5 from .common
import InfoExtractor
20 class WeiboBaseIE(InfoExtractor
):
21 def _update_visitor_cookies(self
, video_id
):
22 visitor_data
= self
._download
_json
(
23 'https://passport.weibo.com/visitor/genvisitor', video_id
,
24 note
='Generating first-visit guest request',
25 transform_source
=strip_jsonp
,
26 data
=urlencode_postdata({
28 'fp': '{"os":"2","browser":"Gecko57,0,0,0","fonts":"undefined","screenInfo":"1440*900*24","plugins":""}',
31 self
._download
_webpage
(
32 'https://passport.weibo.com/visitor/visitor', video_id
,
33 note
='Running first-visit callback to get guest cookies',
36 't': visitor_data
['data']['tid'],
38 'c': '%03d' % visitor_data
['data']['confidence'],
41 '_rand': random
.random(),
44 def _weibo_download_json(self
, url
, video_id
, *args
, fatal
=True, note
='Downloading JSON metadata', **kwargs
):
45 webpage
, urlh
= self
._download
_webpage
_handle
(url
, video_id
, *args
, fatal
=fatal
, note
=note
, **kwargs
)
46 if urllib
.parse
.urlparse(urlh
.url
).netloc
== 'passport.weibo.com':
47 self
._update
_visitor
_cookies
(video_id
)
48 webpage
= self
._download
_webpage
(url
, video_id
, *args
, fatal
=fatal
, note
=note
, **kwargs
)
49 return self
._parse
_json
(webpage
, video_id
, fatal
=fatal
)
51 def _extract_formats(self
, video_info
):
52 media_info
= traverse_obj(video_info
, ('page_info', 'media_info'))
53 formats
= traverse_obj(media_info
, (
54 'playback_list', lambda _
, v
: url_or_none(v
['play_info']['url']), 'play_info', {
56 'format': ('quality_desc', {str}
),
57 'format_id': ('label', {str}
),
58 'ext': ('mime', {mimetype2ext}
),
59 'tbr': ('bitrate', {int_or_none}
, {lambda x: x or None}
),
60 'vcodec': ('video_codecs', {str}
),
61 'fps': ('fps', {int_or_none}
),
62 'width': ('width', {int_or_none}
),
63 'height': ('height', {int_or_none}
),
64 'filesize': ('size', {int_or_none}
),
65 'acodec': ('audio_codecs', {str}
),
66 'asr': ('audio_sample_rate', {int_or_none}
),
67 'audio_channels': ('audio_channels', {int_or_none}
),
69 if not formats
: # fallback, should be barely used
70 for url
in set(traverse_obj(media_info
, (..., {url_or_none}
))):
71 if 'label=' in url
: # filter out non-video urls
72 format_id
, resolution
= self
._search
_regex
(
73 r
'label=(\w+)&template=(\d+x\d+)', url
, 'format info',
74 group
=(1, 2), default
=(None, None))
77 'format_id': format_id
,
78 **parse_resolution(resolution
),
79 **traverse_obj(media_info
, (
80 'video_details', lambda _
, v
: v
['label'].startswith(format_id
), {
81 'size': ('size', {int_or_none}
),
82 'tbr': ('bitrate', {int_or_none}
),
88 def _parse_video_info(self
, video_info
, video_id
=None):
91 'extractor_key': WeiboIE
.ie_key(),
92 'extractor': WeiboIE
.IE_NAME
,
93 'formats': self
._extract
_formats
(video_info
),
94 'http_headers': {'Referer': 'https://weibo.com/'}
,
95 '_old_archive_ids': [make_archive_id('WeiboMobile', video_id
)],
96 **traverse_obj(video_info
, {
97 'id': (('id', 'id_str', 'mid'), {str_or_none}
),
98 'display_id': ('mblogid', {str_or_none}
),
99 'title': ('page_info', 'media_info', ('video_title', 'kol_title', 'name'), {str}
, {lambda x: x or None}
),
100 'description': ('text_raw', {str}
),
101 'duration': ('page_info', 'media_info', 'duration', {int_or_none}
),
102 'timestamp': ('page_info', 'media_info', 'video_publish_time', {int_or_none}
),
103 'thumbnail': ('page_info', 'page_pic', {url_or_none}
),
104 'uploader': ('user', 'screen_name', {str}
),
105 'uploader_id': ('user', ('id', 'id_str'), {str_or_none}
),
106 'uploader_url': ('user', 'profile_url', {lambda x: urljoin('https://weibo.com/', x)}
),
107 'view_count': ('page_info', 'media_info', 'online_users_number', {int_or_none}
),
108 'like_count': ('attitudes_count', {int_or_none}
),
109 'repost_count': ('reposts_count', {int_or_none}
