]>
Commit | Line | Data |
---|---|---|
04d02a9d | 1 | # coding: utf-8 |
ee1e1996 PH |
2 | from __future__ import unicode_literals |
3 | ||
3182f3e2 | 4 | import itertools |
79e93125 | 5 | import re |
f353cbdb | 6 | import random |
79e93125 PH |
7 | |
8 | from .common import InfoExtractor | |
1cc79574 | 9 | from ..compat import ( |
efe470e2 | 10 | compat_HTTPError, |
e704f87f | 11 | compat_parse_qs, |
240b9b7a | 12 | compat_str, |
15707c7e | 13 | compat_urllib_parse_urlencode, |
e704f87f | 14 | compat_urllib_parse_urlparse, |
1cc79574 PH |
15 | ) |
16 | from ..utils import ( | |
efe470e2 | 17 | clean_html, |
79e93125 | 18 | ExtractorError, |
7a6e8a1b | 19 | int_or_none, |
778f9694 | 20 | js_to_json, |
8bbb4b56 | 21 | orderedSet, |
e704f87f | 22 | parse_duration, |
355d074f | 23 | parse_iso8601, |
264e77c4 | 24 | update_url_query, |
6e6bc8da | 25 | urlencode_postdata, |
c64c03be | 26 | urljoin, |
79e93125 PH |
27 | ) |
28 | ||
29 | ||
c5db6bb3 | 30 | class TwitchBaseIE(InfoExtractor): |
240b9b7a | 31 | _VALID_URL_BASE = r'https?://(?:www\.)?twitch\.tv' |
c5db6bb3 | 32 | |
46fd0dd5 | 33 | _API_BASE = 'https://api.twitch.tv' |
9f4576a7 | 34 | _USHER_BASE = 'https://usher.ttvnw.net' |
c64c03be | 35 | _LOGIN_URL = 'https://www.twitch.tv/login' |
95be29e1 | 36 | _CLIENT_ID = 'jzkbprff40iqj646a697cyrvl0zt2m6' |
499bfcbf | 37 | _NETRC_MACHINE = 'twitch' |
79e93125 | 38 | |
355d074f S |
39 | def _handle_error(self, response): |
40 | if not isinstance(response, dict): | |
41 | return | |
42 | error = response.get('error') | |
43 | if error: | |
44 | raise ExtractorError( | |
45 | '%s returned error: %s - %s' % (self.IE_NAME, error, response.get('message')), | |
46 | expected=True) | |
47 | ||
e3f6b569 | 48 | def _call_api(self, path, item_id, note): |
e3f6b569 | 49 | response = self._download_json( |
95be29e1 S |
50 | '%s/%s' % (self._API_BASE, path), item_id, note, |
51 | headers={'Client-ID': self._CLIENT_ID}) | |
355d074f S |
52 | self._handle_error(response) |
53 | return response | |
54 | ||
c5db6bb3 S |
55 | def _real_initialize(self): |
56 | self._login() | |
57 | ||
58 | def _login(self): | |
59 | (username, password) = self._get_login_info() | |
60 | if username is None: | |
61 | return | |
62 | ||
efe470e2 S |
63 | def fail(message): |
64 | raise ExtractorError( | |
65 | 'Unable to login. Twitch said: %s' % message, expected=True) | |
66 | ||
5316566e | 67 | def login_step(page, urlh, note, data): |
c64c03be XDG |
68 | form = self._hidden_inputs(page) |
69 | form.update(data) | |
70 | ||
71 | page_url = urlh.geturl() | |
72 | post_url = self._search_regex( | |
73 | r'<form[^>]+action=(["\'])(?P<url>.+?)\1', page, | |
74 | 'post url', default=page_url, group='url') | |
75 | post_url = urljoin(page_url, post_url) | |
76 | ||
77 | headers = {'Referer': page_url} | |
78 | ||
79 | try: | |
80 | response = self._download_json( | |
81 | post_url, None, note, | |
82 | data=urlencode_postdata(form), | |
83 | headers=headers) | |
84 | except ExtractorError as e: | |
85 | if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400: | |
86 | response = self._parse_json( | |
