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 ..compat import (
   8     compat_str,
   9     compat_urlparse,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     float_or_none,
  14     int_or_none,
  15 )
  16
  17
  18 class BandcampIE(InfoExtractor):
  19     _VALID_URL = r'https?://.*?\.bandcamp\.com/track/(?P<title>.*)'
  20     _TESTS = [{
  21         'url': 'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
  22         'md5': 'c557841d5e50261777a6585648adf439',
  23         'info_dict': {
  24             'id': '1812978515',
  25             'ext': 'mp3',
  26             'title': "youtube-dl  \"'/\\\u00e4\u21ad - youtube-dl test song \"'/\\\u00e4\u21ad",
  27             'duration': 9.8485,
  28         },
  29         '_skip': 'There is a limit of 200 free downloads / month for the test song'
  30     }, {
  31         'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
  32         'md5': '73d0b3171568232574e45652f8720b5c',
  33         'info_dict': {
  34             'id': '2650410135',
  35             'ext': 'mp3',
  36             'title': 'Lanius (Battle)',
  37             'uploader': 'Ben Prunty Music',
  38         },
  39     }]
  40
  41     def _real_extract(self, url):
  42         mobj = re.match(self._VALID_URL, url)
  43         title = mobj.group('title')
  44         webpage = self._download_webpage(url, title)
  45         m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
  46         if not m_download:
  47             m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage)
  48             if m_trackinfo:
  49                 json_code = m_trackinfo.group(1)
  50                 data = json.loads(json_code)[0]
  51                 track_id = compat_str(data['id'])
  52
  53                 if not data.get('file'):
  54                     raise ExtractorError('Not streamable', video_id=track_id, expected=True)
  55
  56                 formats = []
  57                 for format_id, format_url in data['file'].items():
  58                     ext, abr_str = format_id.split('-', 1)
  59                     formats.append({
  60                         'format_id': format_id,
  61                         'url': self._proto_relative_url(format_url, 'http:'),
  62                         'ext': ext,
  63                         'vcodec': 'none',
  64                         'acodec': ext,
  65                         'abr': int_or_none(abr_str),
  66                     })
  67
  68                 self._sort_formats(formats)
  69
  70                 return {
  71                     'id': track_id,
  72                     'title': data['title'],
  73                     'formats': formats,
  74                     'duration': float_or_none(data.get('duration')),
  75                 }
  76             else:
  77                 raise ExtractorError('No free songs found')
  78
  79         download_link = m_download.group(1)
  80         video_id = self._search_regex(
  81             r'(?ms)var TralbumData = .*?[{,]\s*id: (?P<id>\d+),?$',
  82             webpage, 'video id')
  83
  84         download_webpage = self._download_webpage(download_link, video_id, 'Downloading free downloads page')
  85         # We get the dictionary of the track from some javascript code
  86         all_info = self._parse_json(self._search_regex(
  87             r'(?sm)items: (.*?),$', download_webpage, 'items'), video_id)
  88         info = all_info[0]
  89         # We pick mp3-320 for now, until format selection can be easily implemented.
  90         mp3_info = info['downloads']['mp3-320']
  91         # If we try to use this url it says the link has expired
  92         initial_url = mp3_info['url']
  93         m_url = re.match(
  94             r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$',
  95             initial_url)
  96         # We build the url we will use to get the final track url
  97         # This url is build in Bandcamp in the script download_bunde_*.js
  98         request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), video_id, m_url.group('ts'))
  99         final_url_webpage = self._download_webpage(request_url, video_id, 'Requesting download url')
 100         # If we could correctly generate the .rand field the url would be
 101         # in the "download_url" key
 102         final_url = self._proto_relative_url(self._search_regex(
 103             r'"retry_url":"(.+?)"', final_url_webpage, 'final video URL'), 'http:')
 104
 105         return {
 106             'id': video_id,
 107             'title': info['title'],
 108             'ext': 'mp3',
 109             'vcodec': 'none',
 110             'url': final_url,
 111             'thumbnail': info.get('thumb_url'),
 112             'uploader': info.get('artist'),
 113         }
 114
 115
 116 class BandcampAlbumIE(InfoExtractor):
 117     IE_NAME = 'Bandcamp:album'
 118     _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<album_id>[^?#]+)|/?(?:$|[?#]))'
 119
 120     _TESTS = [{
 121         'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
 122         'playlist': [
 123             {
 124                 'md5': '39bc1eded3476e927c724321ddf116cf',
 125                 'info_dict': {
 126                     'id': '1353101989',
 127                     'ext': 'mp3',
 128                     'title': 'Intro',
 129                 }
 130             },
 131             {
 132                 'md5': '1a2c32e2691474643e912cc6cd4bffaa',
 133                 'info_dict': {
 134                     'id': '38097443',
 135                     'ext': 'mp3',
 136                     'title': 'Kero One - Keep It Alive (Blazo remix)',
 137                 }
 138             },
 139         ],
 140         'info_dict': {
 141             'title': 'Jazz Format Mixtape vol.1',
 142             'id': 'jazz-format-mixtape-vol-1',
 143             'uploader_id': 'blazo',
 144         },
 145         'params': {
 146             'playlistend': 2
 147         },
 148         'skip': 'Bandcamp imposes download limits.'
 149     }, {
 150         'url': 'http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave',
 151         'info_dict': {
 152             'title': 'Hierophany of the Open Grave',
 153             'uploader_id': 'nightbringer',
 154             'id': 'hierophany-of-the-open-grave',
 155         },
 156         'playlist_mincount': 9,
 157     }, {
 158         'url': 'http://dotscale.bandcamp.com',
 159         'info_dict': {
 160             'title': 'Loom',
 161             'id': 'dotscale',
 162             'uploader_id': 'dotscale',
 163         },
 164         'playlist_mincount': 7,
 165     }]
 166
 167     def _real_extract(self, url):
 168         mobj = re.match(self._VALID_URL, url)
 169         uploader_id = mobj.group('subdomain')
 170         album_id = mobj.group('album_id')
 171         playlist_id = album_id or uploader_id
 172         webpage = self._download_webpage(url, playlist_id)
 173         tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage)
 174         if not tracks_paths:
 175             raise ExtractorError('The page doesn\'t contain any tracks')
 176         entries = [
 177             self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
 178             for t_path in tracks_paths]
 179         title = self._search_regex(
 180             r'album_title\s*:\s*"(.*?)"', webpage, 'title', fatal=False)
 181         return {
 182             '_type': 'playlist',
 183             'uploader_id': uploader_id,
 184             'id': playlist_id,
 185             'title': title,
 186             'entries': entries,
 187         }