2 from __future__ import unicode_literals
13 from .common import InfoExtractor
14 from ..compat import compat_urllib_parse
21 class IqiyiIE(InfoExtractor):
25 _VALID_URL = r'http://(?:www\.)iqiyi.com/v_.+?\.html'
28 'url': 'http://www.iqiyi.com/v_19rrojlavg.html',
29 'md5': '2cb594dc2781e6c941a110d8f358118b',
31 'id': '9c1fb1b99d192b21c559e5a1a2cb3c73',
32 'title': '美国德州空中惊现奇异云团 酷似UFO',
36 'url': 'http://www.iqiyi.com/v_19rrhnnclk.html',
38 'id': 'e3f585b550a280af23c98b6cb2be19fb',
39 'title': '名侦探柯南第752集',
42 'md5': '7e49376fecaffa115d951634917fe105',
44 'id': 'e3f585b550a280af23c98b6cb2be19fb_part1',
46 'title': '名侦探柯南第752集',
49 'md5': '41b75ba13bb7ac0e411131f92bc4f6ca',
51 'id': 'e3f585b550a280af23c98b6cb2be19fb_part2',
53 'title': '名侦探柯南第752集',
56 'md5': '0cee1dd0a3d46a83e71e2badeae2aab0',
58 'id': 'e3f585b550a280af23c98b6cb2be19fb_part3',
60 'title': '名侦探柯南第752集',
63 'md5': '4f8ad72373b0c491b582e7c196b0b1f9',
65 'id': 'e3f585b550a280af23c98b6cb2be19fb_part4',
67 'title': '名侦探柯南第752集',
70 'md5': 'd89ad028bcfad282918e8098e811711d',
72 'id': 'e3f585b550a280af23c98b6cb2be19fb_part5',
74 'title': '名侦探柯南第752集',
77 'md5': '9cb1e5c95da25dff0660c32ae50903b7',
79 'id': 'e3f585b550a280af23c98b6cb2be19fb_part6',
81 'title': '名侦探柯南第752集',
84 'md5': '155116e0ff1867bbc9b98df294faabc9',
86 'id': 'e3f585b550a280af23c98b6cb2be19fb_part7',
88 'title': '名侦探柯南第752集',
91 'md5': '53f5db77622ae14fa493ed2a278a082b',
93 'id': 'e3f585b550a280af23c98b6cb2be19fb_part8',
95 'title': '名侦探柯南第752集',
109 def construct_video_urls(self, data, video_id, _uuid):
118 def get_encode_code(l):
123 for i in range(c - 1, -1, -1):
124 a = do_xor(int(b[c - i - 1], 16), i)
128 def get_path_key(x, format_id, segment_index):
129 mg = ')(*&^flash@#$%a'
130 tm = self._download_json(
131 'http://data.video.qiyi.com/t?tn=' + str(random.random()), video_id,
132 note='Download path key of segment %d for format %s' % (segment_index + 1, format_id)
134 t = str(int(math.floor(int(tm) / (600.0))))
135 return hashlib.md5((t + mg + x).encode('utf8')).hexdigest()
138 for format_item in data['vp']['tkl'][0]['vs']:
139 if 0 < int(format_item['bid']) <= 10:
140 format_id = self.get_format(format_item['bid'])
146 video_urls_info = format_item['fs']
147 if not format_item['fs'][0]['l'].startswith('/'):
148 t = get_encode_code(format_item['fs'][0]['l'])
149 if t.endswith('mp4'):
150 video_urls_info = format_item['flvs']
152 for segment_index, segment in enumerate(video_urls_info):
154 if not vl.startswith('/'):
155 vl = get_encode_code(vl)
157 vl.split('/')[-1].split('.')[0], format_id, segment_index)
158 filesize = segment['b']
159 base_url = data['vp']['du'].split('/')
160 base_url.insert(-1, key)
161 base_url = '/'.join(base_url)
164 'qyid': uuid.uuid4().hex,
169 'tn': str(int(time.time()))
171 api_video_url = base_url + vl + '?' + \
172 compat_urllib_parse.urlencode(param)
173 js = self._download_json(
174 api_video_url, video_id,
175 note='Download video info of segment %d for format %s' % (segment_index + 1, format_id))
