youtube_dl/extractor/bandcamp.py

   1 from __future__ import unicode_literals
   2
   3 import random
   4 import re
   5 import time
   6
   7 from .common import InfoExtractor
   8 from ..compat import (
   9     compat_str,
  10     compat_urlparse,
  11 )
  12 from ..utils import (
  13     ExtractorError,
  14     float_or_none,
  15     int_or_none,
  16     KNOWN_EXTENSIONS,
  17     parse_filesize,
  18     str_or_none,
  19     try_get,
  20     unescapeHTML,
  21     update_url_query,
  22     unified_strdate,
  23     unified_timestamp,
  24     url_or_none,
  25 )
  26
  27
  28 class BandcampIE(InfoExtractor):
  29     _VALID_URL = r'https?://[^/]+\.bandcamp\.com/track/(?P<title>[^/?#&]+)'
  30     _TESTS = [{
  31         'url': 'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
  32         'md5': 'c557841d5e50261777a6585648adf439',
  33         'info_dict': {
  34             'id': '1812978515',
  35             'ext': 'mp3',
  36             'title': "youtube-dl  \"'/\\\u00e4\u21ad - youtube-dl test song \"'/\\\u00e4\u21ad",
  37             'duration': 9.8485,
  38         },
  39         '_skip': 'There is a limit of 200 free downloads / month for the test song'
  40     }, {
  41         # free download
  42         'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
  43         'md5': '853e35bf34aa1d6fe2615ae612564b36',
  44         'info_dict': {
  45             'id': '2650410135',
  46             'ext': 'aiff',
  47             'title': 'Ben Prunty - Lanius (Battle)',
  48             'thumbnail': r're:^https?://.*\.jpg$',
  49             'uploader': 'Ben Prunty',
  50             'timestamp': 1396508491,
  51             'upload_date': '20140403',
  52             'release_date': '20140403',
  53             'duration': 260.877,
  54             'track': 'Lanius (Battle)',
  55             'track_number': 1,
  56             'track_id': '2650410135',
  57             'artist': 'Ben Prunty',
  58             'album': 'FTL: Advanced Edition Soundtrack',
  59         },
  60     }, {
  61         # no free download, mp3 128
  62         'url': 'https://relapsealumni.bandcamp.com/track/hail-to-fire',
  63         'md5': 'fec12ff55e804bb7f7ebeb77a800c8b7',
  64         'info_dict': {
  65             'id': '2584466013',
  66             'ext': 'mp3',
  67             'title': 'Mastodon - Hail to Fire',
  68             'thumbnail': r're:^https?://.*\.jpg$',
  69             'uploader': 'Mastodon',
  70             'timestamp': 1322005399,
  71             'upload_date': '20111122',
  72             'release_date': '20040207',
  73             'duration': 120.79,
  74             'track': 'Hail to Fire',
  75             'track_number': 5,
  76             'track_id': '2584466013',
  77             'artist': 'Mastodon',
  78             'album': 'Call of the Mastodon',
  79         },
  80     }]
  81
  82     def _real_extract(self, url):
  83         mobj = re.match(self._VALID_URL, url)
  84         title = mobj.group('title')
  85         webpage = self._download_webpage(url, title)
  86         thumbnail = self._html_search_meta('og:image', webpage, default=None)
  87
  88         track_id = None
  89         track = None
  90         track_number = None
  91         duration = None
  92
  93         formats = []
  94         track_info = self._parse_json(
  95             self._search_regex(
  96                 r'trackinfo\s*:\s*\[\s*({.+?})\s*\]\s*,\s*?\n',
  97                 webpage, 'track info', default='{}'), title)
  98         if track_info:
  99             file_ = track_info.get('file')
 100             if isinstance(file_, dict):
 101                 for format_id, format_url in file_.items():
 102                     if not url_or_none(format_url):
 103                         continue
 104                     ext, abr_str = format_id.split('-', 1)
 105                     formats.append({
 106                         'format_id': format_id,
 107                         'url': self._proto_relative_url(format_url, 'http:'),
 108                         'ext': ext,
 109                         'vcodec': 'none',
 110                         'acodec': ext,
 111                         'abr': int_or_none(abr_str),
 112                     })
 113             track = track_info.get('title')
 114             track_id = str_or_none(track_info.get('track_id') or track_info.get('id'))
 115             track_number = int_or_none(track_info.get('track_num'))
 116             duration = float_or_none(track_info.get('duration'))
 117
 118         def extract(key):
 119             return self._search_regex(
 120                 r'\b%s\s*["\']?\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1' % key,
 121                 webpage, key, default=None, group='value')
 122
 123         artist = extract('artist')
 124         album = extract('album_title')
