X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fcnn.py;h=774b7105580f69dedac4318c82aee2ab62c27381;hb=HEAD;hp=220bb55e8f4ea417481f3f207c4112d8c9e13e17;hpb=d13bfc07b79bafa0be6829a18d0d2e273e83339b;p=youtube-dl diff --git a/youtube_dl/extractor/cnn.py b/youtube_dl/extractor/cnn.py index 220bb55e8..774b71055 100644 --- a/youtube_dl/extractor/cnn.py +++ b/youtube_dl/extractor/cnn.py @@ -3,14 +3,11 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..utils import ( - int_or_none, - parse_duration, - url_basename, -) +from .turner import TurnerBaseIE +from ..utils import url_basename -class CNNIE(InfoExtractor): +class CNNIE(TurnerBaseIE): _VALID_URL = r'''(?x)https?://(?:(?Pedition|www|money)\.)?cnn\.com/(?:video/(?:data/.+?|\?)/)?videos?/ (?P.+?/(?P[^/]+?)(?:\.(?:[a-z\-]+)|(?=&)))''' @@ -25,6 +22,7 @@ class CNNIE(InfoExtractor): 'duration': 135, 'upload_date': '20130609', }, + 'expected_warnings': ['Failed to download m3u8 information'], }, { 'url': 'http://edition.cnn.com/video/?/video/us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+rss%2Fcnn_topstories+%28RSS%3A+Top+Stories%29', 'md5': 'b5cc60c60a3477d185af8f19a2a26f4e', @@ -34,7 +32,8 @@ class CNNIE(InfoExtractor): 'title': "Student's epic speech stuns new freshmen", 'description': "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\"", 'upload_date': '20130821', - } + }, + 'expected_warnings': ['Failed to download m3u8 information'], }, { 'url': 'http://www.cnn.com/video/data/2.0/video/living/2014/12/22/growing-america-nashville-salemtown-board-episode-1.hln.html', 'md5': 'f14d02ebd264df951feb2400e2c25a1b', @@ -44,7 +43,8 @@ class CNNIE(InfoExtractor): 'title': 'Nashville Ep. 1: Hand crafted skateboards', 'description': 'md5:e7223a503315c9f150acac52e76de086', 'upload_date': '20141222', - } + }, + 'expected_warnings': ['Failed to download m3u8 information'], }, { 'url': 'http://money.cnn.com/video/news/2016/08/19/netflix-stunning-stats.cnnmoney/index.html', 'md5': '52a515dc1b0f001cd82e4ceda32be9d1', @@ -54,7 +54,11 @@ class CNNIE(InfoExtractor): 'title': '5 stunning stats about Netflix', 'description': 'Did you know that Netflix has more than 80 million members? Here are five facts about the online video distributor that you probably didn\'t know.', 'upload_date': '20160819', - } + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, }, { 'url': 'http://cnn.com/video/?/video/politics/2015/03/27/pkg-arizona-senator-church-attendance-mandatory.ktvk', 'only_matching': True, @@ -79,72 +83,21 @@ class CNNIE(InfoExtractor): }, } + def _extract_timestamp(self, video_data): + # TODO: fix timestamp extraction + return None + def _real_extract(self, url): sub_domain, path, page_title = re.match(self._VALID_URL, url).groups() if sub_domain not in ('money', 'edition'): sub_domain = 'edition' config = self._CONFIG[sub_domain] - info_url = config['data_src'] % path - info = self._download_xml(info_url, page_title) - - formats = [] - rex = re.compile(r'''(?x) - (?P<width>[0-9]+)x(?P<height>[0-9]+) - (?:_(?P<bitrate>[0-9]+)k)? - ''') - for f in info.findall('files/file'): - video_url = config['media_src'] + f.text.strip() - fdct = { - 'format_id': f.attrib['bitrate'], - 'url': video_url, - } - - mf = rex.match(f.attrib['bitrate']) - if mf: - fdct['width'] = int(mf.group('width')) - fdct['height'] = int(mf.group('height')) - fdct['tbr'] = int_or_none(mf.group('bitrate')) - else: - mf = rex.search(f.text) - if mf: - fdct['width'] = int(mf.group('width')) - fdct['height'] = int(mf.group('height')) - fdct['tbr'] = int_or_none(mf.group('bitrate')) - else: - mi = re.match(r'ios_(audio|[0-9]+)$', f.attrib['bitrate']) - if mi: - if mi.group(1) == 'audio': - fdct['vcodec'] = 'none' - fdct['ext'] = 'm4a' - else: - fdct['tbr'] = int(mi.group(1)) - - formats.append(fdct) - - self._sort_formats(formats) - - thumbnails = [{ - 'height': int(t.attrib['height']), - 'width': int(t.attrib['width']), - 'url': t.text, - } for t in info.findall('images/image')] - - metas_el = info.find('metas') - upload_date = ( - metas_el.attrib.get('version') if metas_el is not None else None) - - duration_el = info.find('length') - duration = parse_duration(duration_el.text) - - return { - 'id': info.attrib['id'], - 'title': info.find('headline').text, - 'formats': formats, - 'thumbnails': thumbnails, - 'description': info.find('description').text, - 'duration': duration, - 'upload_date': upload_date, - } + return self._extract_cvp_info( + config['data_src'] % path, page_title, { + 'default': { + 'media_src': config['media_src'], + } + }) class CNNBlogsIE(InfoExtractor): @@ -159,17 +112,14 @@ class CNNBlogsIE(InfoExtractor): 'description': 'Glenn Greenwald responds to comments made this week on Capitol Hill that journalists could be criminal accessories.', 'upload_date': '20140209', }, + 'expected_warnings': ['Failed to download m3u8 information'], 'add_ie': ['CNN'], } def _real_extract(self, url): webpage = self._download_webpage(url, url_basename(url)) cnn_url = self._html_search_regex(r'data-url="(.+?)"', webpage, 'cnn url') - return { - '_type': 'url', - 'url': cnn_url, - 'ie_key': CNNIE.ie_key(), - } + return self.url_result(cnn_url, CNNIE.ie_key()) class CNNArticleIE(InfoExtractor): @@ -181,17 +131,14 @@ class CNNArticleIE(InfoExtractor): 'id': 'bestoftv/2014/12/21/ip-north-korea-obama.cnn', 'ext': 'mp4', 'title': 'Obama: Cyberattack not an act of war', - 'description': 'md5:51ce6750450603795cad0cdfbd7d05c5', + 'description': 'md5:0a802a40d2376f60e6b04c8d5bcebc4b', 'upload_date': '20141221', }, + 'expected_warnings': ['Failed to download m3u8 information'], 'add_ie': ['CNN'], } def _real_extract(self, url): webpage = self._download_webpage(url, url_basename(url)) cnn_url = self._html_search_regex(r"video:\s*'([^']+)'", webpage, 'cnn url') - return { - '_type': 'url', - 'url': 'http://cnn.com/video/?/video/' + cnn_url, - 'ie_key': CNNIE.ie_key(), - } + return self.url_result('http://cnn.com/video/?/video/' + cnn_url, CNNIE.ie_key())