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,
100 return hashlib.md5(text.encode('utf-8')).hexdigest()
102 def construct_video_urls(self, data, video_id, _uuid):
111 def get_encode_code(l):
116 for i in range(c - 1, -1, -1):
117 a = do_xor(int(b[c - i - 1], 16), i)
121 def get_path_key(x, format_id, segment_index):
122 mg = ')(*&^flash@#$%a'
123 tm = self._download_json(
124 'http://data.video.qiyi.com/t?tn=' + str(random.random()), video_id,
125 note='Download path key of segment %d for format %s' % (segment_index + 1, format_id)
127 t = str(int(math.floor(int(tm) / (600.0))))
128 return self.md5_text(t + mg + x)
131 for format_item in data['vp']['tkl'][0]['vs']:
132 if 0 < int(format_item['bid']) <= 10:
133 format_id = self.get_format(format_item['bid'])
139 video_urls_info = format_item['fs']
140 if not format_item['fs'][0]['l'].startswith('/'):
141 t = get_encode_code(format_item['fs'][0]['l'])
142 if t.endswith('mp4'):
143 video_urls_info = format_item['flvs']
145 for segment_index, segment in enumerate(video_urls_info):
147 if not vl.startswith('/'):
148 vl = get_encode_code(vl)
150 vl.split('/')[-1].split('.')[0], format_id, segment_index)
151 filesize = segment['b']
152 base_url = data['vp']['du'].split('/')
153 base_url.insert(-1, key)
154 base_url = '/'.join(base_url)
157 'qyid': uuid.uuid4().hex,
162 'tn': str(int(time.time()))
164 api_video_url = base_url + vl + '?' + \
165 compat_urllib_parse.urlencode(param)
166 js = self._download_json(
167 api_video_url, video_id,
168 note='Download video info of segment %d for format %s' % (segment_index + 1, format_id))
171 (video_url, filesize))
173 video_urls_dict[format_id] = video_urls
174 return video_urls_dict
176 def get_format(self, bid):
177 matched_format_ids = [_format_id for _bid, _format_id in self._FORMATS_MAP if _bid == str(bid)]
178 return matched_format_ids[0] if len(matched_format_ids) else None
180 def get_bid(self, format_id):
181 matched_bids = [_bid for _bid, _format_id in self._FORMATS_MAP if _format_id == format_id]
182 return matched_bids[0] if len(matched_bids) else None
184 def get_raw_data(self, tvid, video_id, enc_key, _uuid):
185 tm = str(int(time.time()))
189 'src': self.md5_text('youtube-dl'),
194 'enc': self.md5_text((enc_key + tail)[1:64:2] + tail),
196 'tn': random.random(),
198 'authkey': self.md5_text(self.md5_text('') + tail),
201 api_url = 'http://cache.video.qiyi.com/vms' + '?' + \
202 compat_urllib_parse.urlencode(param)
203 raw_data = self._download_json(api_url, video_id)
206 def get_enc_key(self, swf_url, video_id):
207 # TODO: automatic key extraction
208 enc_key = 'eac64f22daf001da6ba9aa8da4d501508bbe90a4d4091fea3b0582a85b38c2cc' # last update at 2015-09-23-23 for Zombie::bite
211 def _real_extract(self, url):
212 webpage = self._download_webpage(
213 url, 'temp_id', note='download video page')
214 tvid = self._search_regex(
215 r'data-player-tvid\s*=\s*[\'"](\d+)', webpage, 'tvid')
216 video_id = self._search_regex(
217 r'data-player-videoid\s*=\s*[\'"]([a-f\d]+)', webpage, 'video_id')
218 swf_url = self._search_regex(
219 r'(http://[^\'"]+MainPlayer[^.]+\.swf)', webpage, 'swf player URL')
220 _uuid = uuid.uuid4().hex
222 enc_key = self.get_enc_key(swf_url, video_id)
224 raw_data = self.get_raw_data(tvid, video_id, enc_key, _uuid)
226 if raw_data['code'] != 'A000000':
227 raise ExtractorError('Unable to load data. Error code: ' + raw_data['code'])
229 if not raw_data['data']['vp']['tkl']:
230 raise ExtractorError('No support iQiqy VIP video')
232 data = raw_data['data']
234 title = data['vi']['vn']
236 # generate video_urls_dict
237 video_urls_dict = self.construct_video_urls(
238 data, video_id, _uuid)
242 for format_id in video_urls_dict:
243 video_urls = video_urls_dict[format_id]
244 for i, video_url_info in enumerate(video_urls):
245 if len(entries) < i + 1:
246 entries.append({'formats': []})
247 entries[i]['formats'].append(
249 'url': video_url_info[0],
250 'filesize': video_url_info[-1],
251 'format_id': format_id,
252 'preference': int(self.get_bid(format_id))
256 for i in range(len(entries)):
257 self._sort_formats(entries[i]['formats'])
260 'id': '%s_part%d' % (video_id, i + 1),
267 '_type': 'multi_video',
274 info['id'] = video_id
275 info['title'] = title