Better refactor

tests/a_scripts/data_prep/test_convert_text_to_mds.py

@@ -91,7 +91,7 @@ def test_single_and_multi_process(merge_shard_groups: Mock,
     maybe_create_object_store_from_uri.return_value = mock_object_store
     parse_uri.return_value = ('s3', 'fake-test-bucket', str(remote_folder))
 
-    def call_convert_text_to_mds(processes: int) -> None:
+    def call_convert_text_to_mds() -> None:
         convert_text_to_mds(
             tokenizer_name=tokenizer_name,
             output_folder=f's3://fake-test-output-path',
@@ -106,7 +106,7 @@ def call_convert_text_to_mds(processes: int) -> None:
             reprocess=False,
         )
 
-    call_convert_text_to_mds(processes=processes)
+    call_convert_text_to_mds()
 
     # Check call counts
     assert download_and_convert.call_count == processes  # called once per process
@@ -128,7 +128,7 @@ def call_convert_text_to_mds(processes: int) -> None:
     _assert_files_exist(prefix=remote_folder,
                         files=['index.json', DONE_FILENAME] + shards)
 
-    call_convert_text_to_mds(processes=processes)
+    call_convert_text_to_mds()
 
     # Check call counts
     assert download_and_convert.call_count == processes  # No changes because we shoudn't reprocess
@@ -141,7 +141,7 @@ def call_convert_text_to_mds(processes: int) -> None:
     mock_object_store = Mock(wraps=object_store)
     maybe_create_object_store_from_uri.return_value = mock_object_store
 
-    call_convert_text_to_mds(processes=processes)
+    call_convert_text_to_mds()
 
     # Check call counts
     assert download_and_convert.call_count == processes * 2  # called once per process