Get rid of fluff on fields in a CSV

strip.py

"""
Ensure the new and old fields uses the same CSV quoting conventions and format decimals the same way (15.00 vs 15 and 16.10 vs 16.1), so we can run a diff without being distracted those differences.
"""


import csv

fin = csv.reader(open('../../archives/3_csv_original/2011Q3-summary-sunlight.csv','r'))
fout = csv.writer(open('../../archives/3_csv_original/2011Q3-summary-sunlight-stripped.csv','w'))

for line in fin:
    newline = []
    i = 0
    for field in line:
        field = field.strip().strip('.').replace(',','').strip()
        if i>3 and field not in ["YTD","AMOUNT"]: #number. resolve precision issue
            field = float(field)
        newline.append( field )
        i = i+1
    fout.writerow(newline)


fin = csv.reader(open('2011Q3-house-disburse-summary.csv','r'))
fout = csv.writer(open('2011Q3-house-disburse-summary-stripped.csv','w'))

for line in fin:
    newline = []
    i = 0
    for field in line:
        field = field.strip().strip('.').replace(',','').strip()
        if i>3 and field not in ["YTD","AMOUNT"]: #number. resolve precision issue
            field = float(field)
        newline.append( field )
        i = i+1
    fout.writerow(newline)


"""
run this:
diff --suppress-common-lines -y -W 1500 old-detail-stripped.csv new-detail-stripped.csv > diff.txt

"""