CC BY-SA 3.0 http://creativecommons.org/licenses/by-sa/3.0/deed.en

moonphase_extractor.py

#!/usr/bin/env python
# -*- coding: utf-8 -*-

import urllib
import HTMLParser

OUTPUT_PATH = None
URL = None

# Set URL here
URL = 'http://m-p.sakura.ne.jp/Html/anime.html#anime'

# Set output file path here
OUTPUT_PATH = '/Users/uranusjr/Desktop/o.txt'


class MoonPhaseParser(HTMLParser.HTMLParser, object):
    def __init__(self):
        super(MoonPhaseParser, self).__init__()
        self.table_row = 0
        self.table_column = 0
        self.current_href = ''
        self.output = ''

    def handle_starttag(self, tag, attrs):
        if tag == 'tr':
            self.handle_tr_start(attrs)
        elif tag == 'td':
            self.handle_td_start(attrs)
        elif tag == 'a':
            for attr in attrs:
                if attr[0] == 'href':
                    self.current_href = attr[1]
                    break

    def handle_data(self, data):
        if self.table_row == 1:
            return
        if self.table_column == 1:
            data = data.strip()
            if data:
                self.output += data + ' '
        elif self.table_column == 2:
            data = data.strip()
            data = ' '.join([c for c in data.split()])
            if data:
                self.output += data + '\n' + self.current_href + '\n\n'
        else:
            return

    def handle_tr_start(self, attrs):
        self.table_row += 1
        self.table_column = 0   # reset column count

    def handle_td_start(self, attrs):
        self.table_column += 1


def main():
    content = unicode(urllib.urlopen(URL).read(), 'shift-jis', 'ignore')
    parser = MoonPhaseParser()
    parser.feed(content)
    if not OUTPUT_PATH:
        print parser.output
    else:
        with open(OUTPUT_PATH, 'w+') as f:
            f.write(parser.output.encode('utf-8'))


if __name__ == '__main__':
    main()