[howstuffworks] Update extractor

This commit is contained in:
Tithen-Firion 2014-12-05 19:46:49 +01:00
parent d958fa9ff9
commit e638e83662

View file

@ -1,19 +1,20 @@
from __future__ import unicode_literals from __future__ import unicode_literals
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import find_xpath_attr
class HowStuffWorksIE(InfoExtractor): class HowStuffWorksIE(InfoExtractor):
_VALID_URL = r'https?://[\da-z-]+\.howstuffworks\.com/(?:[^/]+/)*\d+-(?P<id>.+?)-video\.htm' _VALID_URL = r'https?://[\da-z-]+\.howstuffworks\.com/(?:[^/]+/)*\d+-(?P<id>.+?)-video\.htm'
_TESTS = [ _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': { 'info_dict': {
'id': '553475', 'id': '450221',
'ext': 'mp4', 'ext': 'flv',
'title': 'Deadliest Catch: Nautical Collision', 'title': 'Cool Jobs - Iditarod Musher',
'description': 'Check out this clip to get an exclusive look at the Deadliest Catch crew back in action.', '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.',
'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/39521.jpg', 'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/5266.jpg',
}, },
}, },
{ {
@ -27,49 +28,73 @@ class HowStuffWorksIE(InfoExtractor):
}, },
}, },
{ {
'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': { 'info_dict': {
'id': '487036', 'id': '440011',
'ext': 'mp4', 'ext': 'flv',
'title': 'Time Warp: Fire-Breathing Disaster', 'title': 'Sword Swallowing #1 by Dan Meyer',
'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.', 'description': 'Video footage (1 of 3) used by permission of the owner Dan Meyer through Sword Swallowers Association International <www.swordswallow.org>',
'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/34476.jpg', 'thumbnail': 'http://s.hswstatic.com/gif/videos/480x360/118306353233.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): def _real_extract(self, url):
display_id = self._match_id(url) display_id = self._match_id(url)
webpage = self._download_webpage(url, display_id) webpage = self._download_webpage(url, display_id)
clip_info = self._search_regex('(?s)var clip = {(.*?)};', webpage, 'clip info') clip_info = self._search_regex('(?s)var clip = {(.*?)};', webpage, 'clip info')
video_id = self._extract_clip_info('content_id', clip_info, 'video id')
def extract_clip_info(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)
video_id = extract_clip_info('content_id', clip_info, 'video id')
formats = [] formats = []
m3u8_url = self._extract_clip_info('m3u8', clip_info, 'm3u8 url', default=None) m3u8_url = extract_clip_info('m3u8', clip_info, 'm3u8 url', default=None)
if m3u8_url is not None: if m3u8_url is not None:
formats += self._extract_m3u8_formats(m3u8_url , video_id, 'mp4') formats += self._extract_m3u8_formats(m3u8_url , video_id, 'mp4')
mp4 = self._parse_json( mp4 = self._parse_json(
self._extract_clip_info( extract_clip_info(
'mp4', clip_info, 'formats').replace('},]','}]'), video_id) 'mp4', clip_info, 'formats').replace('},]','}]'), video_id)
for video in mp4: for video in mp4:
formats.append({ formats.append({
'url': video['src'], 'url': video['src'],
'format_id': video['bitrate'], '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(video.attrib['system-bitrate']) / 1000
formats.append({
'url': '%s/%s%s' % (http_base, video.attrib['src'], URL_SUFFIX),
'format_id': '%dk' % vbr,
'vbr': vbr,
})
self._sort_formats(formats) self._sort_formats(formats)
return { return {
'id': video_id, 'id': video_id,
'display_id': display_id, 'display_id': display_id,
'title': self._extract_clip_info('clip_title', clip_info, 'title'), 'title': extract_clip_info('clip_title', clip_info, 'title'),
'description': self._extract_clip_info('caption', clip_info, 'description', fatal=False), 'description': extract_clip_info('caption', clip_info, 'description', fatal=False),
'thumbnail': self._extract_clip_info('video_still_url', clip_info, 'thumbnail'), 'thumbnail': extract_clip_info('video_still_url', clip_info, 'thumbnail'),
'duration': self._extract_clip_info('duration', clip_info), 'duration': extract_clip_info('duration', clip_info),
'formats': formats, 'formats': formats,
} }