# coding: utf-8
-
from __future__ import unicode_literals
-from .common import InfoExtractor
+import hashlib
+import math
+import random
+import time
+import uuid
+from .common import InfoExtractor
from ..compat import compat_urllib_parse
-
from ..utils import ExtractorError
-import re
-import time
-import uuid
-import math
-import random
-import zlib
-import hashlib
-
class IqiyiIE(InfoExtractor):
IE_NAME = 'iqiyi'
+ IE_DESC = '爱奇艺'
- _VALID_URL = r'http://(?:www\.)iqiyi.com/.+?\.html'
+ _VALID_URL = r'http://(?:www\.)iqiyi.com/v_.+?\.html'
- _TEST = {
+ _TESTS = [{
'url': 'http://www.iqiyi.com/v_19rrojlavg.html',
'md5': '2cb594dc2781e6c941a110d8f358118b',
'info_dict': {
'title': '美国德州空中惊现奇异云团 酷似UFO',
'ext': 'f4v',
}
- }
-
- def construct_video_urls(self, data, video_id, _uuid, bid):
+ }, {
+ 'url': 'http://www.iqiyi.com/v_19rrhnnclk.html',
+ 'info_dict': {
+ 'id': 'e3f585b550a280af23c98b6cb2be19fb',
+ 'title': '名侦探柯南第752集',
+ },
+ 'playlist': [{
+ 'info_dict': {
+ 'id': 'e3f585b550a280af23c98b6cb2be19fb_part1',
+ 'ext': 'f4v',
+ 'title': '名侦探柯南第752集',
+ },
+ }, {
+ 'info_dict': {
+ 'id': 'e3f585b550a280af23c98b6cb2be19fb_part2',
+ 'ext': 'f4v',
+ 'title': '名侦探柯南第752集',
+ },
+ }, {
+ 'info_dict': {
+ 'id': 'e3f585b550a280af23c98b6cb2be19fb_part3',
+ 'ext': 'f4v',
+ 'title': '名侦探柯南第752集',
+ },
+ }, {
+ 'info_dict': {
+ 'id': 'e3f585b550a280af23c98b6cb2be19fb_part4',
+ 'ext': 'f4v',
+ 'title': '名侦探柯南第752集',
+ },
+ }, {
+ 'info_dict': {
+ 'id': 'e3f585b550a280af23c98b6cb2be19fb_part5',
+ 'ext': 'f4v',
+ 'title': '名侦探柯南第752集',
+ },
+ }, {
+ 'info_dict': {
+ 'id': 'e3f585b550a280af23c98b6cb2be19fb_part6',
+ 'ext': 'f4v',
+ 'title': '名侦探柯南第752集',
+ },
+ }, {
+ 'info_dict': {
+ 'id': 'e3f585b550a280af23c98b6cb2be19fb_part7',
+ 'ext': 'f4v',
+ 'title': '名侦探柯南第752集',
+ },
+ }, {
+ 'info_dict': {
+ 'id': 'e3f585b550a280af23c98b6cb2be19fb_part8',
+ 'ext': 'f4v',
+ 'title': '名侦探柯南第752集',
+ },
+ }],
+ 'params': {
+ 'skip_download': True,
+ },
+ }]
+
+ _FORMATS_MAP = [
+ ('1', 'h6'),
+ ('2', 'h5'),
+ ('3', 'h4'),
+ ('4', 'h3'),
+ ('5', 'h2'),
+ ('10', 'h1'),
+ ]
+
+ @staticmethod
+ def md5_text(text):
+ return hashlib.md5(text.encode('utf-8')).hexdigest()
+
+ def construct_video_urls(self, data, video_id, _uuid):
def do_xor(x, y):
a = y % 3
if a == 1:
s += chr(a)
return s[::-1]
- def get_path_key(x):
+ def get_path_key(x, format_id, segment_index):
mg = ')(*&^flash@#$%a'
tm = self._download_json(
- 'http://data.video.qiyi.com/t?tn=' + str(random.random()), video_id)['t']
+ 'http://data.video.qiyi.com/t?tn=' + str(random.random()), video_id,
+ note='Download path key of segment %d for format %s' % (segment_index + 1, format_id)
+ )['t']
t = str(int(math.floor(int(tm) / (600.0))))
- return hashlib.md5((t + mg + x).encode('utf8')).hexdigest()
-
- # get accept format
- # getting all format will spend minutes for a big video.
