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