yt_dlp/extractor/bandcamp.py

   1 import random
   2 import re
   3 import time
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_str
   7 from ..utils import (
   8     KNOWN_EXTENSIONS,
   9     ExtractorError,
  10     float_or_none,
  11     int_or_none,
  12     parse_filesize,
  13     str_or_none,
  14     try_get,
  15     unified_strdate,
  16     unified_timestamp,
  17     update_url_query,
  18     url_or_none,
  19     urljoin,
  20 )
  21
  22
  23 class BandcampIE(InfoExtractor):
  24     _VALID_URL = r'https?://(?P<uploader>[^/]+)\.bandcamp\.com/track/(?P<id>[^/?#&]+)'
  25     _EMBED_REGEX = [r'<meta property="og:url"[^>]*?content="(?P<url>.*?bandcamp\.com.*?)"']
  26     _TESTS = [{
  27         'url': 'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
  28         'md5': 'c557841d5e50261777a6585648adf439',
  29         'info_dict': {
  30             'id': '1812978515',
  31             'ext': 'mp3',
  32             'title': "youtube-dl  \"'/\\ä↭ - youtube-dl  \"'/\\ä↭ - youtube-dl test song \"'/\\ä↭",
  33             'duration': 9.8485,
  34             'uploader': 'youtube-dl  "\'/\\ä↭',
  35             'upload_date': '20121129',
  36             'timestamp': 1354224127,
  37         },
  38         '_skip': 'There is a limit of 200 free downloads / month for the test song'
  39     }, {
  40         # free download
  41         'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
  42         'info_dict': {
  43             'id': '2650410135',
  44             'ext': 'aiff',
  45             'title': 'Ben Prunty - Lanius (Battle)',
  46             'thumbnail': r're:^https?://.*\.jpg$',
  47             'uploader': 'Ben Prunty',
  48             'timestamp': 1396508491,
  49             'upload_date': '20140403',
  50             'release_timestamp': 1396483200,
  51             'release_date': '20140403',
  52             'duration': 260.877,
  53             'track': 'Lanius (Battle)',
  54             'track_number': 1,
  55             'track_id': '2650410135',
  56             'artist': 'Ben Prunty',
  57             'album': 'FTL: Advanced Edition Soundtrack',
  58         },
  59     }, {
  60         # no free download, mp3 128
  61         'url': 'https://relapsealumni.bandcamp.com/track/hail-to-fire',
  62         'md5': 'fec12ff55e804bb7f7ebeb77a800c8b7',
  63         'info_dict': {
  64             'id': '2584466013',
  65             'ext': 'mp3',
  66             'title': 'Mastodon - Hail to Fire',
  67             'thumbnail': r're:^https?://.*\.jpg$',
  68             'uploader': 'Mastodon',
  69             'timestamp': 1322005399,
  70             'upload_date': '20111122',
  71             'release_timestamp': 1076112000,
  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 _extract_data_attr(self, webpage, video_id, attr='tralbum', fatal=True):
  83         return self._parse_json(self._html_search_regex(
  84             r'data-%s=(["\'])({.+?})\1' % attr, webpage,
  85             attr + ' data', group=2), video_id, fatal=fatal)
  86
  87     def _real_extract(self, url):
  88         title, uploader = self._match_valid_url(url).group('id', 'uploader')
  89         webpage = self._download_webpage(url, title)
  90         tralbum = self._extract_data_attr(webpage, title)
  91         thumbnail = self._og_search_thumbnail(webpage)
  92
  93         track_id = None
  94         track = None
  95         track_number = None
  96         duration = None
  97
  98         formats = []
  99         track_info = try_get(tralbum, lambda x: x['trackinfo'][0], dict)
 100         if track_info:
 101             file_ = track_info.get('file')
 102             if isinstance(file_, dict):
 103                 for format_id, format_url in file_.items():
 104                     if not url_or_none(format_url):
 105                         continue
 106                     ext, abr_str = format_id.split('-', 1)
 107                     formats.append({
 108                         'format_id': format_id,
 109                         'url': self._proto_relative_url(format_url, 'http:'),
 110                         'ext': ext,
 111                         'vcodec': 'none',
 112                         'acodec': ext,
 113                         'abr': int_or_none(abr_str),
 114                     })
 115             track = track_info.get('title')
 116             track_id = str_or_none(
 117                 track_info.get('track_id') or track_info.get('id'))
 118             track_number = int_or_none(track_info.get('track_num'))
 119             duration = float_or_none(track_info.get('duration'))
 120
 121         embed = self._extract_data_attr(webpage, title, 'embed', False)
 122         current = tralbum.get('current') or {}
 123         artist = embed.get('artist') or current.get('artist') or tralbum.get('artist')
 124         timestamp = unified_timestamp(
 125             current.get('publish_date') or tralbum.get('album_publish_date'))
 126
 127         download_link = tralbum.get('freeDownloadPage')
 128         if download_link:
 129             track_id = compat_str(tralbum['id'])
 130
 131             download_webpage = self._download_webpage(
 132                 download_link, track_id, 'Downloading free downloads page')
 133
 134             blob = self._extract_data_attr(download_webpage, track_id, 'blob')
 135
 136             info = try_get(
