]> jfr.im git - yt-dlp.git/blame - youtube_dl/extractor/letv.py
[extractor/generic] Add support for tnaflix network embeds (Closes #7505)
[yt-dlp.git] / youtube_dl / extractor / letv.py
CommitLineData
7f09a662
YCH
1# coding: utf-8
2from __future__ import unicode_literals
3
265bfa2c 4import datetime
57031161 5import re
7f09a662 6import time
5f432ac8 7import base64
7b7507d6 8import hashlib
7f09a662
YCH
9
10from .common import InfoExtractor
265bfa2c 11from ..compat import (
265bfa2c 12 compat_urllib_parse,
1e399778 13 compat_ord,
7b7507d6 14 compat_str,
265bfa2c
PH
15)
16from ..utils import (
17 determine_ext,
18 ExtractorError,
19 parse_iso8601,
5c2266df 20 sanitized_Request,
593ddd85 21 int_or_none,
5f432ac8 22 str_or_none,
1e399778 23 encode_data_uri,
bec30224 24 url_basename,
265bfa2c 25)
7f09a662
YCH
26
27
28class LetvIE(InfoExtractor):
963d0ce7 29 IE_DESC = '乐视网'
57031161 30 _VALID_URL = r'http://www\.letv\.com/ptv/vplay/(?P<id>\d+).html'
7f09a662
YCH
31
32 _TESTS = [{
33 'url': 'http://www.letv.com/ptv/vplay/22005890.html',
1e399778 34 'md5': 'edadcfe5406976f42f9f266057ee5e40',
7f09a662
YCH
35 'info_dict': {
36 'id': '22005890',
37 'ext': 'mp4',
38 'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家',
67706359 39 'description': 'md5:a9cb175fd753e2962176b7beca21a47c',
1e399778
YCH
40 },
41 'params': {
42 'hls_prefer_native': True,
43 },
7f09a662 44 }, {
67706359 45 'url': 'http://www.letv.com/ptv/vplay/1415246.html',
7f09a662 46 'info_dict': {
67706359 47 'id': '1415246',
7f09a662 48 'ext': 'mp4',
67706359
YCH
49 'title': '美人天下01',
50 'description': 'md5:f88573d9d7225ada1359eaf0dbf8bcda',
51 },
1e399778
YCH
52 'params': {
53 'hls_prefer_native': True,
54 },
91410c9b
PH
55 }, {
56 'note': 'This video is available only in Mainland China, thus a proxy is needed',
57 'url': 'http://www.letv.com/ptv/vplay/1118082.html',
1e399778 58 'md5': '2424c74948a62e5f31988438979c5ad1',
91410c9b
PH
59 'info_dict': {
60 'id': '1118082',
61 'ext': 'mp4',
62 'title': '与龙共舞 完整版',
63 'description': 'md5:7506a5eeb1722bb9d4068f85024e3986',
64 },
1e399778
YCH
65 'params': {
66 'hls_prefer_native': True,
67 },
051df9ad 68 'skip': 'Only available in China',
7f09a662
YCH
69 }]
70
71 @staticmethod
72 def urshift(val, n):
73 return val >> n if val >= 0 else (val + 0x100000000) >> n
74
265bfa2c 75 # ror() and calc_time_key() are reversed from a embedded swf file in KLetvPlayer.swf
7f09a662
YCH
76 def ror(self, param1, param2):
77 _loc3_ = 0
78 while _loc3_ < param2:
79 param1 = self.urshift(param1, 1) + ((param1 & 1) << 31)
80 _loc3_ += 1
81 return param1
82
265bfa2c 83 def calc_time_key(self, param1):
7f09a662
YCH
84 _loc2_ = 773625421
85 _loc3_ = self.ror(param1, _loc2_ % 13)
86 _loc3_ = _loc3_ ^ _loc2_
87 _loc3_ = self.ror(_loc3_, _loc2_ % 17)
88 return _loc3_
89
1e399778
YCH
90 # see M3U8Encryption class in KLetvPlayer.swf
91 @staticmethod
92 def decrypt_m3u8(encrypted_data):
93 if encrypted_data[:5].decode('utf-8').lower() != 'vc_01':
94 return encrypted_data