87 | e.cause.read().decode('utf-8'), None) | |
88 | fail(response['message']) | |
89 | raise | |
90 | ||
5316566e S |
91 | redirect_url = urljoin(post_url, response['redirect']) |
92 | return self._download_webpage_handle( | |
93 | redirect_url, None, 'Downloading login redirect page', | |
94 | headers=headers) | |
c64c03be | 95 | |
fbd9f6ea | 96 | login_page, handle = self._download_webpage_handle( |
c5db6bb3 S |
97 | self._LOGIN_URL, None, 'Downloading login page') |
98 | ||
efe470e2 S |
99 | # Some TOR nodes and public proxies are blocked completely |
100 | if 'blacklist_message' in login_page: | |
101 | fail(clean_html(login_page)) | |
102 | ||
5316566e S |
103 | redirect_page, handle = login_step( |
104 | login_page, handle, 'Logging in as %s' % username, { | |
105 | 'username': username, | |
106 | 'password': password, | |
fe646a2f | 107 | }) |
c64c03be XDG |
108 | |
109 | if re.search(r'(?i)<form[^>]+id="two-factor-submit"', redirect_page) is not None: | |
110 | # TODO: Add mechanism to request an SMS or phone call | |
111 | tfa_token = self._get_tfa_info('two-factor authentication token') | |
5316566e | 112 | login_step(redirect_page, handle, 'Submitting TFA token', { |
c64c03be XDG |
113 | 'authy_token': tfa_token, |
114 | 'remember_2fa': 'true', | |
5316566e | 115 | }) |
17b41a33 | 116 | |
d0e958c7 PH |
117 | def _prefer_source(self, formats): |
118 | try: | |
119 | source = next(f for f in formats if f['format_id'] == 'Source') | |
120 | source['preference'] = 10 | |
121 | except StopIteration: | |
122 | pass # No Source stream present | |
123 | self._sort_formats(formats) | |
124 | ||
c5db6bb3 S |
125 | |
126 | class TwitchItemBaseIE(TwitchBaseIE): | |
127 | def _download_info(self, item, item_id): | |
e3f6b569 S |
128 | return self._extract_info(self._call_api( |
129 | 'kraken/videos/%s%s' % (item, item_id), item_id, | |
c5db6bb3 | 130 | 'Downloading %s info JSON' % self._ITEM_TYPE)) |
a22524b0 | 131 | |
c5db6bb3 S |
132 | def _extract_media(self, item_id): |
133 | info = self._download_info(self._ITEM_SHORTCUT, item_id) | |
e3f6b569 S |
134 | response = self._call_api( |
135 | 'api/videos/%s%s' % (self._ITEM_SHORTCUT, item_id), item_id, | |
c5db6bb3 | 136 | 'Downloading %s playlist JSON' % self._ITEM_TYPE) |
355d074f S |
137 | entries = [] |
138 | chunks = response['chunks'] | |
139 | qualities = list(chunks.keys()) | |
140 | for num, fragment in enumerate(zip(*chunks.values()), start=1): | |
141 | formats = [] | |
142 | for fmt_num, fragment_fmt in enumerate(fragment): | |
143 | format_id = qualities[fmt_num] | |
144 | fmt = { | |
145 | 'url': fragment_fmt['url'], | |
146 | 'format_id': format_id, | |
147 | 'quality': 1 if format_id == 'live' else 0, | |
148 | } | |
149 | m = re.search(r'^(?P<height>\d+)[Pp]', format_id) | |
150 | if m: | |
151 | fmt['height'] = int(m.group('height')) | |
152 | formats.append(fmt) | |
153 | self._sort_formats(formats) | |
154 | entry = dict(info) | |
159444a6 | 155 | entry['id'] = '%s_%d' % (entry['id'], num) |
355d074f S |
156 | entry['title'] = '%s part %d' % (entry['title'], num) |