178 (video_url, filesize))
180 video_urls_dict[format_id] = video_urls
181 return video_urls_dict
183 def get_format(self, bid):
184 matched_format_ids = [_format_id for _bid, _format_id in self._FORMATS_MAP if _bid == str(bid)]
185 return matched_format_ids[0] if len(matched_format_ids) else None
187 def get_bid(self, format_id):
188 matched_bids = [_bid for _bid, _format_id in self._FORMATS_MAP if _format_id == format_id]
189 return matched_bids[0] if len(matched_bids) else None
191 def get_raw_data(self, tvid, video_id, enc_key, _uuid):
192 tm = str(int(time.time()))
195 'src': hashlib.md5(b'youtube-dl').hexdigest(),
201 (enc_key + tm + tvid).encode('utf8')).hexdigest(),
203 'tn': random.random(),
205 'authkey': hashlib.md5(
206 (tm + tvid).encode('utf8')).hexdigest()
209 api_url = 'http://cache.video.qiyi.com/vms' + '?' + \
210 compat_urllib_parse.urlencode(param)
211 raw_data = self._download_json(api_url, video_id)
214 def get_enc_key(self, swf_url, video_id):
215 filename, _ = os.path.splitext(url_basename(swf_url))
216 enc_key_json = self._downloader.cache.load('iqiyi-enc-key', filename)
217 if enc_key_json is not None:
218 return enc_key_json[0]
220 req = self._request_webpage(
221 swf_url, video_id, note='download swf content')
223 cn = zlib.decompress(cn[8:])
224 pt = re.compile(b'MixerRemote\x08(?P<enc_key>.+?)\$&vv')
225 enc_key = self._search_regex(pt, cn, 'enc_key').decode('utf8')
227 self._downloader.cache.store('iqiyi-enc-key', filename, [enc_key])
231 def _real_extract(self, url):
232 webpage = self._download_webpage(
233 url, 'temp_id', note='download video page')
234 tvid = self._search_regex(
235 r'data-player-tvid\s*=\s*[\'"](\d+)', webpage, 'tvid')
236 video_id = self._search_regex(
237 r'data-player-videoid\s*=\s*[\'"]([a-f\d]+)', webpage, 'video_id')
238 swf_url = self._search_regex(
239 r'(http://[^\'"]+MainPlayer[^.]+\.swf)', webpage, 'swf player URL')
240 _uuid = uuid.uuid4().hex
242 enc_key = self.get_enc_key(swf_url, video_id)
244 raw_data = self.get_raw_data(tvid, video_id, enc_key, _uuid)
246 if raw_data['code'] != 'A000000':
247 raise ExtractorError('Unable to load data. Error code: ' + raw_data['code'])
249 if not raw_data['data']['vp']['tkl']:
250 raise ExtractorError('No support iQiqy VIP video')
252 data = raw_data['data']
254 title = data['vi']['vn']
256 # generate video_urls_dict
257 video_urls_dict = self.construct_video_urls(
258 data, video_id, _uuid)
262 for format_id in video_urls_dict:
263 video_urls = video_urls_dict[format_id]
264 for i, video_url_info in enumerate(video_urls):
265 if len(entries) < i + 1:
266 entries.append({'formats': []})
267 entries[i]['formats'].append(
269 'url': video_url_info[0],
270 'filesize': video_url_info[-1],
271 'format_id': format_id,
272 'preference': int(self.get_bid(format_id))
276 for i in range(len(entries)):
277 self._sort_formats(entries[i]['formats'])
280 'id': '%s_part%d' % (video_id, i + 1),
287 '_type': 'multi_video',
294 info['id'] = video_id
295 info['title'] = title