_ Git - youtube-dl/blob - youtube_dl/extractor/audiomack.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from .soundcloud import SoundcloudIE
   6 from ..utils import ExtractorError
   7 import datetime
   8 import time
   9
  10
  11 class AudiomackIE(InfoExtractor):
  12     _VALID_URL = r'https?://(?:www\.)?audiomack\.com/song/(?P<id>[\w/-]+)'
  13     IE_NAME = 'audiomack'
  14     _TESTS = [
  15         #hosted on audiomack
  16         {
  17             'url': 'http://www.audiomack.com/song/roosh-williams/extraordinary',
  18             'info_dict':
  19             {
  20                 'id' : 'roosh-williams/extraordinary',
  21                 'ext': 'mp3',
  22                 'title': 'Roosh Williams - Extraordinary'
  23             }
  24         },
  25         #hosted on soundcloud via audiomack
  26         {
  27             'url': 'http://www.audiomack.com/song/xclusiveszone/take-kare',
  28             'file': '172419696.mp3',
  29             'info_dict':
  30             {
  31                 'ext': 'mp3',
  32                 'title': 'Young Thug ft Lil Wayne - Take Kare',
  33                 "upload_date": "20141016",
  34                 "description": "New track produced by London On Da Track called “Take Kare\"\n\nhttp://instagram.com/theyoungthugworld\nhttps://www.facebook.com/ThuggerThuggerCashMoney\n",
  35                 "uploader": "Young Thug World"
  36             }
  37         }
  38     ]
  39
  40     def _real_extract(self, url):
  41         #id is what follows /song/ in url, usually the uploader name + title
  42         id = self._match_id(url)
  43
  44         #Call the api, which gives us a json doc with the real url inside
  45         rightnow = int(time.time())
  46         apiresponse = self._download_json("http://www.audiomack.com/api/music/url/song/"+id+"?_="+str(rightnow), id)
  47
  48         if "url" not in apiresponse:
  49             raise ExtractorError("Unable to deduce api url of song")
  50         realurl = apiresponse["url"]
  51
  52         #Audiomack wraps a lot of soundcloud tracks in their branded wrapper
  53         # - if so, pass the work off to the soundcloud extractor
  54         if SoundcloudIE.suitable(realurl):
  55             return {'_type': 'url', 'url': realurl, 'ie_key': 'Soundcloud'}
  56         else:
  57             #Pull out metadata
  58             page = self._download_webpage(url, id)
  59             artist = self._html_search_regex(r'<span class="artist">(.*)</span>', page, "artist")
  60             songtitle = self._html_search_regex(r'<h1 class="profile-title song-title"><span class="artist">.*</span>(.*)</h1>', page, "title")
  61             title = artist+" - "+songtitle
  62             return {
  63                 'id': id,  # ignore id, which is not useful in song name
  64                 'title': title,
  65                 'url': realurl,
  66                 'ext': 'mp3'
  67             }