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 )
  14
  15
  16 class BandcampIE(InfoExtractor):
  17     _VALID_URL = r'https?://.*?\.bandcamp\.com/track/(?P<title>.*)'
  18     _TESTS = [{
  19         'url': 'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
  20         'md5': 'c557841d5e50261777a6585648adf439',
  21         'info_dict': {
  22             'id': '1812978515',
  23             'ext': 'mp3',
  24             'title': "youtube-dl  \"'/\\\u00e4\u21ad - youtube-dl test song \"'/\\\u00e4\u21ad",
  25             'duration': 9.8485,
  26         },
  27         '_skip': 'There is a limit of 200 free downloads / month for the test song'
  28     }, {
  29         'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
  30         'md5': '2b68e5851514c20efdff2afc5603b8b4',
  31         'info_dict': {
  32             'id': '2650410135',
  33             'ext': 'mp3',
  34             'title': 'Lanius (Battle)',
  35             'uploader': 'Ben Prunty Music',
  36         },
  37     }]
  38
  39     def _real_extract(self, url):
  40         mobj = re.match(self._VALID_URL, url)
  41         title = mobj.group('title')
  42         webpage = self._download_webpage(url, title)
  43         m_download = re.search(r'freeDownloadPage: "(.*?)"', webpage)
  44         if not m_download:
  45             m_trackinfo = re.search(r'trackinfo: (.+),\s*?\n', webpage)
  46             if m_trackinfo:
  47                 json_code = m_trackinfo.group(1)
  48                 data = json.loads(json_code)[0]
  49
  50                 formats = []
  51                 for format_id, format_url in data['file'].items():
  52                     ext, abr_str = format_id.split('-', 1)
  53                     formats.append({
  54                         'format_id': format_id,
  55                         'url': format_url,
  56                         'ext': ext,
  57                         'vcodec': 'none',
  58                         'acodec': ext,
  59                         'abr': int(abr_str),
  60                     })
  61
  62                 self._sort_formats(formats)
  63
  64                 return {
  65                     'id': compat_str(data['id']),
  66                     'title': data['title'],
  67                     'formats': formats,
  68                     'duration': float(data['duration']),
  69                 }
  70             else:
  71                 raise ExtractorError('No free songs found')
  72
  73         download_link = m_download.group(1)
  74         video_id = self._search_regex(
  75             r'(?ms)var TralbumData = .*?[{,]\s*id: (?P<id>\d+),?$',
  76             webpage, 'video id')
  77
  78         download_webpage = self._download_webpage(download_link, video_id, 'Downloading free downloads page')
  79         # We get the dictionary of the track from some javascript code
  80         all_info = self._parse_json(self._search_regex(
  81             r'(?sm)items: (.*?),$', download_webpage, 'items'), video_id)
  82         info = all_info[0]
  83         # We pick mp3-320 for now, until format selection can be easily implemented.
  84         mp3_info = info['downloads']['mp3-320']
  85         # If we try to use this url it says the link has expired
  86         initial_url = mp3_info['url']
  87         m_url = re.match(
  88             r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$',
  89             initial_url)
  90         # We build the url we will use to get the final track url
  91         # This url is build in Bandcamp in the script download_bunde_*.js
  92         request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), video_id, m_url.group('ts'))
  93         final_url_webpage = self._download_webpage(request_url, video_id, 'Requesting download url')
  94         # If we could correctly generate the .rand field the url would be
  95         # in the "download_url" key
  96         final_url = self._search_regex(
  97             r'"retry_url":"(.*?)"', final_url_webpage, 'final video URL')
  98
  99         return {
 100             'id': video_id,
 101             'title': info['title'],
 102             'ext': 'mp3',
 103             'vcodec': 'none',
 104             'url': final_url,
 105             'thumbnail': info.get('thumb_url'),
 106             'uploader': info.get('artist'),
 107         }
 108
 109
 110 class BandcampAlbumIE(InfoExtractor):
 111     IE_NAME = 'Bandcamp:album'
 112     _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<album_id>[^?#]+)|/?(?:$|[?#]))'
 113
 114     _TESTS = [{
 115         'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
 116         'playlist': [
 117             {
 118                 'md5': '39bc1eded3476e927c724321ddf116cf',
 119                 'info_dict': {
 120                     'id': '1353101989',
 121                     'ext': 'mp3',
 122                     'title': 'Intro',
 123                 }
 124             },
 125             {
 126                 'md5': '1a2c32e2691474643e912cc6cd4bffaa',
 127                 'info_dict': {
 128                     'id': '38097443',
 129                     'ext': 'mp3',
 130                     'title': 'Kero One - Keep It Alive (Blazo remix)',
 131                 }
 132             },
 133         ],
 134         'info_dict': {
 135             'title': 'Jazz Format Mixtape vol.1',
 136             'id': 'jazz-format-mixtape-vol-1',
 137             'uploader_id': 'blazo',
 138         },
 139         'params': {
 140             'playlistend': 2
 141         },
 142         'skip': 'Bandcamp imposes download limits.'
 143     }, {
 144         'url': 'http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave',
 145         'info_dict': {
 146             'title': 'Hierophany of the Open Grave',
 147             'uploader_id': 'nightbringer',
 148             'id': 'hierophany-of-the-open-grave',
 149         },
 150         'playlist_mincount': 9,
 151     }, {
 152         'url': 'http://dotscale.bandcamp.com',
 153         'info_dict': {
 154             'title': 'Loom',
 155             'id': 'dotscale',
 156             'uploader_id': 'dotscale',
 157         },
 158         'playlist_mincount': 7,
 159     }]
 160
 161     def _real_extract(self, url):
 162         mobj = re.match(self._VALID_URL, url)
 163         uploader_id = mobj.group('subdomain')
 164         album_id = mobj.group('album_id')
 165         playlist_id = album_id or uploader_id
 166         webpage = self._download_webpage(url, playlist_id)
 167         tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage)
 168         if not tracks_paths:
 169             raise ExtractorError('The page doesn\'t contain any tracks')
 170         entries = [
 171             self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
 172             for t_path in tracks_paths]
 173         title = self._search_regex(
 174             r'album_title\s*:\s*"(.*?)"', webpage, 'title', fatal=False)
 175         return {
 176             '_type': 'playlist',
 177             'uploader_id': uploader_id,
 178             'id': playlist_id,
 179             'title': title,
 180             'entries': entries,
 181         }