Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vidio.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import int_or_none
   8
   9
  10 class VidioIE(InfoExtractor):
  11     _VALID_URL = r'https?://(?:www\.)?vidio\.com/watch/(?P<id>\d+)-(?P<display_id>[^/?#&]+)'
  12     _TESTS = [{
  13         'url': 'http://www.vidio.com/watch/165683-dj_ambred-booyah-live-2015',
  14         'md5': 'cd2801394afc164e9775db6a140b91fe',
  15         'info_dict': {
  16             'id': '165683',
  17             'display_id': 'dj_ambred-booyah-live-2015',
  18             'ext': 'mp4',
  19             'title': 'DJ_AMBRED - Booyah (Live 2015)',
  20             'description': 'md5:27dc15f819b6a78a626490881adbadf8',
  21             'thumbnail': r're:^https?://.*\.jpg$',
  22             'duration': 149,
  23             'like_count': int,
  24         },
  25     }, {
  26         'url': 'https://www.vidio.com/watch/77949-south-korea-test-fires-missile-that-can-strike-all-of-the-north',
  27         'only_matching': True,
  28     }]
  29
  30     def _real_extract(self, url):
  31         mobj = re.match(self._VALID_URL, url)
  32         video_id, display_id = mobj.group('id', 'display_id')
  33
  34         webpage = self._download_webpage(url, display_id)
  35
  36         title = self._og_search_title(webpage)
  37
  38         m3u8_url, duration, thumbnail = [None] * 3
  39
  40         clips = self._parse_json(
  41             self._html_search_regex(
  42                 r'data-json-clips\s*=\s*(["\'])(?P<data>\[.+?\])\1',
  43                 webpage, 'video data', default='[]', group='data'),
  44             display_id, fatal=False)
  45         if clips:
  46             clip = clips[0]
  47             m3u8_url = clip.get('sources', [{}])[0].get('file')
  48             duration = clip.get('clip_duration')
  49             thumbnail = clip.get('image')
  50
  51         m3u8_url = m3u8_url or self._search_regex(
  52             r'data(?:-vjs)?-clip-hls-url=(["\'])(?P<url>(?!\1).+)\1',
  53             webpage, 'hls url')
  54         formats = self._extract_m3u8_formats(
  55             m3u8_url, display_id, 'mp4', entry_protocol='m3u8_native')
  56         self._sort_formats(formats)
  57
  58         duration = int_or_none(duration or self._search_regex(
  59             r'data-video-duration=(["\'])(?P<duartion>\d+)\1', webpage, 'duration'))
  60         thumbnail = thumbnail or self._og_search_thumbnail(webpage)
  61
  62         like_count = int_or_none(self._search_regex(
  63             (r'<span[^>]+data-comment-vote-count=["\'](\d+)',
  64              r'<span[^>]+class=["\'].*?\blike(?:__|-)count\b.*?["\'][^>]*>\s*(\d+)'),
  65             webpage, 'like count', fatal=False))
  66
  67         return {
  68             'id': video_id,
  69             'display_id': display_id,
  70             'title': title,
  71             'description': self._og_search_description(webpage),
  72             'thumbnail': thumbnail,
  73             'duration': duration,
  74             'like_count': like_count,
  75             'formats': formats,
  76         }