentCursorFactory(
-            partial(
-                self.create_concurrent_cursor_from_datetime_based_cursor,
-                state_manager=state_manager,
-                model_type=model_type,
-                component_definition=component_definition,
-                stream_name=stream_name,
-                stream_namespace=stream_namespace,
-                config=config,
-                message_repository=NoopMessageRepository(),
-            )
-        )
-
-        # Return the concurrent cursor and state converter
-        return ConcurrentPerPartitionCursor(
-            cursor_factory=cursor_factory,
-            partition_router=partition_router,
-            stream_name=stream_name,
-            stream_namespace=stream_namespace,
-            stream_state=stream_state,
-            message_repository=self._message_repository,  # type: ignore
-            connector_state_manager=state_manager,
-            cursor_field=cursor_field,
-        )
-
     @staticmethod
     def create_constant_backoff_strategy(
         model: ConstantBackoffStrategyModel, config: Config, **kwargs: Any
@@ -1099,17 +1053,17 @@ class ModelToComponentFactory:
         self, model: CursorPaginationModel, config: Config, decoder: Decoder, **kwargs: Any
     ) -> CursorPaginationStrategy:
         if isinstance(decoder, PaginationDecoderDecorator):
-            if not isinstance(decoder.decoder, (JsonDecoder, XmlDecoder)):
-                raise ValueError(
-                    f"Provided decoder of {type(decoder.decoder)=} is not supported. Please set JsonDecoder or XmlDecoder instead."
-                )
+            inner_decoder = decoder.decoder
+        else:
+            inner_decoder = decoder
+            decoder = PaginationDecoderDecorator(decoder=decoder)
+
+        if self._is_supported_decoder_for_pagination(inner_decoder):
             decoder_to_use = decoder
         else:
-            if not isinstance(decoder, (JsonDecoder, XmlDecoder)):
-                raise ValueError(
-                    f"Provided decoder of {type(decoder)=} is not supported. Please set JsonDecoder or XmlDecoder instead."
-                )
-            decoder_to_use = PaginationDecoderDecorator(decoder=decoder)
+            raise ValueError(
+                self._UNSUPPORTED_DECODER_ERROR.format(decoder_type=type(inner_decoder))
+            )
 
         return CursorPaginationStrategy(
             cursor_value=model.cursor_value,
@@ -1367,15 +1321,18 @@ class ModelToComponentFactory:
                 raise ValueError(
                     "Unsupported Slicer is used. PerPartitionWithGlobalCursor should be used here instead"
                 )
-            cursor = (
-                combined_slicers
-                if isinstance(
-                    combined_slicers, (PerPartitionWithGlobalCursor, GlobalSubstreamCursor)
-                )
-                else self._create_component_from_model(model=model.incremental_sync, config=config)
-            )
-
-            client_side_incremental_sync = {"cursor": cursor}
+            client_side_incremental_sync = {
+                "date_time_based_cursor": self._create_component_from_model(
+                    model=model.incremental_sync, config=config
+                ),
+                "substream_cursor": (
+                    combined_slicers
+                    if isinstance(
+                        combined_slicers, (PerPartitionWithGlobalCursor, GlobalSubstreamCursor)
+                    )
+                    else None
+                ),
+            }
 
         if model.incremental_sync and isinstance(model.incremental_sync, DatetimeBasedCursorModel):
             cursor_model = model.incremental_sync
@@ -1579,11 +1536,10 @@ class ModelToComponentFactory:
         cursor_used_for_stop_condition: Optional[DeclarativeCursor] = None,
     ) -> Union[DefaultPaginator, PaginatorTestReadDecorator]:
         if decoder:
-            if not isinstance(decoder, (JsonDecoder, XmlDecoder)):
-                raise ValueError(
-                    f"Provided decoder of {type(decoder)=} is not supported. Please set JsonDecoder or XmlDecoder instead."
-                )
-            decoder_to_use = PaginationDecoderDecorator(decoder=decoder)
+            if self._is_supported_decoder_for_pagination(decoder):
+                decoder_to_use = PaginationDecoderDecorator(decoder=decoder)
+            else:
+                raise ValueError(self._UNSUPPORTED_DECODER_ERROR.format(decoder_type=type(decoder)))
         else:
             decoder_to_use = PaginationDecoderDecorator(decoder=JsonDecoder(parameters={}))
         page_size_option = (
@@ -1745,7 +1701,11 @@ class ModelToComponentFactory:
 
