import pandas as pd
pd.set_option('display.float_format', lambda x: '%.2f' % x)

%matplotlib inline
import matplotlib.pyplot as plt
import seaborn as sns
colpal = sns.color_palette("hls", 7)
sns.set(palette=colpal, style='ticks', rc={"figure.figsize":(7.75,5),'savefig.dpi':150})

import folium
from IPython.display import HTML

# read data in 
df = pd.read_csv('data/listings.csv',index_col='id')
print('Number of records:',df.shape[0])
print('Number of columns:',df.shape[1])
print(', '.join(df.columns)) #see the columns starting with review_score...

# pre-process the data
revcols = 'host_id	host_listings_count	number_of_reviews	review_scores_rating	review_scores_accuracy	review_scores_cleanliness	review_scores_checkin	review_scores_communication	review_scores_location	review_scores_value'.split('\t')
df = df.dropna(subset=revcols).sort('review_scores_rating',ascending=False)
df = df.rename(columns=dict(zip(revcols[3:],[c.split('_')[-1] for c in revcols[3:]])))
df = df.rename(columns={'neighbourhood_cleansed':'neighborhood'})
revcols = [c.split('_')[-1] for c in revcols[4:]]

# number of listings per neighborhood
grouped = pd.DataFrame(df.pivot_table(index='neighborhood',values='rating',aggfunc='count')).sort('rating',ascending=False).reset_index()
grouped # Appendix A

# number of listings per neighborhood
mapname= 'neighborhoods'
m = folium.Map(location=[38.92,-77.03], zoom_start=12, tiles='OpenStreetMap')
m.geo_json(geo_path='neighbourhoods.geojson', data=grouped, data_out=mapname+'.json',
            columns=['neighborhood','rating'],
            key_on='feature.properties.neighbourhood',
            fill_color='PuRd', fill_opacity=0.7, line_opacity=0.2,
            legend_name='AirBnB listings per neighborhood')
m.create_map(path=mapname+'.html')
HTML('<iframe src='+mapname+'.html style="width: 100%; height: 500px; border: none"></iframe>')

# describe the data
dd = (df[revcols].sum()/len(df)).order(ascending=False)
ax = sns.barplot(y=dd.index,x=dd.values)
#ax = dd.order().plot(kind='barh',color=sns.color_palette('hls'),figsize=(5,3))
ax.set(xlim=(9,9.81),xlabel='Average Review Score',title='AirBnB Listings in Washington, D.C.');
sns.despine(trim=True)
plt.gcf().savefig('docs/review_scores.png',bbox_inches='tight')

df.shape

#export related fields to a csv file
df[revcols] = df[revcols].astype(int)
df[['host_id']+revcols+['number_of_reviews','listing_url','name','neighborhood','room_type','description','latitude','longitude']].to_csv('data/reviewed_listings.csv',encoding='utf8')

# Ratings (stars)...
# <input id="cleanliness" class="rating" data-min="0" data-max="5" data-step="0.5" data-size="xs"
#  data-default-caption="{rating} clean" data-clear-caption="Cleanliness" data-star-captions="{}">

# onclick...
function change() {
    var query = document.getElementById('myTextField').value;
    var cleanliness = $('#cleanliness').val()*2;
    var accuracy = $('#accuracy').val()*2;
    var communication = $('#communication').val()*2;
    var checkin = $('#checkin').val()*2;
    var loc = $('#loc').val()*2;
    var val = $('#val').val()*2;
    console.log(cleanliness);
    if (query.length == 0) {
        alert('Nothing to search for...');
        return;
    }
    group.removeLayer(geojsonlayer);
    viewparam = {
        viewparams: "word:" + query +
        ";cleanliness:" + cleanliness +
        ";accuracy:" + accuracy +
        ";communication:" + communication +
        ";checkin:" + checkin +
        ";loc:" + loc +
        ";val:" + val
    };
    parameters = L.Util.extend(defaultParameters, viewparam);
    URL = owsrootUrl + L.Util.getParamString(parameters);
    console.log("Geoserver URL: " + URL);
    queryWFS(URL);
}

# and the demo...
HTML('<video controls><source src="AirBnB.webm" type="video/webm"></video>')