]>
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 |
48afc6ca | 7 | import json |
79e93125 PH |
8 | |
9 | from .common import InfoExtractor | |
1cc79574 | 10 | from ..compat import ( |
49fa7de3 | 11 | compat_kwargs, |
e704f87f | 12 | compat_parse_qs, |
240b9b7a | 13 | compat_str, |
15707c7e | 14 | compat_urllib_parse_urlencode, |
e704f87f | 15 | compat_urllib_parse_urlparse, |
1cc79574 PH |
16 | ) |
17 | from ..utils import ( | |
efe470e2 | 18 | clean_html, |
79e93125 | 19 | ExtractorError, |
49fa7de3 | 20 | float_or_none, |
7a6e8a1b | 21 | int_or_none, |
8bbb4b56 | 22 | orderedSet, |
e704f87f | 23 | parse_duration, |
355d074f | 24 | parse_iso8601, |
49fa7de3 S |
25 | qualities, |
26 | try_get, | |
27 | unified_timestamp, | |
264e77c4 | 28 | update_url_query, |
3052a30d | 29 | url_or_none, |
c64c03be | 30 | urljoin, |
79e93125 PH |
31 | ) |
32 | ||
33 | ||
c5db6bb3 | 34 | class TwitchBaseIE(InfoExtractor): |
fce79626 | 35 | _VALID_URL_BASE = r'https?://(?:(?:www|go|m)\.)?twitch\.tv' |
c5db6bb3 | 36 | |
46fd0dd5 | 37 | _API_BASE = 'https://api.twitch.tv' |
9f4576a7 | 38 | _USHER_BASE = 'https://usher.ttvnw.net' |
48afc6ca TB |
39 | _LOGIN_FORM_URL = 'https://www.twitch.tv/login' |
40 | _LOGIN_POST_URL = 'https://passport.twitch.tv/login' | |
cb1c3a3c | 41 | _CLIENT_ID = 'kimne78kx3ncx6brgo4mv6wki5h1ko' |
499bfcbf | 42 | _NETRC_MACHINE = 'twitch' |
79e93125 | 43 | |
355d074f S |
44 | def _handle_error(self, response): |
45 | if not isinstance(response, dict): | |
46 | return | |
47 | error = response.get('error') | |
48 | if error: | |
49 | raise ExtractorError( | |
50 | '%s returned error: %s - %s' % (self.IE_NAME, error, response.get('message')), | |
51 | expected=True) | |
52 | ||
49fa7de3 | 53 | def _call_api(self, path, item_id, *args, **kwargs): |
e2f61598 S |
54 | headers = kwargs.get('headers', {}).copy() |
55 | headers['Client-ID'] = self._CLIENT_ID | |
56 | kwargs['headers'] = headers | |
e3f6b569 | 57 | response = self._download_json( |
49fa7de3 S |
58 | '%s/%s' % (self._API_BASE, path), item_id, |
59 | *args, **compat_kwargs(kwargs)) | |
355d074f S |
60 | self._handle_error(response) |
61 | return response | |
62 | ||
c5db6bb3 S |
63 | def _real_initialize(self): |
64 | self._login() | |
65 | ||
66 | def _login(self): | |
68217024 | 67 | username, password = self._get_login_info() |
c5db6bb3 S |
68 | if username is None: |
69 | return | |
70 | ||
efe470e2 S |
71 | def fail(message): |
72 | raise ExtractorError( | |
73 | 'Unable to login. Twitch said: %s' % message, expected=True) | |
74 | ||
5316566e | 75 | def login_step(page, urlh, note, data): |
c64c03be XDG |
76 | form = self._hidden_inputs(page) |
77 | form.update(data) | |
78 | ||
79 | page_url = urlh.geturl() | |
80 | post_url = self._search_regex( | |
81 | r'<form[^>]+action=(["\'])(?P<url>.+?)\1', page, | |
48afc6ca | 82 | 'post url', default=self._LOGIN_POST_URL, group='url') |
c64c03be XDG |
83 | post_url = urljoin(page_url, post_url) |
84 | ||
48afc6ca TB |
85 | headers = { |
86 | 'Referer': page_url, | |
87 | 'Origin': page_url, | |
cb1c3a3c | 88 | 'Content-Type': 'text/plain;charset=UTF-8', |
48afc6ca | 89 | } |
c64c03be | 90 | |
cb1c3a3c S |
91 | response = self._download_json( |
92 | post_url, None, note, data=json.dumps(form).encode(), | |
93 | headers=headers, expected_status=400) | |
94 | error = response.get('error_description') or response.get('error_code') | |
95 | if error: | |
96 | fail(error) | |
c64c03be | 97 | |
1370dba5 S |
98 | if 'Authenticated successfully' in response.get('message', ''): |
99 | return None, None | |
100 | ||
101 | redirect_url = urljoin( | |
