]>
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): | |
660c0c4e | 46 | _VALID_URL = TrovoBaseIE._VALID_URL_BASE + r'(?:s/)?(?!(?:clip|video)/)(?P<id>(?!s/)[^/?&#]+(?![^#]+[?&]vid=))' |
47 | _TESTS = [{ | |
48 | 'url': 'https://trovo.live/Exsl', | |
49 | 'only_matching': True, | |
50 | }, { | |
51 | 'url': 'https://trovo.live/s/SkenonSLive/549759191497', | |
52 | 'only_matching': True, | |
53 | }, { | |
54 | 'url': 'https://trovo.live/s/zijo987/208251706', | |
55 | 'info_dict': { | |
56 | 'id': '104125853_104125853_1656439572', | |
57 | 'ext': 'flv', | |
58 | 'uploader_url': 'https://trovo.live/zijo987', | |
59 | 'uploader_id': '104125853', | |
60 | 'thumbnail': 'https://livecover.trovo.live/screenshot/73846_104125853_104125853-2022-06-29-04-00-22-852x480.jpg', | |
61 | 'uploader': 'zijo987', | |
62 | 'title': 'š„IGRAMO IGRICE UPADAJTEš„2500/5000 2022-06-28 22:01', | |
63 | 'live_status': 'is_live', | |
64 | }, | |
65 | 'skip': 'May not be live' | |
66 | }] | |
a820dc72 RA |
67 | |
68 | def _real_extract(self, url): | |
69 | username = self._match_id(url) | |
6ef5ad9e | 70 | live_info = self._call_api(username, data={ |
71 | 'operationName': 'live_LiveReaderService_GetLiveInfo', | |
72 | 'variables': { | |
73 | 'params': { | |
74 | 'userName': username, | |
75 | }, | |
76 | }, | |
77 | }) | |
a820dc72 RA |
78 | if live_info.get('isLive') == 0: |
79 | raise ExtractorError('%s is offline' % username, expected=True) | |
80 | program_info = live_info['programInfo'] | |
81 | program_id = program_info['id'] | |
39ca3b5c | 82 | title = program_info['title'] |
a820dc72 RA |
83 | |
84 | formats = [] | |
85 | for stream_info in (program_info.get('streamInfo') or []): | |
86 | play_url = stream_info.get('playUrl') | |
87 | if not play_url: | |
88 | continue | |
89 | format_id = stream_info.get('desc') | |
90 | formats.append({ | |
91 | 'format_id': format_id, | |
92 | 'height': int_or_none(format_id[:-1]) if format_id else None, | |
93 | 'url': play_url, | |
660c0c4e | 94 | 'tbr': stream_info.get('bitrate'), |
36147a63 | 95 | 'http_headers': self._HEADERS, |
a820dc72 RA |
96 | }) |
97 | self._sort_formats(formats) | |
98 | ||
99 | info = { | |
100 | 'id': program_id, | |
101 | 'title': title, | |
102 | 'formats': formats, | |
103 | 'thumbnail': program_info.get('coverUrl'), | |
104 | 'is_live': True, | |
105 | } | |
106 | info.update(self._extract_streamer_info(live_info)) | |
107 | return info | |
108 | ||
109 | ||
110 | class TrovoVodIE(TrovoBaseIE): | |
660c0c4e | 111 | _VALID_URL = TrovoBaseIE._VALID_URL_BASE + r'(?:clip|video|s)/(?:[^/]+/\d+[^#]*[?&]vid=)?(?P<id>(?<!/s/)[^/?&#]+)' |
a820dc72 | 112 | _TESTS = [{ |
6ef5ad9e | 113 | 'url': 'https://trovo.live/clip/lc-5285890818705062210?ltab=videos', |
114 | 'params': {'getcomments': True}, | |
a820dc72 | 115 | 'info_dict': { |
6ef5ad9e | 116 | 'id': 'lc-5285890818705062210', |
a820dc72 | 117 | 'ext': 'mp4', |
6ef5ad9e | 118 | 'title': 'fatal moaning for a super goodš¤£š¤£', |
119 | 'uploader': 'OneTappedYou', | |
