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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     determine_ext,
   7     int_or_none,
   8     str_to_int,
   9     urlencode_postdata,
  10 )
  11
  12
  13 class ManyVidsIE(InfoExtractor):
  14     _VALID_URL = r'(?i)https?://(?:www\.)?manyvids\.com/video/(?P<id>\d+)'
  15     _TESTS = [{
  16         # preview video
  17         'url': 'https://www.manyvids.com/Video/133957/everthing-about-me/',
  18         'md5': '03f11bb21c52dd12a05be21a5c7dcc97',
  19         'info_dict': {
  20             'id': '133957',
  21             'ext': 'mp4',
  22             'title': 'everthing about me (Preview)',
  23             'view_count': int,
  24             'like_count': int,
  25         },
  26     }, {
  27         # full video
  28         'url': 'https://www.manyvids.com/Video/935718/MY-FACE-REVEAL/',
  29         'md5': 'f3e8f7086409e9b470e2643edb96bdcc',
  30         'info_dict': {
  31             'id': '935718',
  32             'ext': 'mp4',
  33             'title': 'MY FACE REVEAL',
  34             'view_count': int,
  35             'like_count': int,
  36         },
  37     }]
  38
  39     def _real_extract(self, url):
  40         video_id = self._match_id(url)
  41
  42         webpage = self._download_webpage(url, video_id)
  43
  44         video_url = self._search_regex(
  45             r'data-(?:video-filepath|meta-video)\s*=s*(["\'])(?P<url>(?:(?!\1).)+)\1',
  46             webpage, 'video URL', group='url')
  47
  48         title = self._html_search_regex(
  49             (r'<span[^>]+class=["\']item-title[^>]+>([^<]+)',
  50              r'<h2[^>]+class=["\']h2 m-0["\'][^>]*>([^<]+)'),
  51             webpage, 'title', default=None) or self._html_search_meta(
  52             'twitter:title', webpage, 'title', fatal=True)
  53
  54         if any(p in webpage for p in ('preview_videos', '_preview.mp4')):
  55             title += ' (Preview)'
  56
  57         mv_token = self._search_regex(
  58             r'data-mvtoken=(["\'])(?P<value>(?:(?!\1).)+)\1', webpage,
  59             'mv token', default=None, group='value')
  60
  61         if mv_token:
  62             # Sets some cookies
  63             self._download_webpage(
  64                 'https://www.manyvids.com/includes/ajax_repository/you_had_me_at_hello.php',
  65                 video_id, fatal=False, data=urlencode_postdata({
  66                     'mvtoken': mv_token,
  67                     'vid': video_id,
  68                 }), headers={
  69                     'Referer': url,
  70                     'X-Requested-With': 'XMLHttpRequest'
  71                 })
  72
  73         if determine_ext(video_url) == 'm3u8':
  74             formats = self._extract_m3u8_formats(
  75                 video_url, video_id, 'mp4', entry_protocol='m3u8_native',
  76                 m3u8_id='hls')
  77         else:
  78             formats = [{'url': video_url}]
  79
  80         like_count = int_or_none(self._search_regex(
  81             r'data-likes=["\'](\d+)', webpage, 'like count', default=None))
  82         view_count = str_to_int(self._html_search_regex(
  83             r'(?s)<span[^>]+class="views-wrapper"[^>]*>(.+?)</span', webpage,
  84             'view count', default=None))
  85
  86         return {
  87             'id': video_id,
  88             'title': title,
  89             'view_count': view_count,
  90             'like_count': like_count,
  91             'formats': formats,
  92         }