_ Git - youtube-dl/blob - youtube_dl/extractor/animalplanet.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     parse_duration,
   8     parse_iso8601,
   9 )
  10
  11
  12 class AnimalPlanetIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?animalplanet\.com/([^/]+/)*(?P<id>[^/\?#]+)'
  14     _TESTS = [{
  15         'url': 'http://www.animalplanet.com/tv-shows/i-shouldnt-be-alive/videos/dog-saves-injured-owner/',
  16         'info_dict': {
  17             'id': '10608',
  18             'ext': 'mp4',
  19             'title': 'Dog Saves Injured Owner',
  20             'description': 'A world class athlete is put to the test when she falls into a canyon and breaks her hip. Her only companion is her dog, Taz, who is on a mission to save her!',
  21             'upload_date': '20100410',
  22             'timestamp': 1270857727,
  23             'duration': 220,
  24         },
  25         'params': {
  26             # m3u8 download
  27             'skip_download': True,
  28         }
  29     }, {
  30         'url': 'http://www.animalplanet.com/longfin-eels-maneaters/',
  31         'only_matching': True,
  32     }]
  33
  34     def _real_extract(self, url):
  35         display_id = self._match_id(url)
  36         webpage = self._download_webpage(url, display_id)
  37
  38         video_data = self._parse_json(self._search_regex(
  39             r'initialVideoData\s*=\s*({.+?});',
  40             webpage, 'initialVideoData'), display_id)['playlist'][0]
  41
  42         return {
  43             'id': compat_str(video_data['id']),
  44             'display_id': display_id,
  45             'title': video_data['title'],
  46             'description': video_data.get('description'),
  47             'thumbnail': video_data.get('thumbnailURL'),
  48             'duration': parse_duration(video_data.get('video_length')),
  49             'timestamp': parse_iso8601(video_data.get('publishedDate')),
  50             'formats': self._extract_m3u8_formats(
  51                 video_data['src'], display_id, 'mp4',
  52                 'm3u8_native', m3u8_id='hls')
  53         }