From 1a582dd49d628914fa6a056b490914738f15c56d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jaime=20Marqui=CC=81nez=20Ferra=CC=81ndiz?= Date: Tue, 27 Aug 2013 11:56:48 +0200 Subject: [PATCH] Add an extractor for CNN (closes #1318) --- youtube_dl/extractor/__init__.py | 1 + youtube_dl/extractor/cnn.py | 47 ++++++++++++++++++++++++++++++++ 2 files changed, 48 insertions(+) create mode 100644 youtube_dl/extractor/cnn.py diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index eeeb3db50..ea2af0d0e 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -9,6 +9,7 @@ from .brightcove import BrightcoveIE from .c56 import C56IE from .canalplus import CanalplusIE from .canalc2 import Canalc2IE +from .cnn import CNNIE from .collegehumor import CollegeHumorIE from .comedycentral import ComedyCentralIE from .condenast import CondeNastIE diff --git a/youtube_dl/extractor/cnn.py b/youtube_dl/extractor/cnn.py new file mode 100644 index 000000000..cee78765b --- /dev/null +++ b/youtube_dl/extractor/cnn.py @@ -0,0 +1,47 @@ +import re +import xml.etree.ElementTree + +from .common import InfoExtractor +from ..utils import determine_ext + +class CNNIE(InfoExtractor): + _VALID_URL = r'https?://(edition\.)?cnn\.com/video/(data/.+?|\?)/(?P.+?/(?P[^/]+?)\.cnn)' + + _TEST = { + u'url': u'http://edition.cnn.com/video/?/video/sports/2013/06/09/nadal-1-on-1.cnn', + u'file': u'sports_2013_06_09_nadal-1-on-1.cnn.mp4', + u'md5': u'3e6121ea48df7e2259fe73a0628605c4', + u'info_dict': { + u'title': u'Nadal wins 8th French Open title', + u'description': u'World Sport\'s Amanda Davies chats with 2013 French Open champion Rafael Nadal.', + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + path = mobj.group('path') + page_title = mobj.group('title') + info_xml = self._download_webpage( + 'http://cnn.com/video/data/3.0/%s/index.xml' % path, page_title) + info = xml.etree.ElementTree.fromstring(info_xml.encode('utf-8')) + + formats = [] + for f in info.findall('files/file'): + mf = re.match(r'(\d+)x(\d+)(?:_(.*)k)?',f.attrib['bitrate']) + if mf is not None: + formats.append((int(mf.group(1)), int(mf.group(2)), int(mf.group(3) or 0), f.text)) + formats = sorted(formats) + (_,_,_, video_path) = formats[-1] + video_url = 'http://ht.cdn.turner.com/cnn/big%s' % video_path + + thumbnails = sorted([((int(t.attrib['height']),int(t.attrib['width'])), t.text) for t in info.findall('images/image')]) + thumbs_dict = [{'resolution': res, 'url': t_url} for (res, t_url) in thumbnails] + + return {'id': info.attrib['id'], + 'title': info.find('headline').text, + 'url': video_url, + 'ext': determine_ext(video_url), + 'thumbnail': thumbnails[-1][1], + 'thumbnails': thumbs_dict, + 'description': info.find('description').text, + }