Refactored county places generator

Dockerfile

@@ -12,4 +12,4 @@ COPY data/*.pickle data/
 RUN useradd -m myuser
 USER myuser
 
-CMD gunicorn --bind 0.0.0.0:$PORT app:server
+CMD gunicorn --bind 0.0.0.0:$PORT app:server

app.py

@@ -1,3 +1,4 @@
+from typing import Tuple
 import dash
 import dash_core_components as dcc
 import dash_html_components as html
@@ -6,6 +7,9 @@
 import plotly.express as px
 import pandas as pd
 
+LACDPH = 'lacdph'
+LATIMES = 'latimes'
+
 DATE = 'date'
 COUNTY = 'county'
 NAME = 'name'
@@ -28,8 +32,8 @@
  pd.read_pickle(f'data/lacdph-{x}day.pickle') for x in (7, 14)
 ]
 
-lacdph_csa_list = list(df_dph_7day['csa'].unique())
-lacdph_csa_list.sort()
+LACDPH_CSA_LIST = list(df_dph_7day['csa'].unique())
+LACDPH_CSA_LIST.sort()
 
 external_stylesheets = ['https://codepen.io/chriddyp/pen/bWLwgP.css']
 
@@ -42,6 +46,12 @@
 counties.sort()
 
 
+def latimes_county_places(county: str) -> Tuple[str, ...]:
+ places = list(df_times.loc[df_times[COUNTY] == county, NAME].unique())
+ places.sort()
+ return tuple(places)
+
+
 def place_to_id(county, name):
  id_ = df_times.loc[(df_times[COUNTY] == county) & (df_times[NAME] == name),
  ID].unique()
@@ -118,7 +128,7 @@ def id_to_place(county, id_):
  ]),
  html.Div([
  html.Label('Data Source', htmlFor='selected-data-source'),
- dcc.RadioItems(id='selected-data-source', value='latimes')
+ dcc.RadioItems(id='selected-data-source', value=LATIMES)
  ])
  ],
  style={
@@ -156,40 +166,35 @@ def id_to_place(county, id_):
  Input('selected-place-value', 'value'),
  Input('selected-data-source', 'value'))
 def county_places(county, place_or_id, data_source):
- if data_source == 'latimes':
- places = list(df_times.loc[df_times[COUNTY] == county, NAME].unique())
-  places.sort()
+ places = (LACDPH_CSA_LIST
+  if data_source == LACDPH else latimes_county_places(county))
+ if county == LOS_ANGELES:
  if place_or_id in places:
  selected = place_or_id
- elif county == LOS_ANGELES and place_or_id in lacdph_csa_list:
+ elif data_source == LACDPH and place_or_id in latimes_county_places(
+ LOS_ANGELES):
+ selected = place_to_id(LOS_ANGELES, place_or_id)
+ elif data_source == LATIMES and place_or_id in LACDPH_CSA_LIST:
  selected = id_to_place(LOS_ANGELES, place_or_id)
  else:
  selected = places[0]
  else:
- places = lacdph_csa_list
- if place_or_id in places:
- selected = place_or_id
- elif place_or_id not in places:
- selected = place_to_id(LOS_ANGELES, place_or_id)
- if selected not in places:
- selected = places[0]
- else:
- selected = places[0]
+ selected = place_or_id if place_or_id in places else places[0]
 
  return ('Place'
- if data_source == 'latimes' else 'Countywide Statistical Area', [{
+ if data_source == LATIMES else 'Countywide Statistical Area', [{
  LABEL: x,
  VALUE: x
  } for x in places], selected,
- data_source if county == LOS_ANGELES else 'latimes')
+ data_source if county == LOS_ANGELES else LATIMES)
 
 
 @app.callback(Output('selected-data-source', 'options'),
  Input('selected-county', 'value'))
 def data_source_options(county):
  options = [{LABEL: 'Los Angeles Times', VALUE: 'latimes'}]
  if county == LOS_ANGELES:
- options.append({LABEL: 'LACDPH', VALUE: 'lacdph'})
+ options.append({LABEL: 'LACDPH', VALUE: LACDPH})
  return options
 
 
@@ -199,7 +204,7 @@ def data_source_options(county):
  Input('observational-period', 'value'),
  Input('selected-data-source', 'value'))
 def update_general_graph(county, place, date_range, obs_period, data_source):
- if data_source == 'lacdph':
+ if data_source == LACDPH:
  return update_lacdph_graph(place, date_range, obs_period)
  return update_latimes_graph(county, place, date_range, obs_period)
 
@@ -209,7 +214,9 @@ def update_latimes_graph(county, place, date_range, obs_period):
  NEW_CASES_7DAY, NEW_CASES_14DAY = [f'new_cases_{x}day' for x in (7, 14)]
  CASE_RATE_7DAY, CASE_RATE_14DAY = [f'case_rate_{x}day' for x in (7, 14)]
 
- id_ = place_to_id(county, place)
+ county_places = latimes_county_places(county)
+ id_ = place_to_id(county,
+ place) if place in county_places else county_places[0]
 
  df_place = df_times[df_times[ID] == id_]
  dep_var, dep_var_raw = (CASE_RATE_7DAY,

deploy.sh

@@ -1,5 +1,9 @@
 #!/bin/bash
-bash fetch-latimes-place-totals.sh && bash parse-sources.sh
+if [[ $1 = 'fetch' ]]
+then
+ echo "Fetching LA Times dataset and parsing sources..."
+ # bash fetch-latimes-place-totals.sh && bash parse-sources.sh
+fi
 NAME='ca-local-covid-19-dashboard'
 docker buildx build --platform linux/amd64 --tag $NAME . \
  && docker tag $NAME "registry.heroku.com/$NAME/web" \