yt-dlp/yt_dlp/extractor/sohu.py

290 lines
11 KiB
Python
Raw Normal View History

import base64
2013-08-28 07:59:08 -04:00
import re
import urllib.parse
2013-08-02 05:58:46 -04:00
from .common import InfoExtractor
2017-06-08 13:16:42 -04:00
from ..utils import (
ExtractorError,
float_or_none,
int_or_none,
traverse_obj,
2017-06-08 13:16:42 -04:00
try_get,
unified_timestamp,
url_or_none,
urljoin,
2017-06-08 13:16:42 -04:00
)
2013-08-02 05:58:46 -04:00
class SohuIE(InfoExtractor):
_VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
2013-08-02 05:58:46 -04:00
2016-08-08 06:48:21 -04:00
# Sohu videos give different MD5 sums on Travis CI and my machine
_TESTS = [{
'note': 'This video is available only in Mainland China',
2014-11-26 06:53:55 -05:00
'url': 'http://tv.sohu.com/20130724/n382479172.shtml#super',
'info_dict': {
'id': '382479172',
'ext': 'mp4',
'title': 'MVFar East Movement《The Illest》',
2013-08-02 05:58:46 -04:00
},
'skip': 'On available in China',
}, {
'url': 'http://tv.sohu.com/20150305/n409385080.shtml',
'info_dict': {
'id': '409385080',
'ext': 'mp4',
'title': '《2015湖南卫视羊年元宵晚会》唐嫣《花好月圆》',
},
'skip': 'no longer available',
}, {
'url': 'http://my.tv.sohu.com/us/232799889/78693464.shtml',
'info_dict': {
'id': '78693464',
'ext': 'mp4',
'title': '【爱范品】第31期MWC见不到的奇葩手机',
'uploader': '爱范儿视频',
'duration': 213,
'timestamp': 1425519600,
'upload_date': '20150305',
'thumbnail': 'http://e3f49eaa46b57.cdn.sohucs.com//group1/M10/83/FA/MTAuMTAuODguODA=/6_14cbccdde5eg104SysCutcloud_78693464_7_0b.jpg',
'tags': ['爱范儿', '爱范品', 'MWC', '手机'],
},
}, {
'note': 'Multipart video',
'url': 'http://my.tv.sohu.com/pl/8384802/78910339.shtml',
'info_dict': {
'id': '78910339',
2015-04-19 07:19:44 -04:00
'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
'uploader': '小苍cany',
'duration': 744.0,
'timestamp': 1426269360,
'upload_date': '20150313',
'thumbnail': 'http://e3f49eaa46b57.cdn.sohucs.com//group1/M11/89/57/MTAuMTAuODguODA=/6_14cea022a1dg102SysCutcloud_78910339_8_0b.jpg',
'tags': ['小苍MM', '英雄联盟', '实战秘籍'],
},
'playlist': [{
'info_dict': {
'id': '78910339_part1',
'ext': 'mp4',
'duration': 294,
'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
},
}, {
'info_dict': {
'id': '78910339_part2',
'ext': 'mp4',
'duration': 300,
'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
},
}, {
'info_dict': {
'id': '78910339_part3',
'ext': 'mp4',
'duration': 150,
'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
},
}],
2015-03-14 13:05:01 -04:00
}, {
2015-03-17 11:39:31 -04:00
'note': 'Video with title containing dash',
2015-03-14 13:05:01 -04:00
'url': 'http://my.tv.sohu.com/us/249884221/78932792.shtml',
'info_dict': {
'id': '78932792',
'ext': 'mp4',
'title': 'youtube-dl testing video',
'duration': 360,
'timestamp': 1426348620,
'upload_date': '20150314',
'thumbnail': 'http://e3f49eaa46b57.cdn.sohucs.com//group1/M02/8A/00/MTAuMTAuODguNzk=/6_14cee1be192g102SysCutcloud_78932792_7_7b.jpg',
'tags': [],
2015-03-14 13:05:01 -04:00
},
'params': {
'skip_download': True,
},
}]
2013-08-02 05:58:46 -04:00
