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 (
   8     int_or_none,
   9     parse_duration,
  10     parse_iso8601,
  11 )
  12
  13
  14 class AirMozillaIE(InfoExtractor):
  15     _VALID_URL = r'https?://air\.mozilla\.org/(?P<id>[0-9a-z-]+)/?'
  16     _TEST = {
  17         'url': 'https://air.mozilla.org/privacy-lab-a-meetup-for-privacy-minded-people-in-san-francisco/',
  18         'md5': '2e3e7486ba5d180e829d453875b9b8bf',
  19         'info_dict': {
  20             'id': '6x4q2w',
  21             'ext': 'mp4',
  22             'title': 'Privacy Lab - a meetup for privacy minded people in San Francisco',
  23             'thumbnail': r're:https?://vid\.ly/(?P<id>[0-9a-z-]+)/poster',
  24             '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...',
  25             'timestamp': 1422487800,
  26             'upload_date': '20150128',
  27             'location': 'SFO Commons',
  28             'duration': 3780,
  29             'view_count': int,
  30             'categories': ['Main', 'Privacy'],
  31         }
  32     }
  33
  34     def _real_extract(self, url):
  35         display_id = self._match_id(url)
  36         webpage = self._download_webpage(url, display_id)
  37         video_id = self._html_search_regex(r'//vid.ly/(.*?)/embed', webpage, 'id')
  38
  39         embed_script = self._download_webpage('https://vid.ly/{0}/embed'.format(video_id), video_id)
  40         jwconfig = self._search_regex(r'\svar jwconfig = (\{.*?\});\s', embed_script, 'metadata')
  41         metadata = self._parse_json(jwconfig, video_id)
  42
  43         formats = [{
  44             'url': source['file'],
  45             'ext': source['type'],
  46             'format_id': self._search_regex(r'&format=(.*)$', source['file'], 'video format'),
  47             'format': source['label'],
  48             'height': int(source['label'].rstrip('p')),
  49         } for source in metadata['playlist'][0]['sources']]
  50         self._sort_formats(formats)
  51
  52         view_count = int_or_none(self._html_search_regex(
  53             r'Views since archived: ([0-9]+)',
  54             webpage, 'view count', fatal=False))
  55         timestamp = parse_iso8601(self._html_search_regex(
  56             r'<time datetime="(.*?)"', webpage, 'timestamp', fatal=False))
  57         duration = parse_duration(self._search_regex(
  58             r'Duration:\s*(\d+\s*hours?\s*\d+\s*minutes?)',
  59             webpage, 'duration', fatal=False))
  60
  61         return {
  62             'id': video_id,
  63             'title': self._og_search_title(webpage),
  64             'formats': formats,
  65             'url': self._og_search_url(webpage),
  66             'display_id': display_id,
  67             'thumbnail': metadata['playlist'][0].get('image'),
  68             'description': self._og_search_description(webpage),
  69             'timestamp': timestamp,
  70             'location': self._html_search_regex(r'Location: (.*)', webpage, 'location', default=None),
  71             'duration': duration,
  72             'view_count': view_count,
  73             'categories': re.findall(r'<a href=".*?" class="channel">(.*?)</a>', webpage),
  74         }