2 from __future__ import unicode_literals
13 from .common import InfoExtractor
14 from ..compat import compat_urllib_parse
21 class IqiyiIE(InfoExtractor):
24 _VALID_URL = r'http://(?:www\.)iqiyi.com/v_.+?\.html'
27 'url': 'http://www.iqiyi.com/v_19rrojlavg.html',
28 'md5': '2cb594dc2781e6c941a110d8f358118b',
30 'id': '9c1fb1b99d192b21c559e5a1a2cb3c73',
31 'title': '美国德州空中惊现奇异云团 酷似UFO',
35 'url': 'http://www.iqiyi.com/v_19rrhnnclk.html',
37 'id': 'e3f585b550a280af23c98b6cb2be19fb',
38 'title': '名侦探柯南第752集',
41 'md5': '7e49376fecaffa115d951634917fe105',
43 'id': 'e3f585b550a280af23c98b6cb2be19fb_part1',
45 'title': '名侦探柯南第752集',
48 'md5': '41b75ba13bb7ac0e411131f92bc4f6ca',
50 'id': 'e3f585b550a280af23c98b6cb2be19fb_part2',
52 'title': '名侦探柯南第752集',
55 'md5': '0cee1dd0a3d46a83e71e2badeae2aab0',
57 'id': 'e3f585b550a280af23c98b6cb2be19fb_part3',
59 'title': '名侦探柯南第752集',
62 'md5': '4f8ad72373b0c491b582e7c196b0b1f9',
64 'id': 'e3f585b550a280af23c98b6cb2be19fb_part4',
66 'title': '名侦探柯南第752集',
69 'md5': 'd89ad028bcfad282918e8098e811711d',
71 'id': 'e3f585b550a280af23c98b6cb2be19fb_part5',
73 'title': '名侦探柯南第752集',
76 'md5': '9cb1e5c95da25dff0660c32ae50903b7',
78 'id': 'e3f585b550a280af23c98b6cb2be19fb_part6',
80 'title': '名侦探柯南第752集',
83 'md5': '155116e0ff1867bbc9b98df294faabc9',
85 'id': 'e3f585b550a280af23c98b6cb2be19fb_part7',
87 'title': '名侦探柯南第752集',
90 'md5': '53f5db77622ae14fa493ed2a278a082b',
92 'id': 'e3f585b550a280af23c98b6cb2be19fb_part8',
94 'title': '名侦探柯南第752集',
108 def construct_video_urls(self, data, video_id, _uuid):
117 def get_encode_code(l):
122 for i in range(c - 1, -1, -1):
123 a = do_xor(int(b[c - i - 1], 16), i)
127 def get_path_key(x, format_id, segment_index):
128 mg = ')(*&^flash@#$%a'
129 tm = self._download_json(
130 'http://data.video.qiyi.com/t?tn=' + str(random.random()), video_id,
131 note='Download path key of segment %d for format %s' % (segment_index + 1, format_id)
133 t = str(int(math.floor(int(tm) / (600.0))))
134 return hashlib.md5((t + mg + x).encode('utf8')).hexdigest()
137 for format_item in data['vp']['tkl'][0]['vs']:
138 if 0 < int(format_item['bid']) <= 10:
139 format_id = self.get_format(format_item['bid'])
145 video_urls_info = format_item['fs']
146 if not format_item['fs'][0]['l'].startswith('/'):
147 t = get_encode_code(format_item['fs'][0]['l'])
148 if t.endswith('mp4'):
149 video_urls_info = format_item['flvs']
151 for segment_index, segment in enumerate(video_urls_info):
153 if not vl.startswith('/'):
154 vl = get_encode_code(vl)
156 vl.split('/')[-1].split('.')[0], format_id, segment_index)
157 filesize = segment['b']
158 base_url = data['vp']['du'].split('/')
159 base_url.insert(-1, key)
160 base_url = '/'.join(base_url)
163 'qyid': uuid.uuid4().hex,
168 'tn': str(int(time.time()))
170 api_video_url = base_url + vl + '?' + \
171 compat_urllib_parse.urlencode(param)
172 js = self._download_json(
173 api_video_url, video_id,
174 note='Download video info of segment %d for format %s' % (segment_index + 1, format_id))
