# -*- coding: cp949 -*-
import sys
import urllib2
import csv
from bs4 import BeautifulSoup


def parse_chart(html):
    soup   = BeautifulSoup(html)

    date   = soup.find('div', class_='dateChart clearfix').p.span.text
    tables = soup.select('.boardListWhite')
    top50  = tables[0].tbody.find_all('tr', recursive=False)
    top100 = tables[1].tbody.find_all('tr', recursive=False)

    #with open('top50.html', 'w') as f:
    #    for item in top50:
    #        f.write(item.prettify().encode('utf-8'))

    #with open('top100.html', 'w') as f:
    #    for item in top100:
    #       f.write(item.prettify().encode('utf-8'))

    chart = parse_tr(top50)
    chart += parse_tr(top100)

    return date, tuple(chart)

def parse_tr(trs):

    items = []

    for tr in trs:
        tds = tr.find_all('td', recursive=False)

        # 0th td: Checkbox. No Process

        # 1st td: Rank. If 'alt' has not a number, skip this tds.
        alt  = tds[1].img['alt']
        rank = 0
        if alt.isnumeric() == False:
            continue
        else:
            rank = int(alt)

        # 2nd td: delta
        updn  = tds[2].span['class'][0]
        delta = tds[2].span.text
        if updn == u'up2':
            delta = u'+' + delta
        elif updn == u'down2':
            delta = u'-' + delta

        # 3rd td: song name, artist name, album title.
        links    = tds[3].find_all('a', title=True)
        morelist = tds[3].find('span', class_='overList')

        if morelist:
            song   = links[0]['title']
            artist = ', '.join([a.text for a in morelist.find_all('a')])
            album  = links[-1]['title']
        else:
            song   = links[0]['title']
            artist = links[1]['title']
            album  = links[2]['title']

        items.append((rank, delta, song, artist, album))

    return items

def get_html(url, maxbuf = 10485760):
    res  = urllib2.urlopen(url)
    html = res.read(maxbuf)
    res.close()

    #f = open('melon_weekly.html', 'w')
    #f.write(html)
    #f.close()

    #with open('melon_weekly.html', 'r') as f:
    #    html = f.read()

    return html

def main(argv):

    if len(argv) != 2:
        print u'Usage: python melon_top_100.py [daily|weelky|monthly]'
        return 0    
    
    charturl = {'daily': 'http://www.melon.com/static/cds/chart/web/chartdaily_list.html', \
                'weekly': 'http://www.melon.com/static/cds/chart/web/chartdaily_list.html', \
                'monthly': 'http://www.melon.com/static/cds/chart/web/chartmonthly_list.html'}
        
    html         = get_html(charturl[argv[1]])
    title, chart = parse_chart(html)
    file_name    = title + u'.csv'
    
    with open(file_name, 'wb') as f:
        writer = csv.writer(f, csv.excel)
        for item in chart:
            cp949item = [unicode(i).encode('cp949') for i in item]
            writer.writerow(cp949item)
    
    print u'Chart exported to \'%s\'' % file_name    
    return 0

if __name__ == '__main__':
    sys.exit(main(sys.argv))