wxyc-onnow.py

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

import argparse
import sys

from blessings import Terminal
from lxml.html import parse

RECENTLY_PLAYED_URL = 'http://www.wxyc.info/playlists/recent.html'

def format_row(artist, song, release, label):
    return u'{artist} :: {song} ({release}; {label})'.format(**locals())

def get_recently_played(page, limit=5):
    trs = page.cssselect('tr[bgcolor="#F3F3F3"]')
    return [ tuple(td.text_content().strip() for td in tr.getchildren()[1:-1])
             for tr in trs[:limit] ]

def get_latest_show(page):
    show_gen = (el.text_content().replace('START OF SHOW:', '').strip()
                for el in page.cssselect('tr[bgcolor="#444444"]')
                if el.text_content().strip().startswith('START OF SHOW'))
    try:
        show = next(show_gen)
    except StopIteration:
        show = '(could not determine show name)'
    return show

if __name__ == '__main__':
    parser = argparse.ArgumentParser()
    parser.add_argument('-n', '--number', type=int, default=5,
                        help='Number of recent tracks to show')
    parser.add_argument('--raw', action='store_true',
                        help=('Do not format. Instead print tab-separated '
                              'values for artist, track, album, label'))
    args = parser.parse_args()

    page  = parse(RECENTLY_PLAYED_URL).getroot()
    show_title = get_latest_show(page)
    track_list = get_recently_played(page, args.number)

    if args.raw:
        print('\n'.join(['\t'.join(parts) for parts in track_list]))
    else:
        t = Terminal()
        print('\n' + t.blue_bold(show_title))
        print(t.blue(u'═' * len(show_title)))
        print('\n'.join([format_row(*parts) for parts in track_list]) + '\n')