     @staticmethod
     def create_types_map(model: TypesMapModel, **kwargs: Any) -> TypesMap:
-        return TypesMap(target_type=model.target_type, current_type=model.current_type)
+        return TypesMap(
+            target_type=model.target_type,
+            current_type=model.current_type,
+            condition=model.condition if model.condition is not None else "True",
+        )
 
     def create_schema_type_identifier(
         self, model: SchemaTypeIdentifierModel, config: Config, **kwargs: Any
@@ -1812,6 +1772,11 @@ class ModelToComponentFactory:
     def create_json_decoder(model: JsonDecoderModel, config: Config, **kwargs: Any) -> JsonDecoder:
         return JsonDecoder(parameters={})
 
+    @staticmethod
+    def create_json_parser(model: JsonParserModel, config: Config, **kwargs: Any) -> JsonParser:
+        encoding = model.encoding if model.encoding else "utf-8"
+        return JsonParser(encoding=encoding)
+
     @staticmethod
     def create_jsonl_decoder(
         model: JsonlDecoderModel, config: Config, **kwargs: Any
@@ -1840,6 +1805,12 @@ class ModelToComponentFactory:
     ) -> GzipJsonDecoder:
         return GzipJsonDecoder(parameters={}, encoding=model.encoding)
 
+    def create_zipfile_decoder(
+        self, model: ZipfileDecoderModel, config: Config, **kwargs: Any
+    ) -> ZipfileDecoder:
+        parser = self._create_component_from_model(model=model.parser, config=config)
+        return ZipfileDecoder(parser=parser)
+
     def create_gzip_parser(
         self, model: GzipParserModel, config: Config, **kwargs: Any
     ) -> GzipParser:
@@ -1949,21 +1920,33 @@ class ModelToComponentFactory:
                 expires_in_name=InterpolatedString.create(
                     model.expires_in_name or "expires_in", parameters=model.parameters or {}
                 ).eval(config),
+                client_id_name=InterpolatedString.create(
+                    model.client_id_name or "client_id", parameters=model.parameters or {}
+                ).eval(config),
                 client_id=InterpolatedString.create(
                     model.client_id, parameters=model.parameters or {}
                 ).eval(config),
+                client_secret_name=InterpolatedString.create(
+                    model.client_secret_name or "client_secret", parameters=model.parameters or {}
+                ).eval(config),
                 client_secret=InterpolatedString.create(
                     model.client_secret, parameters=model.parameters or {}
                 ).eval(config),
                 access_token_config_path=model.refresh_token_updater.access_token_config_path,
                 refresh_token_config_path=model.refresh_token_updater.refresh_token_config_path,
                 token_expiry_date_config_path=model.refresh_token_updater.token_expiry_date_config_path,
+                grant_type_name=InterpolatedString.create(
+                    model.grant_type_name or "grant_type", parameters=model.parameters or {}
+                ).eval(config),
                 grant_type=InterpolatedString.create(
                     model.grant_type or "refresh_token", parameters=model.parameters or {}
                 ).eval(config),
                 refresh_request_body=InterpolatedMapping(
                     model.refresh_request_body or {}, parameters=model.parameters or {}
                 ).eval(config),
+                refresh_request_headers=InterpolatedMapping(
+                    model.refresh_request_headers or {}, parameters=model.parameters or {}
+                ).eval(config),
                 scopes=model.scopes,
                 token_expiry_date_format=model.token_expiry_date_format,
                 message_repository=self._message_repository,
@@ -1975,11 +1958,16 @@ class ModelToComponentFactory:
         return DeclarativeOauth2Authenticator(  # type: ignore
             access_token_name=model.access_token_name or "access_token",
             access_token_value=model.access_token_value,
+            client_id_name=model.client_id_name or "client_id",
             client_id=model.client_id,
+            client_secret_name=model.client_secret_name or "client_secret",
             client_secret=model.client_secret,
             expires_in_name=model.expires_in_name or "expires_in",
+            grant_type_name=model.grant_type_name or "grant_type",
             grant_type=model.grant_type or "refresh_token",
             refresh_request_body=model.refresh_request_body,
+            refresh_request_headers=model.refresh_request_headers,
+            refresh_token_name=model.refresh_token_name or "refresh_token",
             refresh_token=model.refresh_token,
             scopes=model.scopes,
             token_expiry_date=model.token_expiry_date,
@@ -1991,22 +1979,22 @@ class ModelToComponentFactory:
             message_repository=self._message_repository,
         )
 
