youtube_dlc/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-dlc.bandcamp.com/track/youtube-dlc-test-song',
  32         'md5': 'c557841d5e50261777a6585648adf439',
  33         'info_dict': {
  34             'id': '1812978515',
  35             'ext': 'mp3',
  36             'title': "youtube-dl  \"'/\\\u00e4\u21ad - youtube-dl  \"'/\\\u00e4\u21ad - youtube-dl test song \"'/\\\u00e4\u21ad",
  37             'duration': 9.8485,
  38             'uploader': "youtube-dl  \"'/\\\u00e4\u21ad",
  39             'timestamp': 1354224127,
  40             'upload_date': '20121129',
  41         },
  42         '_skip': 'There is a limit of 200 free downloads / month for the test song'
  43     }, {
  44         # free download
  45         'url': 'http://benprunty.bandcamp.com/track/lanius-battle',
  46         'md5': '5d92af55811e47f38962a54c30b07ef0',
  47         'info_dict': {
  48             'id': '2650410135',
  49             'ext': 'aiff',
  50             'title': 'Ben Prunty - Lanius (Battle)',
  51             'thumbnail': r're:^https?://.*\.jpg$',
  52             'uploader': 'Ben Prunty',
  53             'timestamp': 1396508491,
  54             'upload_date': '20140403',
  55             'release_date': '20140403',
  56             'duration': 260.877,
  57             'track': 'Lanius (Battle)',
  58             'track_number': 1,
  59             'track_id': '2650410135',
  60             'artist': 'Ben Prunty',
  61             'album': 'FTL: Advanced Edition Soundtrack',
  62         },
  63     }, {
  64         # no free download, mp3 128
  65         'url': 'https://relapsealumni.bandcamp.com/track/hail-to-fire',
  66         'md5': 'fec12ff55e804bb7f7ebeb77a800c8b7',
  67         'info_dict': {
  68             'id': '2584466013',
  69             'ext': 'mp3',
  70             'title': 'Mastodon - Hail to Fire',
  71             'thumbnail': r're:^https?://.*\.jpg$',
  72             'uploader': 'Mastodon',
  73             'timestamp': 1322005399,
  74             'upload_date': '20111122',
  75             'release_date': '20040207',
  76             'duration': 120.79,
  77             'track': 'Hail to Fire',
  78             'track_number': 5,
  79             'track_id': '2584466013',
  80             'artist': 'Mastodon',
  81             'album': 'Call of the Mastodon',
  82         },
  83     }]
  84
  85     def _real_extract(self, url):
  86         mobj = re.match(self._VALID_URL, url)
  87         title = mobj.group('title')
  88         webpage = self._download_webpage(url, title)
  89         thumbnail = self._html_search_meta('og:image', webpage, default=None)
  90
  91         track_id = None
  92         track = None
  93         track_number = None
  94         duration = None
  95
  96         formats = []
  97         trackinfo_block = self._html_search_regex(
  98             r'trackinfo(?:["\']|&quot;):\[\s*({.+?})\s*\],(?:["\']|&quot;)',
  99             webpage, 'track info', default='{}')
 100
 101         track_info = self._parse_json(trackinfo_block, title)
 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
 118             track_id = str_or_none(track_info.get('track_id') or track_info.get('id'))
 119             track_number = int_or_none(track_info.get('track_num'))
 120             duration = float_or_none(track_info.get('duration'))
 121
 122         def extract(key):
 123             data = self._html_search_regex(
 124                 r',(["\']|&quot;)%s\1:\1(?P<value>(?:\\\1|((?!\1).))+)\1' % key,
 125                 webpage, key, default=None, group='value')
 126             return data.replace(r'\"', '"').replace('\\\\', '\\') if data else data
 127
 128         track = extract('title')
 129         artist = extract('artist')
 130         album = extract('album_title')
 131         timestamp = unified_timestamp(
 132             extract('publish_date') or extract('album_publish_date'))
 133         release_date = unified_strdate(extract('album_release_date'))
 134
 135         download_link = self._search_regex(
 136             r'freeDownloadPage(?:["\']|&quot;):\s*(["\']|&quot;)(?P<url>(?:(?!\1).)+)\1', webpage,
 137             'download link', default=None, group='url')
 138         if download_link:
 139             track_id = self._search_regex(
 140                 r'\?id=(?P<id>\d+)&',
 141                 download_link, 'track id')
 142
 143             download_webpage = self._download_webpage(
 144                 download_link, track_id, 'Downloading free downloads page')
 145
 146             blob = self._parse_json(
 147                 self._search_regex(
 148                     r'data-blob=(["\'])(?P<blob>{.+?})\1', download_webpage,
 149                     'blob', group='blob'),
 150                 track_id, transform_source=unescapeHTML)
 151
 152             info = try_get(
 153                 blob, (lambda x: x['digital_items'][0],
 154                        lambda x: x['download_items'][0]), dict)
 155             if info:
 156                 downloads = info.get('downloads')
 157                 if isinstance(downloads, dict):
 158                     if not track:
 159                         track = info.get('title')
 160                     if not artist:
 161                         artist = info.get('artist')
 162                     if not thumbnail:
 163                         thumbnail = info.get('thumb_url')
 164
 165                     download_formats = {}
 166                     download_formats_list = blob.get('download_formats')
 167                     if isinstance(download_formats_list, list):
 168                         for f in blob['download_formats']:
 169                             name, ext = f.get('name'), f.get('file_extension')
 170                             if all(isinstance(x, compat_str) for x in (name, ext)):
 171                                 download_formats[name] = ext.strip('.')
