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