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     ExtractorError,
   9     float_or_none,
  10     int_or_none,
  11     KNOWN_EXTENSIONS,
  12     parse_filesize,
  13     str_or_none,
  14     try_get,
  15     update_url_query,
  16     unified_strdate,
  17     unified_timestamp,
  18     url_or_none,
  19     urljoin,
  20 )
  21
  22
  23 class BandcampIE(InfoExtractor):
  24     _VALID_URL = r'https?://[^/]+\.bandcamp\.com/track/(?P<id>[^/?#&]+)'
  25     _TESTS = [{
  26         'url': 'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
  27         'md5': 'c557841d5e50261777a6585648adf439',
  28         'info_dict': {
  29             'id': '1812978515',
  30             'ext': 'mp3',
  31             'title': "youtube-dl  \"'/\\ä↭ - youtube-dl  \"'/\\ä↭ - youtube-dl test song \"'/\\ä↭",
  32             'duration': 9.8485,
  33             'uploader': 'youtube-dl  "\'/\\ä↭',
  34             'upload_date': '20121129',
  35             'timestamp': 1354224127,
  36         },
  37         '_skip': 'There is a limit of 200 free downloads / month for the test song'
  38     }, {
  39         # free download
  40         'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
  41         'info_dict': {
  42             'id': '2650410135',
  43             'ext': 'aiff',
  44             'title': 'Ben Prunty - Lanius (Battle)',
  45             'thumbnail': r're:^https?://.*\.jpg$',
  46             'uploader': 'Ben Prunty',
  47             'timestamp': 1396508491,
  48             'upload_date': '20140403',
  49             'release_timestamp': 1396483200,
  50             'release_date': '20140403',
  51             'duration': 260.877,
  52             'track': 'Lanius (Battle)',
  53             'track_number': 1,
  54             'track_id': '2650410135',
  55             'artist': 'Ben Prunty',
  56             'album': 'FTL: Advanced Edition Soundtrack',
  57         },
  58     }, {
  59         # no free download, mp3 128
  60         'url': 'https://relapsealumni.bandcamp.com/track/hail-to-fire',
  61         'md5': 'fec12ff55e804bb7f7ebeb77a800c8b7',
  62         'info_dict': {
  63             'id': '2584466013',
  64             'ext': 'mp3',
  65             'title': 'Mastodon - Hail to Fire',
  66             'thumbnail': r're:^https?://.*\.jpg$',
  67             'uploader': 'Mastodon',
  68             'timestamp': 1322005399,
  69             'upload_date': '20111122',
  70             'release_timestamp': 1076112000,
  71             'release_date': '20040207',
  72             'duration': 120.79,
  73             'track': 'Hail to Fire',
  74             'track_number': 5,
  75             'track_id': '2584466013',
  76             'artist': 'Mastodon',
  77             'album': 'Call of the Mastodon',
  78         },
  79     }]
  80
  81     def _extract_data_attr(self, webpage, video_id, attr='tralbum', fatal=True):
  82         return self._parse_json(self._html_search_regex(
  83             r'data-%s=(["\'])({.+?})\1' % attr, webpage,
  84             attr + ' data', group=2), video_id, fatal=fatal)
  85
  86     def _real_extract(self, url):
  87         title = self._match_id(url)
  88         webpage = self._download_webpage(url, title)
  89         tralbum = self._extract_data_attr(webpage, title)
  90         thumbnail = self._og_search_thumbnail(webpage)
  91
  92         track_id = None
  93         track = None
  94         track_number = None
  95         duration = None
  96
  97         formats = []
  98         track_info = try_get(tralbum, lambda x: x['trackinfo'][0], dict)
  99         if track_info:
 100             file_ = track_info.get('file')
 101             if isinstance(file_, dict):
 102                 for format_id, format_url in file_.items():
 103                     if not url_or_none(format_url):
 104                         continue
 105                     ext, abr_str = format_id.split('-', 1)
 106                     formats.append({
 107                         'format_id': format_id,
 108                         'url': self._proto_relative_url(format_url, 'http:'),
 109                         'ext': ext,
 110                         'vcodec': 'none',
 111                         'acodec': ext,
 112                         'abr': int_or_none(abr_str),
 113                     })
 114             track = track_info.get('title')
 115             track_id = str_or_none(
 116                 track_info.get('track_id') or track_info.get('id'))
 117             track_number = int_or_none(track_info.get('track_num'))
 118             duration = float_or_none(track_info.get('duration'))
 119
 120         embed = self._extract_data_attr(webpage, title, 'embed', False)
 121         current = tralbum.get('current') or {}
 122         artist = embed.get('artist') or current.get('artist') or tralbum.get('artist')
 123         timestamp = unified_timestamp(
 124             current.get('publish_date') or tralbum.get('album_publish_date'))
 125
 126         download_link = tralbum.get('freeDownloadPage')
 127         if download_link:
 128             track_id = compat_str(tralbum['id'])
 129
 130             download_webpage = self._download_webpage(
 131                 download_link, track_id, 'Downloading free downloads page')
 132
 133             blob = self._extract_data_attr(download_webpage, track_id, 'blob')