 172
 173                     for format_id, f in downloads.items():
 174                         format_url = f.get('url')
 175                         if not format_url:
 176                             continue
 177                         # Stat URL generation algorithm is reverse engineered from
 178                         # download_*_bundle_*.js
 179                         stat_url = update_url_query(
 180                             format_url.replace('/download/', '/statdownload/'), {
 181                                 '.rand': int(time.time() * 1000 * random.random()),
 182                             })
 183                         format_id = f.get('encoding_name') or format_id
 184                         stat = self._download_json(
 185                             stat_url, track_id, 'Downloading %s JSON' % format_id,
 186                             transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1],
 187                             fatal=False)
 188                         if not stat:
 189                             continue
 190                         retry_url = url_or_none(stat.get('retry_url'))
 191                         if not retry_url:
 192                             continue
 193                         formats.append({
 194                             'url': self._proto_relative_url(retry_url, 'http:'),
 195                             'ext': download_formats.get(format_id),
 196                             'format_id': format_id,
 197                             'format_note': f.get('description'),
 198                             'filesize': parse_filesize(f.get('size_mb')),
 199                             'vcodec': 'none',
 200                         })
 201
 202         self._sort_formats(formats)
 203
 204         title = '%s - %s' % (artist, track) if artist else track
 205
 206         if not duration:
 207             duration = float_or_none(self._html_search_meta(
 208                 'duration', webpage, default=None))
 209
 210         return {
 211             'id': track_id,
 212             'title': title,
 213             'thumbnail': thumbnail,
 214             'uploader': artist,
 215             'timestamp': timestamp,
 216             'release_date': release_date,
 217             'duration': duration,
 218             'track': track,
 219             'track_number': track_number,
 220             'track_id': track_id,
 221             'artist': artist,
 222             'album': album,
 223             'formats': formats,
 224         }
 225
 226
 227 class BandcampAlbumIE(InfoExtractor):
 228     IE_NAME = 'Bandcamp:album'
 229     _VALID_URL = r'https?://(?:(?P<subdomain>[^.]+)\.)?bandcamp\.com(?:/album/(?P<album_id>[^/?#&]+))?'
 230
 231     _TESTS = [{
 232         'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
 233         'playlist': [
 234             {
 235                 'md5': '39bc1eded3476e927c724321ddf116cf',
 236                 'info_dict': {
 237                     'id': '1353101989',
 238                     'ext': 'mp3',
 239                     'title': 'Intro',
 240                 }
 241             },
 242             {
 243                 'md5': '1a2c32e2691474643e912cc6cd4bffaa',
 244                 'info_dict': {
 245                     'id': '38097443',
 246                     'ext': 'mp3',
 247                     'title': 'Kero One - Keep It Alive (Blazo remix)',
 248                 }
 249             },
 250         ],
 251         'info_dict': {
 252             'title': 'Jazz Format Mixtape vol.1',
 253             'id': 'jazz-format-mixtape-vol-1',
 254             'uploader_id': 'blazo',
 255         },
 256         'params': {
 257             'playlistend': 2
 258         },
 259         'skip': 'Bandcamp imposes download limits.'
