yt_dlp/extractor/bandcamp.py

   1 import functools
   2 import json
   3 import random
   4 import re
   5 import time
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     KNOWN_EXTENSIONS,
  10     ExtractorError,
  11     extract_attributes,
  12     float_or_none,
  13     get_element_html_by_id,
  14     int_or_none,
  15     parse_filesize,
  16     str_or_none,
  17     try_get,
  18     unified_strdate,
  19     unified_timestamp,
  20     update_url_query,
  21     url_or_none,
  22     urljoin,
  23 )
  24 from ..utils.traversal import traverse_obj
  25
  26
  27 class BandcampIE(InfoExtractor):
  28     _VALID_URL = r'https?://(?P<uploader>[^/]+)\.bandcamp\.com/track/(?P<id>[^/?#&]+)'
  29     _EMBED_REGEX = [r'<meta property="og:url"[^>]*?content="(?P<url>.*?bandcamp\.com.*?)"']
  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 "\'/\\ä↭ - youtube-dl "\'/\\ä↭ - youtube-dl test song "\'/\\ä↭',
  37             'duration': 9.8485,
  38             'uploader': 'youtube-dl "\'/\\ä↭',
  39             'upload_date': '20121129',
  40             'timestamp': 1354224127,
  41             'track': 'youtube-dl "\'/\\ä↭ - youtube-dl test song "\'/\\ä↭',
  42             'album_artist': 'youtube-dl "\'/\\ä↭',
  43             'track_id': '1812978515',
  44             'artist': 'youtube-dl "\'/\\ä↭',
  45             'uploader_url': 'https://youtube-dl.bandcamp.com',
  46             'uploader_id': 'youtube-dl',
  47             'thumbnail': 'https://f4.bcbits.com/img/a3216802731_5.jpg',
  48         },
  49         'skip': 'There is a limit of 200 free downloads / month for the test song',
  50     }, {
  51         # free download
  52         'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
  53         'info_dict': {
  54             'id': '2650410135',
  55             'ext': 'm4a',
  56             'acodec': r're:[fa]lac',
  57             'title': 'Ben Prunty - Lanius (Battle)',
  58             'thumbnail': r're:^https?://.*\.jpg$',
  59             'uploader': 'Ben Prunty',
  60             'timestamp': 1396508491,
  61             'upload_date': '20140403',
  62             'release_timestamp': 1396483200,
  63             'release_date': '20140403',
  64             'duration': 260.877,
  65             'track': 'Lanius (Battle)',
  66             'track_number': 1,
  67             'track_id': '2650410135',
  68             'artist': 'Ben Prunty',
  69             'album_artist': 'Ben Prunty',
  70             'album': 'FTL: Advanced Edition Soundtrack',
  71             'uploader_url': 'https://benprunty.bandcamp.com',
  72             'uploader_id': 'benprunty',
  73         },
  74     }, {
  75         # no free download, mp3 128
  76         'url': 'https://relapsealumni.bandcamp.com/track/hail-to-fire',
  77         'md5': 'fec12ff55e804bb7f7ebeb77a800c8b7',
  78         'info_dict': {
  79             'id': '2584466013',
  80             'ext': 'mp3',
  81             'title': 'Mastodon - Hail to Fire',
  82             'thumbnail': r're:^https?://.*\.jpg$',
  83             'uploader': 'Mastodon',
  84             'timestamp': 1322005399,
  85             'upload_date': '20111122',
  86             'release_timestamp': 1076112000,
  87             'release_date': '20040207',
  88             'duration': 120.79,
  89             'track': 'Hail to Fire',
  90             'track_number': 5,
  91             'track_id': '2584466013',
  92             'artist': 'Mastodon',
  93             'album_artist': 'Mastodon',
  94             'album': 'Call of the Mastodon',
  95             'uploader_url': 'https://relapsealumni.bandcamp.com',
  96             'uploader_id': 'relapsealumni',
  97         },
  98     }, {
  99         # track from compilation album (artist/album_artist difference)
 100         'url': 'https://diskotopia.bandcamp.com/track/safehouse',
 101         'md5': '19c5337bca1428afa54129f86a2f6a69',
 102         'info_dict': {
 103             'id': '1978174799',
 104             'ext': 'mp3',
 105             'title': 'submerse - submerse - Safehouse',
 106             'thumbnail': r're:^https?://.*\.jpg$',
 107             'uploader': 'submerse',
 108             'timestamp': 1480779297,
 109             'upload_date': '20161203',
 110             'release_timestamp': 1481068800,
 111             'release_date': '20161207',
