[sendtonews] Fix extraction

master
Yen Chi Hsuan 8 years ago
parent b6c4e36728
commit 5c2d087221
No known key found for this signature in database
GPG Key ID: 3FDDD575826C5C30

@ -4,33 +4,43 @@ from __future__ import unicode_literals
import re import re
from .jwplatform import JWPlatformBaseIE from .jwplatform import JWPlatformBaseIE
from ..compat import compat_parse_qs
from ..utils import ( from ..utils import (
ExtractorError, float_or_none,
parse_duration, parse_iso8601,
update_url_query,
) )
class SendtoNewsIE(JWPlatformBaseIE): class SendtoNewsIE(JWPlatformBaseIE):
_VALID_URL = r'https?://embed\.sendtonews\.com/player/embed\.php\?(?P<query>[^#]+)' _VALID_URL = r'https?://embed\.sendtonews\.com/player2/embedplayer\.php\?.*\bSC=(?P<id>[0-9A-Za-z-]+)'
_TEST = { _TEST = {
# From http://cleveland.cbslocal.com/2016/05/16/indians-score-season-high-15-runs-in-blowout-win-over-reds-rapid-reaction/ # From http://cleveland.cbslocal.com/2016/05/16/indians-score-season-high-15-runs-in-blowout-win-over-reds-rapid-reaction/
'url': 'http://embed.sendtonews.com/player/embed.php?SK=GxfCe0Zo7D&MK=175909&PK=5588&autoplay=on&sound=yes', 'url': 'http://embed.sendtonews.com/player2/embedplayer.php?SC=GxfCe0Zo7D-175909-5588&type=single&autoplay=on&sound=YES',
'info_dict': { 'info_dict': {
'id': 'GxfCe0Zo7D-175909-5588', 'id': 'GxfCe0Zo7D-175909-5588'
'ext': 'mp4',
'title': 'Recap: CLE 15, CIN 6',
'description': '5/16/16: Indians\' bats explode for 15 runs in a win',
'duration': 49,
}, },
'playlist_count': 9,
# test the first video only to prevent lengthy tests
'playlist': [{
'info_dict': {
'id': '198180',
'ext': 'mp4',
'title': 'Recap: CLE 5, LAA 4',
'description': '8/14/16: Naquin, Almonte lead Indians in 5-4 win',
'duration': 57.343,
'thumbnail': 're:https?://.*\.jpg$',
'upload_date': '20160815',
'timestamp': 1471221961,
},
}],
'params': { 'params': {
# m3u8 download # m3u8 download
'skip_download': True, 'skip_download': True,
}, },
} }
_URL_TEMPLATE = '//embed.sendtonews.com/player/embed.php?SK=%s&MK=%s&PK=%s' _URL_TEMPLATE = '//embed.sendtonews.com/player2/embedplayer.php?SC=%s'
@classmethod @classmethod
def _extract_url(cls, webpage): def _extract_url(cls, webpage):
@ -39,48 +49,41 @@ class SendtoNewsIE(JWPlatformBaseIE):
.*\bSC=(?P<SC>[0-9a-zA-Z-]+).* .*\bSC=(?P<SC>[0-9a-zA-Z-]+).*
\1>''', webpage) \1>''', webpage)
if mobj: if mobj:
sk, mk, pk = mobj.group('SC').split('-') sc = mobj.group('SC')
return cls._URL_TEMPLATE % (sk, mk, pk) return cls._URL_TEMPLATE % sc
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) playlist_id = self._match_id(url)
params = compat_parse_qs(mobj.group('query'))
data_url = update_url_query(
if 'SK' not in params or 'MK' not in params or 'PK' not in params: url.replace('embedplayer.php', 'data_read.php'),
raise ExtractorError('Invalid URL', expected=True) {'cmd': 'loadInitial'})
playlist_data = self._download_json(data_url, playlist_id)
video_id = '-'.join([params['SK'][0], params['MK'][0], params['PK'][0]])
entries = []
webpage = self._download_webpage(url, video_id) for video in playlist_data['playlistData'][0]:
info_dict = self._parse_jwplayer_data(
jwplayer_data_str = self._search_regex( video['jwconfiguration'],
r'jwplayer\("[^"]+"\)\.setup\((.+?)\);', webpage, 'JWPlayer data') require_title=False, rtmp_params={'no_resume': True})
js_vars = {
'w': 1024, thumbnails = []
'h': 768, if video.get('thumbnailUrl'):
'modeVar': 'html5', thumbnails.append({
} 'id': 'normal',
for name, val in js_vars.items(): 'url': video['thumbnailUrl'],
js_val = '%d' % val if isinstance(val, int) else '"%s"' % val })
jwplayer_data_str = jwplayer_data_str.replace(':%s,' % name, ':%s,' % js_val) if video.get('smThumbnailUrl'):
thumbnails.append({
info_dict = self._parse_jwplayer_data( 'id': 'small',
self._parse_json(jwplayer_data_str, video_id), 'url': video['smThumbnailUrl'],
video_id, require_title=False, rtmp_params={'no_resume': True}) })
info_dict.update({
title = self._html_search_regex( 'title': video['S_headLine'],
r'<div[^>]+class="embedTitle">([^<]+)</div>', webpage, 'title') 'description': video.get('S_fullStory'),
description = self._html_search_regex( 'thumbnails': thumbnails,
r'<div[^>]+class="embedSubTitle">([^<]+)</div>', webpage, 'duration': float_or_none(video.get('SM_length')),
'description', fatal=False) 'timestamp': parse_iso8601(video.get('S_sysDate'), delimiter=' '),
duration = parse_duration(self._html_search_regex( })
r'<div[^>]+class="embedDetails">([0-9:]+)', webpage, entries.append(info_dict)
'duration', fatal=False))
return self.playlist_result(entries, playlist_id)
info_dict.update({
'title': title,
'description': description,
'duration': duration,
})
return info_dict

Loading…
Cancel
Save