_ Git - youtube-dl/blob - youtube_dl/extractor/mixcloud.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_urllib_parse_unquote,
   8     compat_urllib_request
   9 )
  10 from ..utils import (
  11     ExtractorError,
  12     HEADRequest,
  13     NO_DEFAULT,
  14     parse_count,
  15     str_to_int,
  16     clean_html
  17 )
  18
  19
  20 class MixcloudIE(InfoExtractor):
  21     _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/([^/]+)/(?!stream|uploads|favorites|listens|playlists)([^/]+)'
  22     IE_NAME = 'mixcloud'
  23
  24     _TESTS = [{
  25         'url': 'http://www.mixcloud.com/dholbach/cryptkeeper/',
  26         'info_dict': {
  27             'id': 'dholbach-cryptkeeper',
  28             'ext': 'm4a',
  29             'title': 'Cryptkeeper',
  30             'description': 'After quite a long silence from myself, finally another Drum\'n\'Bass mix with my favourite current dance floor bangers.',
  31             'uploader': 'Daniel Holbach',
  32             'uploader_id': 'dholbach',
  33             'thumbnail': 're:https?://.*\.jpg',
  34             'view_count': int,
  35             'like_count': int,
  36         },
  37     }, {
  38         'url': 'http://www.mixcloud.com/gillespeterson/caribou-7-inch-vinyl-mix-chat/',
  39         'info_dict': {
  40             'id': 'gillespeterson-caribou-7-inch-vinyl-mix-chat',
  41             'ext': 'mp3',
  42             'title': 'Caribou 7 inch Vinyl Mix & Chat',
  43             'description': 'md5:2b8aec6adce69f9d41724647c65875e8',
  44             'uploader': 'Gilles Peterson Worldwide',
  45             'uploader_id': 'gillespeterson',
  46             'thumbnail': 're:https?://.*/images/',
  47             'view_count': int,
  48             'like_count': int,
  49         },
  50     }]
  51
  52     def _check_url(self, url, track_id, ext):
  53         try:
  54             # We only want to know if the request succeed
  55             # don't download the whole file
  56             self._request_webpage(
  57                 HEADRequest(url), track_id,
  58                 'Trying %s URL' % ext)
  59             return True
  60         except ExtractorError:
  61             return False
  62
  63     def _real_extract(self, url):
  64         mobj = re.match(self._VALID_URL, url)
  65         uploader = mobj.group(1)
  66         cloudcast_name = mobj.group(2)
  67         track_id = compat_urllib_parse_unquote('-'.join((uploader, cloudcast_name)))
  68
  69         webpage = self._download_webpage(url, track_id)
  70
  71         message = self._html_search_regex(
  72             r'(?s)<div[^>]+class="global-message cloudcast-disabled-notice-light"[^>]*>(.+?)<(?:a|/div)',
  73             webpage, 'error message', default=None)
  74
  75         preview_url = self._search_regex(
  76             r'\s(?:data-preview-url|m-preview)="([^"]+)"',
  77             webpage, 'preview url', default=None if message else NO_DEFAULT)
  78
  79         if message:
  80             raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
  81
  82         song_url = re.sub(r'audiocdn(\d+)', r'stream\1', preview_url)
  83         song_url = song_url.replace('/previews/', '/c/originals/')
  84         if not self._check_url(song_url, track_id, 'mp3'):
  85             song_url = song_url.replace('.mp3', '.m4a').replace('originals/', 'm4a/64/')
  86             if not self._check_url(song_url, track_id, 'm4a'):
  87                 raise ExtractorError('Unable to extract track url')
  88
  89         PREFIX = (
  90             r'm-play-on-spacebar[^>]+'
  91             r'(?:\s+[a-zA-Z0-9-]+(?:="[^"]+")?)*?\s+')
  92         title = self._html_search_regex(
  93             PREFIX + r'm-title="([^"]+)"', webpage, 'title')
  94         thumbnail = self._proto_relative_url(self._html_search_regex(
  95             PREFIX + r'm-thumbnail-url="([^"]+)"', webpage, 'thumbnail',
  96             fatal=False))
  97         uploader = self._html_search_regex(
  98             PREFIX + r'm-owner-name="([^"]+)"',
  99             webpage, 'uploader', fatal=False)
 100         uploader_id = self._search_regex(
 101             r'\s+"profile": "([^"]+)",', webpage, 'uploader id', fatal=False)
 102         description = self._og_search_description(webpage)
 103         like_count = parse_count(self._search_regex(
 104             r'\bbutton-favorite[^>]+>.*?<span[^>]+class=["\']toggle-number[^>]+>\s*([^<]+)',
 105             webpage, 'like count', fatal=False))
 106         view_count = str_to_int(self._search_regex(
 107             [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
 108              r'/listeners/?">([0-9,.]+)</a>'],
 109             webpage, 'play count', fatal=False))
 110
 111         return {
 112             'id': track_id,
 113             'title': title,
 114             'url': song_url,
 115             'description': description,
 116             'thumbnail': thumbnail,
 117             'uploader': uploader,
 118             'uploader_id': uploader_id,
 119             'view_count': view_count,
 120             'like_count': like_count,
 121         }
 122
 123
 124 class MixcloudUserIE(InfoExtractor):
 125     """
