_ Git - youtube-dl/blob - youtube_dl/extractor/einthusan.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..compat import (
   9     compat_b64decode,
  10     compat_str,
  11     compat_urlparse,
  12 )
  13 from ..utils import (
  14     extract_attributes,
  15     ExtractorError,
  16     get_elements_by_class,
  17     urlencode_postdata,
  18 )
  19
  20
  21 class EinthusanIE(InfoExtractor):
  22     _VALID_URL = r'https?://(?P<host>einthusan\.(?:tv|com))/movie/watch/(?P<id>[^/?#&]+)'
  23     _TESTS = [{
  24         'url': 'https://einthusan.tv/movie/watch/9097/',
  25         'md5': 'ff0f7f2065031b8a2cf13a933731c035',
  26         'info_dict': {
  27             'id': '9097',
  28             'ext': 'mp4',
  29             'title': 'Ae Dil Hai Mushkil',
  30             'description': 'md5:33ef934c82a671a94652a9b4e54d931b',
  31             'thumbnail': r're:^https?://.*\.jpg$',
  32         }
  33     }, {
  34         'url': 'https://einthusan.tv/movie/watch/51MZ/?lang=hindi',
  35         'only_matching': True,
  36     }, {
  37         'url': 'https://einthusan.com/movie/watch/9097/',
  38         'only_matching': True,
  39     }]
  40
  41     # reversed from jsoncrypto.prototype.decrypt() in einthusan-PGMovieWatcher.js
  42     def _decrypt(self, encrypted_data, video_id):
  43         return self._parse_json(compat_b64decode((
  44             encrypted_data[:10] + encrypted_data[-1] + encrypted_data[12:-1]
  45         )).decode('utf-8'), video_id)
  46
  47     def _real_extract(self, url):
  48         mobj = re.match(self._VALID_URL, url)
  49         host = mobj.group('host')
  50         video_id = mobj.group('id')
  51
  52         webpage = self._download_webpage(url, video_id)
  53
  54         title = self._html_search_regex(r'<h3>([^<]+)</h3>', webpage, 'title')
  55
  56         player_params = extract_attributes(self._search_regex(
  57             r'(<section[^>]+id="UIVideoPlayer"[^>]+>)', webpage, 'player parameters'))
  58
  59         page_id = self._html_search_regex(
  60             '<html[^>]+data-pageid="([^"]+)"', webpage, 'page ID')
  61         video_data = self._download_json(
  62             'https://%s/ajax/movie/watch/%s/' % (host, video_id), video_id,
  63             data=urlencode_postdata({
  64                 'xEvent': 'UIVideoPlayer.PingOutcome',
  65                 'xJson': json.dumps({
  66                     'EJOutcomes': player_params['data-ejpingables'],
  67                     'NativeHLS': False
  68                 }),
  69                 'arcVersion': 3,
  70                 'appVersion': 59,
  71                 'gorilla.csrf.Token': page_id,
  72             }))['Data']
  73
  74         if isinstance(video_data, compat_str) and video_data.startswith('/ratelimited/'):
  75             raise ExtractorError(
  76                 'Download rate reached. Please try again later.', expected=True)
  77
  78         ej_links = self._decrypt(video_data['EJLinks'], video_id)
  79
  80         formats = []
  81
  82         m3u8_url = ej_links.get('HLSLink')
  83         if m3u8_url:
  84             formats.extend(self._extract_m3u8_formats(
  85                 m3u8_url, video_id, ext='mp4', entry_protocol='m3u8_native'))
  86
  87         mp4_url = ej_links.get('MP4Link')
  88         if mp4_url:
  89             formats.append({
  90                 'url': mp4_url,
  91             })
  92
  93         self._sort_formats(formats)
  94
  95         description = get_elements_by_class('synopsis', webpage)[0]
  96         thumbnail = self._html_search_regex(
  97             r'''<img[^>]+src=(["'])(?P<url>(?!\1).+?/moviecovers/(?!\1).+?)\1''',
  98             webpage, 'thumbnail url', fatal=False, group='url')
  99         if thumbnail is not None:
 100             thumbnail = compat_urlparse.urljoin(url, thumbnail)
 101
 102         return {
 103             'id': video_id,
 104             'title': title,
 105             'formats': formats,
 106             'thumbnail': thumbnail,
 107             'description': description,
 108         }