 134
 135             info = try_get(
 136                 blob, (lambda x: x['digital_items'][0],
 137                        lambda x: x['download_items'][0]), dict)
 138             if info:
 139                 downloads = info.get('downloads')
 140                 if isinstance(downloads, dict):
 141                     if not track:
 142                         track = info.get('title')
 143                     if not artist:
 144                         artist = info.get('artist')
 145                     if not thumbnail:
 146                         thumbnail = info.get('thumb_url')
 147
 148                     download_formats = {}
 149                     download_formats_list = blob.get('download_formats')
 150                     if isinstance(download_formats_list, list):
 151                         for f in blob['download_formats']:
 152                             name, ext = f.get('name'), f.get('file_extension')
 153                             if all(isinstance(x, compat_str) for x in (name, ext)):
 154                                 download_formats[name] = ext.strip('.')
 155
 156                     for format_id, f in downloads.items():
 157                         format_url = f.get('url')
 158                         if not format_url:
 159                             continue
 160                         # Stat URL generation algorithm is reverse engineered from
 161                         # download_*_bundle_*.js
 162                         stat_url = update_url_query(
 163                             format_url.replace('/download/', '/statdownload/'), {
 164                                 '.rand': int(time.time() * 1000 * random.random()),
 165                             })
 166                         format_id = f.get('encoding_name') or format_id
 167                         stat = self._download_json(
 168                             stat_url, track_id, 'Downloading %s JSON' % format_id,
 169                             transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1],
 170                             fatal=False)
 171                         if not stat:
 172                             continue
 173                         retry_url = url_or_none(stat.get('retry_url'))
 174                         if not retry_url:
 175                             continue
 176                         formats.append({
 177                             'url': self._proto_relative_url(retry_url, 'http:'),
 178                             'ext': download_formats.get(format_id),
 179                             'format_id': format_id,
 180                             'format_note': f.get('description'),
 181                             'filesize': parse_filesize(f.get('size_mb')),
 182                             'vcodec': 'none',
 183                             'acodec': format_id.split('-')[0],
 184                         })
 185
 186         self._sort_formats(formats)
 187
 188         title = '%s - %s' % (artist, track) if artist else track
 189
 190         if not duration:
 191             duration = float_or_none(self._html_search_meta(
 192                 'duration', webpage, default=None))
 193
 194         return {
 195             'id': track_id,
 196             'title': title,
 197             'thumbnail': thumbnail,
 198             'uploader': artist,
 199             'timestamp': timestamp,
 200             'release_timestamp': unified_timestamp(tralbum.get('album_release_date')),
 201             'duration': duration,
 202             'track': track,
 203             'track_number': track_number,
 204             'track_id': track_id,
 205             'artist': artist,
 206             'album': embed.get('album_title'),
 207             'formats': formats,
 208         }
 209
 210
 211 class BandcampAlbumIE(BandcampIE):
 212     IE_NAME = 'Bandcamp:album'
 213     _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com/album/(?P<id>[^/?#&]+)'
 214
 215     _TESTS = [{
 216         'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
 217         'playlist': [
 218             {
 219                 'md5': '39bc1eded3476e927c724321ddf116cf',
 220                 'info_dict': {
 221                     'id': '1353101989',
 222                     'ext': 'mp3',
 223                     'title': 'Blazo - Intro',
 224                     'timestamp': 1311756226,
 225                     'upload_date': '20110727',
 226                     'uploader': 'Blazo',
 227                 }
 228             },
 229             {
 230                 'md5': '1a2c32e2691474643e912cc6cd4bffaa',
 231                 'info_dict': {
 232                     'id': '38097443',
 233                     'ext': 'mp3',
 234                     'title': 'Blazo - Kero One - Keep It Alive (Blazo remix)',
 235                     'timestamp': 1311757238,
 236                     'upload_date': '20110727',
 237                     'uploader': 'Blazo',
 238                 }
 239             },
 240         ],
 241         'info_dict': {
 242             'title': 'Jazz Format Mixtape vol.1',
 243             'id': 'jazz-format-mixtape-vol-1',
 244             'uploader_id': 'blazo',
 245         },
 246         'params': {
 247             'playlistend': 2
 248         },
 249         'skip': 'Bandcamp imposes download limits.'
