X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;ds=sidebyside;f=youtube_dl%2Fextractor%2Fliveleak.py;h=4e76c1f4a5f8c0a802ea95eaac6d20eb373998c3;hb=4f879a5be0813ae2c0d756ad52608125117fef58;hp=d01fd01e3a5413b8b98305c94507ce25d8dab44e;hpb=7eeb5bef24b0c76e5d01d8f4ac6f94fb70416bb5;p=youtube-dl diff --git a/youtube_dl/extractor/liveleak.py b/youtube_dl/extractor/liveleak.py index d01fd01e3..4e76c1f4a 100644 --- a/youtube_dl/extractor/liveleak.py +++ b/youtube_dl/extractor/liveleak.py @@ -1,5 +1,6 @@ from __future__ import unicode_literals +import json import re from .common import InfoExtractor @@ -10,7 +11,7 @@ from ..utils import ( class LiveLeakIE(InfoExtractor): _VALID_URL = r'^(?:http://)?(?:\w+\.)?liveleak\.com/view\?(?:.*?)i=(?P[\w_]+)(?:.*)' - _TEST = { + _TESTS = [{ 'url': 'http://www.liveleak.com/view?i=757_1364311680', 'file': '757_1364311680.mp4', 'md5': '0813c2430bea7a46bf13acf3406992f4', @@ -19,15 +20,37 @@ class LiveLeakIE(InfoExtractor): 'uploader': 'ljfriel2', 'title': 'Most unlucky car accident' } - } + }, + { + 'url': 'http://www.liveleak.com/view?i=f93_1390833151', + 'file': 'f93_1390833151.mp4', + 'md5': 'd3f1367d14cc3c15bf24fbfbe04b9abf', + 'info_dict': { + 'description': 'German Television Channel NDR does an exclusive interview with Edward Snowden.\r\nUploaded on LiveLeak cause German Television thinks the rest of the world isn\'t intereseted in Edward Snowden.', + 'uploader': 'ARD_Stinkt', + 'title': 'German Television does first Edward Snowden Interview (ENGLISH)', + } + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('video_id') webpage = self._download_webpage(url, video_id) - video_url = self._search_regex( - r'file: "(.*?)",', webpage, 'video URL') + sources_raw = self._search_regex( + r'(?s)sources:\s*(\[.*?\]),', webpage, 'video URLs', default=None) + if sources_raw is None: + sources_raw = '[{ %s}]' % ( + self._search_regex(r'(file: ".*?"),', webpage, 'video URL')) + + sources_json = re.sub(r'\s([a-z]+):\s', r'"\1": ', sources_raw) + sources = json.loads(sources_json) + + formats = [{ + 'format_note': s.get('label'), + 'url': s['file'], + } for s in sources] + self._sort_formats(formats) video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip() video_description = self._og_search_description(webpage) @@ -36,9 +59,8 @@ class LiveLeakIE(InfoExtractor): return { 'id': video_id, - 'url': video_url, - 'ext': 'mp4', 'title': video_title, 'description': video_description, - 'uploader': video_uploader + 'uploader': video_uploader, + 'formats': formats, }