95 encrypted_data = encrypted_data[5:]
96
2ebd2eac
YCH
97 _loc4_ = bytearray(2 * len(encrypted_data))
98 for idx, val in enumerate(encrypted_data):
99 b = compat_ord(val)
100 _loc4_[2 * idx] = b // 16
101 _loc4_[2 * idx + 1] = b % 16
1e399778
YCH
102 idx = len(_loc4_) - 11
103 _loc4_ = _loc4_[idx:] + _loc4_[:idx]
2ebd2eac
YCH
104 _loc7_ = bytearray(len(encrypted_data))
105 for i in range(len(encrypted_data)):
106 _loc7_[i] = _loc4_[2 * i] * 16 + _loc4_[2 * i + 1]
1e399778
YCH
107
108 return bytes(_loc7_)
109
7f09a662
YCH
110 def _real_extract(self, url):
111 media_id = self._match_id(url)
112 page = self._download_webpage(url, media_id)
113 params = {
114 'id': media_id,
115 'platid': 1,
116 'splatid': 101,
117 'format': 1,
265bfa2c 118 'tkey': self.calc_time_key(int(time.time())),
7f09a662
YCH
119 'domain': 'www.letv.com'
120 }
5c2266df 121 play_json_req = sanitized_Request(
91410c9b
PH
122 'http://api.letv.com/mms/out/video/playJson?' + compat_urllib_parse.urlencode(params)
123 )
63fc8000
YCH
124 cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
125 if cn_verification_proxy:
126 play_json_req.add_header('Ytdl-request-proxy', cn_verification_proxy)
127
7f09a662 128 play_json = self._download_json(
91410c9b 129 play_json_req,
576904bc 130 media_id, 'Downloading playJson data')
7f09a662
YCH
131
132 # Check for errors
133 playstatus = play_json['playstatus']
134 if playstatus['status'] == 0:
135 flag = playstatus['flag']
136 if flag == 1:
137 msg = 'Country %s auth error' % playstatus['country']
138 else:
139 msg = 'Generic error. flag = %d' % flag
140 raise ExtractorError(msg, expected=True)
141
142 playurl = play_json['playurl']
143
144 formats = ['350', '1000', '1300', '720p', '1080p']
145 dispatch = playurl['dispatch']
146
147 urls = []
148 for format_id in formats:
149 if format_id in dispatch:
150 media_url = playurl['domain'][0] + dispatch[format_id][0]
1e399778
YCH
151 media_url += '&' + compat_urllib_parse.urlencode({
152 'm3v': 1,
153 'format': 1,
154 'expect': 3,
155 'rateid': format_id,
7f09a662 156 })
1e399778
YCH
157
158 nodes_data = self._download_json(
159 media_url, media_id,
160 'Download JSON metadata for format %s' % format_id)
161
162 req = self._request_webpage(
163 nodes_data['nodelist'][0]['location'], media_id,
164 note='Downloading m3u8 information for format %s' % format_id)
165
166 m3u8_data = self.decrypt_m3u8(req.read())
7f09a662
YCH
167
168 url_info_dict = {
05a3879f 169 'url': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
91410c9b
PH
170 'ext': determine_ext(dispatch[format_id][1]),
171 'format_id': format_id,
1e399778 172 'protocol': 'm3u8',
7f09a662
YCH
173 }
174
175 if format_id[-1:] == 'p':
593ddd85 176 url_info_dict['height'] = int_or_none(format_id[:-1])
7f09a662
YCH
177
178 urls.append(url_info_dict)
179
180 publish_time = parse_iso8601(self._html_search_regex(
91410c9b 181 r'发布时间&nbsp;([^<>]+) ', page, 'publish time', default=None),
7f09a662 182 delimiter=' ', timezone=datetime.timedelta(hours=8))
