Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/expressen.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     determine_ext,
   9     int_or_none,
  10     unescapeHTML,
  11     unified_timestamp,
  12 )
  13
  14
  15 class ExpressenIE(InfoExtractor):
  16     _VALID_URL = r'''(?x)
  17                     https?://
  18                         (?:www\.)?expressen\.se/
  19                         (?:(?:tvspelare/video|videoplayer/embed)/)?
  20                         tv/(?:[^/]+/)*
  21                         (?P<id>[^/?#&]+)
  22                     '''
  23     _TESTS = [{
  24         'url': 'https://www.expressen.se/tv/ledare/ledarsnack/ledarsnack-om-arbetslosheten-bland-kvinnor-i-speciellt-utsatta-omraden/',
  25         'md5': '2fbbe3ca14392a6b1b36941858d33a45',
  26         'info_dict': {
  27             'id': '8690962',
  28             'ext': 'mp4',
  29             'title': 'Ledarsnack: Om arbetslösheten bland kvinnor i speciellt utsatta områden',
  30             'description': 'md5:f38c81ff69f3de4d269bbda012fcbbba',
  31             'thumbnail': r're:^https?://.*\.jpg$',
  32             'duration': 788,
  33             'timestamp': 1526639109,
  34             'upload_date': '20180518',
  35         },
  36     }, {
  37         'url': 'https://www.expressen.se/tv/kultur/kulturdebatt-med-expressens-karin-olsson/',
  38         'only_matching': True,
  39     }, {
  40         'url': 'https://www.expressen.se/tvspelare/video/tv/ditv/ekonomistudion/experterna-har-ar-fragorna-som-avgor-valet/?embed=true&external=true&autoplay=true&startVolume=0&partnerId=di',
  41         'only_matching': True,
  42     }, {
  43         'url': 'https://www.expressen.se/videoplayer/embed/tv/ditv/ekonomistudion/experterna-har-ar-fragorna-som-avgor-valet/?embed=true&external=true&autoplay=true&startVolume=0&partnerId=di',
  44         'only_matching': True,
  45     }]
  46
  47     @staticmethod
  48     def _extract_urls(webpage):
  49         return [
  50             mobj.group('url') for mobj in re.finditer(
  51                 r'<iframe[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//(?:www\.)?expressen\.se/(?:tvspelare/video|videoplayer/embed)/tv/.+?)\1',
  52                 webpage)]
  53
  54     def _real_extract(self, url):
  55         display_id = self._match_id(url)
  56
  57         webpage = self._download_webpage(url, display_id)
  58
  59         def extract_data(name):
  60             return self._parse_json(
  61                 self._search_regex(
  62                     r'data-%s=(["\'])(?P<value>(?:(?!\1).)+)\1' % name,
  63                     webpage, 'info', group='value'),
  64                 display_id, transform_source=unescapeHTML)
  65
  66         info = extract_data('video-tracking-info')
  67         video_id = info['videoId']
  68
  69         data = extract_data('article-data')
  70         stream = data['stream']
  71
  72         if determine_ext(stream) == 'm3u8':
  73             formats = self._extract_m3u8_formats(
  74                 stream, display_id, 'mp4', entry_protocol='m3u8_native',
  75                 m3u8_id='hls')
  76         else:
  77             formats = [{
  78                 'url': stream,
  79             }]
  80         self._sort_formats(formats)
  81
  82         title = info.get('titleRaw') or data['title']
  83         description = info.get('descriptionRaw')
  84         thumbnail = info.get('socialMediaImage') or data.get('image')
  85         duration = int_or_none(info.get('videoTotalSecondsDuration') or
  86                                data.get('totalSecondsDuration'))
  87         timestamp = unified_timestamp(info.get('publishDate'))
  88
  89         return {
  90             'id': video_id,
  91             'display_id': display_id,
  92             'title': title,
  93             'description': description,
  94             'thumbnail': thumbnail,
  95             'duration': duration,
  96             'timestamp': timestamp,
  97             'formats': formats,
  98         }