def _real_extract(self, url):
2013-08-28 07:59:08 -04:00
def _fetch_data(vid_id, mytv=False):
if mytv:
base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
else:
2014-11-26 06:53:55 -05:00
base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid='
2015-03-17 11:39:31 -04:00
return self._download_json(
base_data_url + vid_id, video_id,
f'Downloading JSON data for {vid_id}',
headers=self.geo_verification_headers())
2013-08-28 07:59:08 -04:00
mobj = self._match_valid_url(url)
2013-08-02 05:58:46 -04:00
video_id = mobj.group('id')
mytv = mobj.group('mytv') is not None
2013-08-28 07:59:08 -04:00
2013-08-02 05:58:46 -04:00
webpage = self._download_webpage(url, video_id)
2015-03-14 13:05:01 -04:00
title = re.sub(r'( - 高清正版在线观看)? - 搜狐视频$', '', self._og_search_title(webpage))
2013-08-02 05:58:46 -04:00
vid = self._html_search_regex(
r'var vid ?= ?["\'](\d+)["\']',
webpage, 'video path')
vid_data = _fetch_data(vid, mytv)
2015-05-09 01:19:54 -04:00
if vid_data['play'] != 1:
if vid_data.get('status') == 12:
raise ExtractorError(
f'{self.IE_NAME} said: There\'s something wrong in the video.',
2015-05-09 01:19:54 -04:00
expected=True)
else:
2017-02-23 05:49:35 -05:00
self.raise_geo_restricted(
f'{self.IE_NAME} said: The video is only licensed to users in Mainland China.')
2013-08-28 07:59:08 -04:00
formats_json = {}
for format_id in ('nor', 'high', 'super', 'ori', 'h2644k', 'h2654k'):
vid_id = vid_data['data'].get(f'{format_id}Vid')
if not vid_id:
continue
vid_id = str(vid_id)
formats_json[format_id] = vid_data if vid == vid_id else _fetch_data(vid_id, mytv)
2013-08-28 07:59:08 -04:00
part_count = vid_data['data']['totalBlocks']
2013-08-28 07:59:08 -04:00
playlist = []
for i in range(part_count):
formats = []
for format_id, format_data in formats_json.items():
allot = format_data['allot']
data = format_data['data']
clip_url = traverse_obj(data, (('clipsURL', 'mp4PlayUrl'), i, {url_or_none}), get_all=False)
if not clip_url:
raise ExtractorError(f'Unable to extract url for clip {i}')
su = data['su']
2015-06-21 12:59:55 -04:00
video_url = 'newflv.sohu.ccgslb.net'
cdn_id = None
2015-06-21 12:59:55 -04:00
retries = 0
2015-06-21 12:59:55 -04:00
while 'newflv.sohu.ccgslb.net' in video_url:
params = {
'prot': 9,
'file': clip_url,
2015-06-21 12:59:55 -04:00
'new': su[i],
'prod': 'h5n',
2015-12-08 01:12:32 -05:00
'rb': 1,
2015-06-21 12:59:55 -04:00
}
if cdn_id is not None:
params['idc'] = cdn_id
2015-06-21 12:59:55 -04:00
download_note = f'Downloading {format_id} video URL part {i + 1} of {part_count}'
2015-06-21 12:59:55 -04:00
if retries > 0:
download_note += f' (retry #{retries})'
2015-06-21 12:59:55 -04:00
part_info = self._parse_json(self._download_webpage(
f'http://{allot}/?{urllib.parse.urlencode(params)}',
2015-06-21 12:59:55 -04:00
video_id, download_note), video_id)
video_url = part_info['url']
cdn_id = part_info.get('nid')
2015-06-21 12:59:55 -04:00
retries += 1
if retries > 5:
raise ExtractorError('Failed to get video URL')
formats.append({
'url': video_url,
'format_id': format_id,
2017-06-08 13:16:42 -04:00