102 | post_url, | |
103 | response.get('redirect') or response['redirect_path']) | |
5316566e S |
104 | return self._download_webpage_handle( |
105 | redirect_url, None, 'Downloading login redirect page', | |
106 | headers=headers) | |
c64c03be | 107 | |
fbd9f6ea | 108 | login_page, handle = self._download_webpage_handle( |
48afc6ca | 109 | self._LOGIN_FORM_URL, None, 'Downloading login page') |
c5db6bb3 | 110 | |
efe470e2 S |
111 | # Some TOR nodes and public proxies are blocked completely |
112 | if 'blacklist_message' in login_page: | |
113 | fail(clean_html(login_page)) | |
114 | ||
5316566e | 115 | redirect_page, handle = login_step( |
e4d95865 | 116 | login_page, handle, 'Logging in', { |
5316566e S |
117 | 'username': username, |
118 | 'password': password, | |
48afc6ca | 119 | 'client_id': self._CLIENT_ID, |
fe646a2f | 120 | }) |
c64c03be | 121 | |
1370dba5 S |
122 | # Successful login |
123 | if not redirect_page: | |
124 | return | |
125 | ||
c64c03be XDG |
126 | if re.search(r'(?i)<form[^>]+id="two-factor-submit"', redirect_page) is not None: |
127 | # TODO: Add mechanism to request an SMS or phone call | |
128 | tfa_token = self._get_tfa_info('two-factor authentication token') | |
5316566e | 129 | login_step(redirect_page, handle, 'Submitting TFA token', { |
c64c03be XDG |
130 | 'authy_token': tfa_token, |
131 | 'remember_2fa': 'true', | |
5316566e | 132 | }) |
17b41a33 | 133 | |
d0e958c7 PH |
134 | def _prefer_source(self, formats): |
135 | try: | |
136 | source = next(f for f in formats if f['format_id'] == 'Source') | |
60e67c5b | 137 | source['quality'] = 10 |
d0e958c7 | 138 | except StopIteration: |
985637cb S |
139 | for f in formats: |
140 | if '/chunked/' in f['url']: | |
141 | f.update({ | |
60e67c5b | 142 | 'quality': 10, |
985637cb S |
143 | 'format_note': 'Source', |
144 | }) | |
d0e958c7 PH |
145 | self._sort_formats(formats) |
146 | ||
c5db6bb3 S |
147 | |
148 | class TwitchItemBaseIE(TwitchBaseIE): | |
149 | def _download_info(self, item, item_id): | |
e3f6b569 S |
150 | return self._extract_info(self._call_api( |
151 | 'kraken/videos/%s%s' % (item, item_id), item_id, | |
c5db6bb3 | 152 | 'Downloading %s info JSON' % self._ITEM_TYPE)) |
a22524b0 | 153 | |
c5db6bb3 S |
154 | def _extract_media(self, item_id): |
155 | info = self._download_info(self._ITEM_SHORTCUT, item_id) | |
e3f6b569 S |
156 | response = self._call_api( |
157 | 'api/videos/%s%s' % (self._ITEM_SHORTCUT, item_id), item_id, | |
c5db6bb3 | 158 | 'Downloading %s playlist JSON' % self._ITEM_TYPE) |
355d074f S |
159 | entries = [] |
160 | chunks = response['chunks'] | |
161 | qualities = list(chunks.keys()) | |
162 | for num, fragment in enumerate(zip(*chunks.values()), start=1): | |
163 | formats = [] | |
164 | for fmt_num, fragment_fmt in enumerate(fragment): | |
165 | format_id = qualities[fmt_num] | |
166 | fmt = { | |
167 | 'url': fragment_fmt['url'], | |
168 | 'format_id': format_id, | |
169 | 'quality': 1 if format_id == 'live' else 0, | |
170 | } | |
171 | m = re.search(r'^(?P<height>\d+)[Pp]', format_id) | |
172 | if m: | |
173 | fmt['height'] = int(m.group('height')) | |
174 | formats.append(fmt) | |
175 | self._sort_formats(formats) | |
176 | entry = dict(info) | |
159444a6 | 177 | entry['id'] = '%s_%d' % (entry['id'], num) |
355d074f S |
178 | entry['title'] = '%s part %d' % (entry['title'], num) |
179 | entry['formats'] = formats | |
180 | entries.append(entry) | |
013bfdd8 | 181 | return self.playlist_result(entries, info['id'], info['title']) |
355d074f S |
182 | |
183 | def _extract_info(self, info): | |
0ff51ada S |
184 | status = info.get('status') |
