Merge remote-tracking branch 'origin/main' into 15-containerization

config/config.yml

@@ -14,7 +14,7 @@ ELASTICSEARCH:
     SCHEME: http
     INDEX: dane-index-k8s
 FILE_SYSTEM:
-    BASE_MOUNT: data # data when running locally
+    BASE_MOUNT: /data # data when running locally
     INPUT_DIR: input-files
     OUTPUT_DIR: output-files
 INPUT:

models.py

@@ -12,8 +12,6 @@ class CallbackResponse(TypedDict):
 
 # These are the types of output this worker (possibly) provides (depending on configuration)
 class OutputType(Enum):
-    # name of output type, should just have a significant name, no other restrictions
-    # (as far as I understand)
     TRANSCRIPT = "transcript"
     PROVENANCE = "provenance"  # produced by provenance.py
 
@@ -23,7 +21,7 @@ class WhisperASRInput:
     state: int  # HTTP status code
     message: str  # error/success message
     source_id: str = ""  # <program ID>__<carrier ID>
-    input_file_path: str = ""  # where the audio was downloaded from
+    input_file_path: str = ""  # where the audio was downloaded to
     provenance: Optional[Provenance] = None  # mostly: how long did it take to download
 
 

worker.py

@@ -79,7 +79,6 @@ def callback(self, task: Task, doc: Document) -> CallbackResponse:
             )
         return processing_result
 
-    # TODO adapt
     def save_to_dane_index(
         self,
         doc: Document,
@@ -88,7 +87,6 @@ def save_to_dane_index(
         provenance: Provenance,
     ) -> None:
         logger.info("saving results to DANE, task id={0}".format(task._id))
-        # TODO figure out the multiple lines per transcript (refresh my memory)
         r = Result(
             self.generator,
             payload={