 125         timestamp = unified_timestamp(
 126             extract('publish_date') or extract('album_publish_date'))
 127         release_date = unified_strdate(extract('album_release_date'))
 128
 129         download_link = self._search_regex(
 130             r'freeDownloadPage\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
 131             'download link', default=None, group='url')
 132         if download_link:
 133             track_id = self._search_regex(
 134                 r'(?ms)var TralbumData = .*?[{,]\s*id: (?P<id>\d+),?$',
 135                 webpage, 'track id')
 136
 137             download_webpage = self._download_webpage(
 138                 download_link, track_id, 'Downloading free downloads page')
 139
 140             blob = self._parse_json(
 141                 self._search_regex(
 142                     r'data-blob=(["\'])(?P<blob>{.+?})\1', download_webpage,
 143                     'blob', group='blob'),
 144                 track_id, transform_source=unescapeHTML)
 145
 146             info = try_get(
 147                 blob, (lambda x: x['digital_items'][0],
 148                        lambda x: x['download_items'][0]), dict)
 149             if info:
 150                 downloads = info.get('downloads')
 151                 if isinstance(downloads, dict):
 152                     if not track:
 153                         track = info.get('title')
 154                     if not artist:
 155                         artist = info.get('artist')
 156                     if not thumbnail:
 157                         thumbnail = info.get('thumb_url')
 158
 159                     download_formats = {}
 160                     download_formats_list = blob.get('download_formats')
 161                     if isinstance(download_formats_list, list):
 162                         for f in blob['download_formats']:
 163                             name, ext = f.get('name'), f.get('file_extension')
 164                             if all(isinstance(x, compat_str) for x in (name, ext)):
 165                                 download_formats[name] = ext.strip('.')
 166
 167                     for format_id, f in downloads.items():
 168                         format_url = f.get('url')
 169                         if not format_url:
 170                             continue
 171                         # Stat URL generation algorithm is reverse engineered from
 172                         # download_*_bundle_*.js
 173                         stat_url = update_url_query(
 174                             format_url.replace('/download/', '/statdownload/'), {
 175                                 '.rand': int(time.time() * 1000 * random.random()),
 176                             })
 177                         format_id = f.get('encoding_name') or format_id
 178                         stat = self._download_json(
 179                             stat_url, track_id, 'Downloading %s JSON' % format_id,
 180                             transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1],
 181                             fatal=False)
 182                         if not stat:
 183                             continue
 184                         retry_url = url_or_none(stat.get('retry_url'))
 185                         if not retry_url:
 186                             continue
 187                         formats.append({
 188                             'url': self._proto_relative_url(retry_url, 'http:'),
 189                             'ext': download_formats.get(format_id),
 190                             'format_id': format_id,
 191                             'format_note': f.get('description'),
 192                             'filesize': parse_filesize(f.get('size_mb')),
 193                             'vcodec': 'none',
 194                         })
 195
 196         self._sort_formats(formats)
 197
 198         title = '%s - %s' % (artist, track) if artist else track
 199
 200         if not duration:
 201             duration = float_or_none(self._html_search_meta(
 202                 'duration', webpage, default=None))
 203
 204         return {
 205             'id': track_id,
 206             'title': title,
 207             'thumbnail': thumbnail,
 208             'uploader': artist,
 209             'timestamp': timestamp,
 210             'release_date': release_date,
 211             'duration': duration,
 212             'track': track,
 213             'track_number': track_number,
 214             'track_id': track_id,
 215             'artist': artist,
 216             'album': album,
 217             'formats': formats,
 218         }
 219
 220
 221 class BandcampAlbumIE(InfoExtractor):
 222     IE_NAME = 'Bandcamp:album'
 223     _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<album_id>[^/?#&]+))?'
