_ 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,
   8 )
   9 from ..utils import (
  10     ExtractorError,
  11     HEADRequest,
  12     int_or_none,
  13     str_to_int,
  14     parse_iso8601,
  15 )
  16
  17
  18 class MixcloudIE(InfoExtractor):
  19     _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/([^/]+)/([^/]+)'
  20     IE_NAME = 'mixcloud'
  21
  22     _TEST = {
  23         'url': 'http://www.mixcloud.com/dholbach/cryptkeeper/',
  24         'info_dict': {
  25             'id': 'dholbach-cryptkeeper',
  26             'ext': 'mp3',
  27             'title': 'Cryptkeeper',
  28             'description': 'After quite a long silence from myself, finally another Drum\'n\'Bass mix with my favourite current dance floor bangers.',
  29             'uploader': 'Daniel Holbach',
  30             'uploader_id': 'dholbach',
  31             'upload_date': '20111115',
  32             'timestamp': 1321359578,
  33             'thumbnail': 're:https?://.*\.jpg',
  34             'view_count': int,
  35             'like_count': int,
  36         },
  37     }
  38
  39     def _get_url(self, track_id, template_url):
  40         server_count = 30
  41         for i in range(server_count):
  42             url = template_url % i
  43             try:
  44                 # We only want to know if the request succeed
  45                 # don't download the whole file
  46                 self._request_webpage(
  47                     HEADRequest(url), track_id,
  48                     'Checking URL %d/%d ...' % (i + 1, server_count + 1))
  49                 return url
  50             except ExtractorError:
  51                 pass
  52
  53         return None
  54
  55     def _real_extract(self, url):
  56         mobj = re.match(self._VALID_URL, url)
  57         uploader = mobj.group(1)
  58         cloudcast_name = mobj.group(2)
  59         track_id = compat_urllib_parse.unquote('-'.join((uploader, cloudcast_name)))
  60
  61         webpage = self._download_webpage(url, track_id)
  62
  63         preview_url = self._search_regex(
  64             r'\s(?:data-preview-url|m-preview)="(.+?)"', webpage, 'preview url')
  65         song_url = preview_url.replace('/previews/', '/c/originals/')
  66         template_url = re.sub(r'(stream\d*)', 'stream%d', song_url)
  67         final_song_url = self._get_url(track_id, template_url)
  68         if final_song_url is None:
  69             self.to_screen('Trying with m4a extension')
  70             template_url = template_url.replace('.mp3', '.m4a').replace('originals/', 'm4a/64/')
  71             final_song_url = self._get_url(track_id, template_url)
  72         if final_song_url is None:
  73             raise ExtractorError('Unable to extract track url')
  74
  75         PREFIX = (
  76             r'<span class="play-button[^"]*?"'
  77             r'(?:\s+[a-zA-Z0-9-]+(?:="[^"]+")?)*?\s+')
  78         title = self._html_search_regex(
  79             PREFIX + r'm-title="([^"]+)"', webpage, 'title')
  80         thumbnail = self._proto_relative_url(self._html_search_regex(
  81             PREFIX + r'm-thumbnail-url="([^"]+)"', webpage, 'thumbnail',
  82             fatal=False))
  83         uploader = self._html_search_regex(
  84             PREFIX + r'm-owner-name="([^"]+)"',
  85             webpage, 'uploader', fatal=False)
  86         uploader_id = self._search_regex(
  87             r'\s+"profile": "([^"]+)",', webpage, 'uploader id', fatal=False)
  88         description = self._og_search_description(webpage)
  89         like_count = str_to_int(self._search_regex(
  90             [r'<meta itemprop="interactionCount" content="UserLikes:([0-9]+)"',
  91              r'/favorites/?">([0-9]+)<'],
  92             webpage, 'like count', fatal=False))
  93         view_count = str_to_int(self._search_regex(
  94             [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
  95              r'/listeners/?">([0-9,.]+)</a>'],
  96             webpage, 'play count', fatal=False))
  97         timestamp = parse_iso8601(self._search_regex(
  98             r'<time itemprop="dateCreated" datetime="([^"]+)">',
  99             webpage, 'upload date', default=None))
 100
 101         return {
 102             'id': track_id,
 103             'title': title,
 104             'url': final_song_url,
 105             'description': description,
 106             'thumbnail': thumbnail,
 107             'uploader': uploader,
 108             'uploader_id': uploader_id,
 109             'timestamp': timestamp,
 110             'view_count': view_count,
 111             'like_count': like_count,
 112         }