 126     Information extractor for Mixcloud users.
 127     It can retrieve a list of a user's uploads, favorites or listens.
 128     """
 129
 130     _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/(?P<type>uploads|favorites|listens)?/?$'
 131     IE_NAME = 'mixcloud:user'
 132
 133     _TESTS = [{
 134         'url': 'http://www.mixcloud.com/dholbach/',
 135         'info_dict': {
 136             'id': 'dholbach/uploads',
 137             'title': 'Daniel Holbach (uploads)',
 138             'description': 'md5:327af72d1efeb404a8216c27240d1370',
 139         },
 140         'playlist_mincount': 11
 141     }, {
 142         'url': 'http://www.mixcloud.com/dholbach/uploads/',
 143         'info_dict': {
 144             'id': 'dholbach/uploads',
 145             'title': 'Daniel Holbach (uploads)',
 146             'description': 'md5:327af72d1efeb404a8216c27240d1370',
 147         },
 148         'playlist_mincount': 11
 149     }, {
 150         'url': 'http://www.mixcloud.com/dholbach/favorites/',
 151         'info_dict': {
 152             'id': 'dholbach/favorites',
 153             'title': 'Daniel Holbach (favorites)',
 154             'description': 'md5:327af72d1efeb404a8216c27240d1370',
 155         },
 156         'playlist_mincount': 244
 157     }, {
 158         'url': 'http://www.mixcloud.com/dholbach/listens/',
 159         'info_dict': {
 160             'id': 'dholbach/listens',
 161             'title': 'Daniel Holbach (listens)',
 162             'description': 'md5:327af72d1efeb404a8216c27240d1370',
 163         },
 164         'playlist_mincount': 846
 165     }]
 166
 167     def _fetch_tracks(self, base_url, video_id, dl_note=None, dl_errnote=None):
 168         # retrieve all fragments of a list of tracks with fake AJAX calls
 169         track_urls = []
 170         current_page = 1
 171         while True:
 172             # fake a AJAX request to retrieve a list fragment
 173             page_url = base_url + "?page=%d&list=main&_ajax=1" % current_page
 174             req = compat_urllib_request.Request(page_url, headers={"X-Requested-With": "XMLHttpRequest"})
 175             resp = self._download_webpage(req, video_id, note=dl_note + " (page %d)" % current_page, errnote=dl_errnote)
 176
 177             # extract all track URLs from fragment
 178             urls = re.findall(r'm-play-button m-url="(?P<url>[^"]+)"', resp)
 179             # clean up URLs
 180             urls = map(clean_html, urls)
 181             # create absolute URLs
 182             urls = map(lambda u: "https://www.mixcloud.com" + u, urls)
 183             track_urls.extend(urls)
 184
 185             # advance to next fragment, if any
 186             if " m-next-page-url=" in resp:
 187                 current_page += 1
 188             else:
 189                 break
 190
 191         return track_urls
 192
 193     def _handle_track_urls(self, urls):
 194         return map(lambda u: self.url_result(u, "Mixcloud"), urls)
 195
 196     def _get_user_description(self, page_content):
 197         return self._html_search_regex(
 198             r'<div class="description-text">.*?<p>(.*?)</p></div></div></div>',
 199             page_content,
 200             "user description",
 201             fatal=False)
 202
 203     def _get_username(self, page_content):
 204         return self._og_search_title(page_content)
 205
 206     def _real_extract(self, url):
 207         mobj = re.match(self._VALID_URL, url)
 208         user_id = mobj.group("user")
 209         list_type = mobj.group("type")
 210
 211         # if only a profile URL was supplied, default to download all uploads
 212         if list_type is None:
 213             list_type = "uploads"
 214
 215         video_id = "%s/%s" % (user_id, list_type)
 216
 217         # download the user's profile to retrieve some metadata
 218         profile = self._download_webpage("https://www.mixcloud.com/%s/" % user_id,