 137                 blob, (lambda x: x['digital_items'][0],
 138                        lambda x: x['download_items'][0]), dict)
 139             if info:
 140                 downloads = info.get('downloads')
 141                 if isinstance(downloads, dict):
 142                     if not track:
 143                         track = info.get('title')
 144                     if not artist:
 145                         artist = info.get('artist')
 146                     if not thumbnail:
 147                         thumbnail = info.get('thumb_url')
 148
 149                     download_formats = {}
 150                     download_formats_list = blob.get('download_formats')
 151                     if isinstance(download_formats_list, list):
 152                         for f in blob['download_formats']:
 153                             name, ext = f.get('name'), f.get('file_extension')
 154                             if all(isinstance(x, compat_str) for x in (name, ext)):
 155                                 download_formats[name] = ext.strip('.')
 156
 157                     for format_id, f in downloads.items():
 158                         format_url = f.get('url')
 159                         if not format_url:
 160                             continue
 161                         # Stat URL generation algorithm is reverse engineered from
 162                         # download_*_bundle_*.js
 163                         stat_url = update_url_query(
 164                             format_url.replace('/download/', '/statdownload/'), {
 165                                 '.rand': int(time.time() * 1000 * random.random()),
 166                             })
 167                         format_id = f.get('encoding_name') or format_id
 168                         stat = self._download_json(
 169                             stat_url, track_id, 'Downloading %s JSON' % format_id,
 170                             transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1],
 171                             fatal=False)
 172                         if not stat:
 173                             continue
 174                         retry_url = url_or_none(stat.get('retry_url'))
 175                         if not retry_url:
 176                             continue
 177                         formats.append({
 178                             'url': self._proto_relative_url(retry_url, 'http:'),
 179                             'ext': download_formats.get(format_id),
 180                             'format_id': format_id,
 181                             'format_note': f.get('description'),
 182                             'filesize': parse_filesize(f.get('size_mb')),
 183                             'vcodec': 'none',
 184                             'acodec': format_id.split('-')[0],
 185                         })
 186
 187         self._sort_formats(formats)
 188
 189         title = '%s - %s' % (artist, track) if artist else track
 190
 191         if not duration:
 192             duration = float_or_none(self._html_search_meta(
 193                 'duration', webpage, default=None))
 194
 195         return {
 196             'id': track_id,
 197             'title': title,
 198             'thumbnail': thumbnail,
 199             'uploader': artist,
 200             'uploader_id': uploader,
 201             'uploader_url': f'https://{uploader}.bandcamp.com',
 202             'timestamp': timestamp,
 203             'release_timestamp': unified_timestamp(tralbum.get('album_release_date')),
 204             'duration': duration,
 205             'track': track,
 206             'track_number': track_number,
 207             'track_id': track_id,
 208             'artist': artist,
 209             'album': embed.get('album_title'),
 210             'formats': formats,
 211         }
 212
 213
 214 class BandcampAlbumIE(BandcampIE):
 215     IE_NAME = 'Bandcamp:album'
 216     _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com/album/(?P<id>[^/?#&]+)'
 217
 218     _TESTS = [{
 219         'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
 220         'playlist': [
 221             {
 222                 'md5': '39bc1eded3476e927c724321ddf116cf',
 223                 'info_dict': {
 224                     'id': '1353101989',
 225                     'ext': 'mp3',
 226                     'title': 'Blazo - Intro',
 227                     'timestamp': 1311756226,
 228                     'upload_date': '20110727',
 229                     'uploader': 'Blazo',
 230                 }
 231             },
 232             {
 233                 'md5': '1a2c32e2691474643e912cc6cd4bffaa',
 234                 'info_dict': {
 235                     'id': '38097443',
 236                     'ext': 'mp3',
 237                     'title': 'Blazo - Kero One - Keep It Alive (Blazo remix)',
 238                     'timestamp': 1311757238,
 239                     'upload_date': '20110727',
 240                     'uploader': 'Blazo',
 241                 }
 242             },
 243         ],
 244         'info_dict': {
 245             'title': 'Jazz Format Mixtape vol.1',
 246             'id': 'jazz-format-mixtape-vol-1',
 247             'uploader_id': 'blazo',
 248         },
 249         'params': {
 250             'playlistend': 2
 251         },
 252         'skip': 'Bandcamp imposes download limits.'
