election2008.py

# -*- coding: utf-8 -*-
"""
Created on Thu Mar 13 08:41:32 2014

@author: Eric

Reads in 2008 county election data (2008 Presidential General Election, County Results, National Atlas of the United States, http://nationalatlas.gov/atlasftp.html?openChapters=chphist#chphist)

Suffixes at the end of variable names:
A: numpy array
B: boolean
D: dictionary
DF: pandas DataFrame
L: list
S: string
SR: pandas Series
T: tuple
Underscores indicate chaining: for instance, "fooT_T" is a tuple of tuples
"""

from matplotlib.collections import PatchCollection
import matplotlib.patches as patches
import matplotlib.pyplot as plt
import numpy as np
import os
import pandas as pd
import shapefile
import sys

import config
reload(config)

sys.path.append(config.GeoDaSandboxPathS)
from pyGDsandbox.dataIO import dbf2df


def main():
    """
    Read in the election data shapefile as a DataFrame
    
    See http://stackoverflow.com/questions/15968762/shapefile-and-matplotlib-plot-polygon-collection-of-shapefile-coordinates for how to quickly plot shapes from a shapefile
    """
    
    filePathS = os.path.join(config.rawDataPathS, 'election_statistics', '2008',
                             'elpo08p020.dbf')
                             
    # Read fields into DataFrame
    fullDF = dbf2df(filePathS)
    fullDF = fullDF.convert_objects(convert_numeric=True)
        
    # Select relevant columns
    finalDF = fullDF.loc[:, ['FIPS', 'TOTAL_VOTE', 'VOTE_DEM', 'VOTE_REP']]
    finalDF.columns = ['FIPS', 'Election2008Total', 'Election2008Dem', 'Election2008Rep']
    shapeIndexL = finalDF.FIPS.tolist()

    # Read in shapefile data
    fullShapeFile = shapefile.Reader(filePathS)
    shapeL = fullShapeFile.shapes()
    del fullShapeFile
    
    # Removing the second (incorrect) entry for Ottawa County, OH
    finalDF = finalDF.loc[~finalDF['FIPS'].isin([39123]) |
                          ~finalDF['Election2008Dem'].isin([12064])]
    
    # Cleaning, sorting, and setting the index
    finalDF = finalDF.drop_duplicates()
    finalDF = finalDF.sort(columns='FIPS')
    finalDF = finalDF.set_index('FIPS')
    
    # This is a work-around for a NaN somewhere in Michigan.
    finalDF = finalDF[pd.notnull(finalDF['Election2008Total'])]
    
    # Correcting the total number of votes cast in Laclede County, MO: the number
    # in the source file is actually the number from the previous county in the
    # list, Knox County. If this error isn't corrected, the Obama vote swing in
    # Laclede County between 2008 and 2012 is -231%. The figure that I'm
    # replacing the vote total with was taken from Wikipedia
    # (http://en.wikipedia.org/wiki/Laclede_County,_Missouri, 2014-04-17).
    finalDF.loc[29105, 'Election2008Total'] = 16379

    # Correcting the number of votes in Washington County, OH, which is
    # erroneous; the correct number can be found at
    # http://en.wikipedia.org/wiki/United_States_presidential_election_in_Ohio,_2008,
    # 2014-04-17.
    finalDF.loc[39167, 'Election2008Total'] = 29932
    finalDF.loc[39167, 'Election2008Dem'] = 12368

    # Correcting the number of votes in LaPorte County, IN, which is
    # erroneous; the correct number can be found at
    # http://en.wikipedia.org/wiki/United_States_presidential_election_in_Indiana,_2008,
    # 2014-04-17.
    finalDF.loc[18091, 'Election2008Total'] = 46919
        
    return (finalDF, shapeIndexL, shapeL)
    
    

def plot_county_results():
    """
    Reads in the shapefile and plots a full map of all counties, colored red or blue according to vote totals
    
    See http://stackoverflow.com/questions/15968762/shapefile-and-matplotlib-plot-polygon-collection-of-shapefile-coordinates for how to quickly plot shapes from a shapefile
    """
    
    # Read in shapedata    
    filePathS = os.path.join(config.rawDataPathS, 'election_statistics', '2008',
                             'elpo08p020')
    fullShapeFile = shapefile.Reader(filePathS)
    shapeL = fullShapeFile.shapes()
    del fullShapeFile
    numShapes = len(shapeL)
    
    # Read in election data
    fullDF = main()
    assert fullDF.shape[0] == numShapes
    fullDF.loc[:, 'DemIsHigher'] = (fullDF.loc[:, u'VOTE_DEM'] >
                                    fullDF.loc[:, u'VOTE_REP'])

    # Plot figure
    fig = plt.figure()
    ax = fig.add_subplot(1, 1, 1)
    shapeBoundsA = np.empty((numShapes, 4))
    shapeBoundsA[:] = np.nan
    for lShape in xrange(numShapes):
        shapeBoundsA[lShape, :] = shapeL[lShape].bbox

        # Find all patches for the given shape
        thisShapesPatches = []
        pointsA = np.array(shapeL[lShape].points)
        shapeFileParts = shapeL[lShape].parts
        allPartsL = list(shapeFileParts) + [pointsA.shape[0]]
        for lPart in xrange(len(shapeFileParts)):
            thisShapesPatches.append(patches.Polygon(
                pointsA[allPartsL[lPart]:allPartsL[lPart+1]]))
        
        # Determine shape color
        if fullDF.loc[lShape, 'DemIsHigher']:
            shapeColorT = (0, 0, 1)
        else:
            shapeColorT = (1, 0, 0)        
        ax.add_collection(PatchCollection(thisShapesPatches,
                                          color=shapeColorT))
    
    # Set axes
    shapeBoundsA = np.empty(4)    
    shapeBoundsA[0] = np.amin(shapeBoundsA[:, 0], 0)
    shapeBoundsA[1] = np.amin(shapeBoundsA[:, 1], 0)
    shapeBoundsA[2] = np.amax(shapeBoundsA[:, 2], 0)
    shapeBoundsA[3] = np.amax(shapeBoundsA[:, 3], 0)    
    ax.set_xlim(shapeBoundsA[0], shapeBoundsA[2])
    ax.set_ylim(shapeBoundsA[1], shapeBoundsA[3])