youtube_dl/extractor/dailymail.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import compat_str
   6 from ..utils import (
   7     int_or_none,
   8     determine_protocol,
   9     try_get,
  10     unescapeHTML,
  11 )
  12
  13
  14 class DailyMailIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/video/[^/]+/video-(?P<id>[0-9]+)'
  16     _TEST = {
  17         'url': 'http://www.dailymail.co.uk/video/tvshowbiz/video-1295863/The-Mountain-appears-sparkling-water-ad-Heavy-Bubbles.html',
  18         'md5': 'f6129624562251f628296c3a9ffde124',
  19         'info_dict': {
  20             'id': '1295863',
  21             'ext': 'mp4',
  22             'title': 'The Mountain appears in sparkling water ad for \'Heavy Bubbles\'',
  23             'description': 'md5:a93d74b6da172dd5dc4d973e0b766a84',
  24         }
  25     }
  26
  27     def _real_extract(self, url):
  28         video_id = self._match_id(url)
  29         webpage = self._download_webpage(url, video_id)
  30         video_data = self._parse_json(self._search_regex(
  31             r"data-opts='({.+?})'", webpage, 'video data'), video_id)
  32         title = unescapeHTML(video_data['title'])
  33
  34         sources_url = (try_get(
  35             video_data,
  36             (lambda x: x['plugins']['sources']['url'],
  37              lambda x: x['sources']['url']), compat_str) or
  38             'http://www.dailymail.co.uk/api/player/%s/video-sources.json' % video_id)
  39
  40         video_sources = self._download_json(sources_url, video_id)
  41
  42         formats = []
  43         for rendition in video_sources['renditions']:
  44             rendition_url = rendition.get('url')
  45             if not rendition_url:
  46                 continue
  47             tbr = int_or_none(rendition.get('encodingRate'), 1000)
  48             container = rendition.get('videoContainer')
  49             is_hls = container == 'M2TS'
  50             protocol = 'm3u8_native' if is_hls else determine_protocol({'url': rendition_url})
  51             formats.append({
  52                 'format_id': ('hls' if is_hls else protocol) + ('-%d' % tbr if tbr else ''),
  53                 'url': rendition_url,
  54                 'width': int_or_none(rendition.get('frameWidth')),
  55                 'height': int_or_none(rendition.get('frameHeight')),
  56                 'tbr': tbr,
  57                 'vcodec': rendition.get('videoCodec'),
  58                 'container': container,
  59                 'protocol': protocol,
  60                 'ext': 'mp4' if is_hls else None,
  61             })
  62         self._sort_formats(formats)
  63
  64         return {
  65             'id': video_id,
  66             'title': title,
  67             'description': unescapeHTML(video_data.get('descr')),
  68             'thumbnail': video_data.get('poster') or video_data.get('thumbnail'),
  69             'formats': formats,
  70         }