youtube_dl/extractor/mixcloud.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     compat_urllib_parse,
   8     ExtractorError,
   9     HEADRequest,
  10     int_or_none,
  11     parse_iso8601,
  12 )
  13
  14
  15 class MixcloudIE(InfoExtractor):
  16     _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/([^/]+)/([^/]+)'
  17     IE_NAME = 'mixcloud'
  18
  19     _TEST = {
  20         'url': 'http://www.mixcloud.com/dholbach/cryptkeeper/',
  21         'info_dict': {
  22             'id': 'dholbach-cryptkeeper',
  23             'ext': 'mp3',
  24             'title': 'Cryptkeeper',
  25             'description': 'After quite a long silence from myself, finally another Drum\'n\'Bass mix with my favourite current dance floor bangers.',
  26             'uploader': 'Daniel Holbach',
  27             'uploader_id': 'dholbach',
  28             'upload_date': '20111115',
  29             'timestamp': 1321359578,
  30             'thumbnail': 're:https?://.*\.jpg',
  31             'view_count': int,
  32             'like_count': int,
  33         },
  34     }
  35
  36     def check_urls(self, url_list):
  37         """Returns 1st active url from list"""
  38         for url in url_list:
  39             try:
  40                 # We only want to know if the request succeed
  41                 # don't download the whole file
  42                 self._request_webpage(HEADRequest(url), None, False)
  43                 return url
  44             except ExtractorError:
  45                 url = None
  46
  47         return None
  48
  49     def _get_url(self, template_url):
  50         return self.check_urls(template_url % i for i in range(30))
  51
  52     def _real_extract(self, url):
  53         mobj = re.match(self._VALID_URL, url)
  54         uploader = mobj.group(1)
  55         cloudcast_name = mobj.group(2)
  56         track_id = compat_urllib_parse.unquote('-'.join((uploader, cloudcast_name)))
  57
  58         webpage = self._download_webpage(url, track_id)
  59
  60         preview_url = self._search_regex(
  61             r'\s(?:data-preview-url|m-preview)="(.+?)"', webpage, 'preview url')
  62         song_url = preview_url.replace('/previews/', '/c/originals/')
  63         template_url = re.sub(r'(stream\d*)', 'stream%d', song_url)
  64         final_song_url = self._get_url(template_url)
  65         if final_song_url is None:
  66             self.to_screen('Trying with m4a extension')
  67             template_url = template_url.replace('.mp3', '.m4a').replace('originals/', 'm4a/64/')
  68             final_song_url = self._get_url(template_url)
  69         if final_song_url is None:
  70             raise ExtractorError('Unable to extract track url')
  71
  72         PREFIX = (
  73             r'<div class="cloudcast-play-button-container"'
  74             r'(?:\s+[a-zA-Z0-9-]+(?:="[^"]+")?)*?\s+')
  75         title = self._html_search_regex(
  76             PREFIX + r'm-title="([^"]+)"', webpage, 'title')
  77         thumbnail = self._proto_relative_url(self._html_search_regex(
  78             PREFIX + r'm-thumbnail-url="([^"]+)"', webpage, 'thumbnail',
  79             fatal=False))
  80         uploader = self._html_search_regex(
  81             PREFIX + r'm-owner-name="([^"]+)"',
  82             webpage, 'uploader', fatal=False)
  83         uploader_id = self._search_regex(
  84             r'\s+"profile": "([^"]+)",', webpage, 'uploader id', fatal=False)
  85         description = self._og_search_description(webpage)
  86         like_count = int_or_none(self._search_regex(
  87             r'<meta itemprop="interactionCount" content="UserLikes:([0-9]+)"',
  88             webpage, 'like count', fatal=False))
  89         view_count = int_or_none(self._search_regex(
  90             r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
  91             webpage, 'play count', fatal=False))
  92         timestamp = parse_iso8601(self._search_regex(
  93             r'<time itemprop="dateCreated" datetime="([^"]+)">',
  94             webpage, 'upload date'))
  95
  96         return {
  97             'id': track_id,
  98             'title': title,
  99             'url': final_song_url,
 100             'description': description,
 101             'thumbnail': thumbnail,
 102             'uploader': uploader,
 103             'uploader_id': uploader_id,
 104             'timestamp': timestamp,
 105             'view_count': view_count,
 106             'like_count': like_count,
 107         }