yt_dlp/extractor/bandcamp.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import random
   5 import re
   6 import time
   7
   8 from .common import InfoExtractor
   9 from ..compat import compat_str
  10 from ..utils import (
  11     ExtractorError,
  12     float_or_none,
  13     int_or_none,
  14     KNOWN_EXTENSIONS,
  15     parse_filesize,
  16     str_or_none,
  17     try_get,
  18     update_url_query,
  19     unified_strdate,
  20     unified_timestamp,
  21     url_or_none,
  22     urljoin,
  23 )
  24
  25
  26 class BandcampIE(InfoExtractor):
  27     _VALID_URL = r'https?://[^/]+\.bandcamp\.com/track/(?P<id>[^/?#&]+)'
  28     _TESTS = [{
  29         'url': 'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
  30         'md5': 'c557841d5e50261777a6585648adf439',
  31         'info_dict': {
  32             'id': '1812978515',
  33             'ext': 'mp3',
  34             'title': "youtube-dl  \"'/\\ä↭ - youtube-dl  \"'/\\ä↭ - youtube-dl test song \"'/\\ä↭",
  35             'duration': 9.8485,
  36             'uploader': 'youtube-dl  "\'/\\ä↭',
  37             'upload_date': '20121129',
  38             'timestamp': 1354224127,
  39         },
  40         '_skip': 'There is a limit of 200 free downloads / month for the test song'
  41     }, {
  42         # free download
  43         'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
  44         'info_dict': {
  45             'id': '2650410135',
  46             'ext': 'aiff',
  47             'title': 'Ben Prunty - Lanius (Battle)',
  48             'thumbnail': r're:^https?://.*\.jpg$',
  49             'uploader': 'Ben Prunty',
  50             'timestamp': 1396508491,
  51             'upload_date': '20140403',
  52             'release_timestamp': 1396483200,
  53             'release_date': '20140403',
  54             'duration': 260.877,
  55             'track': 'Lanius (Battle)',
  56             'track_number': 1,
  57             'track_id': '2650410135',
  58             'artist': 'Ben Prunty',
  59             'album': 'FTL: Advanced Edition Soundtrack',
  60         },
  61     }, {
  62         # no free download, mp3 128
  63         'url': 'https://relapsealumni.bandcamp.com/track/hail-to-fire',
  64         'md5': 'fec12ff55e804bb7f7ebeb77a800c8b7',
  65         'info_dict': {
  66             'id': '2584466013',
  67             'ext': 'mp3',
  68             'title': 'Mastodon - Hail to Fire',
  69             'thumbnail': r're:^https?://.*\.jpg$',
  70             'uploader': 'Mastodon',
  71             'timestamp': 1322005399,
  72             'upload_date': '20111122',
  73             'release_timestamp': 1076112000,
  74             'release_date': '20040207',
  75             'duration': 120.79,
  76             'track': 'Hail to Fire',
  77             'track_number': 5,
  78             'track_id': '2584466013',
  79             'artist': 'Mastodon',
  80             'album': 'Call of the Mastodon',
  81         },
  82     }]
  83
  84     def _extract_data_attr(self, webpage, video_id, attr='tralbum', fatal=True):
  85         return self._parse_json(self._html_search_regex(
  86             r'data-%s=(["\'])({.+?})\1' % attr, webpage,
  87             attr + ' data', group=2), video_id, fatal=fatal)
  88
  89     def _real_extract(self, url):
  90         title = self._match_id(url)
  91         webpage = self._download_webpage(url, title)
  92         tralbum = self._extract_data_attr(webpage, title)
  93         thumbnail = self._og_search_thumbnail(webpage)
  94
  95         track_id = None
  96         track = None
  97         track_number = None
  98         duration = None
  99
 100         formats = []
 101         track_info = try_get(tralbum, lambda x: x['trackinfo'][0], dict)
 102         if track_info:
 103             file_ = track_info.get('file')
 104             if isinstance(file_, dict):
 105                 for format_id, format_url in file_.items():
 106                     if not url_or_none(format_url):
 107                         continue
 108                     ext, abr_str = format_id.split('-', 1)
 109                     formats.append({
 110                         'format_id': format_id,
 111                         'url': self._proto_relative_url(format_url, 'http:'),
 112                         'ext': ext,
 113                         'vcodec': 'none',
 114                         'acodec': ext,
 115                         'abr': int_or_none(abr_str),
 116                     })
 117             track = track_info.get('title')
 118             track_id = str_or_none(
 119                 track_info.get('track_id') or track_info.get('id'))
 120             track_number = int_or_none(track_info.get('track_num'))
