]>
Commit | Line | Data |
---|---|---|
974208e1 | 1 | import itertools |
a820dc72 | 2 | import json |
6ef5ad9e | 3 | import random |
4 | import string | |
a820dc72 RA |
5 | |
6 | from .common import InfoExtractor | |
7 | from ..utils import ( | |
8 | ExtractorError, | |
e0ddbd02 | 9 | format_field, |
a820dc72 RA |
10 | int_or_none, |
11 | str_or_none, | |
12 | try_get, | |
13 | ) | |
14 | ||
15 | ||
16 | class TrovoBaseIE(InfoExtractor): | |
17 | _VALID_URL_BASE = r'https?://(?:www\.)?trovo\.live/' | |
36147a63 | 18 | _HEADERS = {'Origin': 'https://trovo.live'} |
a820dc72 | 19 | |
6ef5ad9e | 20 | def _call_api(self, video_id, data): |
21 | if 'persistedQuery' in data.get('extensions', {}): | |
22 | url = 'https://gql.trovo.live' | |
23 | else: | |
24 | url = 'https://api-web.trovo.live/graphql' | |
25 | ||
26 | resp = self._download_json( | |
27 | url, video_id, data=json.dumps([data]).encode(), headers={'Accept': 'application/json'}, | |
28 | query={ | |
29 | 'qid': ''.join(random.choices(string.ascii_uppercase + string.digits, k=10)), | |
30 | })[0] | |
31 | if 'errors' in resp: | |
32 | raise ExtractorError(f'Trovo said: {resp["errors"][0]["message"]}') | |
33 | return resp['data'][data['operationName']] | |
0cbed930 | 34 | |
a820dc72 RA |
35 | def _extract_streamer_info(self, data): |
36 | streamer_info = data.get('streamerInfo') or {} | |
37 | username = streamer_info.get('userName') | |
38 | return { | |
39 | 'uploader': streamer_info.get('nickName'), | |
40 | 'uploader_id': str_or_none(streamer_info.get('uid')), | |
a70635b8 | 41 | 'uploader_url': format_field(username, None, 'https://trovo.live/%s'), |
a820dc72 RA |
42 | } |
43 | ||
44 | ||
45 | class TrovoIE(TrovoBaseIE): | |
46 | _VALID_URL = TrovoBaseIE._VALID_URL_BASE + r'(?!(?:clip|video)/)(?P<id>[^/?&#]+)' | |
47 | ||
48 | def _real_extract(self, url): | |
49 | username = self._match_id(url) | |
6ef5ad9e | 50 | live_info = self._call_api(username, data={ |
51 | 'operationName': 'live_LiveReaderService_GetLiveInfo', | |
52 | 'variables': { | |
53 | 'params': { | |
54 | 'userName': username, | |
55 | }, | |
56 | }, | |
57 | }) | |
a820dc72 RA |
58 | if live_info.get('isLive') == 0: |
59 | raise ExtractorError('%s is offline' % username, expected=True) | |
60 | program_info = live_info['programInfo'] | |
61 | program_id = program_info['id'] | |
39ca3b5c | 62 | title = program_info['title'] |
a820dc72 RA |
63 | |
64 | formats = [] | |
65 | for stream_info in (program_info.get('streamInfo') or []): | |
66 | play_url = stream_info.get('playUrl') | |
67 | if not play_url: | |
68 | continue | |
69 | format_id = stream_info.get('desc') | |
70 | formats.append({ | |
71 | 'format_id': format_id, | |
72 | 'height': int_or_none(format_id[:-1]) if format_id else None, | |
73 | 'url': play_url, | |
36147a63 | 74 | 'http_headers': self._HEADERS, |
a820dc72 RA |
75 | }) |
76 | self._sort_formats(formats) | |
77 | ||
78 | info = { | |
79 | 'id': program_id, | |
80 | 'title': title, | |
81 | 'formats': formats, | |
82 | 'thumbnail': program_info.get('coverUrl'), | |
83 | 'is_live': True, | |
84 | } | |
85 | info.update(self._extract_streamer_info(live_info)) | |
86 | return info | |
87 | ||
88 | ||
89 | class TrovoVodIE(TrovoBaseIE): | |
90 | _VALID_URL = TrovoBaseIE._VALID_URL_BASE + r'(?:clip|video)/(?P<id>[^/?&#]+)' | |
91 | _TESTS = [{ | |
6ef5ad9e | 92 | 'url': 'https://trovo.live/clip/lc-5285890818705062210?ltab=videos', |
93 | 'params': {'getcomments': True}, | |
a820dc72 | 94 | 'info_dict': { |
6ef5ad9e | 95 | 'id': 'lc-5285890818705062210', |
a820dc72 | 96 | 'ext': 'mp4', |
6ef5ad9e | 97 | 'title': 'fatal moaning for a super good🤣🤣', |
98 | 'uploader': 'OneTappedYou', | |
99 | 'timestamp': 1621628019, | |
100 | 'upload_date': '20210521', | |
101 | 'uploader_id': '100719456', | |
102 | 'duration': 31, | |
a820dc72 RA |
103 | 'view_count': int, |