185 | if status == 'recording': | |
186 | is_live = True | |
187 | elif status == 'recorded': | |
188 | is_live = False | |
189 | else: | |
190 | is_live = None | |
355d074f S |
191 | return { |
192 | 'id': info['_id'], | |
369c12e0 | 193 | 'title': info.get('title') or 'Untitled Broadcast', |
7a6e8a1b S |
194 | 'description': info.get('description'), |
195 | 'duration': int_or_none(info.get('length')), | |
196 | 'thumbnail': info.get('preview'), | |
197 | 'uploader': info.get('channel', {}).get('display_name'), | |
198 | 'uploader_id': info.get('channel', {}).get('name'), | |
199 | 'timestamp': parse_iso8601(info.get('recorded_at')), | |
200 | 'view_count': int_or_none(info.get('views')), | |
0ff51ada | 201 | 'is_live': is_live, |
355d074f S |
202 | } |
203 | ||
c5db6bb3 S |
204 | def _real_extract(self, url): |
205 | return self._extract_media(self._match_id(url)) | |
04d02a9d | 206 | |
04d02a9d | 207 | |
c5db6bb3 S |
208 | class TwitchVideoIE(TwitchItemBaseIE): |
209 | IE_NAME = 'twitch:video' | |
93f78707 | 210 | _VALID_URL = r'%s/[^/]+/b/(?P<id>\d+)' % TwitchBaseIE._VALID_URL_BASE |
c5db6bb3 S |
211 | _ITEM_TYPE = 'video' |
212 | _ITEM_SHORTCUT = 'a' | |
04d02a9d | 213 | |
c5db6bb3 S |
214 | _TEST = { |
215 | 'url': 'http://www.twitch.tv/riotgames/b/577357806', | |
216 | 'info_dict': { | |
217 | 'id': 'a577357806', | |
218 | 'title': 'Worlds Semifinals - Star Horn Royal Club vs. OMG', | |
219 | }, | |
220 | 'playlist_mincount': 12, | |
0db3a661 | 221 | 'skip': 'HTTP Error 404: Not Found', |
c5db6bb3 | 222 | } |
04d02a9d | 223 | |
04d02a9d | 224 | |
c5db6bb3 S |
225 | class TwitchChapterIE(TwitchItemBaseIE): |
226 | IE_NAME = 'twitch:chapter' | |
93f78707 | 227 | _VALID_URL = r'%s/[^/]+/c/(?P<id>\d+)' % TwitchBaseIE._VALID_URL_BASE |
c5db6bb3 S |
228 | _ITEM_TYPE = 'chapter' |
229 | _ITEM_SHORTCUT = 'c' | |
04d02a9d | 230 | |
78111136 | 231 | _TESTS = [{ |
c5db6bb3 S |
232 | 'url': 'http://www.twitch.tv/acracingleague/c/5285812', |
233 | 'info_dict': { | |
234 | 'id': 'c5285812', | |
235 | 'title': 'ACRL Off Season - Sports Cars @ Nordschleife', | |
236 | }, | |
237 | 'playlist_mincount': 3, | |
0db3a661 | 238 | 'skip': 'HTTP Error 404: Not Found', |
78111136 PH |
239 | }, { |
240 | 'url': 'http://www.twitch.tv/tsm_theoddone/c/2349361', | |
241 | 'only_matching': True, | |
242 | }] | |
c5db6bb3 S |
243 | |
244 | ||
245 | class TwitchVodIE(TwitchItemBaseIE): | |
246 | IE_NAME = 'twitch:vod' | |
3f1ce168 S |
247 | _VALID_URL = r'''(?x) |
248 | https?:// | |
249 | (?: | |
644921b3 | 250 | (?:(?:www|go|m)\.)?twitch\.tv/(?:[^/]+/v(?:ideo)?|videos)/| |
bc48773e | 251 | player\.twitch\.tv/\?.*?\bvideo=v? |
3f1ce168 S |
252 | ) |
253 | (?P<id>\d+) | |
254 | ''' | |
c5db6bb3 S |
255 | _ITEM_TYPE = 'vod' |
256 | _ITEM_SHORTCUT = 'v' | |
257 | ||
9c724a98 | 258 | _TESTS = [{ |
e704f87f | 259 | 'url': 'http://www.twitch.tv/riotgames/v/6528877?t=5m10s', |
c5db6bb3 | 260 | 'info_dict': { |
ac0474f8 | 261 | 'id': 'v6528877', |
c5db6bb3 | 262 | 'ext': 'mp4', |
ac0474f8 | 263 | 'title': 'LCK Summer Split - Week 6 Day 1', |
ec85ded8 | 264 | 'thumbnail': r're:^https?://.*\.jpg$', |
ac0474f8 YCH |
265 | 'duration': 17208, |
266 | 'timestamp': 1435131709, | |
267 | 'upload_date': '20150624', | |
268 | 'uploader': 'Riot Games', | |
269 | 'uploader_id': 'riotgames', | |
c5db6bb3 | 270 | 'view_count': int, |
e704f87f | 271 | 'start_time': 310, |
c5db6bb3 S |
272 | }, |
273 | 'params': { | |
274 | # m3u8 download | |
275 | 'skip_download': True, | |
276 | }, | |
9c724a98 S |
277 | }, { |
278 | # Untitled broadcast (title is None) | |
279 | 'url': 'http://www.twitch.tv/belkao_o/v/11230755', | |