 121             duration = float_or_none(track_info.get('duration'))
 122
 123         embed = self._extract_data_attr(webpage, title, 'embed', False)
 124         current = tralbum.get('current') or {}
 125         artist = embed.get('artist') or current.get('artist') or tralbum.get('artist')
 126         timestamp = unified_timestamp(
 127             current.get('publish_date') or tralbum.get('album_publish_date'))
 128
 129         download_link = tralbum.get('freeDownloadPage')
 130         if download_link:
 131             track_id = compat_str(tralbum['id'])
 132
 133             download_webpage = self._download_webpage(
 134                 download_link, track_id, 'Downloading free downloads page')
 135
 136             blob = self._extract_data_attr(download_webpage, track_id, 'blob')
 137
 138             info = try_get(
 139                 blob, (lambda x: x['digital_items'][0],
 140                        lambda x: x['download_items'][0]), dict)
 141             if info:
 142                 downloads = info.get('downloads')
 143                 if isinstance(downloads, dict):
 144                     if not track:
 145                         track = info.get('title')
 146                     if not artist:
 147                         artist = info.get('artist')
 148                     if not thumbnail:
 149                         thumbnail = info.get('thumb_url')
 150
 151                     download_formats = {}
 152                     download_formats_list = blob.get('download_formats')
 153                     if isinstance(download_formats_list, list):
 154                         for f in blob['download_formats']:
 155                             name, ext = f.get('name'), f.get('file_extension')
 156                             if all(isinstance(x, compat_str) for x in (name, ext)):
 157                                 download_formats[name] = ext.strip('.')
 158
 159                     for format_id, f in downloads.items():
 160                         format_url = f.get('url')
 161                         if not format_url:
 162                             continue
 163                         # Stat URL generation algorithm is reverse engineered from
 164                         # download_*_bundle_*.js
 165                         stat_url = update_url_query(
 166                             format_url.replace('/download/', '/statdownload/'), {
 167                                 '.rand': int(time.time() * 1000 * random.random()),
 168                             })
 169                         format_id = f.get('encoding_name') or format_id
 170                         stat = self._download_json(
 171                             stat_url, track_id, 'Downloading %s JSON' % format_id,
 172                             transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1],
 173                             fatal=False)
 174                         if not stat:
 175                             continue
 176                         retry_url = url_or_none(stat.get('retry_url'))
 177                         if not retry_url:
 178                             continue
 179                         formats.append({
 180                             'url': self._proto_relative_url(retry_url, 'http:'),
 181                             'ext': download_formats.get(format_id),
 182                             'format_id': format_id,
 183                             'format_note': f.get('description'),
 184                             'filesize': parse_filesize(f.get('size_mb')),
 185                             'vcodec': 'none',
 186                         })
 187
 188         self._sort_formats(formats)
 189
 190         title = '%s - %s' % (artist, track) if artist else track
 191
 192         if not duration:
 193             duration = float_or_none(self._html_search_meta(
 194                 'duration', webpage, default=None))
 195
 196         return {
 197             'id': track_id,
 198             'title': title,
 199             'thumbnail': thumbnail,
 200             'uploader': artist,
 201             'timestamp': timestamp,
 202             'release_timestamp': unified_timestamp(tralbum.get('album_release_date')),
 203             'duration': duration,
 204             'track': track,
 205             'track_number': track_number,
 206             'track_id': track_id,
 207             'artist': artist,
 208             'album': embed.get('album_title'),
 209             'formats': formats,
 210         }
 211
 212
 213 class BandcampAlbumIE(BandcampIE):
 214     IE_NAME = 'Bandcamp:album'
 215     _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?!/music)(?:/album/(?P<id>[^/?#&]+))?'