 112             'duration': 154.066,
 113             'track': 'submerse - Safehouse',
 114             'track_number': 3,
 115             'track_id': '1978174799',
 116             'artist': 'submerse',
 117             'album_artist': 'Diskotopia',
 118             'album': 'DSK F/W 2016-2017 Free Compilation',
 119             'uploader_url': 'https://diskotopia.bandcamp.com',
 120             'uploader_id': 'diskotopia',
 121         },
 122     }]
 123
 124     def _extract_data_attr(self, webpage, video_id, attr='tralbum', fatal=True):
 125         return self._parse_json(self._html_search_regex(
 126             rf'data-{attr}=(["\'])({{.+?}})\1', webpage,
 127             attr + ' data', group=2), video_id, fatal=fatal)
 128
 129     def _real_extract(self, url):
 130         title, uploader = self._match_valid_url(url).group('id', 'uploader')
 131         webpage = self._download_webpage(url, title)
 132         tralbum = self._extract_data_attr(webpage, title)
 133         thumbnail = self._og_search_thumbnail(webpage)
 134
 135         track_id = None
 136         track = None
 137         track_number = None
 138         duration = None
 139
 140         formats = []
 141         track_info = try_get(tralbum, lambda x: x['trackinfo'][0], dict)
 142         if track_info:
 143             file_ = track_info.get('file')
 144             if isinstance(file_, dict):
 145                 for format_id, format_url in file_.items():
 146                     if not url_or_none(format_url):
 147                         continue
 148                     ext, abr_str = format_id.split('-', 1)
 149                     formats.append({
 150                         'format_id': format_id,
 151                         'url': self._proto_relative_url(format_url, 'http:'),
 152                         'ext': ext,
 153                         'vcodec': 'none',
 154                         'acodec': ext,
 155                         'abr': int_or_none(abr_str),
 156                     })
 157             track = track_info.get('title')
 158             track_id = str_or_none(
 159                 track_info.get('track_id') or track_info.get('id'))
 160             track_number = int_or_none(track_info.get('track_num'))
 161             duration = float_or_none(track_info.get('duration'))
 162
 163         embed = self._extract_data_attr(webpage, title, 'embed', False)
 164         current = tralbum.get('current') or {}
 165         artist = embed.get('artist') or current.get('artist') or tralbum.get('artist')
 166         album_artist = self._html_search_regex(
 167             r'<h3 class="albumTitle">[\S\s]*?by\s*<span>\s*<a href="[^>]+">\s*([^>]+?)\s*</a>',
 168             webpage, 'album artist', fatal=False)
 169         timestamp = unified_timestamp(
 170             current.get('publish_date') or tralbum.get('album_publish_date'))
 171
 172         download_link = tralbum.get('freeDownloadPage')
 173         if download_link:
 174             track_id = str(tralbum['id'])
 175
 176             download_webpage = self._download_webpage(
 177                 download_link, track_id, 'Downloading free downloads page')
 178
 179             blob = self._extract_data_attr(download_webpage, track_id, 'blob')
 180
 181             info = try_get(
 182                 blob, (lambda x: x['digital_items'][0],
 183                        lambda x: x['download_items'][0]), dict)
 184             if info:
 185                 downloads = info.get('downloads')
 186                 if isinstance(downloads, dict):
 187                     if not track:
 188                         track = info.get('title')
 189                     if not artist:
 190                         artist = info.get('artist')
 191                     if not thumbnail:
 192                         thumbnail = info.get('thumb_url')
 193
 194                     download_formats = {}
 195                     download_formats_list = blob.get('download_formats')
 196                     if isinstance(download_formats_list, list):
 197                         for f in blob['download_formats']:
 198                             name, ext = f.get('name'), f.get('file_extension')
 199                             if all(isinstance(x, str) for x in (name, ext)):
 200                                 download_formats[name] = ext.strip('.')