-    @staticmethod
     def create_offset_increment(
-        model: OffsetIncrementModel, config: Config, decoder: Decoder, **kwargs: Any
+        self, model: OffsetIncrementModel, config: Config, decoder: Decoder, **kwargs: Any
     ) -> OffsetIncrement:
         if isinstance(decoder, PaginationDecoderDecorator):
-            if not isinstance(decoder.decoder, (JsonDecoder, XmlDecoder)):
-                raise ValueError(
-                    f"Provided decoder of {type(decoder.decoder)=} is not supported. Please set JsonDecoder or XmlDecoder instead."
-                )
+            inner_decoder = decoder.decoder
+        else:
+            inner_decoder = decoder
+            decoder = PaginationDecoderDecorator(decoder=decoder)
+
+        if self._is_supported_decoder_for_pagination(inner_decoder):
             decoder_to_use = decoder
         else:
-            if not isinstance(decoder, (JsonDecoder, XmlDecoder)):
-                raise ValueError(
-                    f"Provided decoder of {type(decoder)=} is not supported. Please set JsonDecoder or XmlDecoder instead."
-                )
-            decoder_to_use = PaginationDecoderDecorator(decoder=decoder)
+            raise ValueError(
+                self._UNSUPPORTED_DECODER_ERROR.format(decoder_type=type(inner_decoder))
+            )
+
         return OffsetIncrement(
             page_size=model.page_size,
             config=config,
@@ -2191,7 +2179,7 @@ class ModelToComponentFactory:
         if (
             not isinstance(stream_slicer, DatetimeBasedCursor)
             or type(stream_slicer) is not DatetimeBasedCursor
-        ) and not isinstance(stream_slicer, PerPartitionWithGlobalCursor):
+        ):
             # Many of the custom component implementations of DatetimeBasedCursor override get_request_params() (or other methods).
             # Because we're decoupling RequestOptionsProvider from the Cursor, custom components will eventually need to reimplement
             # their own RequestOptionsProvider. However, right now the existing StreamSlicer/Cursor still can act as the SimpleRetriever's
@@ -2351,7 +2339,7 @@ class ModelToComponentFactory:
                 extractor=download_extractor,
                 name=name,
                 record_filter=None,
-                transformations=[],
+                transformations=transformations,
                 schema_normalization=TypeTransformer(TransformConfig.NoTransform),
                 config=config,
                 parameters={},
@@ -2388,6 +2376,16 @@ class ModelToComponentFactory:
             if model.delete_requester
             else None
         )
+        url_requester = (
+            self._create_component_from_model(
+                model=model.url_requester,
+                decoder=decoder,
+                config=config,
+                name=f"job extract_url - {name}",
+            )
+            if model.url_requester
+            else None
+        )
         status_extractor = self._create_component_from_model(
             model=model.status_extractor, decoder=decoder, config=config, name=name
         )
@@ -2398,6 +2396,7 @@ class ModelToComponentFactory:
             creation_requester=creation_requester,
             polling_requester=polling_requester,
             download_retriever=download_retriever,
+            url_requester=url_requester,
             abort_requester=abort_requester,
             delete_requester=delete_requester,
             status_extractor=status_extractor,
@@ -2595,3 +2594,25 @@ class ModelToComponentFactory:
             components_mapping=components_mapping,
             parameters=model.parameters or {},
         )
+
+    _UNSUPPORTED_DECODER_ERROR = (
+        "Specified decoder of {decoder_type} is not supported for pagination."
+        "Please set as `JsonDecoder`, `XmlDecoder`, or a `CompositeRawDecoder` with an inner_parser of `JsonParser` or `GzipParser` instead."
+        "If using `GzipParser`, please ensure that the lowest level inner_parser is a `JsonParser`."
+    )
+
+    def _is_supported_decoder_for_pagination(self, decoder: Decoder) -> bool:
+        if isinstance(decoder, (JsonDecoder, XmlDecoder)):
+            return True
+        elif isinstance(decoder, CompositeRawDecoder):
+            return self._is_supported_parser_for_pagination(decoder.parser)
+        else:
+            return False
+
+    def _is_supported_parser_for_pagination(self, parser: Parser) -> bool:
+        if isinstance(parser, JsonParser):
+            return True
+        elif isinstance(parser, GzipParser):
+            return isinstance(parser.inner_parser, JsonParser)
+        else:
+            return False

