_ 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 compat_urllib_parse_unquote
   7 from ..utils import (
   8     ExtractorError,
   9     HEADRequest,
  10     parse_count,
  11     str_to_int,
  12 )
  13
  14
  15 class MixcloudIE(InfoExtractor):
  16     _VALID_URL = r'^(?:https?://)?(?:www\.)?mixcloud\.com/([^/]+)/([^/]+)'
  17     IE_NAME = 'mixcloud'
  18
  19     _TESTS = [{
  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             'thumbnail': 're:https?://.*\.jpg',
  29             'view_count': int,
  30             'like_count': int,
  31         },
  32     }, {
  33         'url': 'http://www.mixcloud.com/gillespeterson/caribou-7-inch-vinyl-mix-chat/',
  34         'info_dict': {
  35             'id': 'gillespeterson-caribou-7-inch-vinyl-mix-chat',
  36             'ext': 'mp3',
  37             'title': 'Caribou 7 inch Vinyl Mix & Chat',
  38             'description': 'md5:2b8aec6adce69f9d41724647c65875e8',
  39             'uploader': 'Gilles Peterson Worldwide',
  40             'uploader_id': 'gillespeterson',
  41             'thumbnail': 're:https?://.*/images/',
  42             'view_count': int,
  43             'like_count': int,
  44         },
  45     }]
  46
  47     def _check_url(self, url, track_id, ext):
  48         try:
  49             # We only want to know if the request succeed
  50             # don't download the whole file
  51             self._request_webpage(
  52                 HEADRequest(url), track_id,
  53                 'Trying %s URL' % ext)
  54             return True
  55         except ExtractorError:
  56             return False
  57
  58     def _real_extract(self, url):
  59         mobj = re.match(self._VALID_URL, url)
  60         uploader = mobj.group(1)
  61         cloudcast_name = mobj.group(2)
  62         track_id = compat_urllib_parse_unquote('-'.join((uploader, cloudcast_name)))
  63
  64         webpage = self._download_webpage(url, track_id)
  65
  66         preview_url = self._search_regex(
  67             r'\s(?:data-preview-url|m-preview)="([^"]+)"', webpage, 'preview url')
  68         song_url = re.sub(r'audiocdn(\d+)', r'stream\1', preview_url)
  69         song_url = song_url.replace('/previews/', '/c/originals/')
  70         if not self._check_url(song_url, track_id, 'mp3'):
  71             song_url = song_url.replace('.mp3', '.m4a').replace('originals/', 'm4a/64/')
  72             if not self._check_url(song_url, track_id, 'm4a'):
  73                 raise ExtractorError('Unable to extract track url')
  74
  75         PREFIX = (
  76             r'm-play-on-spacebar[^>]+'
  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 = parse_count(self._search_regex(
  90             r'\bbutton-favorite[^>]+>.*?<span[^>]+class=["\']toggle-number[^>]+>\s*([^<]+)',
  91             webpage, 'like count', fatal=False))
  92         view_count = str_to_int(self._search_regex(
  93             [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
  94              r'/listeners/?">([0-9,.]+)</a>'],
  95             webpage, 'play count', fatal=False))
  96
  97         return {
  98             'id': track_id,
  99             'title': title,
 100             'url': song_url,
 101             'description': description,
 102             'thumbnail': thumbnail,
 103             'uploader': uploader,
 104             'uploader_id': uploader_id,
 105             'view_count': view_count,
 106             'like_count': like_count,
 107         }