 260     }, {
 261         'url': 'http://nightbringer.bandcamp.com/album/hierophany-of-the-open-grave',
 262         'info_dict': {
 263             'title': 'Hierophany of the Open Grave',
 264             'uploader_id': 'nightbringer',
 265             'id': 'hierophany-of-the-open-grave',
 266         },
 267         'playlist_mincount': 9,
 268     }, {
 269         'url': 'http://dotscale.bandcamp.com',
 270         'info_dict': {
 271             'title': 'Loom',
 272             'id': 'dotscale',
 273             'uploader_id': 'dotscale',
 274         },
 275         'playlist_mincount': 7,
 276     }, {
 277         # with escaped quote in title
 278         'url': 'https://jstrecords.bandcamp.com/album/entropy-ep',
 279         'info_dict': {
 280             'title': '"Entropy" EP',
 281             'uploader_id': 'jstrecords',
 282             'id': 'entropy-ep',
 283         },
 284         'playlist_mincount': 3,
 285     }, {
 286         # not all tracks have songs
 287         'url': 'https://insulters.bandcamp.com/album/we-are-the-plague',
 288         'info_dict': {
 289             'id': 'we-are-the-plague',
 290             'title': 'WE ARE THE PLAGUE',
 291             'uploader_id': 'insulters',
 292         },
 293         'playlist_count': 2,
 294     }]
 295
 296     @classmethod
 297     def suitable(cls, url):
 298         return (False
 299                 if BandcampWeeklyIE.suitable(url) or BandcampIE.suitable(url)
 300                 else super(BandcampAlbumIE, cls).suitable(url))
 301
 302     def _real_extract(self, url):
 303         mobj = re.match(self._VALID_URL, url)
 304         uploader_id = mobj.group('subdomain')
 305         album_id = mobj.group('album_id')
 306         playlist_id = album_id or uploader_id
 307         webpage = self._download_webpage(url, playlist_id)
 308         track_elements = re.findall(
 309             r'(?s)<div[^>]*>(.*?<a[^>]+href="([^"]+?)"[^>]+itemprop="url"[^>]*>.*?)</div>', webpage)
 310         if not track_elements:
 311             raise ExtractorError('The page doesn\'t contain any tracks')
 312         # Only tracks with duration info have songs
 313         entries = [
 314             self.url_result(
 315                 compat_urlparse.urljoin(url, t_path),
 316                 ie=BandcampIE.ie_key(),
 317                 video_title=self._search_regex(
 318                     r'<span\b[^>]+\bitemprop=["\']name["\'][^>]*>([^<]+)',
 319                     elem_content, 'track title', fatal=False))
 320             for elem_content, t_path in track_elements
 321             if self._html_search_meta('duration', elem_content, default=None)]
 322
 323         title = self._html_search_regex(
 324             r'album_title\s*(?:&quot;|["\']):\s*(&quot;|["\'])(?P<album>(?:\\\1|((?!\1).))+)\1',
 325             webpage, 'title', fatal=False, group='album')
 326
 327         if title:
 328             title = title.replace(r'\"', '"')
 329
 330         return {
 331             '_type': 'playlist',
 332             'uploader_id': uploader_id,
 333             'id': playlist_id,
 334             'title': title,
 335             'entries': entries,
 336         }
 337
 338
 339 class BandcampWeeklyIE(InfoExtractor):
 340     IE_NAME = 'Bandcamp:weekly'
 341     _VALID_URL = r'https?://(?:www\.)?bandcamp\.com/?\?(?:.*?&)?show=(?P<id>\d+)'
 342     _TESTS = [{
 343         'url': 'https://bandcamp.com/?show=224',
 344         'md5': 'b00df799c733cf7e0c567ed187dea0fd',
 345         'info_dict': {
 346             'id': '224',
 347             'ext': 'opus',
 348             'title': 'BC Weekly April 4th 2017 - Magic Moments',
 349             'description': 'md5:5d48150916e8e02d030623a48512c874',
 350             'duration': 5829.77,
 351             'release_date': '20170404',
 352             'series': 'Bandcamp Weekly',
 353             'episode': 'Magic Moments',
 354             'episode_number': 208,
 355             'episode_id': '224',
 356         }
 357     }, {
 358         'url': 'https://bandcamp.com/?blah/blah@&show=228',
 359         'only_matching': True
 360     }]
 361
 362     def _real_extract(self, url):
 363         video_id = self._match_id(url)
 364         webpage = self._download_webpage(url, video_id)
 365
 366         blob = self._parse_json(
 367             self._search_regex(
 368                 r'data-blob=(["\'])(?P<blob>{.+?})\1', webpage,
 369                 'blob', group='blob'),
 370             video_id, transform_source=unescapeHTML)
 371
 372         show = blob['bcw_show']
 373
 374         # This is desired because any invalid show id redirects to `bandcamp.com`
 375         # which happens to expose the latest Bandcamp Weekly episode.
 376         show_id = int_or_none(show.get('show_id')) or int_or_none(video_id)
 377
 378         formats = []
 379         for format_id, format_url in show['audio_stream'].items():
 380             if not url_or_none(format_url):
 381                 continue
 382             for known_ext in KNOWN_EXTENSIONS:
 383                 if known_ext in format_id:
 384                     ext = known_ext
 385                     break
 386             else:
 387                 ext = None
 388             formats.append({
 389                 'format_id': format_id,
 390                 'url': format_url,
 391                 'ext': ext,
 392                 'vcodec': 'none',
 393             })
 394         self._sort_formats(formats)
 395
 396         title = show.get('audio_title') or 'Bandcamp Weekly'
 397         subtitle = show.get('subtitle')
 398         if subtitle:
 399             title += ' - %s' % subtitle
 400
 401         episode_number = None
 402         seq = blob.get('bcw_seq')
 403
 404         if seq and isinstance(seq, list):
 405             try:
 406                 episode_number = next(
 407                     int_or_none(e.get('episode_number'))
 408                     for e in seq
 409                     if isinstance(e, dict) and int_or_none(e.get('id')) == show_id)
 410             except StopIteration:
 411                 pass
 412
 413         return {
 414             'id': video_id,
 415             'title': title,
 416             'description': show.get('desc') or show.get('short_desc'),
 417             'duration': float_or_none(show.get('audio_duration')),
 418             'is_live': False,
 419             'release_date': unified_strdate(show.get('published_date')),
 420             'series': 'Bandcamp Weekly',
 421             'episode': show.get('subtitle'),
 422             'episode_number': episode_number,
 423             'episode_id': compat_str(video_id),
 424             'formats': formats
 425         }