Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/internazionale.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import unified_timestamp
   6
   7
   8 class InternazionaleIE(InfoExtractor):
   9     _VALID_URL = r'https?://(?:www\.)?internazionale\.it/video/(?:[^/]+/)*(?P<id>[^/?#&]+)'
  10     _TEST = {
  11         'url': 'https://www.internazionale.it/video/2015/02/19/richard-linklater-racconta-una-scena-di-boyhood',
  12         'md5': '3e39d32b66882c1218e305acbf8348ca',
  13         'info_dict': {
  14             'id': '265968',
  15             'display_id': 'richard-linklater-racconta-una-scena-di-boyhood',
  16             'ext': 'mp4',
  17             'title': 'Richard Linklater racconta una scena di Boyhood',
  18             'description': 'md5:efb7e5bbfb1a54ae2ed5a4a015f0e665',
  19             'timestamp': 1424354635,
  20             'upload_date': '20150219',
  21             'thumbnail': r're:^https?://.*\.jpg$',
  22         },
  23         'params': {
  24             'format': 'bestvideo',
  25         },
  26     }
  27
  28     def _real_extract(self, url):
  29         display_id = self._match_id(url)
  30
  31         webpage = self._download_webpage(url, display_id)
  32
  33         DATA_RE = r'data-%s=(["\'])(?P<value>(?:(?!\1).)+)\1'
  34
  35         title = self._search_regex(
  36             DATA_RE % 'video-title', webpage, 'title', default=None,
  37             group='value') or self._og_search_title(webpage)
  38
  39         video_id = self._search_regex(
  40             DATA_RE % 'job-id', webpage, 'video id', group='value')
  41         video_path = self._search_regex(
  42             DATA_RE % 'video-path', webpage, 'video path', group='value')
  43
  44         video_base = 'https://video.internazionale.it/%s/%s.' % (video_path, video_id)
  45
  46         formats = self._extract_m3u8_formats(
  47             video_base + 'm3u8', display_id, 'mp4',
  48             entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)
  49         formats.extend(self._extract_mpd_formats(
  50             video_base + 'mpd', display_id, mpd_id='dash', fatal=False))
  51         self._sort_formats(formats)
  52
  53         timestamp = unified_timestamp(self._html_search_meta(
  54             'article:published_time', webpage, 'timestamp'))
  55
  56         return {
  57             'id': video_id,
  58             'display_id': display_id,
  59             'title': title,
  60             'thumbnail': self._og_search_thumbnail(webpage),
  61             'description': self._og_search_description(webpage),
  62             'timestamp': timestamp,
  63             'formats': formats,
  64         }