157 | entry['formats'] = formats | |
158 | entries.append(entry) | |
013bfdd8 | 159 | return self.playlist_result(entries, info['id'], info['title']) |
355d074f S |
160 | |
161 | def _extract_info(self, info): | |
162 | return { | |
163 | 'id': info['_id'], | |
369c12e0 | 164 | 'title': info.get('title') or 'Untitled Broadcast', |
7a6e8a1b S |
165 | 'description': info.get('description'), |
166 | 'duration': int_or_none(info.get('length')), | |
167 | 'thumbnail': info.get('preview'), | |
168 | 'uploader': info.get('channel', {}).get('display_name'), | |
169 | 'uploader_id': info.get('channel', {}).get('name'), | |
170 | 'timestamp': parse_iso8601(info.get('recorded_at')), | |
171 | 'view_count': int_or_none(info.get('views')), | |
355d074f S |
172 | } |
173 | ||
c5db6bb3 S |
174 | def _real_extract(self, url): |
175 | return self._extract_media(self._match_id(url)) | |
04d02a9d | 176 | |
04d02a9d | 177 | |
c5db6bb3 S |
178 | class TwitchVideoIE(TwitchItemBaseIE): |
179 | IE_NAME = 'twitch:video' | |
93f78707 | 180 | _VALID_URL = r'%s/[^/]+/b/(?P<id>\d+)' % TwitchBaseIE._VALID_URL_BASE |
c5db6bb3 S |
181 | _ITEM_TYPE = 'video' |
182 | _ITEM_SHORTCUT = 'a' | |
04d02a9d | 183 | |
c5db6bb3 S |
184 | _TEST = { |
185 | 'url': 'http://www.twitch.tv/riotgames/b/577357806', | |
186 | 'info_dict': { | |
187 | 'id': 'a577357806', | |
188 | 'title': 'Worlds Semifinals - Star Horn Royal Club vs. OMG', | |
189 | }, | |
190 | 'playlist_mincount': 12, | |
0db3a661 | 191 | 'skip': 'HTTP Error 404: Not Found', |
c5db6bb3 | 192 | } |
04d02a9d | 193 | |
04d02a9d | 194 | |
c5db6bb3 S |
195 | class TwitchChapterIE(TwitchItemBaseIE): |
196 | IE_NAME = 'twitch:chapter' | |
93f78707 | 197 | _VALID_URL = r'%s/[^/]+/c/(?P<id>\d+)' % TwitchBaseIE._VALID_URL_BASE |
c5db6bb3 S |
198 | _ITEM_TYPE = 'chapter' |
199 | _ITEM_SHORTCUT = 'c' | |
04d02a9d | 200 | |
78111136 | 201 | _TESTS = [{ |
c5db6bb3 S |
202 | 'url': 'http://www.twitch.tv/acracingleague/c/5285812', |
203 | 'info_dict': { | |
204 | 'id': 'c5285812', | |
205 | 'title': 'ACRL Off Season - Sports Cars @ Nordschleife', | |
206 | }, | |
207 | 'playlist_mincount': 3, | |
0db3a661 | 208 | 'skip': 'HTTP Error 404: Not Found', |
78111136 PH |
209 | }, { |
210 | 'url': 'http://www.twitch.tv/tsm_theoddone/c/2349361', | |
211 | 'only_matching': True, | |
212 | }] | |
c5db6bb3 S |
213 | |
214 | ||
215 | class TwitchVodIE(TwitchItemBaseIE): | |
216 | IE_NAME = 'twitch:vod' | |
3f1ce168 S |
217 | _VALID_URL = r'''(?x) |
218 | https?:// | |
219 | (?: | |
0b23c222 | 220 | (?:www\.)?twitch\.tv/(?:[^/]+/v|videos)/| |
3f1ce168 S |
221 | player\.twitch\.tv/\?.*?\bvideo=v |
222 | ) | |
223 | (?P<id>\d+) | |
224 | ''' | |
c5db6bb3 S |
225 | _ITEM_TYPE = 'vod' |
226 | _ITEM_SHORTCUT = 'v' | |
227 | ||
9c724a98 | 228 | _TESTS = [{ |
e704f87f | 229 | 'url': 'http://www.twitch.tv/riotgames/v/6528877?t=5m10s', |
c5db6bb3 | 230 | 'info_dict': { |
ac0474f8 | 231 | 'id': 'v6528877', |
c5db6bb3 | 232 | 'ext': 'mp4', |
ac0474f8 | 233 | 'title': 'LCK Summer Split - Week 6 Day 1', |
ec85ded8 | 234 | 'thumbnail': r're:^https?://.*\.jpg$', |
