From d7ecee6b95ddcd6c2dbf1fa681c7082de767b3a0 Mon Sep 17 00:00:00 2001
From: David De Martin <59452411+DrDabbidy@users.noreply.github.com>
Date: Sun, 12 Dec 2021 17:32:14 -0500
Subject: [PATCH] Update heatmap_generation.py

---
 heatmap_generation.py | 73 +++++++++++++++++++++++++++++--------------
 1 file changed, 49 insertions(+), 24 deletions(-)

diff --git a/heatmap_generation.py b/heatmap_generation.py
index cdd8253..9d295f9 100644
--- a/heatmap_generation.py
+++ b/heatmap_generation.py
@@ -4,61 +4,86 @@
 import pandas as pd
 import json
 from crime_data import CrimeData
+import dash
+from dash import dcc
+from dash import html
 
 
-def generate_heatmap(data: CrimeData, crime: str) -> None:
+def generate_heatmap(data: CrimeData) -> None:
     """Generate an animated heatmap for the pindexes of the crime over time given the CrimeData,
     data."""
     with open('local-area-boundary.geojson') as file:
         regions = json.load(file)
 
-    unpacked_data = unpack_data(data, crime)
+    unpacked_data = unpack_data(data)
 
-    df = pd.DateFrame({'date': unpacked_data[0],
+    df = pd.DataFrame({'date': unpacked_data[0],
                        'region': unpacked_data[1],
-                       'p-index': unpacked_data[2]})
+                       'p-index': unpacked_data[2],
+                       'crime-type': unpacked_data[3]})
 
-    fig = px.choropleth_mapbox(df, geojson=regions,
-                               locations='region',
-                               color='p-index',
-                               color_continuous_scale="Viridis",
-                               range_color=(-100, 100),
-                               featureidkey="properties.name",
-                               mapbox_style="carto-positron",
-                               opacity=0.5,
-                               center={"lat": 0, "lon": 0},
-                               zoom=1,
-                               animation_frame='date')
+    crime_types = list(data.crime_pindex.keys())
 
-    fig.update_geos(fitbounds="locations", visible=False)
-    fig.update_layout(margin={"r": 0, "t": 0, "l": 0, "b": 0}, )
+    # create a dash app with a dropdown menu so that we can switch between graphs
+    app = dash.Dash()
+    app.layout = html.Div([
+        dcc.Dropdown(
+            id='crime-type-dropdown',
+            options=[{'label': crime, 'value': crime} for crime in crime_types],
+            value=crime_types[0]
+        ),
+        dcc.Graph(id='choropleth-graph')])
 
-    fig.show()
+    @app.callback(
+        dash.dependencies.Output('choropleth-graph', 'figure'),
+        [dash.dependencies.Input('crime-type-dropdown', 'value')])
+    def update_output(crime: str):
+        """Updates which graph is shown in our app."""
+        fig = px.choropleth_mapbox(df[df['crime-type'] == crime], geojson=regions,
+                                   locations='region',
+                                   color='p-index',
+                                   color_continuous_scale=['LawnGreen', 'LightBlue', 'DarkRed'],
+                                   range_color=(-100, 100),
+                                   featureidkey="properties.name",
+                                   mapbox_style="carto-positron",
+                                   opacity=0.5,
+                                   center={"lat": 49.24200376111951, "lon": -123.13312355113719},
+                                   zoom=11,
+                                   animation_frame='date',
+                                   height=750)
 
+        fig.update_geos(fitbounds="locations", visible=False)
+        fig.update_layout(title=f'<b>P-index graph for {crime}</b>')
+        return fig
 
-def unpack_data(data: CrimeData, crime: str) -> tuple[list[str], list[str], list[float]]:
+    app.run_server()
+
+
+def unpack_data(data: CrimeData) -> tuple[list[str], list[str], list[float], list[str]]:
     """Unpack the data in CrimeData into three lists, one corresponding to the dates, one to the
     regions, and one to the pindexes."""
     dates = []
     regions = []
     pindexes = []
-    pindex_data = data.crime_pindex[crime]
+    crime_types = []
 
-    for neighbourhood in pindex_data:
-        for obj in pindex_data[neighbourhood]:
+    for crime in data.crime_pindex:
+        for obj in data.crime_pindex[crime].values():
             years = sorted(list(obj.p_index_dict.keys()))
             for year in years:
                 months = list(obj.p_index_dict[year].keys())
+                # print(months)
                 for month in months:
                     dates.append(month_year_to_str(month, year))
                     regions.append(obj.neighbourhood)
                     pindexes.append(obj.get_data(year, month))
+                    crime_types.append(crime)
 
-    return dates, regions, pindexes
+    return dates, regions, pindexes, crime_types
 
 
 def month_year_to_str(month: int, year: int) -> str:
     """Given the month and year as ints, return the datestring in the form 'month year'"""
 
     months = ['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec']
-    return f'{months[month]} {year}'
+    return f'{months[month - 1]} {year}'