2 from __future__ import unicode_literals
10 from .common import InfoExtractor
11 from ..compat import compat_urllib_parse
12 from ..utils import ExtractorError
15 class IqiyiIE(InfoExtractor):
19 _VALID_URL = r'http://(?:[^.]+\.)?iqiyi\.com/.+\.html'
22 'url': 'http://www.iqiyi.com/v_19rrojlavg.html',
23 'md5': '2cb594dc2781e6c941a110d8f358118b',
25 'id': '9c1fb1b99d192b21c559e5a1a2cb3c73',
26 'title': '美国德州空中惊现奇异云团 酷似UFO',
30 'url': 'http://www.iqiyi.com/v_19rrhnnclk.html',
32 'id': 'e3f585b550a280af23c98b6cb2be19fb',
33 'title': '名侦探柯南第752集',
37 'id': 'e3f585b550a280af23c98b6cb2be19fb_part1',
39 'title': '名侦探柯南第752集',
43 'id': 'e3f585b550a280af23c98b6cb2be19fb_part2',
45 'title': '名侦探柯南第752集',
49 'id': 'e3f585b550a280af23c98b6cb2be19fb_part3',
51 'title': '名侦探柯南第752集',
55 'id': 'e3f585b550a280af23c98b6cb2be19fb_part4',
57 'title': '名侦探柯南第752集',
61 'id': 'e3f585b550a280af23c98b6cb2be19fb_part5',
63 'title': '名侦探柯南第752集',
67 'id': 'e3f585b550a280af23c98b6cb2be19fb_part6',
69 'title': '名侦探柯南第752集',
73 'id': 'e3f585b550a280af23c98b6cb2be19fb_part7',
75 'title': '名侦探柯南第752集',
79 'id': 'e3f585b550a280af23c98b6cb2be19fb_part8',
81 'title': '名侦探柯南第752集',
85 'skip_download': True,
88 'url': 'http://www.iqiyi.com/w_19rt6o8t9p.html',
89 'only_matching': True,
91 'url': 'http://www.iqiyi.com/a_19rrhbc6kt.html',
92 'only_matching': True,
94 'url': 'http://yule.iqiyi.com/pcb.html',
95 'only_matching': True,
109 return hashlib.md5(text.encode('utf-8')).hexdigest()
111 def construct_video_urls(self, data, video_id, _uuid):
120 def get_encode_code(l):
125 for i in range(c - 1, -1, -1):
126 a = do_xor(int(b[c - i - 1], 16), i)
130 def get_path_key(x, format_id, segment_index):
131 mg = ')(*&^flash@#$%a'
132 tm = self._download_json(
133 'http://data.video.qiyi.com/t?tn=' + str(random.random()), video_id,
134 note='Download path key of segment %d for format %s' % (segment_index + 1, format_id)
136 t = str(int(math.floor(int(tm) / (600.0))))
137 return self.md5_text(t + mg + x)
140 for format_item in data['vp']['tkl'][0]['vs']:
141 if 0 < int(format_item['bid']) <= 10:
142 format_id = self.get_format(format_item['bid'])
148 video_urls_info = format_item['fs']
149 if not format_item['fs'][0]['l'].startswith('/'):
150 t = get_encode_code(format_item['fs'][0]['l'])
151 if t.endswith('mp4'):
152 video_urls_info = format_item['flvs']
154 for segment_index, segment in enumerate(video_urls_info):
156 if not vl.startswith('/'):
157 vl = get_encode_code(vl)
159 vl.split('/')[-1].split('.')[0], format_id, segment_index)
160 filesize = segment['b']
161 base_url = data['vp']['du'].split('/')
162 base_url.insert(-1, key)
163 base_url = '/'.join(base_url)
166 'qyid': uuid.uuid4().hex,
171 'tn': str(int(time.time()))
173 api_video_url = base_url + vl + '?' + \
174 compat_urllib_parse.urlencode(param)
175 js = self._download_json(
176 api_video_url, video_id,
177 note='Download video info of segment %d for format %s' % (segment_index + 1, format_id))
180 (video_url, filesize))
182 video_urls_dict[format_id] = video_urls
183 return video_urls_dict
185 def get_format(self, bid):
186 matched_format_ids = [_format_id for _bid, _format_id in self._FORMATS_MAP if _bid == str(bid)]
187 return matched_format_ids[0] if len(matched_format_ids) else None
189 def get_bid(self, format_id):
190 matched_bids = [_bid for _bid, _format_id in self._FORMATS_MAP if _format_id == format_id]
191 return matched_bids[0] if len(matched_bids) else None
193 def get_raw_data(self, tvid, video_id, enc_key, _uuid):
194 tm = str(int(time.time()))
198 'src': self.md5_text('youtube-dl'),
203 'enc': self.md5_text(enc_key + tail),
205 'tn': random.random(),
207 'authkey': self.md5_text(self.md5_text('') + tail),
210 api_url = 'http://cache.video.qiyi.com/vms' + '?' + \
211 compat_urllib_parse.urlencode(param)
212 raw_data = self._download_json(api_url, video_id)
215 def get_enc_key(self, swf_url, video_id):
216 # TODO: automatic key extraction
217 # last update at 2016-01-22 for Zombie::bite
218 enc_key = '6ab6d0280511493ba85594779759d4ed'
221 def _real_extract(self, url):
222 webpage = self._download_webpage(
223 url, 'temp_id', note='download video page')
224 tvid = self._search_regex(
225 r'data-player-tvid\s*=\s*[\'"](\d+)', webpage, 'tvid')
226 video_id = self._search_regex(
227 r'data-player-videoid\s*=\s*[\'"]([a-f\d]+)', webpage, 'video_id')
228 swf_url = self._search_regex(
229 r'(http://[^\'"]+MainPlayer[^.]+\.swf)', webpage, 'swf player URL')
230 _uuid = uuid.uuid4().hex
232 enc_key = self.get_enc_key(swf_url, video_id)
234 raw_data = self.get_raw_data(tvid, video_id, enc_key, _uuid)
236 if raw_data['code'] != 'A000000':
237 raise ExtractorError('Unable to load data. Error code: ' + raw_data['code'])
239 if not raw_data['data']['vp']['tkl']:
240 raise ExtractorError('No support iQiqy VIP video')
242 data = raw_data['data']
244 title = data['vi']['vn']
246 # generate video_urls_dict
247 video_urls_dict = self.construct_video_urls(
248 data, video_id, _uuid)
252 for format_id in video_urls_dict:
253 video_urls = video_urls_dict[format_id]
254 for i, video_url_info in enumerate(video_urls):
255 if len(entries) < i + 1:
256 entries.append({'formats': []})
257 entries[i]['formats'].append(
259 'url': video_url_info[0],
260 'filesize': video_url_info[-1],
261 'format_id': format_id,
262 'preference': int(self.get_bid(format_id))
266 for i in range(len(entries)):
267 self._sort_formats(entries[i]['formats'])
270 'id': '%s_part%d' % (video_id, i + 1),
277 '_type': 'multi_video',
284 info['id'] = video_id
285 info['title'] = title