- if bid == 'best':
- bids = [int(i['bid']) for i in data['vp']['tkl'][0]['vs']
- if 0 < int(i['bid']) <= 10]
- bid = str(max(bids))
+ return self.md5_text(t + mg + x)
video_urls_dict = {}
- for i in data['vp']['tkl'][0]['vs']:
- if 0 < int(i['bid']) <= 10:
- format_id = self.get_format(i['bid'])
+ for format_item in data['vp']['tkl'][0]['vs']:
+ if 0 < int(format_item['bid']) <= 10:
+ format_id = self.get_format(format_item['bid'])
else:
continue
video_urls = []
- video_urls_info = i['fs']
- if not i['fs'][0]['l'].startswith('/'):
- t = get_encode_code(i['fs'][0]['l'])
+ video_urls_info = format_item['fs']
+ if not format_item['fs'][0]['l'].startswith('/'):
+ t = get_encode_code(format_item['fs'][0]['l'])
if t.endswith('mp4'):
- video_urls_info = i['flvs']
-
- if int(i['bid']) != int(bid): # ignore missing match format
- video_urls.extend(
- [('http://example.com/v.flv', ii['b']) for ii in video_urls_info])
- video_urls_dict[format_id] = video_urls
- continue
+ video_urls_info = format_item['flvs']
- for ii in video_urls_info:
- vl = ii['l']
+ for segment_index, segment in enumerate(video_urls_info):
+ vl = segment['l']
if not vl.startswith('/'):
vl = get_encode_code(vl)
key = get_path_key(
- vl.split('/')[-1].split('.')[0])
- filesize = ii['b']
+ vl.split('/')[-1].split('.')[0], format_id, segment_index)
+ filesize = segment['b']
base_url = data['vp']['du'].split('/')
base_url.insert(-1, key)
base_url = '/'.join(base_url)
}
api_video_url = base_url + vl + '?' + \
compat_urllib_parse.urlencode(param)
- js = self._download_json(api_video_url, video_id)
+ js = self._download_json(
+ api_video_url, video_id,
+ note='Download video info of segment %d for format %s' % (segment_index + 1, format_id))
video_url = js['l']
video_urls.append(
(video_url, filesize))
return video_urls_dict
def get_format(self, bid):
- _dict = {
- '1': 'h6',
- '2': 'h5',
- '3': 'h4',
- '4': 'h3',
- '5': 'h2',
- '10': 'h1'
- }
- return _dict.get(str(bid), None)
+ matched_format_ids = [_format_id for _bid, _format_id in self._FORMATS_MAP if _bid == str(bid)]
+ return matched_format_ids[0] if len(matched_format_ids) else None
def get_bid(self, format_id):
- _dict = {
- 'h6': '1',
- 'h5': '2',
- 'h4': '3',
- 'h3': '4',
- 'h2': '5',
- 'h1': '10',
- 'best': 'best'
- }
- return _dict.get(format_id, None)
+ matched_bids = [_bid for _bid, _format_id in self._FORMATS_MAP if _format_id == format_id]
+ return matched_bids[0] if len(matched_bids) else None
def get_raw_data(self, tvid, video_id, enc_key, _uuid):
tm = str(int(time.time()))
+ tail = tm + tvid
param = {
'key': 'fvip',
- 'src': hashlib.md5(b'youtube-dl').hexdigest(),
+ 'src': self.md5_text('youtube-dl'),
'tvId': tvid,
'vid': video_id,
'vinfo': 1,
'tm': tm,
- 'enc': hashlib.md5(
- (enc_key + tm + tvid).encode('utf8')).hexdigest(),
+ 'enc': self.md5_text(enc_key + tail),
'qyid': _uuid,
'tn': random.random(),
'um': 0,
- 'authkey': hashlib.md5(
- (tm + tvid).encode('utf8')).hexdigest()
+ 'authkey': self.md5_text(self.md5_text('') + tail),
}
api_url = 'http://cache.video.qiyi.com/vms' + '?' + \
return raw_data
def get_enc_key(self, swf_url, video_id):
- req = self._request_webpage(
- swf_url, video_id, note='download swf content')
- cn = req.read()
- cn = zlib.decompress(cn[8:])
- pt = re.compile(b'MixerRemote\x08(?P<enc_key>.+?)\$&vv')
- enc_key = self._search_regex(pt, cn, 'enc_key').decode('utf8')
+ # TODO: automatic key extraction
+ # last update at 2015-10-22 for Zombie::bite
+ # '7223c67061dbea1259d0ceb44f44b6d62288f4f80c972170de5201d2321060270e05'[2:66][0::2]
+ enc_key = '2c76de15dcb44bd28ff0927d50d31620'
return enc_key
def _real_extract(self, url):
webpage = self._download_webpage(
url, 'temp_id', note='download video page')
tvid = self._search_regex(
- r'tvId ?= ?(\'|\")(?P<tvid>\d+)', webpage, 'tvid', flags=re.I, group='tvid')
+ r'data-player-tvid\s*=\s*[\'"](\d+)', webpage, 'tvid')
video_id = self._search_regex(
- r'videoId ?= ?(\'|\")(?P<video_id>[a-z\d]+)',
- webpage, 'video_id', flags=re.I, group='video_id')
+ r'data-player-videoid\s*=\s*[\'"]([a-f\d]+)', webpage, 'video_id')
swf_url = self._search_regex(
- r'(?P<swf>http://.+?MainPlayer.+?\.swf)', webpage, 'swf')
+ r'(http://[^\'"]+MainPlayer[^.]+\.swf)', webpage, 'swf player URL')
_uuid = uuid.uuid4().hex
enc_key = self.get_enc_key(swf_url, video_id)
raw_data = self.get_raw_data(tvid, video_id, enc_key, _uuid)
- assert raw_data['code'] == 'A000000'
+
+ if raw_data['code'] != 'A000000':
+ raise ExtractorError('Unable to load data. Error code: ' + raw_data['code'])
+
if not raw_data['data']['vp']['tkl']:
raise ExtractorError('No support iQiqy VIP video')
title = data['vi']['vn']
- format = self._downloader.params.get('format', None)
- bid = self.get_bid(format) if format else 'best'
- if not bid:
- raise ExtractorError('Can\'t get format.')
-
# generate video_urls_dict
video_urls_dict = self.construct_video_urls(
- data, video_id, _uuid, bid)
+ data, video_id, _uuid)
# construct info
entries = []
self._sort_formats(entries[i]['formats'])
entries[i].update(
{
- 'id': '_part%d' % (i + 1),
+ 'id': '%s_part%d' % (video_id, i + 1),
'title': title,
}
)