Refactor and Improvements to #124 (#127)

* Removed artifacts

* Removed artifacts

pages/data.py

@@ -29,7 +29,7 @@
             dcc.Tab(label='Trajectories', value='tab-trajectories-datatable'),
         ]),
         html.Div(id='tabs-content'),
-        dcc.Interval(id='interval-load-more', interval=6000, n_intervals=0), 
+        dcc.Interval(id='interval-load-more', interval=10000, n_intervals=0), # default loading at 10s, can be lowered or hightened based on perf (usual process local is 3s)
         dcc.Store(id='store-uuids', data=[]),  # Store to hold the original UUIDs data
         dcc.Store(id='store-loaded-uuids', data={'data': [], 'loaded': False})  # Store to track loaded data
     ]
@@ -106,10 +106,6 @@ def render_content(tab, store_uuids, store_excluded_uuids, store_trips, store_de
             processed_data = db_utils.add_user_stats(new_data, initial_batch_size)
             loaded_data.extend(processed_data)
 
-            # Create a Patch object to append data progressively
-            patched_data = Patch()
-            patched_data['data'] = processed_data
-
             # Update the store with the new data
             loaded_uuids_store['data'] = loaded_data
             loaded_uuids_store['loaded'] = len(loaded_data) >= len(uuids_list)  # Mark all data as loaded if done
@@ -126,8 +122,7 @@ def render_content(tab, store_uuids, store_excluded_uuids, store_trips, store_de
 
         df = df.drop(columns=[col for col in df.columns if col not in columns])
 
-        # Use the Patch() object to append new data instead of fully replacing the table
-        logging.debug("Returning patched data to update the UI.")
+        logging.debug("Returning appened data to update the UI.")
         return html.Div([populate_datatable(df)]), loaded_uuids_store, False if not loaded_uuids_store['loaded'] else True
 
 

utils/db_utils.py

@@ -235,7 +235,6 @@ def query_demographics():
     return dataframes
 
 def query_trajectories(start_date: str, end_date: str, tz: str):
-
     (start_ts, end_ts) = iso_range_to_ts_range(start_date, end_date, tz)
     ts = esta.TimeSeries.get_aggregate_time_series()
     entries = ts.find_entries(
@@ -255,14 +254,6 @@ def query_trajectories(start_date: str, end_date: str, tz: str):
         df['data.mode_str'] = df['data.mode'].apply(lambda x: ecwm.MotionTypes(x).name if x in set(enum.value for enum in ecwm.MotionTypes) else 'UNKNOWN')
     return df
 
-@lru_cache(maxsize=None)
-def get_time_series_aggregate():
-    return esta.TimeSeries.get_aggregate_time_series()
-
-@lru_cache(maxsize=None)
-def get_user_profile(user_uuid):
-    return edb.get_profile_db().find_one({'user_id': user_uuid})
-
 def add_user_stats(user_data, batch_size=5):
     start_time = time.time()
     time_format = 'YYYY-MM-DD HH:mm:ss'
@@ -271,10 +262,10 @@ def process_user(user):
         user_uuid = UUID(user['user_id'])
 
         # Fetch aggregated data for all users once and cache it
-        ts_aggregate = get_time_series_aggregate()
+        ts_aggregate = esta.TimeSeries.get_aggregate_time_series()
 
         # Fetch data for the user, cached for repeated queries
-        profile_data = get_user_profile(user_uuid)
+        profile_data = edb.get_profile_db().find_one({'user_id': user_uuid})
 
         total_trips = ts_aggregate.find_entries_count(
             key_list=["analysis/confirmed_trip"],