ytdl-nightly/youtube_dl/extractor/audiomack.py

69 lines
2.6 KiB
Python
Raw Normal View History

2014-10-23 23:55:39 +02:00
# coding: utf-8
from __future__ import unicode_literals
from .common import InfoExtractor
2014-10-24 06:54:59 +02:00
from .soundcloud import SoundcloudIE
2014-10-25 04:07:01 +02:00
from ..utils import ExtractorError
2014-10-23 23:55:39 +02:00
import datetime
import time
class AudiomackIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?audiomack\.com/song/(?P<id>[\w/-]+)'
2014-10-24 06:54:59 +02:00
IE_NAME = 'audiomack'
_TESTS = [
#hosted on audiomack
{
'url': 'http://www.audiomack.com/song/roosh-williams/extraordinary',
'info_dict':
{
2014-10-25 04:07:01 +02:00
'id' : 'roosh-williams/extraordinary',
2014-10-24 06:54:59 +02:00
'ext': 'mp3',
'title': 'Roosh Williams - Extraordinary'
}
},
#hosted on soundcloud via audiomack
{
'url': 'http://www.audiomack.com/song/xclusiveszone/take-kare',
'file': '172419696.mp3',
'info_dict':
{
'ext': 'mp3',
'title': 'Young Thug ft Lil Wayne - Take Kare',
"upload_date": "20141016",
"description": "New track produced by London On Da Track called “Take Kare\"\n\nhttp://instagram.com/theyoungthugworld\nhttps://www.facebook.com/ThuggerThuggerCashMoney\n",
"uploader": "Young Thug World"
}
2014-10-23 23:55:39 +02:00
}
2014-10-24 06:54:59 +02:00
]
2014-10-23 23:55:39 +02:00
def _real_extract(self, url):
2014-10-24 06:54:59 +02:00
#id is what follows /song/ in url, usually the uploader name + title
2014-10-25 04:07:01 +02:00
id = self._match_id(url)
2014-10-23 23:55:39 +02:00
2014-10-24 06:54:59 +02:00
#Call the api, which gives us a json doc with the real url inside
2014-10-25 04:07:01 +02:00
rightnow = int(time.time())
apiresponse = self._download_json("http://www.audiomack.com/api/music/url/song/"+id+"?_="+str(rightnow), id)
if "url" not in apiresponse:
raise ExtractorError("Unable to deduce api url of song")
2014-10-24 06:54:59 +02:00
realurl = apiresponse["url"]
#Audiomack wraps a lot of soundcloud tracks in their branded wrapper
# - if so, pass the work off to the soundcloud extractor
if SoundcloudIE.suitable(realurl):
sc = SoundcloudIE(downloader=self._downloader)
return sc._real_extract(realurl)
else:
#Pull out metadata
page = self._download_webpage(url, id)
artist = self._html_search_regex(r'<span class="artist">(.*)</span>', page, "artist")
songtitle = self._html_search_regex(r'<h1 class="profile-title song-title"><span class="artist">.*</span>(.*)</h1>', page, "title")
title = artist+" - "+songtitle
return {
2014-10-25 04:07:01 +02:00
'id': id, # ignore id, which is not useful in song name
2014-10-24 06:54:59 +02:00
'title': title,
'url': realurl,
'ext': 'mp3'
}