67706359 183 description = self._html_search_meta('description', page, fatal=False)
7f09a662
YCH
184
185 return {
186 'id': media_id,
187 'formats': urls,
188 'title': playurl['title'],
189 'thumbnail': playurl['pic'],
67706359 190 'description': description,
7f09a662
YCH
191 'timestamp': publish_time,
192 }
57031161
YCH
193
194
195class LetvTvIE(InfoExtractor):
196 _VALID_URL = r'http://www.letv.com/tv/(?P<id>\d+).html'
197 _TESTS = [{
198 'url': 'http://www.letv.com/tv/46177.html',
199 'info_dict': {
200 'id': '46177',
201 'title': '美人天下',
202 'description': 'md5:395666ff41b44080396e59570dbac01c'
203 },
204 'playlist_count': 35
205 }]
206
207 def _real_extract(self, url):
208 playlist_id = self._match_id(url)
209 page = self._download_webpage(url, playlist_id)
210
211 media_urls = list(set(re.findall(
212 r'http://www.letv.com/ptv/vplay/\d+.html', page)))
213 entries = [self.url_result(media_url, ie='Letv')
214 for media_url in media_urls]
215
67706359
YCH
216 title = self._html_search_meta('keywords', page,
217 fatal=False).split(',')[0]
57031161
YCH
218 description = self._html_search_meta('description', page, fatal=False)
219
220 return self.playlist_result(entries, playlist_id, playlist_title=title,
221 playlist_description=description)
222
223
224class LetvPlaylistIE(LetvTvIE):
225 _VALID_URL = r'http://tv.letv.com/[a-z]+/(?P<id>[a-z]+)/index.s?html'
226 _TESTS = [{
227 'url': 'http://tv.letv.com/izt/wuzetian/index.html',
228 'info_dict': {
229 'id': 'wuzetian',
230 'title': '武媚娘传奇',
231 'description': 'md5:e12499475ab3d50219e5bba00b3cb248'
232 },
67706359
YCH
233 # This playlist contains some extra videos other than the drama itself
234 'playlist_mincount': 96
57031161
YCH
235 }, {
236 'url': 'http://tv.letv.com/pzt/lswjzzjc/index.shtml',
237 'info_dict': {
238 'id': 'lswjzzjc',
67706359
YCH
239 # The title should be "劲舞青春", but I can't find a simple way to
240 # determine the playlist title
57031161
YCH
241 'title': '乐视午间自制剧场',
242 'description': 'md5:b1eef244f45589a7b5b1af9ff25a4489'
243 },
244 'playlist_mincount': 7
245 }]
5f432ac8
FF
246
247
248class LetvCloudIE(InfoExtractor):
249 IE_DESC = '乐视云'
73e74424 250 _VALID_URL = r'https?://yuntv\.letv\.com/bcloud.html\?.+'
5f432ac8
FF
251
252 _TESTS = [{
253 'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=467623dedf',
254 'md5': '26450599afd64c513bc77030ad15db44',
255 'info_dict': {
256 'id': 'p7jnfw5hw9_467623dedf',
257 'ext': 'mp4',
0428106d 258 'title': 'Video p7jnfw5hw9_467623dedf',
5f432ac8
FF
259 },
260 }, {
261 'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=ec93197892&pu=2c7cd40209&auto_play=1&gpcflag=1&width=640&height=360',
7b7507d6 262 'md5': 'e03d9cc8d9c13191e1caf277e42dbd31',
5f432ac8
FF
263 'info_dict': {
264 'id': 'p7jnfw5hw9_ec93197892',
265 'ext': 'mp4',
0428106d 266 'title': 'Video p7jnfw5hw9_ec93197892',
5f432ac8
FF
267 },
268 }, {
269 'url': 'http://yuntv.letv.com/bcloud.html?uu=p7jnfw5hw9&vu=187060b6fd',