),
111 'tags': traverse_obj(video_info
, ('topic_struct', ..., 'topic_title', {str}
)) or None,
115 class WeiboIE(WeiboBaseIE
):
116 _VALID_URL
= r
'https?://(?:m\.weibo\.cn/status|(?:www\.)?weibo\.com/\d+)/(?P<id>[a-zA-Z0-9]+)'
118 'url': 'https://weibo.com/7827771738/N4xlMvjhI',
120 'id': '4910815147462302',
122 'display_id': 'N4xlMvjhI',
123 'title': '【睡前消息暑假版第一期:拉泰国一把 对中国有好处】',
124 'description': 'md5:e2637a7673980d68694ea7c43cf12a5f',
126 'timestamp': 1686312819,
127 'upload_date': '20230609',
128 'thumbnail': r
're:https://.*\.jpg',
129 'uploader': '睡前视频基地',
130 'uploader_id': '7827771738',
131 'uploader_url': 'https://weibo.com/u/7827771738',
135 'tags': ['泰国大选远进党获胜', '睡前消息', '暑期版'],
138 'url': 'https://m.weibo.cn/status/4189191225395228',
140 'id': '4189191225395228',
142 'display_id': 'FBqgOmDxO',
143 'title': '柴犬柴犬的秒拍视频',
144 'description': 'md5:80f461ab5cdae6bbdb70efbf5a1db24f',
146 'timestamp': 1514264429,
147 'upload_date': '20171226',
148 'thumbnail': r
're:https://.*\.jpg',
150 'uploader_id': '5926682210',
151 'uploader_url': 'https://weibo.com/u/5926682210',
157 'url': 'https://weibo.com/0/4224132150961381',
158 'note': 'no playback_list example',
159 'only_matching': True,
162 def _real_extract(self
, url
):
163 video_id
= self
._match
_id
(url
)
165 return self
._parse
_video
_info
(self
._weibo
_download
_json
(
166 f
'https://weibo.com/ajax/statuses/show?id={video_id}', video_id
))
169 class WeiboVideoIE(WeiboBaseIE
):
170 _VALID_URL
= r
'https?://(?:www\.)?weibo\.com/tv/show/(?P<id>\d+:\d+)'
172 'url': 'https://weibo.com/tv/show/1034:4797699866951785?from=old_pc_videoshow',
174 'id': '4797700463137878',
176 'display_id': 'LEZDodaiW',
177 'title': '呃,稍微了解了一下靡烟miya,感觉这东西也太二了',
178 'description': '呃,稍微了解了一下靡烟miya,感觉这东西也太二了 http://t.cn/A6aerGsM ',
180 'timestamp': 1659344278,
181 'upload_date': '20220801',
182 'thumbnail': r
're:https://.*\.jpg',
183 'uploader': '君子爱财陈平安',
184 'uploader_id': '3905382233',
185 'uploader_url': 'https://weibo.com/u/3905382233',
192 def _real_extract(self
, url
):
193 video_id
= self
._match
_id
(url
)
195 post_data
= f
'data={{"Component_Play_Playinfo":{{"oid":"{video_id}"}}}}'.encode()
196 video_info
= self
._weibo
_download
_json
(
197 f
'https://weibo.com/tv/api/component?page=%2Ftv%2Fshow%2F{video_id.replace(":", "%3A")}',
198 video_id
, headers
={'Referer': url}
, data
=post_data
)['data']['Component_Play_Playinfo']
199 return self
.url_result(f
'https://weibo.com/0/{video_info["mid"]}', WeiboIE
)
202 class WeiboUserIE(WeiboBaseIE
):
203 _VALID_URL
= r
'https?://(?:www\.)?weibo\.com/u/(?P<id>\d+)'
205 'url': 'https://weibo.com/u/2066652961?tabtype=video',
209 'description': '萧影殿下的全部视频',
212 'playlist_mincount': 195,
215 def _fetch_page(self
, uid
, cursor
=0, page
=1):
216 return self
._weibo
_download
_json
(
217 'https://weibo.com/ajax/profile/getWaterFallContent',
218 uid
, note
=f
'Downloading videos page {page}',
219 query
={'uid': uid, 'cursor': cursor}
)['data']
221 def _entries(self
, uid
, first_page
):
223 for page
in itertools
.count(1):
224 response
= first_page
if page
== 1 else self
._fetch
_page
(uid
, cursor
, page
)
225 for video_info
in traverse_obj(response
, ('list', ..., {dict}
)):
226 yield self
._parse
_video
_info
(video_info
)
227 cursor
= response
.get('next_cursor')
228 if (int_or_none(cursor
) or -1) < 0:
231 def _real_extract(self
, url
):
232 uid
= self
._match
_id
(url
)
233 first_page
= self
._fetch
_page
(uid
)
234 uploader
= traverse_obj(first_page
, ('list', ..., 'user', 'screen_name', {str}
), get_all
=False)
236 'title': f
'{uploader}的视频',
237 'description': f
'{uploader}的全部视频',
238 'uploader': uploader
,
239 } if uploader
else {}
241 return self
.playlist_result(self
._entries
(uid
, first_page
), uid
, **metainfo
)