280 | 'info_dict': { | |
281 | 'id': 'v11230755', | |
282 | 'ext': 'mp4', | |
283 | 'title': 'Untitled Broadcast', | |
ec85ded8 | 284 | 'thumbnail': r're:^https?://.*\.jpg$', |
9c724a98 S |
285 | 'duration': 1638, |
286 | 'timestamp': 1439746708, | |
287 | 'upload_date': '20150816', | |
288 | 'uploader': 'BelkAO_o', | |
289 | 'uploader_id': 'belkao_o', | |
290 | 'view_count': int, | |
291 | }, | |
292 | 'params': { | |
293 | # m3u8 download | |
294 | 'skip_download': True, | |
295 | }, | |
9bd7bd0b | 296 | 'skip': 'HTTP Error 404: Not Found', |
3f1ce168 S |
297 | }, { |
298 | 'url': 'http://player.twitch.tv/?t=5m10s&video=v6528877', | |
299 | 'only_matching': True, | |
0b23c222 S |
300 | }, { |
301 | 'url': 'https://www.twitch.tv/videos/6528877', | |
302 | 'only_matching': True, | |
fce79626 S |
303 | }, { |
304 | 'url': 'https://m.twitch.tv/beagsandjam/v/247478721', | |
305 | 'only_matching': True, | |
644921b3 S |
306 | }, { |
307 | 'url': 'https://www.twitch.tv/northernlion/video/291940395', | |
308 | 'only_matching': True, | |
bc48773e M |
309 | }, { |
310 | 'url': 'https://player.twitch.tv/?video=480452374', | |
311 | 'only_matching': True, | |
9c724a98 | 312 | }] |
04d02a9d | 313 | |
79e93125 | 314 | def _real_extract(self, url): |
c5db6bb3 | 315 | item_id = self._match_id(url) |
e5e99661 | 316 | |
c5db6bb3 | 317 | info = self._download_info(self._ITEM_SHORTCUT, item_id) |
e3f6b569 S |
318 | access_token = self._call_api( |
319 | 'api/vods/%s/access_token' % item_id, item_id, | |
350c9481 | 320 | 'Downloading %s access token' % self._ITEM_TYPE) |
e5e99661 | 321 | |
c5db6bb3 | 322 | formats = self._extract_m3u8_formats( |
2da4316e | 323 | '%s/vod/%s.m3u8?%s' % ( |
350c9481 | 324 | self._USHER_BASE, item_id, |
15707c7e | 325 | compat_urllib_parse_urlencode({ |
e5e99661 | 326 | 'allow_source': 'true', |
ac455055 | 327 | 'allow_audio_only': 'true', |
e5e99661 S |
328 | 'allow_spectre': 'true', |
329 | 'player': 'twitchweb', | |
330 | 'nauth': access_token['token'], | |
331 | 'nauthsig': access_token['sig'], | |
332 | })), | |
631d4c87 | 333 | item_id, 'mp4', entry_protocol='m3u8_native') |
e5e99661 | 334 | |
d0e958c7 | 335 | self._prefer_source(formats) |
c5db6bb3 | 336 | info['formats'] = formats |
e704f87f NH |
337 | |
338 | parsed_url = compat_urllib_parse_urlparse(url) | |
339 | query = compat_parse_qs(parsed_url.query) | |
340 | if 't' in query: | |
341 | info['start_time'] = parse_duration(query['t'][0]) | |
342 | ||
264e77c4 S |
343 | if info.get('timestamp') is not None: |
344 | info['subtitles'] = { | |
345 | 'rechat': [{ | |
346 | 'url': update_url_query( | |
347 | 'https://rechat.twitch.tv/rechat-messages', { | |
348 | 'video_id': 'v%s' % item_id, | |
349 | 'start': info['timestamp'], | |
350 | }), | |
351 | 'ext': 'json', | |
352 | }], | |
353 | } | |
354 | ||
c5db6bb3 S |
355 | return info |
356 | ||
357 | ||
358 | class TwitchPlaylistBaseIE(TwitchBaseIE): | |
e3f6b569 | 359 | _PLAYLIST_PATH = 'kraken/channels/%s/videos/?offset=%d&limit=%d' |
ececca6c | 360 | _PAGE_LIMIT = 100 |
c5db6bb3 S |
361 | |
362 | def _extract_playlist(self, channel_id): | |
e3f6b569 S |
363 | info = self._call_api( |
364 | 'kraken/channels/%s' % channel_id, | |
c5db6bb3 S |
365 | channel_id, 'Downloading channel info JSON') |
366 | channel_name = info.get('display_name') or info.get('name') | |
367 | entries = [] | |
368 | offset = 0 | |
369 | limit = self._PAGE_LIMIT | |
539a1641 S |
370 | broken_paging_detected = False |
371 | counter_override = None | |
c5db6bb3 | 372 | for counter in itertools.count(1): |
e3f6b569 S |
373 | response = self._call_api( |
