make download_target_extractor optional

Author: dbgold17

airbyte_cdk/sources/declarative/parsers/model_to_component_factory.py

@@ -3603,11 +3603,15 @@ def _get_job_timeout() -> datetime.timedelta:
         status_extractor = self._create_component_from_model(
             model=model.status_extractor, decoder=decoder, config=config, name=name
         )
-        download_target_extractor = self._create_component_from_model(
-            model=model.download_target_extractor,
-            decoder=decoder,
-            config=config,
-            name=name,
+        download_target_extractor = (
+            self._create_component_from_model(
+                model=model.download_target_extractor,
+                decoder=decoder,
+                config=config,
+                name=name,
+            )
+            if model.download_target_extractor
+            else None
         )
 
         job_repository: AsyncJobRepository = AsyncHttpJobRepository(

airbyte_cdk/sources/declarative/requesters/http_job_repository.py

@@ -43,7 +43,7 @@ class AsyncHttpJobRepository(AsyncJobRepository):
     delete_requester: Optional[Requester]
     status_extractor: DpathExtractor
     status_mapping: Mapping[str, AsyncJobStatus]
-    download_target_extractor: DpathExtractor
+    download_target_extractor: Optional[DpathExtractor]
 
     # timeout for the job to be completed, passed from `polling_job_timeout`
     job_timeout: Optional[timedelta] = None
@@ -213,14 +213,14 @@ def fetch_records(self, job: AsyncJob) -> Iterable[Mapping[str, Any]]:
 
         """
 
-        for target_url in self._get_download_targets(job):
+        for download_target in self._get_download_targets(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,
-                    "download_target": target_url,
+                    "download_target": download_target,
                     "creation_response": self._get_creation_response_interpolation_context(job),
                     "polling_response": self._get_polling_response_interpolation_context(job),
                 },
@@ -332,9 +332,18 @@ def _get_create_job_stream_slice(self, job: AsyncJob) -> StreamSlice:
         )
 
     def _get_download_targets(self, job: AsyncJob) -> Iterable[str]:
-        if not self.download_target_requester:
-            url_response = self._polling_job_response_by_id[job.api_job_id()]
-        else:
+        """Returns an iterable of strings to help target requests for downloading async jobs."""
+        # If neither download_target_extractor nor download_target_requester are provided,return a single empty string 
+        # to express the need to make a single download request without any download_target value
+        if not self.download_target_extractor and not self.download_target_requester:
+            lazy_log(
+                LOGGER,
+                logging.DEBUG,
+                lambda: "No download_target_extractor or download_target_requester provided. Using fallback behavior for single download request without download_target.",
+            )
+            return [""]
+
+        if self.download_target_requester:
             stream_slice: StreamSlice = StreamSlice(
                 partition={},
                 cursor_slice={},
@@ -348,5 +357,8 @@ def _get_download_targets(self, job: AsyncJob) -> Iterable[str]:
                     internal_message="Always expect a response or an exception from download_target_requester",
                     failure_type=FailureType.system_error,
                 )
-
+        else:
+            # if no download_target_requester is provided, we extract directly from the polling response
+            url_response = self._polling_job_response_by_id[job.api_job_id()]
+       
         yield from self.download_target_extractor.extract_records(url_response)  # type: ignore # we expect download_target_extractor to always return list of strings