youtube-dl/youtube_dl/extractor/leeco.py

# coding: utf-8
from __future__ import unicode_literals

import base64
import datetime
import hashlib
import re
import time

from .common import InfoExtractor
from ..compat import (
    compat_ord,
    compat_str,
    compat_urllib_parse_urlencode,
)
from ..utils import (
    determine_ext,
    encode_data_uri,
    ExtractorError,
    int_or_none,
    orderedSet,
    parse_iso8601,
    str_or_none,
    url_basename,
    urshift,
    update_url_query,
)


class LeIE(InfoExtractor):
    IE_DESC = '乐视网'
    _VALID_URL = r'https?://(?:www\.le\.com/ptv/vplay|(?:sports\.le|(?:www\.)?lesports)\.com/(?:match|video))/(?P<id>\d+)\.html'

    _URL_TEMPLATE = 'http://www.le.com/ptv/vplay/%s.html'

    _TESTS = [{
        'url': 'http://www.le.com/ptv/vplay/22005890.html',
        'md5': 'edadcfe5406976f42f9f266057ee5e40',
        'info_dict': {
            'id': '22005890',
            'ext': 'mp4',
            'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家',
            'description': 'md5:a9cb175fd753e2962176b7beca21a47c',
        },
        'params': {
            'hls_prefer_native': True,
        },
    }, {
        'url': 'http://www.le.com/ptv/vplay/1415246.html',
        'info_dict': {
            'id': '1415246',
            'ext': 'mp4',
            'title': '美人天下01',
            'description': 'md5:f88573d9d7225ada1359eaf0dbf8bcda',
        },
        'params': {
            'hls_prefer_native': True,
        },
    }, {
        'note': 'This video is available only in Mainland China, thus a proxy is needed',
        'url': 'http://www.le.com/ptv/vplay/1118082.html',
        'md5': '2424c74948a62e5f31988438979c5ad1',
        'info_dict': {
            'id': '1118082',
            'ext': 'mp4',
            'title': '与龙共舞 完整版',
            'description': 'md5:7506a5eeb1722bb9d4068f85024e3986',
        },
        'params': {
            'hls_prefer_native': True,
        },
        'skip': 'Only available in China',
    }, {
        'url': 'http://sports.le.com/video/25737697.html',
        'only_matching': True,
    }, {
        'url': 'http://www.lesports.com/match/1023203003.html',
        'only_matching': True,
    }, {
        'url': 'http://sports.le.com/match/1023203003.html',
        'only_matching': True,
    }]

    # ror() and calc_time_key() are reversed from a embedded swf file in KLetvPlayer.swf
    def ror(self, param1, param2):
        _loc3_ = 0
        while _loc3_ < param2:
            param1 = urshift(param1, 1) + ((param1 & 1) << 31)
            _loc3_ += 1
        return param1

    def calc_time_key(self, param1):
        _loc2_ = 773625421
        _loc3_ = self.ror(param1, _loc2_ % 13)
        _loc3_ = _loc3_ ^ _loc2_
        _loc3_ = self.ror(_loc3_, _loc2_ % 17)
        return _loc3_

    # reversed from http://jstatic.letvcdn.com/sdk/player.js
    def get_mms_key(self, time):
        return self.ror(time, 8) ^ 185025305

    # see M3U8Encryption class in KLetvPlayer.swf
    @staticmethod
    def decrypt_m3u8(encrypted_data):
        if encrypted_data[:5].decode('utf-8').lower() != 'vc_01':
            return encrypted_data
        encrypted_data = encrypted_data[5:]

        _loc4_ = bytearray(2 * len(encrypted_data))
        for idx, val in enumerate(encrypted_data):
            b = compat_ord(val)
            _loc4_[2 * idx] = b // 16
            _loc4_[2 * idx + 1] = b % 16
        idx = len(_loc4_) - 11
        _loc4_ = _loc4_[idx:] + _loc4_[:idx]
        _loc7_ = bytearray(len(encrypted_data))
        for i in range(len(encrypted_data)):
            _loc7_[i] = _loc4_[2 * i] * 16 + _loc4_[2 * i + 1]

        return bytes(_loc7_)

    def _check_errors(self, play_json):
        # Check for errors
        playstatus = play_json['playstatus']
        if playstatus['status'] == 0:
            flag = playstatus['flag']
            if flag == 1:
                msg = 'Country %s auth error' % playstatus['country']
            else:
                msg = 'Generic error. flag = %d' % flag
            raise ExtractorError(msg, expected=True)

    def _real_extract(self, url):
        media_id = self._match_id(url)
        page = self._download_webpage(url, media_id)

        play_json_h5 = self._download_json(
            'http://api.le.com/mms/out/video/playJsonH5',
            media_id, 'Downloading html5 playJson data', query={
                'id': media_id,
                'platid': 3,
                'splatid': 304,
                'format': 1,
                'tkey': self.get_mms_key(int(time.time())),
                'domain': 'www.le.com',
                'tss': 'no',
            },
            headers=self.geo_verification_headers())
        self._check_errors(play_json_h5)

        play_json_flash = self._download_json(
            'http://api.le.com/mms/out/video/playJson',
            media_id, 'Downloading flash playJson data', query={
                'id': media_id,
                'platid': 1,
                'splatid': 101,
                'format': 1,
                'tkey': self.calc_time_key(int(time.time())),
                'domain': 'www.le.com',
            },
            headers=self.geo_verification_headers())
        self._check_errors(play_json_flash)

        def get_h5_urls(media_url, format_id):
            location = self._download_json(
                media_url, media_id,
                'Download JSON metadata for format %s' % format_id, query={
                    'format': 1,
                    'expect': 3,
                    'tss': 'no',
                })['location']

            return {
                'http': update_url_query(location, {'tss': 'no'}),
                'hls': update_url_query(location, {'tss': 'ios'}),
            }

        def get_flash_urls(media_url, format_id):
            media_url += '&' + compat_urllib_parse_urlencode({
                'm3v': 1,
                'format': 1,
                'expect': 3,
                'rateid': format_id,
            })

            nodes_data = self._download_json(
                media_url, media_id,
                'Download JSON metadata for format %s' % format_id)

            req = self._request_webpage(
                nodes_data['nodelist'][0]['location'], media_id,
                note='Downloading m3u8 information for format %s' % format_id)

            m3u8_data = self.decrypt_m3u8(req.read())

            return {
                'hls': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
            }

        extracted_formats = []
        formats = []
        for play_json, get_urls in ((play_json_h5, get_h5_urls), (play_json_flash, get_flash_urls)):
            playurl = play_json['playurl'