120 | 'timestamp': 1621628019, | |
121 | 'upload_date': '20210521', | |
122 | 'uploader_id': '100719456', | |
123 | 'duration': 31, | |
a820dc72 RA |
124 | 'view_count': int, |
125 | 'like_count': int, | |
126 | 'comment_count': int, | |
6ef5ad9e | 127 | 'comments': 'mincount:1', |
128 | 'categories': ['Call of Duty: Mobile'], | |
129 | 'uploader_url': 'https://trovo.live/OneTappedYou', | |
130 | 'thumbnail': r're:^https?://.*\.jpg', | |
a820dc72 | 131 | }, |
660c0c4e | 132 | }, { |
133 | 'url': 'https://trovo.live/s/SkenonSLive/549759191497?vid=ltv-100829718_100829718_387702301737980280', | |
134 | 'info_dict': { | |
135 | 'id': 'ltv-100829718_100829718_387702301737980280', | |
136 | 'ext': 'mp4', | |
137 | 'timestamp': 1654909624, | |
138 | 'thumbnail': 'http://vod.trovo.live/1f09baf0vodtransger1301120758/ef9ea3f0387702301737980280/coverBySnapshot/coverBySnapshot_10_0.jpg', | |
139 | 'uploader_id': '100829718', | |
140 | 'uploader': 'SkenonSLive', | |
141 | 'title': 'Trovo u secanju, uz par modova i muzike :)', | |
142 | 'uploader_url': 'https://trovo.live/SkenonSLive', | |
143 | 'duration': 10830, | |
144 | 'view_count': int, | |
145 | 'like_count': int, | |
146 | 'upload_date': '20220611', | |
147 | 'comment_count': int, | |
148 | 'categories': ['Minecraft'], | |
149 | } | |
a820dc72 | 150 | }, { |
6ef5ad9e | 151 | 'url': 'https://trovo.live/video/ltv-100095501_100095501_1609596043', |
a820dc72 | 152 | 'only_matching': True, |
660c0c4e | 153 | }, { |
154 | 'url': 'https://trovo.live/s/SkenonSLive/549759191497?foo=bar&vid=ltv-100829718_100829718_387702301737980280', | |
155 | 'only_matching': True, | |
a820dc72 RA |
156 | }] |
157 | ||
158 | def _real_extract(self, url): | |
159 | vid = self._match_id(url) | |
6ef5ad9e | 160 | |
161 | # NOTE: It is also possible to extract this info from the Nuxt data on the website, | |
162 | # however that seems unreliable - sometimes it randomly doesn't return the data, | |
163 | # at least when using a non-residential IP. | |
164 | resp = self._call_api(vid, data={ | |
165 | 'operationName': 'batchGetVodDetailInfo', | |
166 | 'variables': { | |
167 | 'params': { | |
168 | 'vids': [vid], | |
169 | }, | |
170 | }, | |
171 | 'extensions': { | |
172 | 'persistedQuery': { | |
173 | 'version': 1, | |
174 | 'sha256Hash': 'ceae0355d66476e21a1dd8e8af9f68de95b4019da2cda8b177c9a2255dad31d0', | |
175 | }, | |
176 | }, | |
177 | }) | |
178 | vod_detail_info = resp['VodDetailInfos'][vid] | |
a820dc72 RA |
179 | vod_info = vod_detail_info['vodInfo'] |
180 | title = vod_info['title'] | |
181 | ||
6ef5ad9e | 182 | if try_get(vod_info, lambda x: x['playbackRights']['playbackRights'] != 'Normal'): |
183 | playback_rights_setting = vod_info['playbackRights']['playbackRightsSetting'] | |
184 | if playback_rights_setting == 'SubscriberOnly': | |
185 | raise ExtractorError('This video is only available for subscribers', expected=True) | |
186 | else: | |
187 | raise ExtractorError(f'This video is not available ({playback_rights_setting})', expected=True) | |
