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://(?:www\.)iqiyi.com/v_.+?\.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,
98 def construct_video_urls(self, data, video_id, _uuid):
107 def get_encode_code(l):
112 for i in range(c - 1, -1, -1):
113 a = do_xor(int(b[c - i - 1], 16), i)
117 def get_path_key(x, format_id, segment_index):
118 mg = ')(*&^flash@#$%a'
119 tm = self._download_json(
120 'http://data.video.qiyi.com/t?tn=' + str(random.random()), video_id,
121 note='Download path key of segment %d for format %s' % (segment_index + 1, format_id)
123 t = str(int(math.floor(int(tm) / (600.0))))
124 return hashlib.md5((t + mg + x).encode('utf8')).hexdigest()
127 for format_item in data['vp']['tkl'][0]['vs']:
128 if 0 < int(format_item['bid']) <= 10:
129 format_id = self.get_format(format_item['bid'])
135 video_urls_info = format_item['fs']
136 if not format_item['fs'][0]['l'].startswith('/'):
137 t = get_encode_code(format_item['fs'][0]['l'])
138 if t.endswith('mp4'):
139 video_urls_info = format_item['flvs']
141 for segment_index, segment in enumerate(video_urls_info):
143 if not vl.startswith('/'):
144 vl = get_encode_code(vl)
146 vl.split('/')[-1].split('.')[0], format_id, segment_index)
147 filesize = segment['b']
148 base_url = data['vp']['du'].split('/')
149 base_url.insert(-1, key)
150 base_url = '/'.join(base_url)
153 'qyid': uuid.uuid4().hex,
158 'tn': str(int(time.time()))
160 api_video_url = base_url + vl + '?' + \
161 compat_urllib_parse.urlencode(param)
162 js = self._download_json(
163 api_video_url, video_id,
164 note='Download video info of segment %d for format %s' % (segment_index + 1, format_id))
167 (video_url, filesize))
169 video_urls_dict[format_id] = video_urls
170 return video_urls_dict
172 def get_format(self, bid):
173 matched_format_ids = [_format_id for _bid, _format_id in self._FORMATS_MAP if _bid == str(bid)]
174 return matched_format_ids[0] if len(matched_format_ids) else None
176 def get_bid(self, format_id):
177 matched_bids = [_bid for _bid, _format_id in self._FORMATS_MAP if _format_id == format_id]
178 return matched_bids[0] if len(matched_bids) else None
180 def get_raw_data(self, tvid, video_id, enc_key, _uuid):
181 tm = str(int(time.time()))
184 'src': hashlib.md5(b'youtube-dl').hexdigest(),
190 (enc_key + tm + tvid).encode('utf8')).hexdigest(),
192 'tn': random.random(),
194 'authkey': hashlib.md5(
195 (tm + tvid).encode('utf8')).hexdigest()
198 api_url = 'http://cache.video.qiyi.com/vms' + '?' + \
199 compat_urllib_parse.urlencode(param)
200 raw_data = self._download_json(api_url, video_id)
203 def get_enc_key(self, swf_url, video_id):
204 enc_key = '8e29ab5666d041c3a1ea76e06dabdffb'
207 def _real_extract(self, url):
208 webpage = self._download_webpage(
209 url, 'temp_id', note='download video page')
210 tvid = self._search_regex(
211 r'data-player-tvid\s*=\s*[\'"](\d+)', webpage, 'tvid')
212 video_id = self._search_regex(
213 r'data-player-videoid\s*=\s*[\'"]([a-f\d]+)', webpage, 'video_id')
214 swf_url = self._search_regex(
215 r'(http://[^\'"]+MainPlayer[^.]+\.swf)', webpage, 'swf player URL')
216 _uuid = uuid.uuid4().hex
218 enc_key = self.get_enc_key(swf_url, video_id)
220 raw_data = self.get_raw_data(tvid, video_id, enc_key, _uuid)
222 if raw_data['code'] != 'A000000':
223 raise ExtractorError('Unable to load data. Error code: ' + raw_data['code'])
225 if not raw_data['data']['vp']['tkl']:
226 raise ExtractorError('No support iQiqy VIP video')
228 data = raw_data['data']
230 title = data['vi']['vn']
232 # generate video_urls_dict
233 video_urls_dict = self.construct_video_urls(
234 data, video_id, _uuid)
238 for format_id in video_urls_dict:
239 video_urls = video_urls_dict[format_id]
240 for i, video_url_info in enumerate(video_urls):
241 if len(entries) < i + 1:
242 entries.append({'formats': []})
243 entries[i]['formats'].append(
245 'url': video_url_info[0],
246 'filesize': video_url_info[-1],
247 'format_id': format_id,
248 'preference': int(self.get_bid(format_id))
252 for i in range(len(entries)):
253 self._sort_formats(entries[i]['formats'])
256 'id': '%s_part%d' % (video_id, i + 1),
263 '_type': 'multi_video',
270 info['id'] = video_id
271 info['title'] = title