_ Git - youtube-dl/blob - 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     determine_ext,
   9     int_or_none,
  10     qualities,
  11     unescapeHTML,
  12 )
  13
  14
  15 class GiantBombIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?giantbomb\.com/(?:videos|shows)/(?P<display_id>[^/]+)/(?P<id>\d+-\d+)'
  17     _TESTS = [{
  18         'url': 'http://www.giantbomb.com/videos/quick-look-destiny-the-dark-below/2300-9782/',
  19         'md5': '132f5a803e7e0ab0e274d84bda1e77ae',
  20         'info_dict': {
  21             'id': '2300-9782',
  22             'display_id': 'quick-look-destiny-the-dark-below',
  23             'ext': 'mp4',
  24             'title': 'Quick Look: Destiny: The Dark Below',
  25             'description': 'md5:0aa3aaf2772a41b91d44c63f30dfad24',
  26             'duration': 2399,
  27             'thumbnail': r're:^https?://.*\.jpg$',
  28         }
  29     }, {
  30         'url': 'https://www.giantbomb.com/shows/ben-stranding/2970-20212',
  31         'only_matching': True,
  32     }]
  33
  34     def _real_extract(self, url):
  35         mobj = re.match(self._VALID_URL, url)
  36         video_id = mobj.group('id')
  37         display_id = mobj.group('display_id')
  38
  39         webpage = self._download_webpage(url, display_id)
  40
  41         title = self._og_search_title(webpage)
  42         description = self._og_search_description(webpage)
  43         thumbnail = self._og_search_thumbnail(webpage)
  44
  45         video = json.loads(unescapeHTML(self._search_regex(
  46             r'data-video="([^"]+)"', webpage, 'data-video')))
  47
  48         duration = int_or_none(video.get('lengthSeconds'))
  49
  50         quality = qualities([
  51             'f4m_low', 'progressive_low', 'f4m_high',
  52             'progressive_high', 'f4m_hd', 'progressive_hd'])
  53
  54         formats = []
  55         for format_id, video_url in video['videoStreams'].items():
  56             if format_id == 'f4m_stream':
  57                 continue
  58             ext = determine_ext(video_url)
  59             if ext == 'f4m':
  60                 f4m_formats = self._extract_f4m_formats(video_url + '?hdcore=3.3.1', display_id)
  61                 if f4m_formats:
  62                     f4m_formats[0]['quality'] = quality(format_id)
  63                     formats.extend(f4m_formats)
  64             elif ext == 'm3u8':
  65                 formats.extend(self._extract_m3u8_formats(
  66                     video_url, display_id, ext='mp4', entry_protocol='m3u8_native',
  67                     m3u8_id='hls', fatal=False))
  68             else:
  69                 formats.append({
  70                     'url': video_url,
  71                     'format_id': format_id,
  72                     'quality': quality(format_id),
  73                 })
  74
  75         if not formats:
  76             youtube_id = video.get('youtubeID')
  77             if youtube_id:
  78                 return self.url_result(youtube_id, 'Youtube')
  79
  80         self._sort_formats(formats)
  81
  82         return {
  83             'id': video_id,
  84             'display_id': display_id,
  85             'title': title,
  86             'description': description,
  87             'thumbnail': thumbnail,
  88             'duration': duration,
  89             'formats': formats,
  90         }