 216
 217     _TESTS = [{
 218         'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
 219         'playlist': [
 220             {
 221                 'md5': '39bc1eded3476e927c724321ddf116cf',
 222                 'info_dict': {
 223                     'id': '1353101989',
 224                     'ext': 'mp3',
 225                     'title': 'Blazo - Intro',
 226                     'timestamp': 1311756226,
 227                     'upload_date': '20110727',
 228                     'uploader': 'Blazo',
 229                 }
 230             },
 231             {
 232                 'md5': '1a2c32e2691474643e912cc6cd4bffaa',
 233                 'info_dict': {
 234                     'id': '38097443',
 235                     'ext': 'mp3',
 236                     'title': 'Blazo - Kero One - Keep It Alive (Blazo remix)',
 237                     'timestamp': 1311757238,
 238                     'upload_date': '20110727',
 239                     'uploader': 'Blazo',
 240                 }
 241             },
 242         ],
 243         'info_dict': {
 244             'title': 'Jazz Format Mixtape vol.1',
 245             'id': 'jazz-format-mixtape-vol-1',
 246             'uploader_id': 'blazo',
 247         },
 248         'params': {
 249             'playlistend': 2
 250         },
 251         'skip': 'Bandcamp imposes download limits.'
 252     }, {
 253         'url': 'http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave',
 254         'info_dict': {
 255             'title': 'Hierophany of the Open Grave',
 256             'uploader_id': 'nightbringer',
 257             'id': 'hierophany-of-the-open-grave',
 258         },
 259         'playlist_mincount': 9,
 260     }, {
 261         'url': 'http://dotscale.bandcamp.com',
 262         'info_dict': {
 263             'title': 'Loom',
 264             'id': 'dotscale',
 265             'uploader_id': 'dotscale',
 266         },
 267         'playlist_mincount': 7,
 268     }, {
 269         # with escaped quote in title
 270         'url': 'https://jstrecords.bandcamp.com/album/entropy-ep',
 271         'info_dict': {
 272             'title': '"Entropy" EP',
 273             'uploader_id': 'jstrecords',
 274             'id': 'entropy-ep',
 275             'description': 'md5:0ff22959c943622972596062f2f366a5',
 276         },
 277         'playlist_mincount': 3,
 278     }, {
 279         # not all tracks have songs
 280         'url': 'https://insulters.bandcamp.com/album/we-are-the-plague',
 281         'info_dict': {
 282             'id': 'we-are-the-plague',
 283             'title': 'WE ARE THE PLAGUE',
 284             'uploader_id': 'insulters',
 285             'description': 'md5:b3cf845ee41b2b1141dc7bde9237255f',
 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         uploader_id, album_id = re.match(self._VALID_URL, url).groups()
 298         playlist_id = album_id or uploader_id
 299         webpage = self._download_webpage(url, playlist_id)
 300         tralbum = self._extract_data_attr(webpage, playlist_id)
 301         track_info = tralbum.get('trackinfo')
 302         if not track_info:
 303             raise ExtractorError('The page doesn\'t contain any tracks')
 304         # Only tracks with duration info have songs
 305         entries = [
 306             self.url_result(
 307                 urljoin(url, t['title_link']), BandcampIE.ie_key(),
 308                 str_or_none(t.get('track_id') or t.get('id')), t.get('title'))
 309             for t in track_info
 310             if t.get('duration')]
 311
 312         current = tralbum.get('current') or {}
 313
 314         return {
 315             '_type': 'playlist',
 316             'uploader_id': uploader_id,
 317             'id': playlist_id,
 318             'title': current.get('title'),
 319             'description': current.get('about'),
 320             'entries': entries,
 321         }
 322
 323
 324 class BandcampWeeklyIE(BandcampIE):