7b7507d6 270 'md5': 'cb988699a776b22d4a41b9d43acfb3ac',
5f432ac8
FF
271 'info_dict': {
272 'id': 'p7jnfw5hw9_187060b6fd',
273 'ext': 'mp4',
0428106d 274 'title': 'Video p7jnfw5hw9_187060b6fd',
5f432ac8
FF
275 },
276 }]
277
7b7507d6
YCH
278 @staticmethod
279 def sign_data(obj):
280 if obj['cf'] == 'flash':
281 salt = '2f9d6924b33a165a6d8b5d3d42f4f987'
282 items = ['cf', 'format', 'ran', 'uu', 'ver', 'vu']
283 elif obj['cf'] == 'html5':
284 salt = 'fbeh5player12c43eccf2bec3300344'
285 items = ['cf', 'ran', 'uu', 'bver', 'vu']
286 input_data = ''.join([item + obj[item] for item in items]) + salt
287 obj['sign'] = hashlib.md5(input_data.encode('utf-8')).hexdigest()
288
289 def _get_formats(self, cf, uu, vu, media_id):
290 def get_play_json(cf, timestamp):
291 data = {
292 'cf': cf,
293 'ver': '2.2',
294 'bver': 'firefox44.0',
295 'format': 'json',
296 'uu': uu,
297 'vu': vu,
298 'ran': compat_str(timestamp),
299 }
300 self.sign_data(data)
301 return self._download_json(
302 'http://api.letvcloud.com/gpc.php?' + compat_urllib_parse.urlencode(data),
303 media_id, 'Downloading playJson data for type %s' % cf)
304
305 play_json = get_play_json(cf, time.time())
306 # The server time may be different from local time
307 if play_json.get('code') == 10071:
308 play_json = get_play_json(cf, play_json['timestamp'])
5f432ac8 309
26de1bba
YCH
310 if not play_json.get('data'):
311 if play_json.get('message'):
312 raise ExtractorError('Letv cloud said: %s' % play_json['message'], expected=True)
313 elif play_json.get('code'):
314 raise ExtractorError('Letv cloud returned error %d' % play_json['code'], expected=True)
315 else:
316 raise ExtractorError('Letv cloud returned an unknwon error')
317
bec30224
YCH
318 def b64decode(s):
319 return base64.b64decode(s.encode('utf-8')).decode('utf-8')
320
10defdd0
YCH
321 formats = []
322 for media in play_json['data']['video_info']['media'].values():
323 play_url = media['play_url']
bec30224
YCH
324 url = b64decode(play_url['main_url'])
325 decoded_url = b64decode(url_basename(url))
10defdd0 326 formats.append({
bec30224
YCH
327 'url': url,
328 'ext': determine_ext(decoded_url),
10defdd0
YCH
329 'format_id': int_or_none(play_url.get('vtype')),
330 'format_note': str_or_none(play_url.get('definition')),
331 'width': int_or_none(play_url.get('vwidth')),
332 'height': int_or_none(play_url.get('vheight')),
333 })
7b7507d6
YCH
334
335 return formats
336
337 def _real_extract(self, url):
338 uu_mobj = re.search('uu=([\w]+)', url)
339 vu_mobj = re.search('vu=([\w]+)', url)
340
341 if not uu_mobj or not vu_mobj:
342 raise ExtractorError('Invalid URL: %s' % url, expected=True)
343
344 uu = uu_mobj.group(1)
345 vu = vu_mobj.group(1)
346 media_id = uu + '_' + vu
347
348 formats = self._get_formats('flash', uu, vu, media_id) + self._get_formats('html5', uu, vu, media_id)
5f432ac8
FF
349 self._sort_formats(formats)
350
351 return {
352 'id': media_id,
0428106d 353 'title': 'Video %s' % media_id,
5f432ac8
FF
354 'formats': formats,
355 }