Try a smaller chunk size - for testing

django/parse_m2/initiate_post_parsing.py

@@ -20,8 +20,8 @@ def calculate_date_range(event: Metro2Event):
 def associate_previous_records(event: Metro2Event):
     logger = logging.getLogger('parse_m2.associate_previous_records')
 
-    logger.info("First, make sure all previous_values pointers are empty")
-    event.get_all_account_activity().update(previous_values_id=None)
+    # logger.info("First, make sure all previous_values pointers are empty")
+    # event.get_all_account_activity().update(previous_values_id=None)
 
     logger.info(f"Beginning to update all records for event: {event.id}")
     query_sql = """

django/parse_m2/m2_parser.py

@@ -22,9 +22,9 @@
 class M2FileParser:
     # Parser version is saved on each file record.
     # Increment this version for all updates to parser functionality.
-    parser_version = "3.1"
+    parser_version = "3.1.2"
 
-    chunk_size = 2000  # TODO: determine a good number for this
+    chunk_size = 500  # Try a smaller chunk size
     any_non_whitespace = r'\S'
 
     activity_date = None

django/parse_m2/parse_utils.py

@@ -117,7 +117,7 @@ def decode_if_needed(input: any) -> str:
         return input
     elif isinstance(input, bytes):
         try:
-            return input.decode('utf-8', errors='replace').replace('\x00', '\uFFFD')
+            return input.decode('utf-8', errors='replace')
         except (UnicodeDecodeError, AttributeError) as e:
             raise UnreadableLineException(f"Decode failed: {e}") from e
     else: