youtube_dl/extractor/dailymail.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_str
   8 from ..utils import (
   9     int_or_none,
  10     determine_protocol,
  11     try_get,
  12     unescapeHTML,
  13 )
  14
  15
  16 class DailyMailIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/(?:video/[^/]+/video-|embed/video/)(?P<id>[0-9]+)'
  18     _TESTS = [{
  19         'url': 'http://www.dailymail.co.uk/video/tvshowbiz/video-1295863/The-Mountain-appears-sparkling-water-ad-Heavy-Bubbles.html',
  20         'md5': 'f6129624562251f628296c3a9ffde124',
  21         'info_dict': {
  22             'id': '1295863',
  23             'ext': 'mp4',
  24             'title': 'The Mountain appears in sparkling water ad for \'Heavy Bubbles\'',
  25             'description': 'md5:a93d74b6da172dd5dc4d973e0b766a84',
  26         }
  27     }, {
  28         'url': 'http://www.dailymail.co.uk/embed/video/1295863.html',
  29         'only_matching': True,
  30     }]
  31
  32     @staticmethod
  33     def _extract_urls(webpage):
  34         return re.findall(
  35             r'<iframe\b[^>]+\bsrc=["\'](?P<url>(?:https?:)?//(?:www\.)?dailymail\.co\.uk/embed/video/\d+\.html)',
  36             webpage)
  37
  38     def _real_extract(self, url):
  39         video_id = self._match_id(url)
  40         webpage = self._download_webpage(url, video_id)
  41         video_data = self._parse_json(self._search_regex(
  42             r"data-opts='({.+?})'", webpage, 'video data'), video_id)
  43         title = unescapeHTML(video_data['title'])
  44
  45         sources_url = (try_get(
  46             video_data,
  47             (lambda x: x['plugins']['sources']['url'],
  48              lambda x: x['sources']['url']), compat_str)
  49             or 'http://www.dailymail.co.uk/api/player/%s/video-sources.json' % video_id)
  50
  51         video_sources = self._download_json(sources_url, video_id)
  52         body = video_sources.get('body')
  53         if body:
  54             video_sources = body
  55
  56         formats = []
  57         for rendition in video_sources['renditions']:
  58             rendition_url = rendition.get('url')
  59             if not rendition_url:
  60                 continue
  61             tbr = int_or_none(rendition.get('encodingRate'), 1000)
  62             container = rendition.get('videoContainer')
  63             is_hls = container == 'M2TS'
  64             protocol = 'm3u8_native' if is_hls else determine_protocol({'url': rendition_url})
  65             formats.append({
  66                 'format_id': ('hls' if is_hls else protocol) + ('-%d' % tbr if tbr else ''),
  67                 'url': rendition_url,
  68                 'width': int_or_none(rendition.get('frameWidth')),
  69                 'height': int_or_none(rendition.get('frameHeight')),
  70                 'tbr': tbr,
  71                 'vcodec': rendition.get('videoCodec'),
  72                 'container': container,
  73                 'protocol': protocol,
  74                 'ext': 'mp4' if is_hls else None,
  75             })
  76         self._sort_formats(formats)
  77
  78         return {
  79             'id': video_id,
  80             'title': title,
  81             'description': unescapeHTML(video_data.get('descr')),
  82             'thumbnail': video_data.get('poster') or video_data.get('thumbnail'),
  83             'formats': formats,
  84         }