104 | 'like_count': int, | |
105 | 'comment_count': int, | |
6ef5ad9e | 106 | 'comments': 'mincount:1', |
107 | 'categories': ['Call of Duty: Mobile'], | |
108 | 'uploader_url': 'https://trovo.live/OneTappedYou', | |
109 | 'thumbnail': r're:^https?://.*\.jpg', | |
a820dc72 RA |
110 | }, |
111 | }, { | |
6ef5ad9e | 112 | 'url': 'https://trovo.live/video/ltv-100095501_100095501_1609596043', |
a820dc72 RA |
113 | 'only_matching': True, |
114 | }] | |
115 | ||
116 | def _real_extract(self, url): | |
117 | vid = self._match_id(url) | |
6ef5ad9e | 118 | |
119 | # NOTE: It is also possible to extract this info from the Nuxt data on the website, | |
120 | # however that seems unreliable - sometimes it randomly doesn't return the data, | |
121 | # at least when using a non-residential IP. | |
122 | resp = self._call_api(vid, data={ | |
123 | 'operationName': 'batchGetVodDetailInfo', | |
124 | 'variables': { | |
125 | 'params': { | |
126 | 'vids': [vid], | |
127 | }, | |
128 | }, | |
129 | 'extensions': { | |
130 | 'persistedQuery': { | |
131 | 'version': 1, | |
132 | 'sha256Hash': 'ceae0355d66476e21a1dd8e8af9f68de95b4019da2cda8b177c9a2255dad31d0', | |
133 | }, | |
134 | }, | |
135 | }) | |
136 | vod_detail_info = resp['VodDetailInfos'][vid] | |
a820dc72 RA |
137 | vod_info = vod_detail_info['vodInfo'] |
138 | title = vod_info['title'] | |
139 | ||
6ef5ad9e | 140 | if try_get(vod_info, lambda x: x['playbackRights']['playbackRights'] != 'Normal'): |
141 | playback_rights_setting = vod_info['playbackRights']['playbackRightsSetting'] | |
142 | if playback_rights_setting == 'SubscriberOnly': | |
143 | raise ExtractorError('This video is only available for subscribers', expected=True) | |
144 | else: | |
145 | raise ExtractorError(f'This video is not available ({playback_rights_setting})', expected=True) | |
146 | ||
a820dc72 RA |
147 | language = vod_info.get('languageName') |
148 | formats = [] | |
149 | for play_info in (vod_info.get('playInfos') or []): | |
150 | play_url = play_info.get('playUrl') | |
151 | if not play_url: | |
152 | continue | |
153 | format_id = play_info.get('desc') | |
154 | formats.append({ | |
155 | 'ext': 'mp4', | |
156 | 'filesize': int_or_none(play_info.get('fileSize')), | |
157 | 'format_id': format_id, | |
158 | 'height': int_or_none(format_id[:-1]) if format_id else None, | |
159 | 'language': language, | |
160 | 'protocol': 'm3u8_native', | |
161 | 'tbr': int_or_none(play_info.get('bitrate')), | |
162 | 'url': play_url, | |
36147a63 | 163 | 'http_headers': self._HEADERS, |
a820dc72 RA |
164 | }) |
165 | self._sort_formats(formats) | |
166 | ||
167 | category = vod_info.get('categoryName') | |
168 | get_count = lambda x: int_or_none(vod_info.get(x + 'Num')) | |
169 | ||
a820dc72 RA |
170 | info = { |
171 | 'id': vid, | |
172 | 'title': title, | |
173 | 'formats': formats, | |
174 | 'thumbnail': vod_info.get('coverUrl'), | |
175 | 'timestamp': int_or_none(vod_info.get('publishTs')), | |
176 | 'duration': int_or_none(vod_info.get('duration')), | |
177 | 'view_count': get_count('watch'), | |
178 | 'like_count': get_count('like'), | |
179 | 'comment_count': get_count('comment'), | |
a820dc72 | 180 | 'categories': [category] if category else None, |
6ef5ad9e | 181 | '__post_extractor': self.extract_comments(vid), |
a820dc72 RA |
182 | } |
183 | info.update(self._extract_streamer_info(vod_detail_info)) | |
184 | return info | |
974208e1 | 185 | |
6ef5ad9e | 186 | def _get_comments(self, vid): |
187 | for page in itertools.count(1): | |
188 | comments_json = self._call_api(vid, data={ | |
189 | 'operationName': 'getCommentList', | |
190 | 'variables': { | |
191 | 'params': { | |
192 | 'appInfo': { | |
193 | 'postID': vid, | |
194 | }, | |
195 | 'preview': {}, | |
196 | 'pageSize': 99, | |
197 | 'page': page, | |
198 | }, | |
199 | }, | |
200 | 'extensions': { | |
201 | 'persistedQuery': { | |
202 | 'version': 1, | |
