X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fhowstuffworks.py;h=e9733912132798d99be18bb935dcd3c3b190525d;hb=d4f64cabf4ede444b390bb71b90ad4103ce572c0;hp=c46e8b911abc47c679db8d2788fb959c54c5e488;hpb=d958fa9ff961fb17b55aba0d70e5175e2c63f631;p=youtube-dl diff --git a/youtube_dl/extractor/howstuffworks.py b/youtube_dl/extractor/howstuffworks.py index c46e8b911..e97339121 100644 --- a/youtube_dl/extractor/howstuffworks.py +++ b/youtube_dl/extractor/howstuffworks.py @@ -1,19 +1,27 @@ from __future__ import unicode_literals from .common import InfoExtractor +from ..utils import ( + find_xpath_attr, + int_or_none, + js_to_json, + unescapeHTML, +) class HowStuffWorksIE(InfoExtractor): _VALID_URL = r'https?://[\da-z-]+\.howstuffworks\.com/(?:[^/]+/)*\d+-(?P.+?)-video\.htm' _TESTS = [ { - 'url': 'http://adventure.howstuffworks.com/39521-deadliest-catch-nautical-collision-video.htm', + 'url': 'http://adventure.howstuffworks.com/5266-cool-jobs-iditarod-musher-video.htm', 'info_dict': { - 'id': '553475', - 'ext': 'mp4', - 'title': 'Deadliest Catch: Nautical Collision', - 'description': 'Check out this clip to get an exclusive look at the Deadliest Catch crew back in action.', - 'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/39521.jpg', + 'id': '450221', + 'ext': 'flv', + 'title': 'Cool Jobs - Iditarod Musher', + 'description': 'Cold sleds, freezing temps and warm dog breath... an Iditarod musher\'s dream. Kasey-Dee Gardner jumps on a sled to find out what the big deal is.', + 'display_id': 'cool-jobs-iditarod-musher', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 161, }, }, { @@ -23,53 +31,73 @@ class HowStuffWorksIE(InfoExtractor): 'ext': 'mp4', 'title': 'Survival Zone: Food and Water In the Savanna', 'description': 'Learn how to find both food and water while trekking in the African savannah. In this video from the Discovery Channel.', - 'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/7199.jpg', + 'display_id': 'survival-zone-food-and-water-in-the-savanna', + 'thumbnail': 're:^https?://.*\.jpg$', }, }, { - 'url': 'http://entertainment.howstuffworks.com/arts/34476-time-warp-fire-breathing-disaster-video.htm', + 'url': 'http://entertainment.howstuffworks.com/arts/2706-sword-swallowing-1-by-dan-meyer-video.htm', 'info_dict': { - 'id': '487036', - 'ext': 'mp4', - 'title': 'Time Warp: Fire-Breathing Disaster', - 'description': 'Fire-breathers are safe from flames when they angle their heads up as they blow -- see what would happen if they looked down instead. Check out this clip from Discovery\'s "Time Warp" series to learn more.', - 'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/34476.jpg', + 'id': '440011', + 'ext': 'flv', + 'title': 'Sword Swallowing #1 by Dan Meyer', + 'description': 'Video footage (1 of 3) used by permission of the owner Dan Meyer through Sword Swallowers Association International ', + 'display_id': 'sword-swallowing-1-by-dan-meyer', + 'thumbnail': 're:^https?://.*\.jpg$', }, }, ] - def _extract_clip_info(self, key, clip_info, name=None, **kargs): - if name is None: - name = key - return self._html_search_regex( - r"\s*%s\s*: '?([^'\n]*[^'\n,])" % key, clip_info, name, **kargs) - def _real_extract(self, url): display_id = self._match_id(url) webpage = self._download_webpage(url, display_id) - clip_info = self._search_regex('(?s)var clip = {(.*?)};', webpage, 'clip info') - video_id = self._extract_clip_info('content_id', clip_info, 'video id') + clip_js = self._search_regex( + r'(?s)var clip = ({.*?});', webpage, 'clip info') + clip_info = self._parse_json( + clip_js, display_id, transform_source=js_to_json) + + video_id = clip_info['content_id'] formats = [] - m3u8_url = self._extract_clip_info('m3u8', clip_info, 'm3u8 url', default=None) - if m3u8_url is not None: - formats += self._extract_m3u8_formats(m3u8_url , video_id, 'mp4') - mp4 = self._parse_json( - self._extract_clip_info( - 'mp4', clip_info, 'formats').replace('},]','}]'), video_id) - for video in mp4: + m3u8_url = clip_info.get('m3u8') + if m3u8_url: + formats += self._extract_m3u8_formats(m3u8_url, video_id, 'mp4') + for video in clip_info.get('mp4', []): formats.append({ 'url': video['src'], 'format_id': video['bitrate'], - 'tbr': int(video['bitrate'].rstrip('k')), + 'vbr': int(video['bitrate'].rstrip('k')), }) + + if not formats: + smil = self._download_xml( + 'http://services.media.howstuffworks.com/videos/%s/smil-service.smil' % video_id, + video_id, 'Downloading video SMIL') + + http_base = find_xpath_attr( + smil, + './{0}head/{0}meta'.format('{http://www.w3.org/2001/SMIL20/Language}'), + 'name', + 'httpBase').get('content') + + URL_SUFFIX = '?v=2.11.3&fp=LNX 11,2,202,356&r=A&g=A' + + for video in smil.findall( + './{0}body/{0}switch/{0}video'.format('{http://www.w3.org/2001/SMIL20/Language}')): + vbr = int_or_none(video.attrib['system-bitrate'], scale=1000) + formats.append({ + 'url': '%s/%s%s' % (http_base, video.attrib['src'], URL_SUFFIX), + 'format_id': '%dk' % vbr, + 'vbr': vbr, + }) + self._sort_formats(formats) return { - 'id': video_id, + 'id': '%s' % video_id, 'display_id': display_id, - 'title': self._extract_clip_info('clip_title', clip_info, 'title'), - 'description': self._extract_clip_info('caption', clip_info, 'description', fatal=False), - 'thumbnail': self._extract_clip_info('video_still_url', clip_info, 'thumbnail'), - 'duration': self._extract_clip_info('duration', clip_info), + 'title': unescapeHTML(clip_info['clip_title']), + 'description': unescapeHTML(clip_info.get('caption')), + 'thumbnail': clip_info.get('video_still_url'), + 'duration': clip_info.get('duration'), 'formats': formats, }