374 | self._PLAYLIST_PATH % (channel_id, offset, limit), | |
539a1641 | 375 | channel_id, |
93753aad | 376 | 'Downloading %s JSON page %s' |
539a1641 | 377 | % (self._PLAYLIST_TYPE, counter_override or counter)) |
c2e64f71 S |
378 | page_entries = self._extract_playlist_page(response) |
379 | if not page_entries: | |
c5db6bb3 | 380 | break |
539a1641 S |
381 | total = int_or_none(response.get('_total')) |
382 | # Since the beginning of March 2016 twitch's paging mechanism | |
383 | # is completely broken on the twitch side. It simply ignores | |
384 | # a limit and returns the whole offset number of videos. | |
385 | # Working around by just requesting all videos at once. | |
dec2cae0 | 386 | # Upd: pagination bug was fixed by twitch on 15.03.2016. |
539a1641 S |
387 | if not broken_paging_detected and total and len(page_entries) > limit: |
388 | self.report_warning( | |
dec2cae0 | 389 | 'Twitch pagination is broken on twitch side, requesting all videos at once', |
539a1641 S |
390 | channel_id) |
391 | broken_paging_detected = True | |
a8276b26 | 392 | offset = total |
539a1641 S |
393 | counter_override = '(all at once)' |
394 | continue | |
c2e64f71 | 395 | entries.extend(page_entries) |
539a1641 S |
396 | if broken_paging_detected or total and len(page_entries) >= total: |
397 | break | |
c5db6bb3 | 398 | offset += limit |
c2e64f71 | 399 | return self.playlist_result( |
f0c6c2bc | 400 | [self._make_url_result(entry) for entry in orderedSet(entries)], |
c2e64f71 S |
401 | channel_id, channel_name) |
402 | ||
f0c6c2bc LR |
403 | def _make_url_result(self, url): |
404 | try: | |
405 | video_id = 'v%s' % TwitchVodIE._match_id(url) | |
406 | return self.url_result(url, TwitchVodIE.ie_key(), video_id=video_id) | |
407 | except AssertionError: | |
408 | return self.url_result(url) | |
409 | ||
c2e64f71 S |
410 | def _extract_playlist_page(self, response): |
411 | videos = response.get('videos') | |
412 | return [video['url'] for video in videos] if videos else [] | |
c5db6bb3 S |
413 | |
414 | def _real_extract(self, url): | |
415 | return self._extract_playlist(self._match_id(url)) | |
416 | ||
417 | ||
418 | class TwitchProfileIE(TwitchPlaylistBaseIE): | |
419 | IE_NAME = 'twitch:profile' | |
420 | _VALID_URL = r'%s/(?P<id>[^/]+)/profile/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE | |
421 | _PLAYLIST_TYPE = 'profile' | |
422 | ||
fce79626 | 423 | _TESTS = [{ |
c5db6bb3 S |
424 | 'url': 'http://www.twitch.tv/vanillatv/profile', |
425 | 'info_dict': { | |
426 | 'id': 'vanillatv', | |
427 | 'title': 'VanillaTV', | |
428 | }, | |
429 | 'playlist_mincount': 412, | |
fce79626 S |
430 | }, { |
431 | 'url': 'http://m.twitch.tv/vanillatv/profile', | |
432 | 'only_matching': True, | |
433 | }] | |
c5db6bb3 S |
434 | |
435 | ||
93753aad S |
436 | class TwitchVideosBaseIE(TwitchPlaylistBaseIE): |
437 | _VALID_URL_VIDEOS_BASE = r'%s/(?P<id>[^/]+)/videos' % TwitchBaseIE._VALID_URL_BASE | |
438 | _PLAYLIST_PATH = TwitchPlaylistBaseIE._PLAYLIST_PATH + '&broadcast_type=' | |
439 | ||
440 | ||
441 | class TwitchAllVideosIE(TwitchVideosBaseIE): | |
442 | IE_NAME = 'twitch:videos:all' | |
443 | _VALID_URL = r'%s/all' % TwitchVideosBaseIE._VALID_URL_VIDEOS_BASE | |
444 | _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'archive,upload,highlight' | |
445 | _PLAYLIST_TYPE = 'all videos' | |
446 | ||
fce79626 | 447 | _TESTS = [{ |
93753aad S |
448 | 'url': 'https://www.twitch.tv/spamfish/videos/all', |
449 | 'info_dict': { | |
450 | 'id': 'spamfish', | |
451 | 'title': 'Spamfish', | |
452 | }, | |
453 | 'playlist_mincount': 869, | |
fce79626 S |
454 | }, { |
