Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/bandcamp.py

   1 from __future__ import unicode_literals
   2
   3 import json
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     compat_str,
   9     compat_urlparse,
  10     ExtractorError,
  11 )
  12
  13
  14 class BandcampIE(InfoExtractor):
  15     _VALID_URL = r'https?://.*?\.bandcamp\.com/track/(?P<title>.*)'
  16     _TESTS = [{
  17         'url': 'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
  18         'md5': 'c557841d5e50261777a6585648adf439',
  19         'info_dict': {
  20             'id': '1812978515',
  21             'ext': 'mp3',
  22             'title': "youtube-dl  \"'/\\\u00e4\u21ad - youtube-dl test song \"'/\\\u00e4\u21ad",
  23             'duration': 9.8485,
  24         },
  25         '_skip': 'There is a limit of 200 free downloads / month for the test song'
  26     }, {
  27         'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
  28         'md5': '2b68e5851514c20efdff2afc5603b8b4',
  29         'info_dict': {
  30             'id': '2650410135',
  31             'ext': 'mp3',
  32             'title': 'Lanius (Battle)',
  33             'uploader': 'Ben Prunty Music',
  34         },
  35     }]
  36
  37     def _real_extract(self, url):
  38         mobj = re.match(self._VALID_URL, url)
  39         title = mobj.group('title')
  40         webpage = self._download_webpage(url, title)
  41         m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
  42         if not m_download:
  43             m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage)
  44             if m_trackinfo:
  45                 json_code = m_trackinfo.group(1)
  46                 data = json.loads(json_code)[0]
  47
  48                 formats = []
  49                 for format_id, format_url in data['file'].items():
  50                     ext, abr_str = format_id.split('-', 1)
  51                     formats.append({
  52                         'format_id': format_id,
  53                         'url': format_url,
  54                         'ext': ext,
  55                         'vcodec': 'none',
  56                         'acodec': ext,
  57                         'abr': int(abr_str),
  58                     })
  59
  60                 self._sort_formats(formats)
  61
  62                 return {
  63                     'id': compat_str(data['id']),
  64                     'title': data['title'],
  65                     'formats': formats,
  66                     'duration': float(data['duration']),
  67                 }
  68             else:
  69                 raise ExtractorError('No free songs found')
  70
  71         download_link = m_download.group(1)
  72         video_id = self._search_regex(
  73             r'var TralbumData = {.*?id: (?P<id>\d+),?$',
  74             webpage, 'video id', flags=re.MULTILINE | re.DOTALL)
  75
  76         download_webpage = self._download_webpage(download_link, video_id, 'Downloading free downloads page')
  77         # We get the dictionary of the track from some javascript code
  78         info = re.search(r'items: (.*?),$', download_webpage, re.MULTILINE).group(1)
  79         info = json.loads(info)[0]
  80         # We pick mp3-320 for now, until format selection can be easily implemented.
  81         mp3_info = info['downloads']['mp3-320']
  82         # If we try to use this url it says the link has expired
  83         initial_url = mp3_info['url']
  84         re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$'
  85         m_url = re.match(re_url, initial_url)
  86         #We build the url we will use to get the final track url
  87         # This url is build in Bandcamp in the script download_bunde_*.js
  88         request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), video_id, m_url.group('ts'))
  89         final_url_webpage = self._download_webpage(request_url, video_id, 'Requesting download url')
  90         # If we could correctly generate the .rand field the url would be
  91         #in the "download_url" key
  92         final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1)
  93
  94         return {
  95             'id': video_id,
  96             'title': info['title'],
  97             'ext': 'mp3',
  98             'vcodec': 'none',
  99             'url': final_url,
 100             'thumbnail': info.get('thumb_url'),
 101             'uploader': info.get('artist'),
 102         }
 103
 104
 105 class BandcampAlbumIE(InfoExtractor):
 106     IE_NAME = 'Bandcamp:album'
 107     _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<title>[^?#]+))'
 108
 109     _TESTS = [{
 110         'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
 111         'playlist': [
 112             {
 113                 'file': '1353101989.mp3',
 114                 'md5': '39bc1eded3476e927c724321ddf116cf',
 115                 'info_dict': {
 116                     'title': 'Intro',
 117                 }
 118             },
 119             {
 120                 'file': '38097443.mp3',
 121                 'md5': '1a2c32e2691474643e912cc6cd4bffaa',
 122                 'info_dict': {
 123                     'title': 'Kero One - Keep It Alive (Blazo remix)',
 124                 }
 125             },
 126         ],
 127         'params': {
 128             'playlistend': 2
 129         },
 130         'skip': 'Bandcamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test'
 131     }, {
 132         'url': 'http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave',
 133         'info_dict': {
 134             'title': 'Hierophany of the Open Grave',
 135         },
 136         'playlist_mincount': 9,
 137     }]
 138
 139     def _real_extract(self, url):
 140         mobj = re.match(self._VALID_URL, url)
 141         playlist_id = mobj.group('subdomain')
 142         title = mobj.group('title')
 143         display_id = title or playlist_id
 144         webpage = self._download_webpage(url, display_id)
 145         tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage)
 146         if not tracks_paths:
 147             raise ExtractorError('The page doesn\'t contain any tracks')
 148         entries = [
 149             self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
 150             for t_path in tracks_paths]
 151         title = self._search_regex(r'album_title : "(.*?)"', webpage, 'title')
 152         return {
 153             '_type': 'playlist',
 154             'id': playlist_id,
 155             'display_id': display_id,
 156             'title': title,
 157             'entries': entries,
 158         }