X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fhotnewhiphop.py;h=4703e1894f08f70057ace62fc1411980ea7728b4;hb=HEAD;hp=0ee74fb38410a4acce1e15c7a9ce98d80409012e;hpb=bfb9f7bc4c5c6fd9b2d3d46be133988f70534d26;p=youtube-dl diff --git a/youtube_dl/extractor/hotnewhiphop.py b/youtube_dl/extractor/hotnewhiphop.py index 0ee74fb38..4703e1894 100644 --- a/youtube_dl/extractor/hotnewhiphop.py +++ b/youtube_dl/extractor/hotnewhiphop.py @@ -1,44 +1,66 @@ -import re -import base64 +from __future__ import unicode_literals from .common import InfoExtractor +from ..compat import compat_b64decode +from ..utils import ( + ExtractorError, + HEADRequest, + sanitized_Request, + urlencode_postdata, +) class HotNewHipHopIE(InfoExtractor): - _VALID_URL = r'http://www\.hotnewhiphop.com/.*\.(?P.*)\.html' + _VALID_URL = r'https?://(?:www\.)?hotnewhiphop\.com/.*\.(?P.*)\.html' _TEST = { - u'url': u"http://www.hotnewhiphop.com/freddie-gibbs-lay-it-down-song.1435540.html", - u'file': u'1435540.mp3', - u'md5': u'2c2cd2f76ef11a9b3b581e8b232f3d96', - u'info_dict': { - u"title": u'Freddie Gibbs "Lay It Down"' + 'url': 'http://www.hotnewhiphop.com/freddie-gibbs-lay-it-down-song.1435540.html', + 'md5': '2c2cd2f76ef11a9b3b581e8b232f3d96', + 'info_dict': { + 'id': '1435540', + 'ext': 'mp3', + 'title': 'Freddie Gibbs - Lay It Down' } } def _real_extract(self, url): - m = re.match(self._VALID_URL, url) - video_id = m.group('id') + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) - webpage_src = self._download_webpage(url, video_id) + video_url_base64 = self._search_regex( + r'data-path="(.*?)"', webpage, 'video URL', default=None) - video_url_base64 = self._search_regex(r'data-path="(.*?)"', - webpage_src, u'video URL', fatal=False) - - if video_url_base64 == None: - video_url = self._search_regex(r'"contentUrl" content="(.*?)"', webpage_src, - u'video URL') + if video_url_base64 is None: + video_url = self._search_regex( + r'"contentUrl" content="(.*?)"', webpage, 'content URL') return self.url_result(video_url, ie='Youtube') - video_url = base64.b64decode(video_url_base64).decode('utf-8') + reqdata = urlencode_postdata([ + ('mediaType', 's'), + ('mediaId', video_id), + ]) + r = sanitized_Request( + 'http://www.hotnewhiphop.com/ajax/media/getActions/', data=reqdata) + r.add_header('Content-Type', 'application/x-www-form-urlencoded') + mkd = self._download_json( + r, video_id, note='Requesting media key', + errnote='Could not download media key') + if 'mediaKey' not in mkd: + raise ExtractorError('Did not get a media key') + + redirect_url = compat_b64decode(video_url_base64).decode('utf-8') + redirect_req = HEADRequest(redirect_url) + req = self._request_webpage( + redirect_req, video_id, + note='Resolving final URL', errnote='Could not resolve final URL') + video_url = req.geturl() + if video_url.endswith('.html'): + raise ExtractorError('Redirect failed') - video_title = self._html_search_regex(r"(.*)", - webpage_src, u'title') + video_title = self._og_search_title(webpage).strip() - results = [{ - 'id': video_id, - 'url' : video_url, - 'title' : video_title, - 'thumbnail' : self._og_search_thumbnail(webpage_src), - 'ext' : 'mp3', - }] - return results + return { + 'id': video_id, + 'url': video_url, + 'title': video_title, + 'thumbnail': self._og_search_thumbnail(webpage), + }