youtube_dl/extractor/dailymail.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     int_or_none,
   7     determine_protocol,
   8     unescapeHTML,
   9 )
  10
  11
  12 class DailyMailIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/video/[^/]+/video-(?P<id>[0-9]+)'
  14     _TEST = {
  15         'url': 'http://www.dailymail.co.uk/video/tvshowbiz/video-1295863/The-Mountain-appears-sparkling-water-ad-Heavy-Bubbles.html',
  16         'md5': 'f6129624562251f628296c3a9ffde124',
  17         'info_dict': {
  18             'id': '1295863',
  19             'ext': 'mp4',
  20             'title': 'The Mountain appears in sparkling water ad for \'Heavy Bubbles\'',
  21             'description': 'md5:a93d74b6da172dd5dc4d973e0b766a84',
  22         }
  23     }
  24
  25     def _real_extract(self, url):
  26         video_id = self._match_id(url)
  27         webpage = self._download_webpage(url, video_id)
  28         video_data = self._parse_json(self._search_regex(
  29             r"data-opts='({.+?})'", webpage, 'video data'), video_id)
  30         title = unescapeHTML(video_data['title'])
  31         video_sources = self._download_json(video_data.get(
  32             'sources', {}).get('url') or 'http://www.dailymail.co.uk/api/player/%s/video-sources.json' % video_id, video_id)
  33
  34         formats = []
  35         for rendition in video_sources['renditions']:
  36             rendition_url = rendition.get('url')
  37             if not rendition_url:
  38                 continue
  39             tbr = int_or_none(rendition.get('encodingRate'), 1000)
  40             container = rendition.get('videoContainer')
  41             is_hls = container == 'M2TS'
  42             protocol = 'm3u8_native' if is_hls else determine_protocol({'url': rendition_url})
  43             formats.append({
  44                 'format_id': ('hls' if is_hls else protocol) + ('-%d' % tbr if tbr else ''),
  45                 'url': rendition_url,
  46                 'width': int_or_none(rendition.get('frameWidth')),
  47                 'height': int_or_none(rendition.get('frameHeight')),
  48                 'tbr': tbr,
  49                 'vcodec': rendition.get('videoCodec'),
  50                 'container': container,
  51                 'protocol': protocol,
  52                 'ext': 'mp4' if is_hls else None,
  53             })
  54         self._sort_formats(formats)
  55
  56         return {
  57             'id': video_id,
  58             'title': title,
  59             'description': unescapeHTML(video_data.get('descr')),
  60             'thumbnail': video_data.get('poster') or video_data.get('thumbnail'),
  61             'formats': formats,
  62         }