ac0474f8 YCH |
235 | 'duration': 17208, |
236 | 'timestamp': 1435131709, | |
237 | 'upload_date': '20150624', | |
238 | 'uploader': 'Riot Games', | |
239 | 'uploader_id': 'riotgames', | |
c5db6bb3 | 240 | 'view_count': int, |
e704f87f | 241 | 'start_time': 310, |
c5db6bb3 S |
242 | }, |
243 | 'params': { | |
244 | # m3u8 download | |
245 | 'skip_download': True, | |
246 | }, | |
9c724a98 S |
247 | }, { |
248 | # Untitled broadcast (title is None) | |
249 | 'url': 'http://www.twitch.tv/belkao_o/v/11230755', | |
250 | 'info_dict': { | |
251 | 'id': 'v11230755', | |
252 | 'ext': 'mp4', | |
253 | 'title': 'Untitled Broadcast', | |
ec85ded8 | 254 | 'thumbnail': r're:^https?://.*\.jpg$', |
9c724a98 S |
255 | 'duration': 1638, |
256 | 'timestamp': 1439746708, | |
257 | 'upload_date': '20150816', | |
258 | 'uploader': 'BelkAO_o', | |
259 | 'uploader_id': 'belkao_o', | |
260 | 'view_count': int, | |
261 | }, | |
262 | 'params': { | |
263 | # m3u8 download | |
264 | 'skip_download': True, | |
265 | }, | |
9bd7bd0b | 266 | 'skip': 'HTTP Error 404: Not Found', |
3f1ce168 S |
267 | }, { |
268 | 'url': 'http://player.twitch.tv/?t=5m10s&video=v6528877', | |
269 | 'only_matching': True, | |
0b23c222 S |
270 | }, { |
271 | 'url': 'https://www.twitch.tv/videos/6528877', | |
272 | 'only_matching': True, | |
9c724a98 | 273 | }] |
04d02a9d | 274 | |
79e93125 | 275 | def _real_extract(self, url): |
c5db6bb3 | 276 | item_id = self._match_id(url) |
e5e99661 | 277 | |
c5db6bb3 | 278 | info = self._download_info(self._ITEM_SHORTCUT, item_id) |
e3f6b569 S |
279 | access_token = self._call_api( |
280 | 'api/vods/%s/access_token' % item_id, item_id, | |
350c9481 | 281 | 'Downloading %s access token' % self._ITEM_TYPE) |
e5e99661 | 282 | |
c5db6bb3 | 283 | formats = self._extract_m3u8_formats( |
350c9481 S |
284 | '%s/vod/%s?%s' % ( |
285 | self._USHER_BASE, item_id, | |
15707c7e | 286 | compat_urllib_parse_urlencode({ |
e5e99661 | 287 | 'allow_source': 'true', |
ac455055 | 288 | 'allow_audio_only': 'true', |
e5e99661 S |
289 | 'allow_spectre': 'true', |
290 | 'player': 'twitchweb', | |
291 | 'nauth': access_token['token'], | |
292 | 'nauthsig': access_token['sig'], | |
293 | })), | |
631d4c87 | 294 | item_id, 'mp4', entry_protocol='m3u8_native') |
e5e99661 | 295 | |
d0e958c7 | 296 | self._prefer_source(formats) |
c5db6bb3 | 297 | info['formats'] = formats |
e704f87f NH |
298 | |
299 | parsed_url = compat_urllib_parse_urlparse(url) | |
300 | query = compat_parse_qs(parsed_url.query) | |
301 | if 't' in query: | |
302 | info['start_time'] = parse_duration(query['t'][0]) | |
303 | ||
264e77c4 S |
304 | if info.get('timestamp') is not None: |
305 | info['subtitles'] = { | |
306 | 'rechat': [{ | |
307 | 'url': update_url_query( | |
308 | 'https://rechat.twitch.tv/rechat-messages', { | |
309 | 'video_id': 'v%s' % item_id, | |
310 | 'start': info['timestamp'], | |
311 | }), | |
312 | 'ext': 'json', | |
313 | }], | |
314 | } | |
315 | ||
c5db6bb3 S |
316 | return info |
317 | ||
318 | ||
319 | class TwitchPlaylistBaseIE(TwitchBaseIE): | |
