youtube_dl/extractor/giantbomb.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     unescapeHTML,
   9     qualities,
  10     int_or_none,
  11 )
  12
  13
  14 class GiantBombIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?giantbomb\.com/videos/(?P<display_id>[^/]+)/(?P<id>\d+-\d+)'
  16     _TEST = {
  17         'url': 'http://www.giantbomb.com/videos/quick-look-destiny-the-dark-below/2300-9782/',
  18         'md5': '57badeface303ecf6b98b812de1b9018',
  19         'info_dict': {
  20             'id': '2300-9782',
  21             'display_id': 'quick-look-destiny-the-dark-below',
  22             'ext': 'mp4',
  23             'title': 'Quick Look: Destiny: The Dark Below',
  24             'description': 'md5:0aa3aaf2772a41b91d44c63f30dfad24',
  25             'duration': 2399,
  26             'thumbnail': 're:^https?://.*\.jpg$',
  27         }
  28     }
  29
  30     def _real_extract(self, url):
  31         mobj = re.match(self._VALID_URL, url)
  32         video_id = mobj.group('id')
  33         display_id = mobj.group('display_id')
  34
  35         webpage = self._download_webpage(url, display_id)
  36
  37         title = self._og_search_title(webpage)
  38         description = self._og_search_description(webpage)
  39         thumbnail = self._og_search_thumbnail(webpage)
  40
  41         video = json.loads(unescapeHTML(self._search_regex(
  42             r'data-video="([^"]+)"', webpage, 'data-video')))
  43
  44         duration = int_or_none(video.get('lengthSeconds'))
  45
  46         quality = qualities([
  47             'f4m_low', 'progressive_low', 'f4m_high',
  48             'progressive_high', 'f4m_hd', 'progressive_hd'])
  49
  50         formats = []
  51         for format_id, video_url in video['videoStreams'].items():
  52             if format_id == 'f4m_stream':
  53                 continue
  54             if video_url.endswith('.f4m'):
  55                 f4m_formats = self._extract_f4m_formats(video_url + '?hdcore=3.3.1', display_id)
  56                 if f4m_formats:
  57                     f4m_formats[0]['quality'] = quality(format_id)
  58                     formats.extend(f4m_formats)
  59             else:
  60                 formats.append({
  61                     'url': video_url,
  62                     'format_id': format_id,
  63                     'quality': quality(format_id),
  64                 })
  65
  66         if not formats:
  67             youtube_id = video.get('youtubeID')
  68             if youtube_id:
  69                 return self.url_result(youtube_id, 'Youtube')
  70
  71         self._sort_formats(formats)
  72
  73         return {
  74             'id': video_id,
  75             'display_id': display_id,
  76             'title': title,
  77             'description': description,
  78             'thumbnail': thumbnail,
  79             'duration': duration,
  80             'formats': formats,
  81         }