airbyte_cdk/sources/declarative/requesters/README.md

@@ -0,0 +1,56 @@
+# AsyncHttpJobRepository sequence diagram
+
+- Components marked as optional are not required and can be ignored.
+- if `url_requester` is not provided, `urls_extractor` will get urls from the `polling_job_response`
+- interpolation_context, e.g. `create_job_response` or `polling_job_response` can be obtained from stream_slice
+
+```mermaid
+---
+title: AsyncHttpJobRepository Sequence Diagram
+---
+sequenceDiagram
+    participant AsyncHttpJobRepository as AsyncOrchestrator
+    participant CreationRequester as creation_requester
+    participant PollingRequester as polling_requester
+    participant UrlRequester as url_requester (Optional)
+    participant DownloadRetriever as download_retriever
+    participant AbortRequester as abort_requester (Optional)
+    participant DeleteRequester as delete_requester (Optional)
+    participant Reporting Server as Async Reporting Server
+
+    AsyncHttpJobRepository ->> CreationRequester: Initiate job creation
+    CreationRequester ->> Reporting Server: Create job request
+    Reporting Server -->> CreationRequester: Job ID response
+    CreationRequester -->> AsyncHttpJobRepository: Job ID
+
+    loop Poll for job status
+        AsyncHttpJobRepository ->> PollingRequester: Check job status
+        PollingRequester ->> Reporting Server: Status request (interpolation_context: `create_job_response`)
+        Reporting Server -->> PollingRequester: Status response
+        PollingRequester -->> AsyncHttpJobRepository: Job status
+    end
+
+    alt Status: Ready
+        AsyncHttpJobRepository ->> UrlRequester: Request download URLs (if applicable)
+        UrlRequester ->> Reporting Server: URL request (interpolation_context: `polling_job_response`)
+        Reporting Server -->> UrlRequester: Download URLs
+        UrlRequester -->> AsyncHttpJobRepository: Download URLs
+
+        AsyncHttpJobRepository ->> DownloadRetriever: Download reports
+        DownloadRetriever ->> Reporting Server: Retrieve report data (interpolation_context: `url`)
+        Reporting Server -->> DownloadRetriever: Report data
+        DownloadRetriever -->> AsyncHttpJobRepository: Report data
+    else Status: Failed
+        AsyncHttpJobRepository ->> AbortRequester: Send abort request
+        AbortRequester ->> Reporting Server: Abort job
+        Reporting Server -->> AbortRequester: Abort confirmation
+        AbortRequester -->> AsyncHttpJobRepository: Confirmation
+    end
+
+    AsyncHttpJobRepository ->> DeleteRequester: Send delete job request
+    DeleteRequester ->> Reporting Server: Delete job
+    Reporting Server -->> DeleteRequester: Deletion confirmation
+    DeleteRequester -->> AsyncHttpJobRepository: Confirmation
+
+
+```

airbyte_cdk/sources/declarative/requesters/http_job_repository.py

@@ -31,6 +31,10 @@ LOGGER = logging.getLogger("airbyte")
 
 @dataclass
 class AsyncHttpJobRepository(AsyncJobRepository):
+    """
+    See Readme file for more details about flow.
+    """
+
     creation_requester: Requester
     polling_requester: Requester
     download_retriever: SimpleRetriever
@@ -44,6 +48,9 @@ class AsyncHttpJobRepository(AsyncJobRepository):
     record_extractor: RecordExtractor = field(
         init=False, repr=False, default_factory=lambda: ResponseToFileExtractor({})
     )
+    url_requester: Optional[Requester] = (
+        None  # use it in case polling_requester provides some <id> and extra request is needed to obtain list of urls to download from
+    )
 
     def __post_init__(self) -> None:
         self._create_job_response_by_id: Dict[str, Response] = {}
@@ -186,10 +193,13 @@ class AsyncHttpJobRepository(AsyncJobRepository):
 
         """
 
