From 9e9bc793f3abddc4824cfcb13f569163fb0a4ba7 Mon Sep 17 00:00:00 2001 From: xavier Date: Thu, 23 Oct 2014 23:54:59 -0500 Subject: [PATCH] Finished audiomack extractor --- youtube_dl/extractor/audiomack.py | 82 ++++++++++++++++++++----------- 1 file changed, 53 insertions(+), 29 deletions(-) diff --git a/youtube_dl/extractor/audiomack.py b/youtube_dl/extractor/audiomack.py index c5214f401..2ececa998 100644 --- a/youtube_dl/extractor/audiomack.py +++ b/youtube_dl/extractor/audiomack.py @@ -1,43 +1,67 @@ +# Xavier Beynon 2014 # coding: utf-8 from __future__ import unicode_literals from .common import InfoExtractor +from .soundcloud import SoundcloudIE import datetime import time -import urllib.request -import json class AudiomackIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?audiomack\.com/song/(?P[\w/-]+)' - _TEST = { - 'url': 'https://www.audiomack.com/song/crewneckkramer/story-i-tell', - 'info_dict': { - 'id': 'story-i-tell', - 'ext': 'mp3', - 'title': 'story-i-tell' + IE_NAME = 'audiomack' + _TESTS = [ + #hosted on audiomack + { + 'url': 'http://www.audiomack.com/song/roosh-williams/extraordinary', + 'file': 'Roosh Williams - Extraordinary.mp3', + 'info_dict': + { + '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" + } } - } + ] def _real_extract(self, url): - # TODO more code goes here, for example ... - #webpage = self._download_webpage(url, video_id) - #title = self._html_search_regex(r'

(.*?)

', webpage, 'title') - - assert("/song/" in url) - songurl = url[url.index("/song/")+5:] - title = songurl[songurl.rindex("/")+1:] - video_id = title - t = int(time.mktime(datetime.datetime.now().timetuple())) - s = "http://www.audiomack.com/api/music/url/song"+songurl+"?_="+str(t) - f = urllib.request.urlopen(s) - j = f.read(1000).decode("utf-8") - data = json.loads(j) + #id is what follows /song/ in url, usually the uploader name + title + id = url[url.index("/song/")+5:] - return { - 'id': video_id, - 'title': title, - 'url' : data["url"], - 'ext' : 'mp3' - # TODO more properties (see youtube_dl/extractor/common.py) - } + #Call the api, which gives us a json doc with the real url inside + rightnow = int(time.mktime(datetime.datetime.now().timetuple())) + apiresponse = self._download_json("http://www.audiomack.com/api/music/url/song"+id+"?_="+str(rightnow), id) + if not url in apiresponse: + raise Exception("Unable to deduce api url of song") + 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'(.*)', page, "artist") + songtitle = self._html_search_regex(r'

.*(.*)

', page, "title") + title = artist+" - "+songtitle + return { + 'id': title, # ignore id, which is not useful in song name + 'title': title, + 'url': realurl, + 'ext': 'mp3' + } -- 2.22.2