mirror of
https://github.com/ytdl-org/youtube-dl
synced 2024-11-24 00:36:52 +01:00
[iqiyi] Add new extractor for iqiyi.com
This commit is contained in:
PeterDing
parent
f0bfaa2d7d
commit
605ec701b7
@ -229,6 +229,7 @@ from .infoq import InfoQIE
|
||||
from .instagram import InstagramIE, InstagramUserIE
|
||||
from .internetvideoarchive import InternetVideoArchiveIE
|
||||
from .iprima import IPrimaIE
|
||||
from .iqiyi import IqiyiIE
|
||||
from .ivi import (
|
||||
IviIE,
|
||||
IviCompilationIE
|
||||
|
||||
214
youtube_dl/extractor/iqiyi.py
Normal file
214
youtube_dl/extractor/iqiyi.py
Normal file
@ -0,0 +1,214 @@
|
||||
# coding: utf-8
|
||||
|
||||
from __future__ import unicode_literals
|
||||
|
||||
from .common import InfoExtractor
|
||||
|
||||
from ..compat import (
|
||||
compat_chr,
|
||||
compat_parse_qs,
|
||||
compat_urllib_parse,
|
||||
compat_urllib_request,
|
||||
compat_urlparse,
|
||||
compat_str,
|
||||
)
|
||||
|
||||
from ..utils import ExtractorError
|
||||
|
||||
import re
|
||||
import time
|
||||
import json
|
||||
import uuid
|
||||
import math
|
||||
import random
|
||||
import zlib
|
||||
import hashlib
|
||||
|
||||
class IqiyiIE(InfoExtractor):
|
||||
IE_NAME = 'iqiyi'
|
||||
|
||||
_VALID_URL = r'http://(?:www\.)iqiyi.com/.+?\.html'
|
||||
|
||||
_TEST = {
|
||||
'url': 'http://www.iqiyi.com/v_19rrojlavg.html',
|
||||
'md5': '260f0f59686e65e886995d0ba791ab83',
|
||||
'info_dict': {
|
||||
'id': '9c1fb1b99d192b21c559e5a1a2cb3c73',
|
||||
'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):
|
||||
def do_xor(x, y):
|
||||
a = y % 3
|
||||
if a == 1:
|
||||
return x ^ 121
|
||||
if a == 2:
|
||||
return x ^ 72
|
||||
return x ^ 103
|
||||
|
||||
def get_encode_code(l):
|
||||
a = 0
|
||||
b = l.split('-')
|
||||
c = len(b)
|
||||
s = ''
|
||||
for i in range(c - 1, -1, -1):
|
||||
a = do_xor(int(b[c-i-1], 16), i)
|
||||
s += chr(a)
|
||||
return s[::-1]
|
||||
|
||||
def get_path_key(x):
|
||||
mg = ')(*&^flash@#$%a'
|
||||
tm = self._download_json(
|
||||
'http://data.video.qiyi.com/t?tn=' + str(random.random()), video_id)['t']
|
||||
t = str(int(math.floor(int(tm)/(600.0))))
|
||||
return hashlib.md5(
|
||||
(t+mg+x).encode('utf8')).hexdigest()
|
||||
|
||||
video_urls_dict = {}
|
||||
for i in data['vp']['tkl'][0]['vs']:
|
||||
if 0 < int(i['bid']) <= 10:
|
||||
format_id = self.get_format(i['bid'])
|
||||
|
||||
video_urls_info = i['fs']
|
||||
if not i['fs'][0]['l'].startswith('/'):
|
||||
t = get_encode_code(i['fs'][0]['l'])
|
||||
if t.endswith('mp4'):
|
||||
video_urls_info = i['flvs']
|
||||
|
||||
video_urls = []
|
||||
for ii in video_urls_info:
|
||||
vl = ii['l']
|
||||
if not vl.startswith('/'):
|
||||
vl = get_encode_code(vl)
|
||||
key = get_path_key(
|
||||
vl.split('/')[-1].split('.')[0])
|
||||
filesize = ii['b']
|
||||
base_url = data['vp']['du'].split('/')
|
||||
base_url.insert(-1, key)
|
||||
base_url = '/'.join(base_url)
|
||||
param = {
|
||||
'su': _uuid,
|
||||
'qyid': uuid.uuid4().hex,