 201
 202                     for format_id, f in downloads.items():
 203                         format_url = f.get('url')
 204                         if not format_url:
 205                             continue
 206                         # Stat URL generation algorithm is reverse engineered from
 207                         # download_*_bundle_*.js
 208                         stat_url = update_url_query(
 209                             format_url.replace('/download/', '/statdownload/'), {
 210                                 '.rand': int(time.time() * 1000 * random.random()),
 211                             })
 212                         format_id = f.get('encoding_name') or format_id
 213                         stat = self._download_json(
 214                             stat_url, track_id, f'Downloading {format_id} JSON',
 215                             transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1],
 216                             fatal=False)
 217                         if not stat:
 218                             continue
 219                         retry_url = url_or_none(stat.get('retry_url'))
 220                         if not retry_url:
 221                             continue
 222                         formats.append({
 223                             'url': self._proto_relative_url(retry_url, 'http:'),
 224                             'ext': download_formats.get(format_id),
 225                             'format_id': format_id,
 226                             'format_note': f.get('description'),
 227                             'filesize': parse_filesize(f.get('size_mb')),
 228                             'vcodec': 'none',
 229                             'acodec': format_id.split('-')[0],
 230                         })
 231
 232         title = f'{artist} - {track}' if artist else track
 233
 234         if not duration:
 235             duration = float_or_none(self._html_search_meta(
 236                 'duration', webpage, default=None))
 237
 238         return {
 239             'id': track_id,
 240             'title': title,
 241             'thumbnail': thumbnail,
 242             'uploader': artist,
 243             'uploader_id': uploader,
 244             'uploader_url': f'https://{uploader}.bandcamp.com',
 245             'timestamp': timestamp,
 246             'release_timestamp': unified_timestamp(tralbum.get('album_release_date')),
 247             'duration': duration,
 248             'track': track,
 249             'track_number': track_number,
 250             'track_id': track_id,
 251             'artist': artist,
 252             'album': embed.get('album_title'),
 253             'album_artist': album_artist,
 254             'formats': formats,
 255         }
 256
 257
 258 class BandcampAlbumIE(BandcampIE):  # XXX: Do not subclass from concrete IE
 259     IE_NAME = 'Bandcamp:album'
 260     _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com/album/(?P<id>[^/?#&]+)'
 261
 262     _TESTS = [{
 263         'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
 264         'playlist': [
 265             {
 266                 'md5': '39bc1eded3476e927c724321ddf116cf',
 267                 'info_dict': {
 268                     'id': '1353101989',
 269                     'ext': 'mp3',
 270                     'title': 'Blazo - Intro',
 271                     'timestamp': 1311756226,
 272                     'upload_date': '20110727',
 273                     'uploader': 'Blazo',
 274                 },
 275             },
 276             {
 277                 'md5': '1a2c32e2691474643e912cc6cd4bffaa',
 278                 'info_dict': {
 279                     'id': '38097443',
 280                     'ext': 'mp3',
 281                     'title': 'Blazo - Kero One - Keep It Alive (Blazo remix)',
 282                     'timestamp': 1311757238,
 283                     'upload_date': '20110727',
 284                     'uploader': 'Blazo',
 285                 },
 286             },
 287         ],
 288         'info_dict': {