455 | 'url': 'https://m.twitch.tv/spamfish/videos/all', | |
456 | 'only_matching': True, | |
457 | }] | |
93753aad S |
458 | |
459 | ||
460 | class TwitchUploadsIE(TwitchVideosBaseIE): | |
461 | IE_NAME = 'twitch:videos:uploads' | |
462 | _VALID_URL = r'%s/uploads' % TwitchVideosBaseIE._VALID_URL_VIDEOS_BASE | |
463 | _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'upload' | |
464 | _PLAYLIST_TYPE = 'uploads' | |
465 | ||
fce79626 | 466 | _TESTS = [{ |
93753aad S |
467 | 'url': 'https://www.twitch.tv/spamfish/videos/uploads', |
468 | 'info_dict': { | |
469 | 'id': 'spamfish', | |
470 | 'title': 'Spamfish', | |
471 | }, | |
472 | 'playlist_mincount': 0, | |
fce79626 S |
473 | }, { |
474 | 'url': 'https://m.twitch.tv/spamfish/videos/uploads', | |
475 | 'only_matching': True, | |
476 | }] | |
93753aad S |
477 | |
478 | ||
479 | class TwitchPastBroadcastsIE(TwitchVideosBaseIE): | |
480 | IE_NAME = 'twitch:videos:past-broadcasts' | |
481 | _VALID_URL = r'%s/past-broadcasts' % TwitchVideosBaseIE._VALID_URL_VIDEOS_BASE | |
482 | _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'archive' | |
c5db6bb3 S |
483 | _PLAYLIST_TYPE = 'past broadcasts' |
484 | ||
fce79626 | 485 | _TESTS = [{ |
93753aad S |
486 | 'url': 'https://www.twitch.tv/spamfish/videos/past-broadcasts', |
487 | 'info_dict': { | |
488 | 'id': 'spamfish', | |
489 | 'title': 'Spamfish', | |
490 | }, | |
491 | 'playlist_mincount': 0, | |
fce79626 S |
492 | }, { |
493 | 'url': 'https://m.twitch.tv/spamfish/videos/past-broadcasts', | |
494 | 'only_matching': True, | |
495 | }] | |
93753aad S |
496 | |
497 | ||
498 | class TwitchHighlightsIE(TwitchVideosBaseIE): | |
499 | IE_NAME = 'twitch:videos:highlights' | |
500 | _VALID_URL = r'%s/highlights' % TwitchVideosBaseIE._VALID_URL_VIDEOS_BASE | |
501 | _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'highlight' | |
502 | _PLAYLIST_TYPE = 'highlights' | |
503 | ||
fce79626 | 504 | _TESTS = [{ |
93753aad | 505 | 'url': 'https://www.twitch.tv/spamfish/videos/highlights', |
c5db6bb3 S |
506 | 'info_dict': { |
507 | 'id': 'spamfish', | |
508 | 'title': 'Spamfish', | |
509 | }, | |
93753aad | 510 | 'playlist_mincount': 805, |
fce79626 S |
511 | }, { |
512 | 'url': 'https://m.twitch.tv/spamfish/videos/highlights', | |
513 | 'only_matching': True, | |
514 | }] | |
240b9b7a S |
515 | |
516 | ||
517 | class TwitchStreamIE(TwitchBaseIE): | |
518 | IE_NAME = 'twitch:stream' | |
9db8f6c5 S |
519 | _VALID_URL = r'''(?x) |
520 | https?:// | |
521 | (?: | |
fce79626 | 522 | (?:(?:www|go|m)\.)?twitch\.tv/| |
9db8f6c5 S |
523 | player\.twitch\.tv/\?.*?\bchannel= |
524 | ) | |
525 | (?P<id>[^/#?]+) | |
526 | ''' | |
240b9b7a | 527 | |
faa1f83a | 528 | _TESTS = [{ |
240b9b7a S |
529 | 'url': 'http://www.twitch.tv/shroomztv', |
530 | 'info_dict': { | |
531 | 'id': '12772022048', | |
532 | 'display_id': 'shroomztv', | |
533 | 'ext': 'mp4', | |
534 | 'title': 're:^ShroomzTV [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', | |
535 | 'description': 'H1Z1 - lonewolfing with ShroomzTV | A3 Battle Royale later - @ShroomzTV', | |
536 | 'is_live': True, | |
537 | 'timestamp': 1421928037, | |
538 | 'upload_date': '20150122', | |
539 | 'uploader': 'ShroomzTV', | |
540 | 'uploader_id': 'shroomztv', | |
541 | 'view_count': int, | |
542 | }, | |
543 | 'params': { | |
544 | # m3u8 download | |
545 | 'skip_download': True, | |
546 | }, | |
faa1f83a S |
547 | }, { |
548 | 'url': 'http://www.twitch.tv/miracle_doto#profile-0', | |
549 | 'only_matching': True, | |
9db8f6c5 S |
550 | }, { |
551 | 'url': 'https://player.twitch.tv/?channel=lotsofs', | |
552 | 'only_matching': True, | |
b763e1d6 S |
