G09 demo.py

#! /bin/python3
#  Feb 21 (PJW)

import csv

#
#  Set up a function for reading files. Returns a list of dictionaries
#

def read_file(filename):
    fh = open(filename)
    reader = csv.DictReader(fh)
    records = []    
    for rec in reader:
        records.append( rec )
    fh.close()
    return records

#%%
#
#  Read the files of state names and populations
#

name_list = read_file('state_name.csv')
pop_list = read_file('state_pop.csv')

#%%
#
#  Print a couple of entries. Notice that everything is a string.
#

print( '\nname_list[0]:\n  ',name_list[0] )
print( '\npop_list[0]:\n  ',pop_list[0] )

#%%
#
#  Now build a dictionary of name information with FIPS codes as keys
#

name_by_fips = {}

for rec in name_list:
    fips = rec['State']
    if fips != '00':
        name_by_fips[ fips ] = rec

print( '\nname_by_fips["36"]:\n   ', name_by_fips["36"] )

#%%
#
#  Build a dictionary of population information with FIPS codes as keys
#

pop_by_fips = {}

for rec in pop_list:
    fips = rec['STATEFP']
    pop_by_fips[ fips ] = rec

print( '\npop_by_fips["36"]:\n  ', pop_by_fips["36"] )

#%%
#
#  Do a left join of the population data onto the name data. Walk 
#  through the name data, look up the population information, and 
#  then add it to the name dictionary. Make the population numeric
#  in the process.
#

for fips in name_by_fips.keys():
    
    name_rec = name_by_fips[fips]
    pop_rec = pop_by_fips[fips]
    
    name_rec['pop'] = float(pop_rec['pop'])

print( '\nname_by_fips["36"]:\n   ', name_by_fips["36"] )

#%%
#
#  Print a couple of states
#

print( )
print( name_by_fips['06'] )
print( name_by_fips['48'] )
print( name_by_fips['36'] )

#%%
#
#  Aggregate the data by division
#

division_total = {}

for rec in name_by_fips.values():

    #  Get the division and population for this state
    
    div = rec['Division']
    pop = rec['pop']

    #  Add the population to the division's total
    
    if div in division_total:
        division_total[div] = division_total[div] + pop
    else:
        division_total[div] = pop

#%%
#
#  Compute each state's share in its division population
#

for rec in name_by_fips.values():

    #  Get the division total 
    
    div = rec['Division']
    div_pop = division_total[div]
    
    #  Compute the percentage and add it to the state record
    
    pct = 100*rec['pop']/div_pop 
    rec['percent'] = pct
    
#%%
#
#  Write out the merged data. First step is to get the list of fields
#  from one of the entries in the dictionary. Any will do so use New York, 
#  which is FIPS 36.
#

fields = list( name_by_fips["36"].keys() )
print( fields )

#
#  Open the output file. The newline argument is needed to avoid extra
#  blank lines because DictWriter() adds them itself.

fh = open('merged.csv','w',newline='')

#
#  Set up the writer object
#

writer = csv.DictWriter(fh,fields)

#
#  Write the field names to the first line
#

writer.writeheader()

# 
#  Write the rest of the entries
#

for name_rec in name_by_fips.values():
    writer.writerow(name_rec)

fh.close()