#!/usr/bin/env python
# coding: utf-8

# # Map

# In[5]:


import fiona

#This is a canned demo - I happen to have the Local Authority Code for the Isle of Wight...
#...and copies of ward geojson files by LA from https://github.com/martinjc/UK-GeoJSON
geojson_local='IWgeodata/wards_by_lad/E06000046.json'
fi=fiona.open(geojson_local)
centre_lat,centre_lon=((fi.bounds[0]+fi.bounds[2])/2,(fi.bounds[1]+fi.bounds[3])/2)


# In[6]:


import json

with open(geojson_local) as f:    
    jdata = json.load(f)
    
for j in jdata['features']:
    print(j['properties']['WD13CD'],j['properties']['WD13NM'])


# In[7]:


import pandas as pd


# In[8]:


#previous results from https://blog.ouseful.info/2013/05/15/asking-questions-of-data-contained-in-a-google-spreadsheet-using-a-basic-structured-query-language/
url="https://docs.google.com/spreadsheets/d/1oyoF9Tgc2n0q2YksNACVf2AY4crOnJIhndTXyEoTJmQ/export?format=csv"
df=pd.read_csv("https://docs.google.com/spreadsheets/d/1oyoF9Tgc2n0q2YksNACVf2AY4crOnJIhndTXyEoTJmQ/export?format=csv")
df.head(10)


# In[9]:


#Compare the official ward names with ones in result file
set(df['Electoral Division'].unique()) - set(j['properties']['WD13NM'] for j in jdata['features']),set(j['properties']['WD13NM'] for j in jdata['features']) - set(df['Electoral Division'].unique())


# In[10]:


#Clean the previous results data
df['Electoral Division']=df['Electoral Division'].str.replace('&','and')
df['Electoral Division'].unique()


# In[11]:


#Find winning percentages
winners=df[ ~df['Candidate'].isin(['Total votes cast',
                                   'Postal votes',
                                   'Postal votes'])].sort_values('Electorate', ascending=False).groupby('Electoral Division', as_index=False).first()
winners['pc']=100*winners['Electorate']/winners['NoOnRoll']
winners


# In[29]:


#Percentage of no on roll voting for winner
#crib? https://blog.ouseful.info/2015/04/17/creating-interactive-election-maps-using-folium-and-ipython-notebooks/
#http://nbviewer.jupyter.org/gist/psychemedia/fbcd7cf1daabe0004e27/folium_shapefiles.ipynb
import folium
iwmap=folium.Map([centre_lon,centre_lat], zoom_start=11)
iwmap.choropleth(
    geo_path=geojson_local,
    data=winners,
    columns=['Electoral Division', 'pc'],
    key_on='feature.properties.WD13NM',
    fill_color='PuBuGn', fill_opacity=0.7
)
iwmap


# In[37]:


turnout=df[df['Candidate'].isin(['Total votes cast'])][['Electoral Division','Candidate','Electorate','NoOnRoll']]
turnout['turnout']=turnout['Electorate']/turnout['NoOnRoll']
turnout


# In[38]:


#Percentage of no on roll voting
#crib? https://blog.ouseful.info/2015/04/17/creating-interactive-election-maps-using-folium-and-ipython-notebooks/
#http://nbviewer.jupyter.org/gist/psychemedia/fbcd7cf1daabe0004e27/folium_shapefiles.ipynb
import folium
iwmap=folium.Map([centre_lon,centre_lat], zoom_start=11)
iwmap.choropleth(
    geo_path=geojson_local,
    data=turnout,
    columns=['Electoral Division', 'turnout'],
    key_on='feature.properties.WD13NM',
    fill_color='PuBuGn', fill_opacity=0.7
)
iwmap


# In[12]:


winners['Party'].unique()


# In[13]:


colmap={'Island Independents':'lightblue', 'Conservative':'blue', 'Independent':'black', 'Labour':'red',
       'none':'grey', 'Sandown Independents':'darkgrey', 'UKIP':'purple'}


# In[15]:


winners['col']=winners['Party'].map(colmap)
winners.head()


# In[25]:


#Ward by Party

#There has to be an easier way of doing qualitative choropleths using folium
#This hack is one I struggled towards finding before...
#http://nbviewer.jupyter.org/gist/psychemedia/fbcd7cf1daabe0004e27/folium_shapefiles.ipynb
jj=json.load(open(geojson_local))
results_map = folium.Map([centre_lon,centre_lat], zoom_start=11)
for c in jj['features']:
    #The choropleth format requires that features are provided in a list and a FeatureCollection defined as the type
    #So we wrap the feature definition for each constituency in the necessary format
    geodata= {"type": "FeatureCollection", "features": [c]}
    #Get the name of the seat for the current constituency
    ward=c['properties']['WD13NM']
    #We can now lookup the colour 
    colour= colmap[winners[winners['Electoral Division']==ward]['Party'].iloc[0]]
    results_map.choropleth(geo_str= json.dumps(geodata),fill_color=colour,fill_opacity=1)
results_map


# In[ ]:





# In[ ]: