medicafe 0.250410.0__tar.gz → 0.250430.0__tar.gz

{medicafe-0.250410.0 → medicafe-0.250430.0}/MediBot/MediBot_Preprocessor_lib.py

@@ -436,12 +436,17 @@ def update_diagnosis_codes(csv_data):
         min_surgery_date = min(surgery_dates)
         max_surgery_date = max(surgery_dates)
         
-        # Apply a ±5-day margin to the surgery dates
-        margin = timedelta(days=5)
+        # Apply a ±8-day margin to the surgery dates... Increased from 5 days.
+        margin = timedelta(days=8)
         threshold_start = min_surgery_date - margin
         threshold_end = max_surgery_date + margin
         
-        MediLink_ConfigLoader.log("Processing DOCX files modified between {} and {}.".format(threshold_start, threshold_end), level="INFO")
+        # TODO (Low) This is a bad idea. We need a better way to handle this because it leaves 
+        # us with a situation where if we take 'too long' to download the DOCX files, it will presume that the DOCX files are out of range because 
+        # the modfied date is a bad proxy for the date of the surgery which would be contained inside the DOCX file. The processing overhead for extracting the
+        # date of the surgery from the DOCX file is non-trivial and computationally expensive so we need a smarter way to handle this.
+
+        MediLink_ConfigLoader.log("BAD IDEA: Processing DOCX files modified between {} and {}.".format(threshold_start, threshold_end), level="INFO")
 
         # Gather all relevant DOCX files in the specified directory
         docx_files = [