Update app.py

app.py

@@ -4,38 +4,44 @@ import hopsworks
 import math
 import os
 
-def download_data(path):
-    os.path.abspath(path)
-    df = pd.read_csv(path)
-    wanted_df = df[['airport', 'flight_number', 'ontime', 'delayed']].copy()
-    return wanted_df
-
-def access_online_dataframe():
-    # Get data from online source
-    hopsworks_api_key  = os.environ['HOPSWORKS_API_KEY']
-    project            = hopsworks.login(api_key_value = hopsworks_api_key)
-    dataset_api        = project.get_dataset_api()
-    today_path         = dataset_api.download("Resources/today_timetable_prediction/today_timetable_prediction.csv", overwrite=True)
-    tomorrow_path      = dataset_api.download("Resources/tomorrow_timetable_prediction/tomorrow_timetable_prediction.csv", overwrite=True)
-    today_df           = download_data(today_path)
-    tomorrow_df        = download_data(tomorrow_path)
-    return today_df, tomorrow_df
+hopsworks_today_path    = "Resources/today_timetable_prediction/today_timetable_prediction.csv"
+hopsworks_tomorrow_path = "Resources/tomorrow_timetable_prediction/tomorrow_timetable_prediction.csv"
 
 
+def get_dataframe(online_dataframe_path):
+    # Connect to Hopsworks File System
+    project     = hopsworks.login(api_key_value = os.environ['HOPSWORKS_API_KEY'])
+    dataset_api = project.get_dataset_api()
+
+    # Download online dataframe and get path
+    dataframe_path = os.path.abspath(dataset_api.download(online_dataframe_path, overwrite = True))
+
+    # Read dataframe from local path, drop duplicates, return
+    dataframe = pd.read_csv(dataframe_path)
+    dataframe.drop_duplicates(inplace=True)
+    return dataframe
+
+def get_tomorrow_dataframe():
+    return get_dataframe(hopsworks_today_path)
+
+def get_today_dataframe():
+    return get_dataframe(hopsworks_tomorrow_path)
+
+def get_dataframe_of(day):
+    if   (day == 'today'):
+        return get_today_dataframe()
+    elif (day == 'tomorrow'):
+        return get_tomorrow_dataframe()
+
 def get_possible_destinations():
-    today_df, tomorrow_df = access_online_dataframe()
-    total_df = pd.DataFrame({'airport': pd.concat([today_df['airport'], tomorrow_df['airport']]).drop_duplicates().reset_index(drop=True)})
+    today_df, tomorrow_df = get_today_dataframe(), get_tomorrow_dataframe()
+    total_df = pd.DataFrame({'airport': pd.concat([today_df['airport'], tomorrow_df['airport']]).drop_duplicates().reset_index(drop=True).sort_values()})
     total_dest = (total_df['airport']).tolist()
     return total_dest
 
 
 def get_specific_flights(day, max_delay, departure_hour, ampm, weather, destinations, yes):
-    today_df, tomorrow_df = access_online_dataframe()
-    df = pd.DataFrame()
-    if(day == 'today'):
-        df = today_df
-    else:
-        df = tomorrow_df
+    df = get_dataframe_of(day)
 
     # Remove unwanted destinations
     destinations = [dest for dest in destinations if dest not in ["That's a reason why I travel alone...", "I prefer not to say"]]
@@ -61,14 +67,7 @@ def get_specific_flights(day, max_delay, departure_hour, ampm, weather, destinat
     return filtered_df
 
 def full_day_departure(day):
-    today_df, tomorrow_df = access_online_dataframe()
-    df = pd.DataFrame()
-    if(day == 'today'):
-        df = today_df
-    else:
-        df = tomorrow_df
-
-    return df
+    return get_dataframe_of(day)