e3f6b569 | 320 | _PLAYLIST_PATH = 'kraken/channels/%s/videos/?offset=%d&limit=%d' |
ececca6c | 321 | _PAGE_LIMIT = 100 |
c5db6bb3 S |
322 | |
323 | def _extract_playlist(self, channel_id): | |
e3f6b569 S |
324 | info = self._call_api( |
325 | 'kraken/channels/%s' % channel_id, | |
c5db6bb3 S |
326 | channel_id, 'Downloading channel info JSON') |
327 | channel_name = info.get('display_name') or info.get('name') | |
328 | entries = [] | |
329 | offset = 0 | |
330 | limit = self._PAGE_LIMIT | |
539a1641 S |
331 | broken_paging_detected = False |
332 | counter_override = None | |
c5db6bb3 | 333 | for counter in itertools.count(1): |
e3f6b569 S |
334 | response = self._call_api( |
335 | self._PLAYLIST_PATH % (channel_id, offset, limit), | |
539a1641 | 336 | channel_id, |
93753aad | 337 | 'Downloading %s JSON page %s' |
539a1641 | 338 | % (self._PLAYLIST_TYPE, counter_override or counter)) |
c2e64f71 S |
339 | page_entries = self._extract_playlist_page(response) |
340 | if not page_entries: | |
c5db6bb3 | 341 | break |
539a1641 S |
342 | total = int_or_none(response.get('_total')) |
343 | # Since the beginning of March 2016 twitch's paging mechanism | |
344 | # is completely broken on the twitch side. It simply ignores | |
345 | # a limit and returns the whole offset number of videos. | |
346 | # Working around by just requesting all videos at once. | |
dec2cae0 | 347 | # Upd: pagination bug was fixed by twitch on 15.03.2016. |
539a1641 S |
348 | if not broken_paging_detected and total and len(page_entries) > limit: |
349 | self.report_warning( | |
dec2cae0 | 350 | 'Twitch pagination is broken on twitch side, requesting all videos at once', |
539a1641 S |
351 | channel_id) |
352 | broken_paging_detected = True | |
a8276b26 | 353 | offset = total |
539a1641 S |
354 | counter_override = '(all at once)' |
355 | continue | |
c2e64f71 | 356 | entries.extend(page_entries) |
539a1641 S |
357 | if broken_paging_detected or total and len(page_entries) >= total: |
358 | break | |
c5db6bb3 | 359 | offset += limit |
c2e64f71 | 360 | return self.playlist_result( |
8bbb4b56 | 361 | [self.url_result(entry) for entry in orderedSet(entries)], |
c2e64f71 S |
362 | channel_id, channel_name) |
363 | ||
364 | def _extract_playlist_page(self, response): | |
365 | videos = response.get('videos') | |
366 | return [video['url'] for video in videos] if videos else [] | |
c5db6bb3 S |
367 | |
368 | def _real_extract(self, url): | |
369 | return self._extract_playlist(self._match_id(url)) | |
370 | ||
371 | ||
372 | class TwitchProfileIE(TwitchPlaylistBaseIE): | |
373 | IE_NAME = 'twitch:profile' | |
374 | _VALID_URL = r'%s/(?P<id>[^/]+)/profile/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE | |
375 | _PLAYLIST_TYPE = 'profile' | |
376 | ||
377 | _TEST = { | |
378 | 'url': 'http://www.twitch.tv/vanillatv/profile', | |
379 | 'info_dict': { | |
380 | 'id': 'vanillatv', | |
381 | 'title': 'VanillaTV', | |
382 | }, | |
383 | 'playlist_mincount': 412, | |
384 | } | |
385 | ||
386 | ||
93753aad S |
387 | class TwitchVideosBaseIE(TwitchPlaylistBaseIE): |
388 | _VALID_URL_VIDEOS_BASE = r'%s/(?P<id>[^/]+)/videos' % TwitchBaseIE._VALID_URL_BASE | |