 250     }, {
 251         'url': 'http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave',
 252         'info_dict': {
 253             'title': 'Hierophany of the Open Grave',
 254             'uploader_id': 'nightbringer',
 255             'id': 'hierophany-of-the-open-grave',
 256         },
 257         'playlist_mincount': 9,
 258     }, {
 259         # with escaped quote in title
 260         'url': 'https://jstrecords.bandcamp.com/album/entropy-ep',
 261         'info_dict': {
 262             'title': '"Entropy" EP',
 263             'uploader_id': 'jstrecords',
 264             'id': 'entropy-ep',
 265             'description': 'md5:0ff22959c943622972596062f2f366a5',
 266         },
 267         'playlist_mincount': 3,
 268     }, {
 269         # not all tracks have songs
 270         'url': 'https://insulters.bandcamp.com/album/we-are-the-plague',
 271         'info_dict': {
 272             'id': 'we-are-the-plague',
 273             'title': 'WE ARE THE PLAGUE',
 274             'uploader_id': 'insulters',
 275             'description': 'md5:b3cf845ee41b2b1141dc7bde9237255f',
 276         },
 277         'playlist_count': 2,
 278     }]
 279
 280     @classmethod
 281     def suitable(cls, url):
 282         return (False
 283                 if BandcampWeeklyIE.suitable(url) or BandcampIE.suitable(url)
 284                 else super(BandcampAlbumIE, cls).suitable(url))
 285
 286     def _real_extract(self, url):
 287         uploader_id, album_id = self._match_valid_url(url).groups()
 288         playlist_id = album_id or uploader_id
 289         webpage = self._download_webpage(url, playlist_id)
 290         tralbum = self._extract_data_attr(webpage, playlist_id)
 291         track_info = tralbum.get('trackinfo')
 292         if not track_info:
 293             raise ExtractorError('The page doesn\'t contain any tracks')
 294         # Only tracks with duration info have songs
 295         entries = [
 296             self.url_result(
 297                 urljoin(url, t['title_link']), BandcampIE.ie_key(),
 298                 str_or_none(t.get('track_id') or t.get('id')), t.get('title'))
 299             for t in track_info
 300             if t.get('duration')]
 301
 302         current = tralbum.get('current') or {}
 303
 304         return {
 305             '_type': 'playlist',
 306             'uploader_id': uploader_id,
 307             'id': playlist_id,
 308             'title': current.get('title'),
 309             'description': current.get('about'),
 310             'entries': entries,
 311         }
 312
 313
 314 class BandcampWeeklyIE(BandcampIE):
 315     IE_NAME = 'Bandcamp:weekly'
 316     _VALID_URL = r'https?://(?:www\.)?bandcamp\.com/?\?(?:.*?&)?show=(?P<id>\d+)'
 317     _TESTS = [{
 318         'url': 'https://bandcamp.com/?show=224',
 319         'md5': 'b00df799c733cf7e0c567ed187dea0fd',
 320         'info_dict': {
 321             'id': '224',
 322             'ext': 'opus',
 323             'title': 'BC Weekly April 4th 2017 - Magic Moments',
 324             'description': 'md5:5d48150916e8e02d030623a48512c874',
 325             'duration': 5829.77,
 326             'release_date': '20170404',
 327             'series': 'Bandcamp Weekly',
 328             'episode': 'Magic Moments',
 329             'episode_id': '224',
 330         },
 331         'params': {
 332             'format': 'opus-lo',
 333         },
 334     }, {
 335         'url': 'https://bandcamp.com/?blah/blah@&show=228',
 336         'only_matching': True
 337     }]
 338
 339     def _real_extract(self, url):
 340         show_id = self._match_id(url)
 341         webpage = self._download_webpage(url, show_id)
 342
 343         blob = self._extract_data_attr(webpage, show_id, 'blob')
 344
 345         show = blob['bcw_data'][show_id]
 346