188 | ||
a820dc72 RA |
189 | language = vod_info.get('languageName') |
190 | formats = [] | |
191 | for play_info in (vod_info.get('playInfos') or []): | |
192 | play_url = play_info.get('playUrl') | |
193 | if not play_url: | |
194 | continue | |
195 | format_id = play_info.get('desc') | |
196 | formats.append({ | |
197 | 'ext': 'mp4', | |
198 | 'filesize': int_or_none(play_info.get('fileSize')), | |
199 | 'format_id': format_id, | |
200 | 'height': int_or_none(format_id[:-1]) if format_id else None, | |
201 | 'language': language, | |
202 | 'protocol': 'm3u8_native', | |
203 | 'tbr': int_or_none(play_info.get('bitrate')), | |
204 | 'url': play_url, | |
36147a63 | 205 | 'http_headers': self._HEADERS, |
a820dc72 RA |
206 | }) |
207 | self._sort_formats(formats) | |
208 | ||
209 | category = vod_info.get('categoryName') | |
210 | get_count = lambda x: int_or_none(vod_info.get(x + 'Num')) | |
211 | ||
a820dc72 RA |
212 | info = { |
213 | 'id': vid, | |
214 | 'title': title, | |
215 | 'formats': formats, | |
216 | 'thumbnail': vod_info.get('coverUrl'), | |
217 | 'timestamp': int_or_none(vod_info.get('publishTs')), | |
218 | 'duration': int_or_none(vod_info.get('duration')), | |
219 | 'view_count': get_count('watch'), | |
220 | 'like_count': get_count('like'), | |
221 | 'comment_count': get_count('comment'), | |
a820dc72 | 222 | 'categories': [category] if category else None, |
6ef5ad9e | 223 | '__post_extractor': self.extract_comments(vid), |
a820dc72 RA |
224 | } |
225 | info.update(self._extract_streamer_info(vod_detail_info)) | |
226 | return info | |
974208e1 | 227 | |
6ef5ad9e | 228 | def _get_comments(self, vid): |
229 | for page in itertools.count(1): | |
230 | comments_json = self._call_api(vid, data={ | |
231 | 'operationName': 'getCommentList', | |
232 | 'variables': { | |
233 | 'params': { | |
234 | 'appInfo': { | |
235 | 'postID': vid, | |
236 | }, | |
237 | 'preview': {}, | |
238 | 'pageSize': 99, | |
239 | 'page': page, | |
240 | }, | |
241 | }, | |
242 | 'extensions': { | |
243 | 'persistedQuery': { | |
244 | 'version': 1, | |
245 | 'sha256Hash': 'be8e5f9522ddac7f7c604c0d284fd22481813263580849926c4c66fb767eed25', | |
246 | }, | |
247 | }, | |
248 | }) | |
249 | for comment in comments_json['commentList']: | |
250 | content = comment.get('content') | |
251 | if not content: | |
252 | continue | |
253 | author = comment.get('author') or {} | |
254 | parent = comment.get('parentID') | |
255 | yield { | |
256 | 'author': author.get('nickName'), | |
257 | 'author_id': str_or_none(author.get('uid')), | |
258 | 'id': str_or_none(comment.get('commentID')), | |
259 | 'text': content, | |
260 | 'timestamp': int_or_none(comment.get('createdAt')), | |
261 | 'parent': 'root' if parent == 0 else str_or_none(parent), | |
262 | } | |
263 | ||
264 | if comments_json['lastPage']: | |
265 | break | |
266 | ||
974208e1 | 267 | |
3262f8ab | 268 | class TrovoChannelBaseIE(TrovoBaseIE): |
974208e1 AG |