 224
 225     _TESTS = [{
 226         'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
 227         'playlist': [
 228             {
 229                 'md5': '39bc1eded3476e927c724321ddf116cf',
 230                 'info_dict': {
 231                     'id': '1353101989',
 232                     'ext': 'mp3',
 233                     'title': 'Intro',
 234                 }
 235             },
 236             {
 237                 'md5': '1a2c32e2691474643e912cc6cd4bffaa',
 238                 'info_dict': {
 239                     'id': '38097443',
 240                     'ext': 'mp3',
 241                     'title': 'Kero One - Keep It Alive (Blazo remix)',
 242                 }
 243             },
 244         ],
 245         'info_dict': {
 246             'title': 'Jazz Format Mixtape vol.1',
 247             'id': 'jazz-format-mixtape-vol-1',
 248             'uploader_id': 'blazo',
 249         },
 250         'params': {
 251             'playlistend': 2
 252         },
 253         'skip': 'Bandcamp imposes download limits.'
 254     }, {
 255         'url': 'http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave',
 256         'info_dict': {
 257             'title': 'Hierophany of the Open Grave',
 258             'uploader_id': 'nightbringer',
 259             'id': 'hierophany-of-the-open-grave',
 260         },
 261         'playlist_mincount': 9,
 262     }, {
 263         'url': 'http://dotscale.bandcamp.com',
 264         'info_dict': {
 265             'title': 'Loom',
 266             'id': 'dotscale',
 267             'uploader_id': 'dotscale',
 268         },
 269         'playlist_mincount': 7,
 270     }, {
 271         # with escaped quote in title
 272         'url': 'https://jstrecords.bandcamp.com/album/entropy-ep',
 273         'info_dict': {
 274             'title': '"Entropy" EP',
 275             'uploader_id': 'jstrecords',
 276             'id': 'entropy-ep',
 277         },
 278         'playlist_mincount': 3,
 279     }, {
 280         # not all tracks have songs
 281         'url': 'https://insulters.bandcamp.com/album/we-are-the-plague',
 282         'info_dict': {
 283             'id': 'we-are-the-plague',
 284             'title': 'WE ARE THE PLAGUE',
 285             'uploader_id': 'insulters',
 286         },
 287         'playlist_count': 2,
 288     }]
 289
 290     @classmethod
 291     def suitable(cls, url):
 292         return (False
 293                 if BandcampWeeklyIE.suitable(url) or BandcampIE.suitable(url)
 294                 else super(BandcampAlbumIE, cls).suitable(url))
 295
 296     def _real_extract(self, url):
 297         mobj = re.match(self._VALID_URL, url)
 298         uploader_id = mobj.group('subdomain')
 299         album_id = mobj.group('album_id')
 300         playlist_id = album_id or uploader_id
 301         webpage = self._download_webpage(url, playlist_id)
 302         track_elements = re.findall(
 303             r'(?s)<div[^>]*>(.*?<a[^>]+href="([^"]+?)"[^>]+itemprop="url"[^>]*>.*?)</div>', webpage)
 304         if not track_elements:
 305             raise ExtractorError('The page doesn\'t contain any tracks')
 306         # Only tracks with duration info have songs
 307         entries = [
 308             self.url_result(
 309                 compat_urlparse.urljoin(url, t_path),
 310                 ie=BandcampIE.ie_key(),
 311                 video_title=self._search_regex(
 312                     r'<span\b[^>]+\bitemprop=["\']name["\'][^>]*>([^<]+)',
 313                     elem_content, 'track title', fatal=False))
 314             for elem_content, t_path in track_elements
 315             if self._html_search_meta('duration', elem_content, default=None)]
 316
 317         title = self._html_search_regex(
 318             r'album_title\s*:\s*"((?:\\.|[^"\\])+?)"',