553 | }, { |
554 | 'url': 'https://go.twitch.tv/food', | |
555 | 'only_matching': True, | |
fce79626 S |
556 | }, { |
557 | 'url': 'https://m.twitch.tv/food', | |
558 | 'only_matching': True, | |
faa1f83a | 559 | }] |
240b9b7a | 560 | |
9db8f6c5 S |
561 | @classmethod |
562 | def suitable(cls, url): | |
563 | return (False | |
564 | if any(ie.suitable(url) for ie in ( | |
565 | TwitchVideoIE, | |
566 | TwitchChapterIE, | |
567 | TwitchVodIE, | |
568 | TwitchProfileIE, | |
569 | TwitchAllVideosIE, | |
570 | TwitchUploadsIE, | |
571 | TwitchPastBroadcastsIE, | |
db348e88 S |
572 | TwitchHighlightsIE, |
573 | TwitchClipsIE)) | |
9db8f6c5 S |
574 | else super(TwitchStreamIE, cls).suitable(url)) |
575 | ||
240b9b7a S |
576 | def _real_extract(self, url): |
577 | channel_id = self._match_id(url) | |
578 | ||
e3f6b569 | 579 | stream = self._call_api( |
9aa929d3 | 580 | 'kraken/streams/%s?stream_type=all' % channel_id, channel_id, |
240b9b7a S |
581 | 'Downloading stream JSON').get('stream') |
582 | ||
240b9b7a | 583 | if not stream: |
cc764a6d | 584 | raise ExtractorError('%s is offline' % channel_id, expected=True) |
240b9b7a | 585 | |
06966677 S |
586 | # Channel name may be typed if different case than the original channel name |
587 | # (e.g. http://www.twitch.tv/TWITCHPLAYSPOKEMON) that will lead to constructing | |
588 | # an invalid m3u8 URL. Working around by use of original channel name from stream | |
589 | # JSON and fallback to lowercase if it's not available. | |
1793d71d S |
590 | channel_id = stream.get('channel', {}).get('name') or channel_id.lower() |
591 | ||
e3f6b569 S |
592 | access_token = self._call_api( |
593 | 'api/channels/%s/access_token' % channel_id, channel_id, | |
240b9b7a S |
594 | 'Downloading channel access token') |
595 | ||
596 | query = { | |
597 | 'allow_source': 'true', | |
ac455055 | 598 | 'allow_audio_only': 'true', |
9aa929d3 | 599 | 'allow_spectre': 'true', |
f353cbdb | 600 | 'p': random.randint(1000000, 10000000), |
240b9b7a S |
601 | 'player': 'twitchweb', |
602 | 'segment_preference': '4', | |
d34e7949 S |
603 | 'sig': access_token['sig'].encode('utf-8'), |
604 | 'token': access_token['token'].encode('utf-8'), | |
240b9b7a | 605 | } |
240b9b7a S |
606 | formats = self._extract_m3u8_formats( |
607 | '%s/api/channel/hls/%s.m3u8?%s' | |
15707c7e | 608 | % (self._USHER_BASE, channel_id, compat_urllib_parse_urlencode(query)), |
240b9b7a | 609 | channel_id, 'mp4') |
d0e958c7 | 610 | self._prefer_source(formats) |
240b9b7a S |
611 | |
612 | view_count = stream.get('viewers') | |
613 | timestamp = parse_iso8601(stream.get('created_at')) | |
614 | ||
615 | channel = stream['channel'] | |
616 | title = self._live_title(channel.get('display_name') or channel.get('name')) | |
617 | description = channel.get('status') | |
618 | ||
619 | thumbnails = [] | |
620 | for thumbnail_key, thumbnail_url in stream['preview'].items(): | |
621 | m = re.search(r'(?P<width>\d+)x(?P<height>\d+)\.jpg$', thumbnail_key) | |
622 | if not m: | |
623 | continue | |
624 | thumbnails.append({ | |
625 | 'url': thumbnail_url, | |
626 | 'width': int(m.group('width')), | |
627 | 'height': int(m.group('height')), | |
628 | }) | |
629 | ||
630 | return { | |
631 | 'id': compat_str(stream['_id']), | |
632 | 'display_id': channel_id, | |
633 | 'title': title, | |
634 | 'description': description, | |
635 | 'thumbnails': thumbnails, | |
636 | 'uploader': channel.get('display_name'), | |
637 | 'uploader_id': channel.get('name'), | |
638 | 'timestamp': timestamp, | |
639 | 'view_count': view_count, | |
640 | 'formats': formats, | |
641 | 'is_live': True, | |