269 | def _get_vod_json(self, page, uid): |
270 | raise NotImplementedError('This method must be implemented by subclasses') | |
271 | ||
272 | def _entries(self, uid): | |
273 | for page in itertools.count(1): | |
274 | vod_json = self._get_vod_json(page, uid) | |
275 | vods = vod_json.get('vodInfos', []) | |
276 | for vod in vods: | |
277 | yield self.url_result( | |
278 | 'https://trovo.live/%s/%s' % (self._TYPE, vod.get('vid')), | |
279 | ie=TrovoVodIE.ie_key()) | |
280 | has_more = vod_json['hasMore'] | |
281 | if not has_more: | |
282 | break | |
283 | ||
284 | def _real_extract(self, url): | |
285 | id = self._match_id(url) | |
6ef5ad9e | 286 | live_info = self._call_api(id, data={ |
287 | 'operationName': 'live_LiveReaderService_GetLiveInfo', | |
288 | 'variables': { | |
289 | 'params': { | |
290 | 'userName': id, | |
291 | }, | |
292 | }, | |
293 | }) | |
294 | uid = str(live_info['streamerInfo']['uid']) | |
974208e1 AG |
295 | return self.playlist_result(self._entries(uid), playlist_id=uid) |
296 | ||
297 | ||
298 | class TrovoChannelVodIE(TrovoChannelBaseIE): | |
299 | _VALID_URL = r'trovovod:(?P<id>[^\s]+)' | |
96565c7e | 300 | IE_DESC = 'All VODs of a trovo.live channel; "trovovod:" prefix' |
974208e1 AG |
301 | |
302 | _TESTS = [{ | |
303 | 'url': 'trovovod:OneTappedYou', | |
304 | 'playlist_mincount': 24, | |
305 | 'info_dict': { | |
306 | 'id': '100719456', | |
307 | }, | |
308 | }] | |
309 | ||
974208e1 AG |
310 | _TYPE = 'video' |
311 | ||
312 | def _get_vod_json(self, page, uid): | |
6ef5ad9e | 313 | return self._call_api(uid, data={ |
314 | 'operationName': 'getChannelLtvVideoInfos', | |
315 | 'variables': { | |
316 | 'params': { | |
317 | 'channelID': int(uid), | |
318 | 'pageSize': 99, | |
319 | 'currPage': page, | |
320 | }, | |
321 | }, | |
322 | 'extensions': { | |
323 | 'persistedQuery': { | |
324 | 'version': 1, | |
325 | 'sha256Hash': '78fe32792005eab7e922cafcdad9c56bed8bbc5f5df3c7cd24fcb84a744f5f78', | |
326 | }, | |
327 | }, | |
328 | }) | |
974208e1 AG |
329 | |
330 | ||
331 | class TrovoChannelClipIE(TrovoChannelBaseIE): | |
332 | _VALID_URL = r'trovoclip:(?P<id>[^\s]+)' | |
96565c7e | 333 | IE_DESC = 'All Clips of a trovo.live channel; "trovoclip:" prefix' |
974208e1 AG |
334 | |
335 | _TESTS = [{ | |
336 | 'url': 'trovoclip:OneTappedYou', | |
337 | 'playlist_mincount': 29, | |
338 | 'info_dict': { | |
339 | 'id': '100719456', | |
340 | }, | |
341 | }] | |
342 | ||
974208e1 AG |
343 | _TYPE = 'clip' |
344 | ||
345 | def _get_vod_json(self, page, uid): | |
6ef5ad9e | 346 | return self._call_api(uid, data={ |
347 | 'operationName': 'getChannelClipVideoInfos', | |
348 | 'variables': { | |
349 | 'params': { | |
350 | 'channelID': int(uid), | |
351 | 'pageSize': 99, | |
352 | 'currPage': page, | |
353 | }, | |
354 | }, | |
355 | 'extensions': { | |
356 | 'persistedQuery': { | |
357 | 'version': 1, | |
358 | 'sha256Hash': 'e7924bfe20059b5c75fc8ff9e7929f43635681a7bdf3befa01072ed22c8eff31', | |
359 | }, | |
360 | }, | |
361 | }) |