"""Given an HTML-barf listing, generate a CSV.
"""
import csv
from datetime import date

from BeautifulSoup import BeautifulSoup
soup1 = BeautifulSoup(open('1-50.htm'))
soup2 = BeautifulSoup(open('51-89.htm'))

months = dict()
months['Jan'] = 1
months['Feb'] = 2
months['Mar'] = 3
months['Apr'] = 4
months['May'] = 5
months['Jun'] = 6
months['Jul'] = 7
months['Aug'] = 8
months['Sep'] = 9
months['Oct'] = 10
months['Nov'] = 11
months['Dec'] = 12

want1 = [5, 9, 12, 16, 19, 26, 29, 34, 37, 41, 43]
want2 = [1, 3, 10, 16, 20, 24, 27, 30, 32, 34, 35, 36, 37]

out = csv.writer(open('newsletters.csv', 'w+'))
out.writerow(['mid', 'postdate'])

for (soup, want) in ((soup1, want1), (soup2, want2)):
    for row in soup('tr', bgcolor=['#E0E0E0', '#EAEAEA']):
        cols = row('td')
        rownum = int(cols[0].font.string)
        if rownum not in want:
            continue
        mid = row['id'][3:] # id="row9999999"
        month, day, year = [x.strip(',') for x in cols[4].font.string.split()]
        postdate = date(int(year), months[month], int(day))
        out.writerow([mid, postdate])