12d1fb5a | 642 | } |
778f9694 S |
643 | |
644 | ||
49fa7de3 | 645 | class TwitchClipsIE(TwitchBaseIE): |
778f9694 | 646 | IE_NAME = 'twitch:clips' |
db348e88 | 647 | _VALID_URL = r'https?://(?:clips\.twitch\.tv/(?:[^/]+/)*|(?:www\.)?twitch\.tv/[^/]+/clip/)(?P<id>[^/?#&]+)' |
778f9694 | 648 | |
74ba450a | 649 | _TESTS = [{ |
49fa7de3 | 650 | 'url': 'https://clips.twitch.tv/FaintLightGullWholeWheat', |
778f9694 S |
651 | 'md5': '761769e1eafce0ffebfb4089cb3847cd', |
652 | 'info_dict': { | |
49fa7de3 | 653 | 'id': '42850523', |
778f9694 S |
654 | 'ext': 'mp4', |
655 | 'title': 'EA Play 2016 Live from the Novo Theatre', | |
ec85ded8 | 656 | 'thumbnail': r're:^https?://.*\.jpg', |
49fa7de3 S |
657 | 'timestamp': 1465767393, |
658 | 'upload_date': '20160612', | |
778f9694 S |
659 | 'creator': 'EA', |
660 | 'uploader': 'stereotype_', | |
49fa7de3 | 661 | 'uploader_id': '43566419', |
778f9694 | 662 | }, |
74ba450a S |
663 | }, { |
664 | # multiple formats | |
665 | 'url': 'https://clips.twitch.tv/rflegendary/UninterestedBeeDAESuppy', | |
666 | 'only_matching': True, | |
db348e88 S |
667 | }, { |
668 | 'url': 'https://www.twitch.tv/sergeynixon/clip/StormyThankfulSproutFutureMan', | |
669 | 'only_matching': True, | |
74ba450a | 670 | }] |
778f9694 S |
671 | |
672 | def _real_extract(self, url): | |
673 | video_id = self._match_id(url) | |
674 | ||
49fa7de3 S |
675 | status = self._download_json( |
676 | 'https://clips.twitch.tv/api/v2/clips/%s/status' % video_id, | |
677 | video_id) | |
778f9694 | 678 | |
49fa7de3 | 679 | formats = [] |
778f9694 | 680 | |
49fa7de3 S |
681 | for option in status['quality_options']: |
682 | if not isinstance(option, dict): | |
683 | continue | |
3052a30d S |
684 | source = url_or_none(option.get('source')) |
685 | if not source: | |
49fa7de3 S |
686 | continue |
687 | formats.append({ | |
688 | 'url': source, | |
689 | 'format_id': option.get('quality'), | |
690 | 'height': int_or_none(option.get('quality')), | |
691 | 'fps': int_or_none(option.get('frame_rate')), | |
692 | }) | |
74ba450a | 693 | |
07acdc5a S |
694 | self._sort_formats(formats) |
695 | ||
49fa7de3 S |
696 | info = { |
697 | 'formats': formats, | |
698 | } | |
74ba450a | 699 | |
49fa7de3 S |
700 | clip = self._call_api( |
701 | 'kraken/clips/%s' % video_id, video_id, fatal=False, headers={ | |
702 | 'Accept': 'application/vnd.twitchtv.v5+json', | |
703 | }) | |
778f9694 | 704 | |
49fa7de3 S |
705 | if clip: |
706 | quality_key = qualities(('tiny', 'small', 'medium')) | |
707 | thumbnails = [] | |
708 | thumbnails_dict = clip.get('thumbnails') | |
709 | if isinstance(thumbnails_dict, dict): | |
710 | for thumbnail_id, thumbnail_url in thumbnails_dict.items(): | |
711 | thumbnails.append({ | |
712 | 'id': thumbnail_id, | |
713 | 'url': thumbnail_url, | |
714 | 'preference': quality_key(thumbnail_id), | |
715 | }) | |
716 | ||
717 | info.update({ | |
718 | 'id': clip.get('tracking_id') or video_id, | |
719 | 'title': clip.get('title') or video_id, | |
720 | 'duration': float_or_none(clip.get('duration')), | |
721 | 'views': int_or_none(clip.get('views')), | |
722 | 'timestamp': unified_timestamp(clip.get('created_at')), | |
723 | 'thumbnails': thumbnails, | |
724 | 'creator': try_get(clip, lambda x: x['broadcaster']['display_name'], compat_str), | |
725 | 'uploader': try_get(clip, lambda x: x['curator']['display_name'], compat_str), | |
726 | 'uploader_id': try_get(clip, lambda x: x['curator']['id'], compat_str), | |
727 | }) | |
728 | else: | |
729 | info.update({ | |
730 | 'title': video_id, | |
731 | 'id': video_id, | |
732 | }) | |
0cacae28 | 733 | |
49fa7de3 | 734 | return info |