 289             'title': 'Jazz Format Mixtape vol.1',
 290             'id': 'jazz-format-mixtape-vol-1',
 291             'uploader_id': 'blazo',
 292         },
 293         'params': {
 294             'playlistend': 2,
 295         },
 296         'skip': 'Bandcamp imposes download limits.',
 297     }, {
 298         'url': 'http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave',
 299         'info_dict': {
 300             'title': 'Hierophany of the Open Grave',
 301             'uploader_id': 'nightbringer',
 302             'id': 'hierophany-of-the-open-grave',
 303         },
 304         'playlist_mincount': 9,
 305     }, {
 306         # with escaped quote in title
 307         'url': 'https://jstrecords.bandcamp.com/album/entropy-ep',
 308         'info_dict': {
 309             'title': '"Entropy" EP',
 310             'uploader_id': 'jstrecords',
 311             'id': 'entropy-ep',
 312             'description': 'md5:0ff22959c943622972596062f2f366a5',
 313         },
 314         'playlist_mincount': 3,
 315     }, {
 316         # not all tracks have songs
 317         'url': 'https://insulters.bandcamp.com/album/we-are-the-plague',
 318         'info_dict': {
 319             'id': 'we-are-the-plague',
 320             'title': 'WE ARE THE PLAGUE',
 321             'uploader_id': 'insulters',
 322             'description': 'md5:b3cf845ee41b2b1141dc7bde9237255f',
 323         },
 324         'playlist_count': 2,
 325     }]
 326
 327     @classmethod
 328     def suitable(cls, url):
 329         return (False
 330                 if BandcampWeeklyIE.suitable(url) or BandcampIE.suitable(url)
 331                 else super().suitable(url))
 332
 333     def _real_extract(self, url):
 334         uploader_id, album_id = self._match_valid_url(url).groups()
 335         playlist_id = album_id or uploader_id
 336         webpage = self._download_webpage(url, playlist_id)
 337         tralbum = self._extract_data_attr(webpage, playlist_id)
 338         track_info = tralbum.get('trackinfo')
 339         if not track_info:
 340             raise ExtractorError('The page doesn\'t contain any tracks')
 341         # Only tracks with duration info have songs
 342         entries = [
 343             self.url_result(
 344                 urljoin(url, t['title_link']), BandcampIE.ie_key(),
 345                 str_or_none(t.get('track_id') or t.get('id')), t.get('title'))
 346             for t in track_info
 347             if t.get('duration')]
 348
 349         current = tralbum.get('current') or {}
 350
 351         return {
 352             '_type': 'playlist',
 353             'uploader_id': uploader_id,
 354             'id': playlist_id,
 355             'title': current.get('title'),
 356             'description': current.get('about'),
 357             'entries': entries,
 358         }
 359
 360
 361 class BandcampWeeklyIE(BandcampIE):  # XXX: Do not subclass from concrete IE
 362     IE_NAME = 'Bandcamp:weekly'
 363     _VALID_URL = r'https?://(?:www\.)?bandcamp\.com/?\?(?:.*?&)?show=(?P<id>\d+)'
 364     _TESTS = [{
 365         'url': 'https://bandcamp.com/?show=224',
 366         'md5': 'b00df799c733cf7e0c567ed187dea0fd',
 367         'info_dict': {
 368             'id': '224',
 369             'ext': 'opus',
 370             'title': 'BC Weekly April 4th 2017 - Magic Moments',
 371             'description': 'md5:5d48150916e8e02d030623a48512c874',
 372             'duration': 5829.77,
 373             'release_date': '20170404',
 374             'series': 'Bandcamp Weekly',
 375             'episode': 'Magic Moments',
 376             'episode_id': '224',
 377         },
 378         'params': {
 379             'format': 'opus-lo',
 380         },
 381     }, {
 382         'url': 'https://bandcamp.com/?blah/blah@&show=228',
 383         'only_matching': True,
 384     }]
 385
 386     def _real_extract(self, url):
 387         show_id = self._match_id(url)
 388         webpage = self._download_webpage(url, show_id)
 389
 390         blob = self._extract_data_attr(webpage, show_id, 'blob')
 391
 392         show = blob['bcw_data'][show_id]