'filesize': int_or_none(
try_get(data, lambda x: x['clipsBytes'][i])),
'width': int_or_none(data.get('width')),
'height': int_or_none(data.get('height')),
'fps': int_or_none(data.get('fps')),
})
playlist.append({
'id': f'{video_id}_part{i + 1}',
2013-08-02 05:58:46 -04:00
'title': title,
'duration': vid_data['data']['clipsDuration'][i],
'formats': formats,
})
2013-08-28 07:59:08 -04:00
if len(playlist) == 1:
info = playlist[0]
2013-08-02 22:29:58 -04:00
info['id'] = video_id
2013-08-28 07:59:08 -04:00
else:
info = {
2015-04-19 07:19:44 -04:00
'_type': 'multi_video',
2013-08-28 07:59:08 -04:00
'entries': playlist,
'id': video_id,
2015-04-19 07:19:44 -04:00
'title': title,
'duration': traverse_obj(vid_data, ('data', 'totalDuration', {float_or_none})),
2013-08-28 07:59:08 -04:00
}
if mytv:
publish_time = unified_timestamp(self._search_regex(
r'publishTime:\s*["\'](\d+-\d+-\d+ \d+:\d+)["\']', webpage, 'publish time', fatal=False))
else:
publish_time = traverse_obj(vid_data, ('tv_application_time', {unified_timestamp}))
return {
'timestamp': publish_time - 8 * 3600 if publish_time else None,
**traverse_obj(vid_data, {
'alt_title': ('data', 'subName', {str}),
'uploader': ('wm_data', 'wm_username', {str}),
'thumbnail': ('data', 'coverImg', {url_or_none}),
'tags': ('data', 'tag', {str.split}),
}),
**info,
}
class SohuVIE(InfoExtractor):
_VALID_URL = r'https?://tv\.sohu\.com/v/(?P<id>[\w=-]+)\.html(?:$|[#?])'
_TESTS = [{
'note': 'Multipart video',
'url': 'https://tv.sohu.com/v/MjAyMzA2MTQvbjYwMTMxNTE5Mi5zaHRtbA==.html',
'info_dict': {
'id': '601315192',
'title': '《淬火丹心》第1集',
'alt_title': '“点天灯”发生事故',
'duration': 2701.692,
'timestamp': 1686758040,
'upload_date': '20230614',
'thumbnail': 'http://photocdn.tv.sohu.com/img/20230614/vrsa_hor_1686738763256_454010551.jpg',
},
'playlist_mincount': 9,
'skip': 'Only available in China',
}, {
'url': 'https://tv.sohu.com/v/dXMvMjMyNzk5ODg5Lzc4NjkzNDY0LnNodG1s.html',
'info_dict': {
'id': '78693464',
'ext': 'mp4',
'title': '【爱范品】第31期MWC见不到的奇葩手机',
'uploader': '爱范儿视频',
'duration': 213,
'timestamp': 1425519600,
'upload_date': '20150305',
'thumbnail': 'http://e3f49eaa46b57.cdn.sohucs.com//group1/M10/83/FA/MTAuMTAuODguODA=/6_14cbccdde5eg104SysCutcloud_78693464_7_0b.jpg',
'tags': ['爱范儿', '爱范品', 'MWC', '手机'],
},
}, {
'note': 'Multipart video',
'url': 'https://tv.sohu.com/v/dXMvMjQyNTYyMTYzLzc4OTEwMzM5LnNodG1s.html?src=pl',
'info_dict': {
'id': '78910339',
'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
'uploader': '小苍cany',
'duration': 744.0,
'timestamp': 1426269360,
'upload_date': '20150313',
'thumbnail': 'http://e3f49eaa46b57.cdn.sohucs.com//group1/M11/89/57/MTAuMTAuODguODA=/6_14cea022a1dg102SysCutcloud_78910339_8_0b.jpg',
'tags': ['小苍MM', '英雄联盟', '实战秘籍'],
},
'playlist_mincount': 3,
}]
def _real_extract(self, url):
encoded_id = self._match_id(url)
path = base64.urlsafe_b64decode(encoded_id).decode()
subdomain = 'tv' if re.match(r'\d+/n\d+\.shtml', path) else 'my.tv'
return self.url_result(urljoin(f'http://{subdomain}.sohu.com/', path), SohuIE)