updated the processor.py to fix the dates format error (#2)

- updated the processor.py to fix the dates format error (ed5bf38a155bae67cb3fbcb8861bb22955ed16ad)


Co-authored-by: Bismark Bantar <afanyu237@users.noreply.huggingface.co>

preprocessor.py

@@ -85,8 +85,39 @@ def preprocess(data):
             if sender and sender.strip().lower() != "system":
                 filtered_messages.append(f"{sender.strip()}: {entry['Message']}")
                 valid_dates.append(f"{entry['Date']}, {entry['Time'].replace(' ', ' ')}")
+    print("-_____--------------__________----------_____________----------______________")
+    def convert_to_target_format(date_str):
+        try:
+            # Attempt to parse the original date string
+            dt = datetime.strptime(date_str, '%d/%m/%Y, %H:%M')
+        except ValueError:
+            # Return the original date string if parsing fails
+            return date_str
+
+        # Extract components without leading zeros
+        month = dt.month
+        day = dt.day
+        year_short = dt.strftime('%y')  # Last two digits of the year
+        
+        # Convert to 12-hour format and determine AM/PM
+        hour_12 = dt.hour % 12
+        if hour_12 == 0:
+            hour_12 = 12  # Adjust 0 (from 12 AM/PM) to 12
+        hour_str = str(hour_12)
+        
+        # Format minute with leading zero if necessary
+        minute_str = f"{dt.minute:02d}"
+        
+        # Get AM/PM designation
+        am_pm = dt.strftime('%p')
+        
+        # Construct the formatted date string with Unicode narrow space
+        return f"{month}/{day}/{year_short}, {hour_str}:{minute_str}\u202f{am_pm}"
+  
+    converted_dates = [convert_to_target_format(date) for date in valid_dates]
+
 
-    df = pd.DataFrame({'user_message': filtered_messages, 'message_date': valid_dates})
+    df = pd.DataFrame({'user_message': filtered_messages, 'message_date': converted_dates})
     df['message_date'] = pd.to_datetime(df['message_date'], format='%m/%d/%y, %I:%M %p', errors='coerce')
     df.rename(columns={'message_date': 'date'}, inplace=True)