389 | _PLAYLIST_PATH = TwitchPlaylistBaseIE._PLAYLIST_PATH + '&broadcast_type=' | |
390 | ||
391 | ||
392 | class TwitchAllVideosIE(TwitchVideosBaseIE): | |
393 | IE_NAME = 'twitch:videos:all' | |
394 | _VALID_URL = r'%s/all' % TwitchVideosBaseIE._VALID_URL_VIDEOS_BASE | |
395 | _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'archive,upload,highlight' | |
396 | _PLAYLIST_TYPE = 'all videos' | |
397 | ||
398 | _TEST = { | |
399 | 'url': 'https://www.twitch.tv/spamfish/videos/all', | |
400 | 'info_dict': { | |
401 | 'id': 'spamfish', | |
402 | 'title': 'Spamfish', | |
403 | }, | |
404 | 'playlist_mincount': 869, | |
405 | } | |
406 | ||
407 | ||
408 | class TwitchUploadsIE(TwitchVideosBaseIE): | |
409 | IE_NAME = 'twitch:videos:uploads' | |
410 | _VALID_URL = r'%s/uploads' % TwitchVideosBaseIE._VALID_URL_VIDEOS_BASE | |
411 | _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'upload' | |
412 | _PLAYLIST_TYPE = 'uploads' | |
413 | ||
414 | _TEST = { | |
415 | 'url': 'https://www.twitch.tv/spamfish/videos/uploads', | |
416 | 'info_dict': { | |
417 | 'id': 'spamfish', | |
418 | 'title': 'Spamfish', | |
419 | }, | |
420 | 'playlist_mincount': 0, | |
421 | } | |
422 | ||
423 | ||
424 | class TwitchPastBroadcastsIE(TwitchVideosBaseIE): | |
425 | IE_NAME = 'twitch:videos:past-broadcasts' | |
426 | _VALID_URL = r'%s/past-broadcasts' % TwitchVideosBaseIE._VALID_URL_VIDEOS_BASE | |
427 | _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'archive' | |
c5db6bb3 S |
428 | _PLAYLIST_TYPE = 'past broadcasts' |
429 | ||
430 | _TEST = { | |
93753aad S |
431 | 'url': 'https://www.twitch.tv/spamfish/videos/past-broadcasts', |
432 | 'info_dict': { | |
433 | 'id': 'spamfish', | |
434 | 'title': 'Spamfish', | |
435 | }, | |
436 | 'playlist_mincount': 0, | |
437 | } | |
438 | ||
439 | ||
440 | class TwitchHighlightsIE(TwitchVideosBaseIE): | |
441 | IE_NAME = 'twitch:videos:highlights' | |
442 | _VALID_URL = r'%s/highlights' % TwitchVideosBaseIE._VALID_URL_VIDEOS_BASE | |
443 | _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'highlight' | |
444 | _PLAYLIST_TYPE = 'highlights' | |
445 | ||
446 | _TEST = { | |
447 | 'url': 'https://www.twitch.tv/spamfish/videos/highlights', | |
c5db6bb3 S |
448 | 'info_dict': { |
449 | 'id': 'spamfish', | |
450 | 'title': 'Spamfish', | |
451 | }, | |
93753aad | 452 | 'playlist_mincount': 805, |
c5db6bb3 | 453 | } |
240b9b7a S |
454 | |
455 | ||
456 | class TwitchStreamIE(TwitchBaseIE): | |
457 | IE_NAME = 'twitch:stream' | |
9db8f6c5 S |
458 | _VALID_URL = r'''(?x) |
459 | https?:// | |
460 | (?: | |
461 | (?:www\.)?twitch\.tv/| | |
462 | player\.twitch\.tv/\?.*?\bchannel= | |
463 | ) | |
464 | (?P<id>[^/#?]+) | |
465 | ''' | |
240b9b7a | 466 | |
faa1f83a | 467 | _TESTS = [{ |
240b9b7a S |
468 | 'url': 'http://www.twitch.tv/shroomztv', |
469 | 'info_dict': { | |
470 | 'id': '12772022048', | |
471 | 'display_id': 'shroomztv', | |
472 | 'ext': 'mp4', | |
473 | 'title': 're:^ShroomzTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', | |