177 (video_url, filesize))
179 video_urls_dict[format_id] = video_urls
180 return video_urls_dict
182 def get_format(self, bid):
183 matched_format_ids = [_format_id for _bid, _format_id in self._FORMATS_MAP if _bid == str(bid)]
184 return matched_format_ids[0] if len(matched_format_ids) else None
186 def get_bid(self, format_id):
187 matched_bids = [_bid for _bid, _format_id in self._FORMATS_MAP if _format_id == format_id]
188 return matched_bids[0] if len(matched_bids) else None
190 def get_raw_data(self, tvid, video_id, enc_key, _uuid):
191 tm = str(int(time.time()))
194 'src': hashlib.md5(b'youtube-dl').hexdigest(),
200 (enc_key + tm + tvid).encode('utf8')).hexdigest(),
202 'tn': random.random(),
204 'authkey': hashlib.md5(
205 (tm + tvid).encode('utf8')).hexdigest()
208 api_url = 'http://cache.video.qiyi.com/vms' + '?' + \
209 compat_urllib_parse.urlencode(param)
210 raw_data = self._download_json(api_url, video_id)
213 def get_enc_key(self, swf_url, video_id):
214 filename, _ = os.path.splitext(url_basename(swf_url))
215 enc_key_json = self._downloader.cache.load('iqiyi-enc-key', filename)
216 if enc_key_json is not None:
217 return enc_key_json[0]
219 req = self._request_webpage(
220 swf_url, video_id, note='download swf content')
222 cn = zlib.decompress(cn[8:])
223 pt = re.compile(b'MixerRemote\x08(?P<enc_key>.+?)\$&vv')
224 enc_key = self._search_regex(pt, cn, 'enc_key').decode('utf8')
226 self._downloader.cache.store('iqiyi-enc-key', filename, [enc_key])
230 def _real_extract(self, url):
231 webpage = self._download_webpage(
232 url, 'temp_id', note='download video page')
233 tvid = self._search_regex(
234 r'data-player-tvid\s*=\s*[\'"](\d+)', webpage, 'tvid')
235 video_id = self._search_regex(
236 r'data-player-videoid\s*=\s*[\'"]([a-f\d]+)', webpage, 'video_id')
237 swf_url = self._search_regex(
238 r'(http://[^\'"]+MainPlayer[^.]+\.swf)', webpage, 'swf player URL')
239 _uuid = uuid.uuid4().hex
241 enc_key = self.get_enc_key(swf_url, video_id)
243 raw_data = self.get_raw_data(tvid, video_id, enc_key, _uuid)
245 if raw_data['code'] != 'A000000':
246 raise ExtractorError('Unable to load data. Error code: ' + raw_data['code'])
248 if not raw_data['data']['vp']['tkl']:
249 raise ExtractorError('No support iQiqy VIP video')
251 data = raw_data['data']
253 title = data['vi']['vn']
255 # generate video_urls_dict
256 video_urls_dict = self.construct_video_urls(
257 data, video_id, _uuid)
261 for format_id in video_urls_dict:
262 video_urls = video_urls_dict[format_id]
263 for i, video_url_info in enumerate(video_urls):
264 if len(entries) < i + 1:
265 entries.append({'formats': []})
266 entries[i]['formats'].append(
268 'url': video_url_info[0],
269 'filesize': video_url_info[-1],
270 'format_id': format_id,
271 'preference': int(self.get_bid(format_id))
275 for i in range(len(entries)):
276 self._sort_formats(entries[i]['formats'])
279 'id': '%s_part%d' % (video_id, i + 1),
286 '_type': 'multi_video',
293 info['id'] = video_id
294 info['title'] = title