Pin to newer version of poetry, and cleanup package pinning

README.md

@@ -12,7 +12,7 @@ This is the repository for the TransitMatters Data Dashboard. Client code is wri
 - node 18.x and npm 9.x required
   - With `nvm` installed, use `nvm install && nvm use`
   - verify with `node -v`
-- Python 3.10 with recent poetry
+- Python 3.10 with recent poetry (1.5.0 or later)
   - Verify with `python --version && poetry --version`
   - `poetry self update` to update poetry
 

package.json

@@ -103,7 +103,7 @@
     "generate-oss-list": "generate-attribution --outputDir common/constants/licenses",
     "lint": "npm run lint-frontend && npm run lint-backend",
     "lint-frontend": "tsc --noEmit && eslint . && next lint",
-    "lint-backend": "cd server && poetry run flake8",
+    "lint-backend": "cd server && poetry run flake8 && poetry run black . --check",
     "test": "react-scripts test",
     "eject": "react-scripts eject",
     "postinstall": "cd server && poetry install",

server/app.py

@@ -48,9 +48,7 @@ def healthcheck():
         "API Key Present": (lambda: len(secrets.MBTA_V2_API_KEY) > 0),
         "S3 Headway Fetching": (
             lambda: "2020-11-07 10:33:40"
-            in json.dumps(
-                data_funcs.headways(date(year=2020, month=11, day=7), ["70061"])
-            )
+            in json.dumps(data_funcs.headways(date(year=2020, month=11, day=7), ["70061"]))
         ),
         "Performance API Check": (
             lambda: MbtaPerformanceAPI.get_api_data(
@@ -112,9 +110,7 @@ def traveltime_route(user_date):
 @app.route("/api/alerts/{user_date}", cors=cors_config)
 def alerts_route(user_date):
     date = parse_user_date(user_date)
-    return json.dumps(
-        data_funcs.alerts(date, mutlidict_to_dict(app.current_request.query_params))
-    )
+    return json.dumps(data_funcs.alerts(date, mutlidict_to_dict(app.current_request.query_params)))
 
 
 @app.route("/api/aggregate/traveltimes", cors=cors_config)
@@ -163,11 +159,7 @@ def dwells_aggregate_route():
 def get_git_id():
     # Only do this on localhost
     if TM_FRONTEND_HOST == "localhost":
-        git_id = str(
-            subprocess.check_output(
-                ["git", "describe", "--always", "--dirty", "--abbrev=10"]
-            )
-        )[2:-3]
+        git_id = str(subprocess.check_output(["git", "describe", "--always", "--dirty", "--abbrev=10"]))[2:-3]
         return json.dumps({"git_id": git_id})
     else:
         raise ConflictError("Cannot get git id from serverless host")

server/chalicelib/constants.py

@@ -6,5 +6,5 @@
     "line-red": ["line-red-a", "line-red-b"],
     "line-green": ["line-green-b", "line-green-c", "line-green-d", "line-green-e"],
     "line-blue": ["line-blue"],
-    "line-orange": ["line-orange"]
+    "line-orange": ["line-orange"],
 }

server/chalicelib/dynamo.py

@@ -18,7 +18,10 @@ def query_daily_trips_on_route(table_name, route, start_date, end_date):
 def query_daily_trips_on_line(table_name, line, start_date, end_date):
     route_keys = constants.LINE_TO_ROUTE_MAP[line]
     with concurrent.futures.ThreadPoolExecutor(max_workers=4) as executor:
-        futures = [executor.submit(query_daily_trips_on_route, table_name, route_key, start_date, end_date) for route_key in route_keys]
+        futures = [
+            executor.submit(query_daily_trips_on_route, table_name, route_key, start_date, end_date)
+            for route_key in route_keys
+        ]
         results = []
         for future in concurrent.futures.as_completed(futures):
             result = future.result()

server/chalicelib/secrets.py

@@ -3,7 +3,4 @@
 MBTA_V2_API_KEY = os.environ.get("MBTA_V2_API_KEY", "")
 MBTA_V3_API_KEY = os.environ.get("MBTA_V3_API_KEY", "")
 