 325     IE_NAME = 'Bandcamp:weekly'
 326     _VALID_URL = r'https?://(?:www\.)?bandcamp\.com/?\?(?:.*?&)?show=(?P<id>\d+)'
 327     _TESTS = [{
 328         'url': 'https://bandcamp.com/?show=224',
 329         'md5': 'b00df799c733cf7e0c567ed187dea0fd',
 330         'info_dict': {
 331             'id': '224',
 332             'ext': 'opus',
 333             'title': 'BC Weekly April 4th 2017 - Magic Moments',
 334             'description': 'md5:5d48150916e8e02d030623a48512c874',
 335             'duration': 5829.77,
 336             'release_date': '20170404',
 337             'series': 'Bandcamp Weekly',
 338             'episode': 'Magic Moments',
 339             'episode_id': '224',
 340         },
 341         'params': {
 342             'format': 'opus-lo',
 343         },
 344     }, {
 345         'url': 'https://bandcamp.com/?blah/blah@&show=228',
 346         'only_matching': True
 347     }]
 348
 349     def _real_extract(self, url):
 350         show_id = self._match_id(url)
 351         webpage = self._download_webpage(url, show_id)
 352
 353         blob = self._extract_data_attr(webpage, show_id, 'blob')
 354
 355         show = blob['bcw_data'][show_id]
 356
 357         formats = []
 358         for format_id, format_url in show['audio_stream'].items():
 359             if not url_or_none(format_url):
 360                 continue
 361             for known_ext in KNOWN_EXTENSIONS:
 362                 if known_ext in format_id:
 363                     ext = known_ext
 364                     break
 365             else:
 366                 ext = None
 367             formats.append({
 368                 'format_id': format_id,
 369                 'url': format_url,
 370                 'ext': ext,
 371                 'vcodec': 'none',
 372             })
 373         self._sort_formats(formats)
 374
 375         title = show.get('audio_title') or 'Bandcamp Weekly'
 376         subtitle = show.get('subtitle')
 377         if subtitle:
 378             title += ' - %s' % subtitle
 379
 380         return {
 381             'id': show_id,
 382             'title': title,
 383             'description': show.get('desc') or show.get('short_desc'),
 384             'duration': float_or_none(show.get('audio_duration')),
 385             'is_live': False,
 386             'release_date': unified_strdate(show.get('published_date')),
 387             'series': 'Bandcamp Weekly',
 388             'episode': show.get('subtitle'),
 389             'episode_id': show_id,
 390             'formats': formats
 391         }
 392
 393
 394 class BandcampMusicIE(InfoExtractor):
 395     _VALID_URL = r'https?://(?P<id>[^/]+)\.bandcamp\.com/music'
 396     _TESTS = [{
 397         'url': 'https://steviasphere.bandcamp.com/music',
 398         'playlist_mincount': 47,
 399         'info_dict': {
 400             'id': 'steviasphere',
 401         },
 402     }, {
 403         'url': 'https://coldworldofficial.bandcamp.com/music',
 404         'playlist_mincount': 10,
 405         'info_dict': {
 406             'id': 'coldworldofficial',
 407         },
 408     }, {
 409         'url': 'https://nuclearwarnowproductions.bandcamp.com/music',
 410         'playlist_mincount': 399,
 411         'info_dict': {
 412             'id': 'nuclearwarnowproductions',
 413         },
 414     }
 415     ]
 416
 417     _TYPE_IE_DICT = {
 418         'album': BandcampAlbumIE.ie_key(),
 419         'track': BandcampIE.ie_key()
 420     }
 421
 422     def _real_extract(self, url):
 423         id = self._match_id(url)
 424         webpage = self._download_webpage(url, id)
 425         items = re.findall(r'href\=\"\/(?P<path>(?P<type>album|track)+/[^\"]+)', webpage)
 426         entries = [
 427             self.url_result(
 428                 f'https://{id}.bandcamp.com/{item[0]}',
 429                 ie=self._TYPE_IE_DICT[item[1]])
 430             for item in items]
 431         return self.playlist_result(entries, id)