_ Git - youtube-dl/blob - youtube_dl/extractor/airmozilla.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import parse_iso8601
   8
   9
  10 class AirMozillaIE(InfoExtractor):
  11     _VALID_URL = r'https?://air\.mozilla\.org/(?P<id>[0-9a-z-]+)/?'
  12     _TEST = {
  13         'url': 'https://air.mozilla.org/privacy-lab-a-meetup-for-privacy-minded-people-in-san-francisco/',
  14         'md5': '2e3e7486ba5d180e829d453875b9b8bf',
  15         'info_dict': {
  16             'id': '6x4q2w',
  17             'ext': 'mp4',
  18             'title': 'Privacy Lab - a meetup for privacy minded people in San Francisco',
  19             'thumbnail': 're:https://\w+\.cloudfront\.net/6x4q2w/poster\.jpg\?t=\d+',
  20             'description': 'Brings together privacy professionals and others interested in privacy at for-profits, non-profits, and NGOs in an effort to contribute to the state of the ecosystem...',
  21             'timestamp': 1422487800,
  22             'upload_date': '20150128',
  23             'location': 'SFO Commons',
  24             'duration': 3780,
  25             'view_count': int,
  26             'categories': ['Main'],
  27         }
  28     }
  29
  30     _QUALITY_MAP = {
  31         '360p': 0,
  32         '576p': 1,
  33         '640p': 2,
  34         '720p': 3,
  35     }
  36
  37     def _real_extract(self, url):
  38         display_id = self._match_id(url)
  39         webpage = self._download_webpage(url, display_id)
  40         video_id = self._html_search_regex(r'//vid.ly/(.*?)/embed', webpage, 'id')
  41
  42         embed_script = self._download_webpage('https://vid.ly/{0}/embed'.format(video_id), video_id)
  43         jwconfig = self._search_regex(r'\svar jwconfig = (\{.*?\});\s', embed_script, 'metadata')
  44         metadata = self._parse_json(jwconfig, video_id)
  45
  46         formats = []
  47         for source in metadata['playlist'][0]['sources']:
  48             fmt = {
  49                 'url': source['file'],
  50                 'ext': source['type'],
  51                 'format_id': self._search_regex(r'&format=(.*)$', source['file'], 'video format'),
  52                 'resolution': source['label'],
  53                 'quality': self._QUALITY_MAP.get(source['label'], -1),
  54             }
  55             formats.append(fmt)
  56         self._sort_formats(formats)
  57
  58         duration_match = re.search(r'Duration:(?: (?P<H>\d+) hours?)?(?: (?P<M>\d+) minutes?)?', webpage)
  59
  60         return {
  61             'id': video_id,
  62             'title': self._og_search_title(webpage),
  63             'formats': formats,
  64             'url': self._og_search_url(webpage),
  65             'display_id': display_id,
  66             'thumbnail': metadata['playlist'][0]['image'],
  67             'description': self._og_search_description(webpage),
  68             'timestamp': parse_iso8601(self._html_search_regex(r'<time datetime="(.*?)"', webpage, 'timestamp')),
  69             'location': self._html_search_regex(r'Location: (.*)', webpage, 'location', default=None),
  70             'duration': int(duration_match.groupdict()['H'] or 0) * 3600 + int(duration_match.groupdict()['M'] or 0) * 60,
  71             'view_count': int(self._html_search_regex(r'Views since archived: ([0-9]+)', webpage, 'view count')),
  72             'categories': re.findall(r'<a href=".*?" class="channel">(.*?)</a>', webpage),
  73         }