-        for url in self.urls_extractor.extract_records(
-            self._polling_job_response_by_id[job.api_job_id()]
-        ):
-            stream_slice: StreamSlice = StreamSlice(partition={"url": url}, cursor_slice={})
+        for url in self._get_download_url(job):
+            job_slice = job.job_parameters()
+            stream_slice = StreamSlice(
+                partition=job_slice.partition,
+                cursor_slice=job_slice.cursor_slice,
+                extra_fields={**job_slice.extra_fields, "url": url},
+            )
             for message in self.download_retriever.read_records({}, stream_slice):
                 if isinstance(message, Record):
                     yield message.data
@@ -226,3 +236,22 @@ class AsyncHttpJobRepository(AsyncJobRepository):
             cursor_slice={},
         )
         return stream_slice
+
+    def _get_download_url(self, job: AsyncJob) -> Iterable[str]:
+        if not self.url_requester:
+            url_response = self._polling_job_response_by_id[job.api_job_id()]
+        else:
+            stream_slice: StreamSlice = StreamSlice(
+                partition={
+                    "polling_job_response": self._polling_job_response_by_id[job.api_job_id()]
+                },
+                cursor_slice={},
+            )
+            url_response = self.url_requester.send_request(stream_slice=stream_slice)  # type: ignore # we expect url_requester to always be presented, otherwise raise an exception as we cannot proceed with the report
+            if not url_response:
+                raise AirbyteTracedException(
+                    internal_message="Always expect a response or an exception from url_requester",
+                    failure_type=FailureType.system_error,
+                )
+
+        yield from self.urls_extractor.extract_records(url_response)  # type: ignore # we expect urls_extractor to always return list of strings

airbyte_cdk/sources/declarative/retrievers/simple_retriever.py

@@ -160,7 +160,7 @@ class SimpleRetriever(Retriever):
             stream_slice,
             next_page_token,
             self._paginator.get_request_headers,
-            self.request_option_provider.get_request_headers,
+            self.stream_slicer.get_request_headers,
         )
         if isinstance(headers, str):
             raise ValueError("Request headers cannot be a string")

airbyte_cdk/sources/declarative/schema/dynamic_schema_loader.py

@@ -10,6 +10,7 @@ from typing import Any, List, Mapping, MutableMapping, Optional, Union
 import dpath
 from typing_extensions import deprecated
 
+from airbyte_cdk.sources.declarative.interpolation.interpolated_boolean import InterpolatedBoolean
 from airbyte_cdk.sources.declarative.interpolation.interpolated_string import InterpolatedString
 from airbyte_cdk.sources.declarative.retrievers.retriever import Retriever
 from airbyte_cdk.sources.declarative.schema.schema_loader import SchemaLoader
@@ -53,6 +54,7 @@ class TypesMap:
 
     target_type: Union[List[str], str]
     current_type: Union[List[str], str]
+    condition: Optional[str]
 
 
 @deprecated("This class is experimental. Use at your own risk.", category=ExperimentalClassWarning)
@@ -177,7 +179,7 @@ class DynamicSchemaLoader(SchemaLoader):
             if field_type_path
             else "string"
         )
-        mapped_field_type = self._replace_type_if_not_valid(raw_field_type)
+        mapped_field_type = self._replace_type_if_not_valid(raw_field_type, raw_schema)
         if (
             isinstance(mapped_field_type, list)
             and len(mapped_field_type) == 2
@@ -194,14 +196,22 @@ class DynamicSchemaLoader(SchemaLoader):
             )
 
     def _replace_type_if_not_valid(
-        self, field_type: Union[List[str], str]
+        self,
+        field_type: Union[List[str], str],
+        raw_schema: MutableMapping[str, Any],
     ) -> Union[List[str], str]:
         """
         Replaces a field type if it matches a type mapping in `types_map`.
         """
         if self.schema_type_identifier.types_mapping:
             for types_map in self.schema_type_identifier.types_mapping:
-                if field_type == types_map.current_type:
+                # conditional is optional param, setting to true if not provided
+                condition = InterpolatedBoolean(
+                    condition=types_map.condition if types_map.condition is not None else "True",
+                    parameters={},
+                ).eval(config=self.config, raw_schema=raw_schema)
+
+                if field_type == types_map.current_type and condition:
                     return types_map.target_type
         return field_type
 

airbyte_cdk/sources/file_based/config/abstract_file_based_spec.py

@@ -31,6 +31,17 @@ class DeliverRawFiles(BaseModel):
 
     delivery_type: Literal["use_file_transfer"] = Field("use_file_transfer", const=True)
 
+    preserve_directory_structure: bool = Field(
+        title="Preserve Sub-Directories in File Paths",
+        description=(
+            "If enabled, sends subdirectory folder structure "
+            "along with source file names to the destination. "
+            "Otherwise, files will be synced by their names only. "
+            "This option is ignored when file-based replication is not enabled."
+        ),
+        default=True,
+    )
+
 
 class AbstractFileBasedSpec(BaseModel):
     """