|
||||
'client': '',
|
||||
'z': '',
|
||||
'bt': '',
|
||||
'ct': '',
|
||||
'tn': str(int(time.time()))
|
||||
}
|
||||
api_video_url = base_url + vl + '?' + \
|
||||
compat_urllib_parse.urlencode(param)
|
||||
js = self._download_json(api_video_url, video_id)
|
||||
video_url = js['l']
|
||||
video_urls.append(
|
||||
(video_url, filesize))
|
||||
|
||||
video_urls_dict[format_id] = video_urls
|
||||
return video_urls_dict
|
||||
|
||||
def get_format(self, bid):
|
||||
bid_dict = {
|
||||
'1': 'standard',
|
||||
'2': 'high',
|
||||
'3': 'super',
|
||||
'4': 'suprt-high',
|
||||
'5': 'fullhd',
|
||||
'10': '4k'
|
||||
}
|
||||
return bid_dict[str(bid)]
|
||||
|
||||
def get_raw_data(self, tvid, video_id, enc_key, _uuid):
|
||||
tm = str(int(time.time()))
|
||||
param = {
|
||||
'key': 'fvip',
|
||||
'src': hashlib.md5(b'youtube-dl').hexdigest(),
|
||||
'tvId': tvid,
|
||||
'vid': video_id,
|
||||
'vinfo': 1,
|
||||
'tm': tm,
|
||||
'enc': hashlib.md5(
|
||||
(enc_key + tm + tvid).encode('utf8')).hexdigest(),
|
||||
'qyid': _uuid,
|
||||
'tn': random.random(),
|
||||
'um': 0,
|
||||
'authkey': hashlib.md5(
|
||||
(tm + tvid).encode('utf8')).hexdigest()
|
||||
}
|
||||
|
||||
api_url = 'http://cache.video.qiyi.com/vms' + '?' + \
|
||||
compat_urllib_parse.urlencode(param)
|
||||
raw_data = self._download_json(api_url, video_id)
|
||||
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')
|
||||
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')
|
||||
video_id = self._search_regex(
|
||||
r'videoId ?= ?(\'|\")(?P<video_id>[a-z\d]+)',
|
||||
webpage, 'video_id', flags=re.I, group='video_id')
|
||||
swf_url = self._search_regex(
|
||||
r'(?P<swf>http://.+?MainPlayer.+?\.swf)', webpage, 'swf')
|
||||
_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 not raw_data['data']['vp']['tkl']:
|
||||
raise ExtractorError('No support iQiqy VIP video')
|
||||
|
||||
data = raw_data['data']
|
||||
|
||||
title = data['vi']['vn']
|
||||
|
||||
# generate video_urls_dict
|
||||
video_urls_dict = self.construct_video_urls(data, video_id, _uuid)
|
||||
|
||||
# construct info
|
||||
entries = []
|
||||
for format_id in video_urls_dict:
|
||||
video_urls = video_urls_dict[format_id]
|
||||
for i, video_url_info in enumerate(video_urls):
|
||||
if len(entries) < i+1:
|
||||
entries.append({'formats': []})
|
||||
entries[i]['formats'].append(
|
||||
{
|
||||
'url': video_url_info[0],
|
||||
'filesize': video_url_info[-1],
|
||||
'format_id': format_id,
|
||||
}
|
||||
)
|
||||
|
||||
for i in range(len(entries)):
|
||||
entries[i].update(
|
||||
{
|
||||
'id': '_part%d' % (i+1),
|
||||
'title': title,
|
||||
}
|
||||
)
|
||||
|
||||
if len(entries) > 1:
|
||||
info = {
|
||||
'_type': 'multi_video',
|
||||
'id': video_id,
|
||||
'title': title,
|
||||
'entries': entries,
|
||||
}
|
||||
else:
|
||||
info = entries[0]
|
||||
info['id'] = video_id
|
||||
info['title'] = title
|
||||
|
||||
return info
|
||||
Loading…
Reference in New Issue
Block a user