 219                                          video_id,
 220                                          note="Downloading user profile",
 221                                          errnote="Unable to download user profile")
 222
 223         username = self._get_username(profile)
 224         description = self._get_user_description(profile)
 225
 226         # retrieve all page fragments of uploads, favorites or listens
 227         track_urls = self._fetch_tracks(
 228             "https://www.mixcloud.com/%s/%s/" % (user_id, list_type),
 229             video_id,
 230             dl_note="Downloading list of %s" % list_type,
 231             dl_errnote="Unable to download list of %s" % list_type)
 232
 233         # let MixcloudIE handle each track URL
 234         entries = self._handle_track_urls(track_urls)
 235
 236         return {
 237             '_type': 'playlist',
 238             'entries': entries,
 239             'title': "%s (%s)" % (username, list_type),
 240             'id': video_id,
 241             "description": description
 242         }
 243
 244
 245 class MixcloudPlaylistIE(MixcloudUserIE):
 246     """
 247     Information extractor for Mixcloud playlists.
 248     """
 249
 250     _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/(?P<user>[^/]+)/playlists/(?P<playlist>[^/]+)/?$'
 251     IE_NAME = 'mixcloud:playlist'
 252
 253     _TESTS = [{
 254         'url': 'https://www.mixcloud.com/RedBullThre3style/playlists/tokyo-finalists-2015/',
 255         'info_dict': {
 256             'id': 'RedBullThre3style/playlists/tokyo-finalists-2015',
 257             'title': 'National Champions 2015',
 258             'description': 'md5:6ff5fb01ac76a31abc9b3939c16243a3',
 259         },
 260         'playlist_mincount': 16
 261     }, {
 262         'url': 'https://www.mixcloud.com/maxvibes/playlists/jazzcat-on-ness-radio/',
 263         'info_dict': {
 264             'id': 'maxvibes/playlists/jazzcat-on-ness-radio',
 265             'title': 'Jazzcat on Ness Radio',
 266             'description': 'md5:7bbbf0d6359a0b8cda85224be0f8f263',
 267         },
 268         'playlist_mincount': 23
 269     }]
 270
 271     def _get_playlist_title(self, page_content):
 272         return self._html_search_regex(
 273             r'<span class="main-list-title list-playlist-title ">(?P<title>.*?)</span>',
 274             page_content,
 275             "playlist title",
 276             group="title",
 277             fatal=True
 278         )
 279
 280     def _real_extract(self, url):
 281         mobj = re.match(self._VALID_URL, url)
 282         user_id = mobj.group("user")
 283         playlist_id = mobj.group("playlist")
 284         video_id = "%s/playlists/%s" % (user_id, playlist_id)
 285
 286         # download the playlist page to retrieve some metadata
 287         profile = self._download_webpage(url,
 288                                          user_id,
 289                                          note="Downloading playlist page",
 290                                          errnote="Unable to download playlist page")
 291
 292         description = self._get_user_description(profile)
 293         playlist_title = self._get_playlist_title(profile)
 294
 295         # retrieve all page fragments of playlist
 296         track_urls = self._fetch_tracks(
 297             "https://www.mixcloud.com/%s/playlists/%s/" % (user_id, playlist_id),
 298             video_id,
 299             dl_note="Downloading tracklist of %s" % playlist_title,
 300             dl_errnote="Unable to tracklist of %s" % playlist_title)
 301
 302         # let MixcloudIE handle each track
 303         entries = self._handle_track_urls(track_urls)
 304
 305         return {
 306             '_type': 'playlist',
 307             'entries': entries,
 308             'title': playlist_title,
 309             'id': video_id,
 310             "description": description
 311         }