474 | 'description': 'H1Z1 - lonewolfing with ShroomzTV | A3 Battle Royale later - @ShroomzTV', | |
475 | 'is_live': True, | |
476 | 'timestamp': 1421928037, | |
477 | 'upload_date': '20150122', | |
478 | 'uploader': 'ShroomzTV', | |
479 | 'uploader_id': 'shroomztv', | |
480 | 'view_count': int, | |
481 | }, | |
482 | 'params': { | |
483 | # m3u8 download | |
484 | 'skip_download': True, | |
485 | }, | |
faa1f83a S |
486 | }, { |
487 | 'url': 'http://www.twitch.tv/miracle_doto#profile-0', | |
488 | 'only_matching': True, | |
9db8f6c5 S |
489 | }, { |
490 | 'url': 'https://player.twitch.tv/?channel=lotsofs', | |
491 | 'only_matching': True, | |
faa1f83a | 492 | }] |
240b9b7a | 493 | |
9db8f6c5 S |
494 | @classmethod |
495 | def suitable(cls, url): | |
496 | return (False | |
497 | if any(ie.suitable(url) for ie in ( | |
498 | TwitchVideoIE, | |
499 | TwitchChapterIE, | |
500 | TwitchVodIE, | |
501 | TwitchProfileIE, | |
502 | TwitchAllVideosIE, | |
503 | TwitchUploadsIE, | |
504 | TwitchPastBroadcastsIE, | |
505 | TwitchHighlightsIE)) | |
506 | else super(TwitchStreamIE, cls).suitable(url)) | |
507 | ||
240b9b7a S |
508 | def _real_extract(self, url): |
509 | channel_id = self._match_id(url) | |
510 | ||
e3f6b569 | 511 | stream = self._call_api( |
9aa929d3 | 512 | 'kraken/streams/%s?stream_type=all' % channel_id, channel_id, |
240b9b7a S |
513 | 'Downloading stream JSON').get('stream') |
514 | ||
240b9b7a | 515 | if not stream: |
cc764a6d | 516 | raise ExtractorError('%s is offline' % channel_id, expected=True) |
240b9b7a | 517 | |
06966677 S |
518 | # Channel name may be typed if different case than the original channel name |
519 | # (e.g. http://www.twitch.tv/TWITCHPLAYSPOKEMON) that will lead to constructing | |
520 | # an invalid m3u8 URL. Working around by use of original channel name from stream | |
521 | # JSON and fallback to lowercase if it's not available. | |
1793d71d S |
522 | channel_id = stream.get('channel', {}).get('name') or channel_id.lower() |
523 | ||
e3f6b569 S |
524 | access_token = self._call_api( |
525 | 'api/channels/%s/access_token' % channel_id, channel_id, | |
240b9b7a S |
526 | 'Downloading channel access token') |
527 | ||
528 | query = { | |
529 | 'allow_source': 'true', | |
ac455055 | 530 | 'allow_audio_only': 'true', |
9aa929d3 | 531 | 'allow_spectre': 'true', |
f353cbdb | 532 | 'p': random.randint(1000000, 10000000), |
240b9b7a S |
533 | 'player': 'twitchweb', |
534 | 'segment_preference': '4', | |
d34e7949 S |
535 | 'sig': access_token['sig'].encode('utf-8'), |
536 | 'token': access_token['token'].encode('utf-8'), | |
240b9b7a | 537 | } |
240b9b7a S |
538 | formats = self._extract_m3u8_formats( |
539 | '%s/api/channel/hls/%s.m3u8?%s' | |
15707c7e | 540 | % (self._USHER_BASE, channel_id, compat_urllib_parse_urlencode(query)), |
240b9b7a | 541 | channel_id, 'mp4') |
d0e958c7 | 542 | self._prefer_source(formats) |
240b9b7a S |
543 | |
544 | view_count = stream.get('viewers') | |
545 | timestamp = parse_iso8601(stream.get('created_at')) | |
546 | ||
547 | channel = stream['channel'] | |
548 | title = self._live_title(channel.get('display_name') or channel.get('name')) | |