203 | 'sha256Hash': 'be8e5f9522ddac7f7c604c0d284fd22481813263580849926c4c66fb767eed25', | |
204 | }, | |
205 | }, | |
206 | }) | |
207 | for comment in comments_json['commentList']: | |
208 | content = comment.get('content') | |
209 | if not content: | |
210 | continue | |
211 | author = comment.get('author') or {} | |
212 | parent = comment.get('parentID') | |
213 | yield { | |
214 | 'author': author.get('nickName'), | |
215 | 'author_id': str_or_none(author.get('uid')), | |
216 | 'id': str_or_none(comment.get('commentID')), | |
217 | 'text': content, | |
218 | 'timestamp': int_or_none(comment.get('createdAt')), | |
219 | 'parent': 'root' if parent == 0 else str_or_none(parent), | |
220 | } | |
221 | ||
222 | if comments_json['lastPage']: | |
223 | break | |
224 | ||
974208e1 | 225 | |
3262f8ab | 226 | class TrovoChannelBaseIE(TrovoBaseIE): |
974208e1 AG |
227 | def _get_vod_json(self, page, uid): |
228 | raise NotImplementedError('This method must be implemented by subclasses') | |
229 | ||
230 | def _entries(self, uid): | |
231 | for page in itertools.count(1): | |
232 | vod_json = self._get_vod_json(page, uid) | |
233 | vods = vod_json.get('vodInfos', []) | |
234 | for vod in vods: | |
235 | yield self.url_result( | |
236 | 'https://trovo.live/%s/%s' % (self._TYPE, vod.get('vid')), | |
237 | ie=TrovoVodIE.ie_key()) | |
238 | has_more = vod_json['hasMore'] | |
239 | if not has_more: | |
240 | break | |
241 | ||
242 | def _real_extract(self, url): | |
243 | id = self._match_id(url) | |
6ef5ad9e | 244 | live_info = self._call_api(id, data={ |
245 | 'operationName': 'live_LiveReaderService_GetLiveInfo', | |
246 | 'variables': { | |
247 | 'params': { | |
248 | 'userName': id, | |
249 | }, | |
250 | }, | |
251 | }) | |
252 | uid = str(live_info['streamerInfo']['uid']) | |
974208e1 AG |
253 | return self.playlist_result(self._entries(uid), playlist_id=uid) |
254 | ||
255 | ||
256 | class TrovoChannelVodIE(TrovoChannelBaseIE): | |
257 | _VALID_URL = r'trovovod:(?P<id>[^\s]+)' | |
96565c7e | 258 | IE_DESC = 'All VODs of a trovo.live channel; "trovovod:" prefix' |
974208e1 AG |
259 | |
260 | _TESTS = [{ | |
261 | 'url': 'trovovod:OneTappedYou', | |
262 | 'playlist_mincount': 24, | |
263 | 'info_dict': { | |
264 | 'id': '100719456', | |
265 | }, | |
266 | }] | |
267 | ||
974208e1 AG |
268 | _TYPE = 'video' |
269 | ||
270 | def _get_vod_json(self, page, uid): | |
6ef5ad9e | 271 | return self._call_api(uid, data={ |
272 | 'operationName': 'getChannelLtvVideoInfos', | |
273 | 'variables': { | |
274 | 'params': { | |
275 | 'channelID': int(uid), | |
276 | 'pageSize': 99, | |
277 | 'currPage': page, | |
278 | }, | |
279 | }, | |
280 | 'extensions': { | |
281 | 'persistedQuery': { | |
282 | 'version': 1, | |
283 | 'sha256Hash': '78fe32792005eab7e922cafcdad9c56bed8bbc5f5df3c7cd24fcb84a744f5f78', | |
284 | }, | |
285 | }, | |
286 | }) | |
974208e1 AG |
287 | |
288 | ||
289 | class TrovoChannelClipIE(TrovoChannelBaseIE): | |
290 | _VALID_URL = r'trovoclip:(?P<id>[^\s]+)' | |
96565c7e | 291 | IE_DESC = 'All Clips of a trovo.live channel; "trovoclip:" prefix' |
974208e1 AG |
292 | |
293 | _TESTS = [{ | |
294 | 'url': 'trovoclip:OneTappedYou', | |
295 | 'playlist_mincount': 29, | |
296 | 'info_dict': { | |
297 | 'id': '100719456', | |
298 | }, | |
299 | }] | |
300 | ||
974208e1 AG |
301 | _TYPE = 'clip' |
302 | ||
303 | def _get_vod_json(self, page, uid): | |
6ef5ad9e | 304 | return self._call_api(uid, data={ |
305 | 'operationName': 'getChannelClipVideoInfos', | |
306 | 'variables': { | |
307 | 'params': { | |
308 | 'channelID': int(uid), | |
309 | 'pageSize': 99, | |
310 | 'currPage': page, | |
311 | }, | |
312 | }, | |
313 | 'extensions': { | |
314 | 'persistedQuery': { | |
315 | 'version': 1, | |
316 | 'sha256Hash': 'e7924bfe20059b5c75fc8ff9e7929f43635681a7bdf3befa01072ed22c8eff31', | |
317 | }, | |
318 | }, | |
319 | }) |