import pandas as pd
import os
%matplotlib inline
import matplotlib.pyplot as plt
import numpy as np

cur_dir = os.path.dirname(os.path.realpath('__file__'))

df = pd.read_csv(cur_dir + '/data/cross_raw_data/' + 'bc_full_crossing_data.csv')

df.head()

df['Port Name'][1]

df_douglas = df[df['Port Name'] == 'AZ: Douglas                                                 '][:]

df_douglas.head()

df_douglas['Total Crossers'] = df_douglas['Train Passengers'] + df_douglas['Bus Passengers'] + df_douglas['Personal Vehicle Passengers'] + df_douglas['Pedestrians']

df_douglas_annual_sum = df_douglas.groupby(df_douglas['Year']).sum()

df_douglas_annual_sum.head()

df_douglas_annual_sum['Bus Passengers'].plot(kind='area', grid=False)

df_douglas_annual_sum['Personal Vehicle Passengers'].plot(kind='area', grid=False)

df_douglas_annual_sum['Pedestrians'].plot(kind='area', grid=False)

# Create a figure with a single subplot
f, ax = plt.subplots(1, figsize=(10,5))

# Set bar width at 1
bar_width = 1

# positions of the left bar-boundaries
bar_l = [i for i in range(len(df_douglas_annual_sum['Bus Passengers']))] 

# positions of the x-axis ticks (center of the bars as bar labels)
tick_pos = [i+(bar_width/2) for i in bar_l] 

# Create the total score for each participant
totals = [i+j+k for i,j,k in zip(df_douglas_annual_sum['Bus Passengers'], 
                                 df_douglas_annual_sum['Personal Vehicle Passengers'], 
                                 df_douglas_annual_sum['Pedestrians'])]

# Create the percentage of the total score the pre_score value for each participant was
pre_rel = [i / j * 100 for  i,j in zip(df_douglas_annual_sum['Bus Passengers'], totals)]

# Create the percentage of the total score the mid_score value for each participant was
mid_rel = [i / j * 100 for  i,j in zip(df_douglas_annual_sum['Personal Vehicle Passengers'], totals)]

# Create the percentage of the total score the post_score value for each participant was
post_rel = [i / j * 100 for  i,j in zip(df_douglas_annual_sum['Pedestrians'], totals)]

# Create a bar chart in position bar_1
ax.bar(bar_l, 
       # using pre_rel data
       pre_rel, 
       # labeled 
       label='Bus Passengers', 
       # with alpha
       alpha=0.9, 
       # with color
       color='#019600',
       # with bar width
       width=bar_width,
       # with border color
       edgecolor='white'
       )

# Create a bar chart in position bar_1
ax.bar(bar_l, 
       # using mid_rel data
       mid_rel, 
       # with pre_rel
       bottom=pre_rel, 
       # labeled 
       label='Personal Vehicle Passengers', 
       # with alpha
       alpha=0.9, 
       # with color
       color='#3C5F5A', 
       # with bar width
       width=bar_width,
       # with border color
       edgecolor='white'
       )

# Create a bar chart in position bar_1
ax.bar(bar_l, 
       # using post_rel data
       post_rel, 
       # with pre_rel and mid_rel on bottom
       bottom=[i+j for i,j in zip(pre_rel, mid_rel)], 
       # labeled 
       label='Pedestrians',
       # with alpha
       alpha=0.9, 
       # with color
       color='#219AD8', 
       # with bar width
       width=bar_width,
       # with border color
       edgecolor='white'
       )

# Set the ticks to be first names
plt.xticks(tick_pos, df_douglas_annual_sum.index)
ax.set_ylabel("Percent Of Total")
ax.set_xlabel("Year")

# Let the borders of the graphic
plt.xlim([min(tick_pos)-bar_width, max(tick_pos)+bar_width])
plt.ylim(-10, 110)

# rotate axis labels
plt.setp(plt.gca().get_xticklabels(), rotation=45, horizontalalignment='right')

# shot plot
plt.show()

# Create a figure with a single subplot
f, ax = plt.subplots(1, figsize=(10,5))

# Create a scatterplot of,
            # x axis: Latency, and 
plt.bar(df_douglas_annual_sum.index, 
            # y axis: Download, with
            df_douglas_annual_sum['Personal Vehicle Passengers'], 
            # the color assigned as blue
            color='b')

plt.bar(df_douglas_annual_sum.index, 
            # y axis: Download, with
            df_douglas_annual_sum['Pedestrians'], 
            # the color assigned as blue
            color='r')

plt.bar(df_douglas_annual_sum.index, 
            # y axis: Download, with
            df_douglas_annual_sum['Bus Passengers'], 
            # the color assigned as blue
            color='g')

# Chart title
plt.title('Types Of Border Crossers')

# y label
plt.ylabel('Total People')

# x label
plt.xlabel('Year')

df.columns

df_crossing = df.groupby('Port Name').sum()

df_crossing.sort('Trains', ascending=False).head()

df_crossing['container_total'] = df_crossing['Loaded Truck Containers'] + df_crossing['Empty Truck Containers']

df_crossing = df_crossing.sort(columns='container_total')

# input data, specifically the second and 
# third rows, skipping the first column
x1 = df_crossing['Loaded Truck Containers']
x2 = df_crossing['Empty Truck Containers']

# Create the bar labels
bar_labels = df_crossing.index

# Create a figure
fig = plt.figure(figsize=(10,8))

# Set the y position
y_pos = np.arange(len(x1))
y_pos = [x for x in y_pos]
plt.yticks(y_pos, bar_labels, fontsize=10)

# Create a horizontal bar in the position y_pos
plt.barh(y_pos, 
         # using x1 data
         x1, 
         # that is centered
         align='center', 
         # with alpha 0.4
         alpha=0.4, 
         # and color green
         color='#263F13')

# Create a horizontal bar in the position y_pos
plt.barh(y_pos, 
         # using NEGATIVE x2 data
         -x2,
         # that is centered
         align='center', 
         # with alpha 0.4
         alpha=0.4, 
         # and color green
         color='#77A61D')

# annotation and labels
plt.xlabel('Empty Containers: Light Green. Full Containers: Dark Green')
t = plt.title('Comparison of Empty And Full Truck Containers Per Border Crossing')
plt.ylim([-1,len(x1)+0.1])
plt.grid()

plt.show()