]>
jfr.im git - yt-dlp.git/blob - yt_dlp/extractor/sina.py
1 from .common
import InfoExtractor
8 get_element_by_attribute
,
13 class SinaIE(InfoExtractor
):
14 _VALID_URL
= r
'''(?x)https?://(?:.*?\.)?video\.sina\.com\.cn/
16 (?:view/|.*\#)(?P<id>\d+)|
17 .+?/(?P<pseudo_id>[^/?#]+)(?:\.s?html)|
18 # This is used by external sites like Weibo
19 api/sinawebApi/outplay.php/(?P<token>.+?)\.swf
25 'url': 'http://video.sina.com.cn/news/spj/topvideoes20160504/?opsubject_id=top1#250576622',
26 'md5': 'd38433e2fc886007729735650ae4b3e9',
30 'title': '现场:克鲁兹宣布退选 特朗普将稳获提名',
34 'url': 'http://video.sina.com.cn/v/b/101314253-1290078633.html',
38 'title': '军方提高对朝情报监视级别',
40 'skip': 'the page does not exist or has been deleted',
43 'url': 'http://video.sina.com.cn/view/250587748.html',
44 'md5': '3d1807a25c775092aab3bc157fff49b4',
48 'title': '瞬间泪目:8年前汶川地震珍贵视频首曝光',
53 def _real_extract(self
, url
):
54 mobj
= self
._match
_valid
_url
(url
)
56 video_id
= mobj
.group('id')
58 if mobj
.group('token') is not None:
59 # The video id is in the redirected url
60 self
.to_screen('Getting video id')
61 request
= HEADRequest(url
)
62 _
, urlh
= self
._download
_webpage
_handle
(request
, 'NA', False)
63 return self
._real
_extract
(urlh
.geturl())
65 pseudo_id
= mobj
.group('pseudo_id')
66 webpage
= self
._download
_webpage
(url
, pseudo_id
)
67 error
= get_element_by_attribute('class', 'errtitle', webpage
)
69 raise ExtractorError('%s said: %s' % (
70 self
.IE_NAME
, clean_html(error
)), expected
=True)
71 video_id
= self
._search
_regex
(
72 r
"video_id\s*:\s*'(\d+)'", webpage
, 'video id')
74 video_data
= self
._download
_json
(
75 'http://s.video.sina.com.cn/video/h5play',
76 video_id
, query
={'video_id': video_id}
)
77 if video_data
['code'] != 1:
78 raise ExtractorError('%s said: %s' % (
79 self
.IE_NAME
, video_data
['message']), expected
=True)
81 video_data
= video_data
['data']
82 title
= video_data
['title']
83 description
= video_data
.get('description')
85 description
= description
.strip()
87 preference
= qualities(['cif', 'sd', 'hd', 'fhd', 'ffd'])
89 for quality_id
, quality
in video_data
.get('videos', {}).get('mp4', {}
).items():
90 file_api
= quality
.get('file_api')
91 file_id
= quality
.get('file_id')
92 if not file_api
or not file_id
:
95 'format_id': quality_id
,
96 'url': update_url_query(file_api
, {'vid': file_id}
),
97 'quality': preference(quality_id
),
104 'description': description
,
105 'thumbnail': video_data
.get('image'),
106 'duration': int_or_none(video_data
.get('length')),
107 'timestamp': int_or_none(video_data
.get('create_time')),