 253     }, {
 254         'url': 'http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave',
 255         'info_dict': {
 256             'title': 'Hierophany of the Open Grave',
 257             'uploader_id': 'nightbringer',
 258             'id': 'hierophany-of-the-open-grave',
 259         },
 260         'playlist_mincount': 9,
 261     }, {
 262         # with escaped quote in title
 263         'url': 'https://jstrecords.bandcamp.com/album/entropy-ep',
 264         'info_dict': {
 265             'title': '"Entropy" EP',
 266             'uploader_id': 'jstrecords',
 267             'id': 'entropy-ep',
 268             'description': 'md5:0ff22959c943622972596062f2f366a5',
 269         },
 270         'playlist_mincount': 3,
 271     }, {
 272         # not all tracks have songs
 273         'url': 'https://insulters.bandcamp.com/album/we-are-the-plague',
 274         'info_dict': {
 275             'id': 'we-are-the-plague',
 276             'title': 'WE ARE THE PLAGUE',
 277             'uploader_id': 'insulters',
 278             'description': 'md5:b3cf845ee41b2b1141dc7bde9237255f',
 279         },
 280         'playlist_count': 2,
 281     }]
 282
 283     @classmethod
 284     def suitable(cls, url):
 285         return (False
 286                 if BandcampWeeklyIE.suitable(url) or BandcampIE.suitable(url)
 287                 else super(BandcampAlbumIE, cls).suitable(url))
 288
 289     def _real_extract(self, url):
 290         uploader_id, album_id = self._match_valid_url(url).groups()
 291         playlist_id = album_id or uploader_id
 292         webpage = self._download_webpage(url, playlist_id)
 293         tralbum = self._extract_data_attr(webpage, playlist_id)
 294         track_info = tralbum.get('trackinfo')
 295         if not track_info:
 296             raise ExtractorError('The page doesn\'t contain any tracks')
 297         # Only tracks with duration info have songs
 298         entries = [
 299             self.url_result(
 300                 urljoin(url, t['title_link']), BandcampIE.ie_key(),
 301                 str_or_none(t.get('track_id') or t.get('id')), t.get('title'))
 302             for t in track_info
 303             if t.get('duration')]
 304
 305         current = tralbum.get('current') or {}
 306
 307         return {
 308             '_type': 'playlist',
 309             'uploader_id': uploader_id,
 310             'id': playlist_id,
 311             'title': current.get('title'),
 312             'description': current.get('about'),
 313             'entries': entries,
 314         }
 315
 316
 317 class BandcampWeeklyIE(BandcampIE):
 318     IE_NAME = 'Bandcamp:weekly'
 319     _VALID_URL = r'https?://(?:www\.)?bandcamp\.com/?\?(?:.*?&)?show=(?P<id>\d+)'
 320     _TESTS = [{
 321         'url': 'https://bandcamp.com/?show=224',
 322         'md5': 'b00df799c733cf7e0c567ed187dea0fd',
 323         'info_dict': {
 324             'id': '224',
 325             'ext': 'opus',
 326             'title': 'BC Weekly April 4th 2017 - Magic Moments',
 327             'description': 'md5:5d48150916e8e02d030623a48512c874',
 328             'duration': 5829.77,
 329             'release_date': '20170404',
 330             'series': 'Bandcamp Weekly',
 331             'episode': 'Magic Moments',
 332             'episode_id': '224',
 333         },
 334         'params': {
 335             'format': 'opus-lo',
 336         },
 337     }, {
 338         'url': 'https://bandcamp.com/?blah/blah@&show=228',
 339         'only_matching': True
 340     }]
 341
 342     def _real_extract(self, url):
 343         show_id = self._match_id(url)
 344         webpage = self._download_webpage(url, show_id)
 345
 346         blob = self._extract_data_attr(webpage, show_id, 'blob')
 347
 348         show = blob['bcw_data'][show_id]
 349