549 | description = channel.get('status') | |
550 | ||
551 | thumbnails = [] | |
552 | for thumbnail_key, thumbnail_url in stream['preview'].items(): | |
553 | m = re.search(r'(?P<width>\d+)x(?P<height>\d+)\.jpg$', thumbnail_key) | |
554 | if not m: | |
555 | continue | |
556 | thumbnails.append({ | |
557 | 'url': thumbnail_url, | |
558 | 'width': int(m.group('width')), | |
559 | 'height': int(m.group('height')), | |
560 | }) | |
561 | ||
562 | return { | |
563 | 'id': compat_str(stream['_id']), | |
564 | 'display_id': channel_id, | |
565 | 'title': title, | |
566 | 'description': description, | |
567 | 'thumbnails': thumbnails, | |
568 | 'uploader': channel.get('display_name'), | |
569 | 'uploader_id': channel.get('name'), | |
570 | 'timestamp': timestamp, | |
571 | 'view_count': view_count, | |
572 | 'formats': formats, | |
573 | 'is_live': True, | |
12d1fb5a | 574 | } |
778f9694 S |
575 | |
576 | ||
577 | class TwitchClipsIE(InfoExtractor): | |
578 | IE_NAME = 'twitch:clips' | |
579 | _VALID_URL = r'https?://clips\.twitch\.tv/(?:[^/]+/)*(?P<id>[^/?#&]+)' | |
580 | ||
74ba450a | 581 | _TESTS = [{ |
778f9694 S |
582 | 'url': 'https://clips.twitch.tv/ea/AggressiveCobraPoooound', |
583 | 'md5': '761769e1eafce0ffebfb4089cb3847cd', | |
584 | 'info_dict': { | |
585 | 'id': 'AggressiveCobraPoooound', | |
586 | 'ext': 'mp4', | |
587 | 'title': 'EA Play 2016 Live from the Novo Theatre', | |
ec85ded8 | 588 | 'thumbnail': r're:^https?://.*\.jpg', |
778f9694 S |
589 | 'creator': 'EA', |
590 | 'uploader': 'stereotype_', | |
591 | 'uploader_id': 'stereotype_', | |
592 | }, | |
74ba450a S |
593 | }, { |
594 | # multiple formats | |
595 | 'url': 'https://clips.twitch.tv/rflegendary/UninterestedBeeDAESuppy', | |
596 | 'only_matching': True, | |
597 | }] | |
778f9694 S |
598 | |
599 | def _real_extract(self, url): | |
600 | video_id = self._match_id(url) | |
601 | ||
602 | webpage = self._download_webpage(url, video_id) | |
603 | ||
604 | clip = self._parse_json( | |
605 | self._search_regex( | |
606 | r'(?s)clipInfo\s*=\s*({.+?});', webpage, 'clip info'), | |
607 | video_id, transform_source=js_to_json) | |
608 | ||
74ba450a S |
609 | title = clip.get('channel_title') or self._og_search_title(webpage) |
610 | ||
611 | formats = [{ | |
612 | 'url': option['source'], | |
613 | 'format_id': option.get('quality'), | |
614 | 'height': int_or_none(option.get('quality')), | |
615 | } for option in clip.get('quality_options', []) if option.get('source')] | |
616 | ||
617 | if not formats: | |
618 | formats = [{ | |
619 | 'url': clip['clip_video_url'], | |
620 | }] | |
778f9694 | 621 | |
0cacae28 S |
622 | self._sort_formats(formats) |
623 | ||
778f9694 S |
624 | return { |
625 | 'id': video_id, | |
778f9694 S |
626 | 'title': title, |
627 | 'thumbnail': self._og_search_thumbnail(webpage), | |
628 | 'creator': clip.get('broadcaster_display_name') or clip.get('broadcaster_login'), | |
629 | 'uploader': clip.get('curator_login'), | |
630 | 'uploader_id': clip.get('curator_display_name'), | |
74ba450a | 631 | 'formats': formats, |
778f9694 | 632 | } |