Put columns for multiple geographies of ACS 2010 comparison profiles into more usable file, including combining with annotated file.

flattenfactfinder.py

import csv

fout = csv.writer( open('cpflat.csv','wU') )  
    
def process(i):    
    fin = csv.reader( open('ACS_10_1YR_CP0%s.csv' % i,'r') )   
    fin_ann = csv.reader( open('ACS_10_1YR_CP0%s_ann.csv' % i,'r') ) 

    fin.next()
    headers = fin.next()[3:]
    headers2 = fin.next()[3:]
    headers3 = fin.next()[3:]
    
    fin_ann.next()
    fin_ann.next()
    fin_ann.next()
    fin_ann.next()
    
    headers4 = [None for x in range(len(headers))]
    if i in [2,3]:
        headers4 = fin.next()[3:]
        fin_ann.next()
    headers5 = [None for x in range(len(headers))]
    if i in [3,]:
        headers5 = fin.next()[3:]
        fin_ann.next()
    colseries = fin.next()[3:]
    coltype = fin.next()[3:]
    fin_ann.next()
    fin_ann.next()
    
    if i==2:
        writeheaders = ['cp','geo1','geo2','geo3','h1','h2','h3','h4','h5'] + coltype[:9]
        fout.writerow(writeheaders)
        print writeheaders
    else:
        print i
    
    hedindexes = []
    
    oldhed = None
    hedstart = 0
    for j in range(len(headers)):
        hed = (headers[j], headers2[j], headers3[j], headers4[j], headers5[j])
        if oldhed and oldhed!=hed:
            hedindexes.append( (oldhed, hedstart, j) )
            hedstart = j
        oldhed = hed

    """VC03 #CP2 header row starts on line 6; CP3 on line 7; CP4 on line 5
    HOUSING OCCUPANCY

    Total housing units
    HC01
    2010 Estimate"""

    for line in fin:
        line_ann = fin_ann.next()
        geo = ['CP0%s'%i,] + line[:3]
        geo_ann = ['CP0%s'%i,] + line[:3]
        if geo != geo_ann: print geo, ' != ', geo_ann
        for h in hedindexes:
            hed = h[0]
            start = h[1]
            stop = h[2]
            vals = [ line[start+3], line[start+4], line_ann[start+5],
                line[start+6], line_ann[start+7],
                line[start+8], line_ann[start+9],
                line[start+10], line_ann[start+11],
                ]

            outline = geo + list(hed) + vals
            fout.writerow( outline )

for i in [2,3,4]: #i didn't use CP_05, but it should work in addition to 2, 3 and 4
    process(i)