 347         formats = []
 348         for format_id, format_url in show['audio_stream'].items():
 349             if not url_or_none(format_url):
 350                 continue
 351             for known_ext in KNOWN_EXTENSIONS:
 352                 if known_ext in format_id:
 353                     ext = known_ext
 354                     break
 355             else:
 356                 ext = None
 357             formats.append({
 358                 'format_id': format_id,
 359                 'url': format_url,
 360                 'ext': ext,
 361                 'vcodec': 'none',
 362             })
 363         self._sort_formats(formats)
 364
 365         title = show.get('audio_title') or 'Bandcamp Weekly'
 366         subtitle = show.get('subtitle')
 367         if subtitle:
 368             title += ' - %s' % subtitle
 369
 370         return {
 371             'id': show_id,
 372             'title': title,
 373             'description': show.get('desc') or show.get('short_desc'),
 374             'duration': float_or_none(show.get('audio_duration')),
 375             'is_live': False,
 376             'release_date': unified_strdate(show.get('published_date')),
 377             'series': 'Bandcamp Weekly',
 378             'episode': show.get('subtitle'),
 379             'episode_id': show_id,
 380             'formats': formats
 381         }
 382
 383
 384 class BandcampUserIE(InfoExtractor):
 385     IE_NAME = 'Bandcamp:user'
 386     _VALID_URL = r'https?://(?!www\.)(?P<id>[^.]+)\.bandcamp\.com(?:/music)?/?(?:[#?]|$)'
 387
 388     _TESTS = [{
 389         # Type 1 Bandcamp user page.
 390         'url': 'https://adrianvonziegler.bandcamp.com',
 391         'info_dict': {
 392             'id': 'adrianvonziegler',
 393             'title': 'Discography of adrianvonziegler',
 394         },
 395         'playlist_mincount': 23,
 396     }, {
 397         # Bandcamp user page with only one album
 398         'url': 'http://dotscale.bandcamp.com',
 399         'info_dict': {
 400             'id': 'dotscale',
 401             'title': 'Discography of dotscale'
 402         },
 403         'playlist_count': 1,
 404     }, {
 405         # Type 2 Bandcamp user page.
 406         'url': 'https://nightcallofficial.bandcamp.com',
 407         'info_dict': {
 408             'id': 'nightcallofficial',
 409             'title': 'Discography of nightcallofficial',
 410         },
 411         'playlist_count': 4,
 412     }, {
 413         'url': 'https://steviasphere.bandcamp.com/music',
 414         'playlist_mincount': 47,
 415         'info_dict': {
 416             'id': 'steviasphere',
 417             'title': 'Discography of steviasphere',
 418         },
 419     }, {
 420         'url': 'https://coldworldofficial.bandcamp.com/music',
 421         'playlist_mincount': 10,
 422         'info_dict': {
 423             'id': 'coldworldofficial',
 424             'title': 'Discography of coldworldofficial',
 425         },
 426     }, {
 427         'url': 'https://nuclearwarnowproductions.bandcamp.com/music',
 428         'playlist_mincount': 399,
 429         'info_dict': {
 430             'id': 'nuclearwarnowproductions',
 431             'title': 'Discography of nuclearwarnowproductions',
 432         },
 433     }]
 434
 435     def _real_extract(self, url):
 436         uploader = self._match_id(url)
 437         webpage = self._download_webpage(url, uploader)
 438
 439         discography_data = (re.findall(r'<li data-item-id=["\'][^>]+>\s*<a href=["\'](?![^"\'/]*?/merch)([^"\']+)', webpage)
 440                             or re.findall(r'<div[^>]+trackTitle["\'][^"\']+["\']([^"\']+)', webpage))
 441
 442         return self.playlist_from_matches(
 443             discography_data, uploader, f'Discography of {uploader}', getter=lambda x: urljoin(url, x))