-HEALTHCHECK_HIDE_SECRETS = [
-    MBTA_V2_API_KEY,
-    MBTA_V3_API_KEY
-]
+HEALTHCHECK_HIDE_SECRETS = [MBTA_V2_API_KEY, MBTA_V3_API_KEY]

server/chalicelib/service_levels.py

@@ -67,21 +67,21 @@ def get_service_levels(
 
 
 def get_weekly_scheduled_service(scheduled_service_arr, start_date, end_date):
-    df = pd.DataFrame({'value': scheduled_service_arr}, index=pd.date_range(start_date, end_date))
-    weekly_df = df.resample('W-SUN').median()
+    df = pd.DataFrame({"value": scheduled_service_arr}, index=pd.date_range(start_date, end_date))
+    weekly_df = df.resample("W-SUN").median()
     # Drop the first week if it is incomplete
     if datetime.fromisoformat(start_date.isoformat()).weekday() != 6:
         weekly_df = weekly_df[1:]
-    return weekly_df['value'].tolist()
+    return weekly_df["value"].tolist()
 
 
 def get_monthly_scheduled_service(scheduled_service_arr, start_date, end_date):
-    df = pd.DataFrame({'value': scheduled_service_arr}, index=pd.date_range(start_date, end_date))
-    monthly_df = df.resample('M').median()
+    df = pd.DataFrame({"value": scheduled_service_arr}, index=pd.date_range(start_date, end_date))
+    monthly_df = df.resample("M").median()
     # Drop the first month if it is incomplete
     if datetime.fromisoformat(start_date.isoformat()).day != 1:
         monthly_df = monthly_df[1:]
-    return monthly_df['value'].tolist()
+    return monthly_df["value"].tolist()
 
 
 def get_scheduled_service(
@@ -105,11 +105,11 @@ def get_scheduled_service(
             scheduled_service_count = None
         scheduled_service_arr.append(scheduled_service_count)
     counts = []
-    if agg == 'daily':
+    if agg == "daily":
         counts = scheduled_service_arr
-    if agg == 'weekly':
+    if agg == "weekly":
         counts = get_weekly_scheduled_service(scheduled_service_arr, start_date, end_date)
-    if agg == 'monthly':
+    if agg == "monthly":
         counts = get_monthly_scheduled_service(scheduled_service_arr, start_date, end_date)
 
     return {

server/chalicelib/speed.py

@@ -19,12 +19,19 @@ def aggregate_actual_trips(actual_trips, agg, start_date):
     # Create a DataFrame from the flattened data
     df = pd.DataFrame(flat_data)
     # Set miles_covered to NaN for each date with any entry having miles_covered as nan
-    df.loc[df.groupby('date')['miles_covered'].transform(lambda x: (np.isnan(x)).any()), ['count', 'total_time', 'miles_covered']] = np.nan
+    df.loc[
+        df.groupby("date")["miles_covered"].transform(lambda x: (np.isnan(x)).any()),
+        ["count", "total_time", "miles_covered"],
+    ] = np.nan
     # Group each branch into one entry. Keep NaN entries as NaN
-    df_grouped = df.groupby('date').agg({'miles_covered': 'sum', 'total_time': 'sum', 'count': 'sum', 'line': 'first'}).reset_index()
+    df_grouped = (
+        df.groupby("date")
+        .agg({"miles_covered": "sum", "total_time": "sum", "count": "sum", "line": "first"})
+        .reset_index()
+    )
     # set index to use datetime object.
-    df_grouped.set_index(pd.to_datetime(df_grouped['date']), inplace=True)
-    return df_grouped.to_dict(orient='records')
+    df_grouped.set_index(pd.to_datetime(df_grouped["date"]), inplace=True)
+    return df_grouped.to_dict(orient="records")
 
 
 def trip_metrics_by_line(params):
@@ -53,7 +60,7 @@ def trip_metrics_by_line(params):
 
 
 def is_invalid_range(start_date, end_date, max_delta):
-    '''Check if number of requested entries is more than maximum for the table'''
+    """Check if number of requested entries is more than maximum for the table"""
     start_datetime = datetime.strptime(start_date, DATE_FORMAT_BACKEND)
     end_datetime = datetime.strptime(end_date, DATE_FORMAT_BACKEND)
     return start_datetime + timedelta(days=max_delta) < end_datetime