 393
 394         formats = []
 395         for format_id, format_url in show['audio_stream'].items():
 396             if not url_or_none(format_url):
 397                 continue
 398             for known_ext in KNOWN_EXTENSIONS:
 399                 if known_ext in format_id:
 400                     ext = known_ext
 401                     break
 402             else:
 403                 ext = None
 404             formats.append({
 405                 'format_id': format_id,
 406                 'url': format_url,
 407                 'ext': ext,
 408                 'vcodec': 'none',
 409             })
 410
 411         title = show.get('audio_title') or 'Bandcamp Weekly'
 412         subtitle = show.get('subtitle')
 413         if subtitle:
 414             title += f' - {subtitle}'
 415
 416         return {
 417             'id': show_id,
 418             'title': title,
 419             'description': show.get('desc') or show.get('short_desc'),
 420             'duration': float_or_none(show.get('audio_duration')),
 421             'is_live': False,
 422             'release_date': unified_strdate(show.get('published_date')),
 423             'series': 'Bandcamp Weekly',
 424             'episode': show.get('subtitle'),
 425             'episode_id': show_id,
 426             'formats': formats,
 427         }
 428
 429
 430 class BandcampUserIE(InfoExtractor):
 431     IE_NAME = 'Bandcamp:user'
 432     _VALID_URL = r'https?://(?!www\.)(?P<id>[^.]+)\.bandcamp\.com(?:/music)?/?(?:[#?]|$)'
 433
 434     _TESTS = [{
 435         # Type 1 Bandcamp user page.
 436         'url': 'https://adrianvonziegler.bandcamp.com',
 437         'info_dict': {
 438             'id': 'adrianvonziegler',
 439             'title': 'Discography of adrianvonziegler',
 440         },
 441         'playlist_mincount': 23,
 442     }, {
 443         # Bandcamp user page with only one album
 444         'url': 'http://dotscale.bandcamp.com',
 445         'info_dict': {
 446             'id': 'dotscale',
 447             'title': 'Discography of dotscale',
 448         },
 449         'playlist_count': 1,
 450     }, {
 451         # Type 2 Bandcamp user page.
 452         'url': 'https://nightcallofficial.bandcamp.com',
 453         'info_dict': {
 454             'id': 'nightcallofficial',
 455             'title': 'Discography of nightcallofficial',
 456         },
 457         'playlist_count': 4,
 458     }, {
 459         'url': 'https://steviasphere.bandcamp.com/music',
 460         'playlist_mincount': 47,
 461         'info_dict': {
 462             'id': 'steviasphere',
 463             'title': 'Discography of steviasphere',
 464         },
 465     }, {
 466         'url': 'https://coldworldofficial.bandcamp.com/music',
 467         'playlist_mincount': 7,
 468         'info_dict': {
 469             'id': 'coldworldofficial',
 470             'title': 'Discography of coldworldofficial',
 471         },
 472     }, {
 473         'url': 'https://nuclearwarnowproductions.bandcamp.com/music',
 474         'playlist_mincount': 399,
 475         'info_dict': {
 476             'id': 'nuclearwarnowproductions',
 477             'title': 'Discography of nuclearwarnowproductions',
 478         },
 479     }]
 480
 481     def _yield_items(self, webpage):
 482         yield from (
 483             re.findall(r'<li data-item-id=["\'][^>]+>\s*<a href=["\'](?![^"\'/]*?/merch)([^"\']+)', webpage)
 484             or re.findall(r'<div[^>]+trackTitle["\'][^"\']+["\']([^"\']+)', webpage))
 485
 486         yield from traverse_obj(webpage, (
 487             {functools.partial(get_element_html_by_id, 'music-grid')}, {extract_attributes},
 488             'data-client-items', {json.loads}, ..., 'page_url', {str}))
 489
 490     def _real_extract(self, url):
 491         uploader = self._match_id(url)
 492         webpage = self._download_webpage(url, uploader)
 493
 494         return self.playlist_from_matches(
 495             self._yield_items(webpage), uploader, f'Discography of {uploader}',
 496             getter=functools.partial(urljoin, url))