 319             webpage, 'title', fatal=False)
 320         if title:
 321             title = title.replace(r'\"', '"')
 322         return {
 323             '_type': 'playlist',
 324             'uploader_id': uploader_id,
 325             'id': playlist_id,
 326             'title': title,
 327             'entries': entries,
 328         }
 329
 330
 331 class BandcampWeeklyIE(InfoExtractor):
 332     IE_NAME = 'Bandcamp:weekly'
 333     _VALID_URL = r'https?://(?:www\.)?bandcamp\.com/?\?(?:.*?&)?show=(?P<id>\d+)'
 334     _TESTS = [{
 335         'url': 'https://bandcamp.com/?show=224',
 336         'md5': 'b00df799c733cf7e0c567ed187dea0fd',
 337         'info_dict': {
 338             'id': '224',
 339             'ext': 'opus',
 340             'title': 'BC Weekly April 4th 2017 - Magic Moments',
 341             'description': 'md5:5d48150916e8e02d030623a48512c874',
 342             'duration': 5829.77,
 343             'release_date': '20170404',
 344             'series': 'Bandcamp Weekly',
 345             'episode': 'Magic Moments',
 346             'episode_number': 208,
 347             'episode_id': '224',
 348         }
 349     }, {
 350         'url': 'https://bandcamp.com/?blah/blah@&show=228',
 351         'only_matching': True
 352     }]
 353
 354     def _real_extract(self, url):
 355         video_id = self._match_id(url)
 356         webpage = self._download_webpage(url, video_id)
 357
 358         blob = self._parse_json(
 359             self._search_regex(
 360                 r'data-blob=(["\'])(?P<blob>{.+?})\1', webpage,
 361                 'blob', group='blob'),
 362             video_id, transform_source=unescapeHTML)
 363
 364         show = blob['bcw_show']
 365
 366         # This is desired because any invalid show id redirects to `bandcamp.com`
 367         # which happens to expose the latest Bandcamp Weekly episode.
 368         show_id = int_or_none(show.get('show_id')) or int_or_none(video_id)
 369
 370         formats = []
 371         for format_id, format_url in show['audio_stream'].items():
 372             if not url_or_none(format_url):
 373                 continue
 374             for known_ext in KNOWN_EXTENSIONS:
 375                 if known_ext in format_id:
 376                     ext = known_ext
 377                     break
 378             else:
 379                 ext = None
 380             formats.append({
 381                 'format_id': format_id,
 382                 'url': format_url,
 383                 'ext': ext,
 384                 'vcodec': 'none',
 385             })
 386         self._sort_formats(formats)
 387
 388         title = show.get('audio_title') or 'Bandcamp Weekly'
 389         subtitle = show.get('subtitle')
 390         if subtitle:
 391             title += ' - %s' % subtitle
 392
 393         episode_number = None
 394         seq = blob.get('bcw_seq')
 395
 396         if seq and isinstance(seq, list):
 397             try:
 398                 episode_number = next(
 399                     int_or_none(e.get('episode_number'))
 400                     for e in seq
 401                     if isinstance(e, dict) and int_or_none(e.get('id')) == show_id)
 402             except StopIteration:
 403                 pass
 404
 405         return {
 406             'id': video_id,
 407             'title': title,
 408             'description': show.get('desc') or show.get('short_desc'),
 409             'duration': float_or_none(show.get('audio_duration')),
 410             'is_live': False,
 411             'release_date': unified_strdate(show.get('published_date')),
 412             'series': 'Bandcamp Weekly',
 413             'episode': show.get('subtitle'),
 414             'episode_number': episode_number,
 415             'episode_id': compat_str(video_id),
 416             'formats': formats
 417         }