]>
Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/sina.py
2 from __future__
import unicode_literals
6 from .common
import InfoExtractor
13 get_element_by_attribute
,
18 class SinaIE(InfoExtractor
):
19 _VALID_URL
= r
'''(?x)https?://(?:.*?\.)?video\.sina\.com\.cn/
21 (?:view/|.*\#)(?P<video_id>\d+)|
22 .+?/(?P<pseudo_id>[^/?#]+)(?:\.s?html)|
23 # This is used by external sites like Weibo
24 api/sinawebApi/outplay.php/(?P<token>.+?)\.swf
30 'url': 'http://video.sina.com.cn/news/spj/topvideoes20160504/?opsubject_id=top1#250576622',
31 'md5': 'd38433e2fc886007729735650ae4b3e9',
35 'title': '现场:克鲁兹宣布退选 特朗普将稳获提名',
39 'url': 'http://video.sina.com.cn/v/b/101314253-1290078633.html',
43 'title': '军方提高对朝情报监视级别',
45 'skip': 'the page does not exist or has been deleted',
48 'url': 'http://video.sina.com.cn/view/250587748.html',
49 'md5': '3d1807a25c775092aab3bc157fff49b4',
53 'title': '瞬间泪目:8年前汶川地震珍贵视频首曝光',
58 def _real_extract(self
, url
):
59 mobj
= re
.match(self
._VALID
_URL
, url
)
61 video_id
= mobj
.group('video_id')
63 if mobj
.group('token') is not None:
64 # The video id is in the redirected url
65 self
.to_screen('Getting video id')
66 request
= HEADRequest(url
)
67 _
, urlh
= self
._download
_webpage
_handle
(request
, 'NA', False)
68 return self
._real
_extract
(urlh
.geturl())
70 pseudo_id
= mobj
.group('pseudo_id')
71 webpage
= self
._download
_webpage
(url
, pseudo_id
)
72 error
= get_element_by_attribute('class', 'errtitle', webpage
)
74 raise ExtractorError('%s said: %s' % (
75 self
.IE_NAME
, clean_html(error
)), expected
=True)
76 video_id
= self
._search
_regex
(
77 r
"video_id\s*:\s*'(\d+)'", webpage
, 'video id')
79 video_data
= self
._download
_json
(
80 'http://s.video.sina.com.cn/video/h5play',
81 video_id
, query
={'video_id': video_id
})
82 if video_data
['code'] != 1:
83 raise ExtractorError('%s said: %s' % (
84 self
.IE_NAME
, video_data
['message']), expected
=True)
86 video_data
= video_data
['data']
87 title
= video_data
['title']
88 description
= video_data
.get('description')
90 description
= description
.strip()
92 preference
= qualities(['cif', 'sd', 'hd', 'fhd', 'ffd'])
94 for quality_id
, quality
in video_data
.get('videos', {}).get('mp4', {}).items():
95 file_api
= quality
.get('file_api')
96 file_id
= quality
.get('file_id')
97 if not file_api
or not file_id
:
100 'format_id': quality_id
,
101 'url': update_url_query(file_api
, {'vid': file_id
}),
102 'preference': preference(quality_id
),
105 self
._sort
_formats
(formats
)
110 'description': description
,
111 'thumbnail': video_data
.get('image'),
112 'duration': int_or_none(video_data
.get('length')),
113 'timestamp': int_or_none(video_data
.get('create_time')),