 350         formats = []
 351         for format_id, format_url in show['audio_stream'].items():
 352             if not url_or_none(format_url):
 353                 continue
 354             for known_ext in KNOWN_EXTENSIONS:
 355                 if known_ext in format_id:
 356                     ext = known_ext
 357                     break
 358             else:
 359                 ext = None
 360             formats.append({
 361                 'format_id': format_id,
 362                 'url': format_url,
 363                 'ext': ext,
 364                 'vcodec': 'none',
 365             })
 366         self._sort_formats(formats)
 367
 368         title = show.get('audio_title') or 'Bandcamp Weekly'
 369         subtitle = show.get('subtitle')
 370         if subtitle:
 371             title += ' - %s' % subtitle
 372
 373         return {
 374             'id': show_id,
 375             'title': title,
 376             'description': show.get('desc') or show.get('short_desc'),
 377             'duration': float_or_none(show.get('audio_duration')),
 378             'is_live': False,
 379             'release_date': unified_strdate(show.get('published_date')),
 380             'series': 'Bandcamp Weekly',
 381             'episode': show.get('subtitle'),
 382             'episode_id': show_id,
 383             'formats': formats
 384         }
 385
 386
 387 class BandcampUserIE(InfoExtractor):
 388     IE_NAME = 'Bandcamp:user'
 389     _VALID_URL = r'https?://(?!www\.)(?P<id>[^.]+)\.bandcamp\.com(?:/music)?/?(?:[#?]|$)'
 390
 391     _TESTS = [{
 392         # Type 1 Bandcamp user page.
 393         'url': 'https://adrianvonziegler.bandcamp.com',
 394         'info_dict': {
 395             'id': 'adrianvonziegler',
 396             'title': 'Discography of adrianvonziegler',
 397         },
 398         'playlist_mincount': 23,
 399     }, {
 400         # Bandcamp user page with only one album
 401         'url': 'http://dotscale.bandcamp.com',
 402         'info_dict': {
 403             'id': 'dotscale',
 404             'title': 'Discography of dotscale'
 405         },
 406         'playlist_count': 1,
 407     }, {
 408         # Type 2 Bandcamp user page.
 409         'url': 'https://nightcallofficial.bandcamp.com',
 410         'info_dict': {
 411             'id': 'nightcallofficial',
 412             'title': 'Discography of nightcallofficial',
 413         },
 414         'playlist_count': 4,
 415     }, {
 416         'url': 'https://steviasphere.bandcamp.com/music',
 417         'playlist_mincount': 47,
 418         'info_dict': {
 419             'id': 'steviasphere',
 420             'title': 'Discography of steviasphere',
 421         },
 422     }, {
 423         'url': 'https://coldworldofficial.bandcamp.com/music',
 424         'playlist_mincount': 10,
 425         'info_dict': {
 426             'id': 'coldworldofficial',
 427             'title': 'Discography of coldworldofficial',
 428         },
 429     }, {
 430         'url': 'https://nuclearwarnowproductions.bandcamp.com/music',
 431         'playlist_mincount': 399,
 432         'info_dict': {
 433             'id': 'nuclearwarnowproductions',
 434             'title': 'Discography of nuclearwarnowproductions',
 435         },
 436     }]
 437
 438     def _real_extract(self, url):
 439         uploader = self._match_id(url)
 440         webpage = self._download_webpage(url, uploader)
 441
 442         discography_data = (re.findall(r'<li data-item-id=["\'][^>]+>\s*<a href=["\'](?![^"\'/]*?/merch)([^"\']+)', webpage)
 443                             or re.findall(r'<div[^>]+trackTitle["\'][^"\']+["\']([^"\']+)', webpage))
 444
 445         return self.playlist_from_matches(
 446             discography_data, uploader, f'Discography of {uploader}', getter=lambda x: urljoin(url, x))