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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7
   8
   9 class TeleBruxellesIE(InfoExtractor):
  10     _VALID_URL = r'https?://(?:www\.)?(?:telebruxelles|bx1)\.be/(?:[^/]+/)*(?P<id>[^/#?]+)'
  11     _TESTS = [{
  12         'url': 'http://bx1.be/news/que-risque-lauteur-dune-fausse-alerte-a-la-bombe/',
  13         'md5': 'a2a67a5b1c3e8c9d33109b902f474fd9',
  14         'info_dict': {
  15             'id': '158856',
  16             'display_id': 'que-risque-lauteur-dune-fausse-alerte-a-la-bombe',
  17             'ext': 'mp4',
  18             'title': 'Que risque l’auteur d’une fausse alerte à la bombe ?',
  19             'description': 'md5:3cf8df235d44ebc5426373050840e466',
  20         },
  21     }, {
  22         'url': 'http://bx1.be/sport/futsal-schaerbeek-sincline-5-3-a-thulin/',
  23         'md5': 'dfe07ecc9c153ceba8582ac912687675',
  24         'info_dict': {
  25             'id': '158433',
  26             'display_id': 'futsal-schaerbeek-sincline-5-3-a-thulin',
  27             'ext': 'mp4',
  28             'title': 'Futsal : Schaerbeek s’incline 5-3 à Thulin',
  29             'description': 'md5:fd013f1488d5e2dceb9cebe39e2d569b',
  30         },
  31     }, {
  32         'url': 'http://bx1.be/emission/bxenf1-gastronomie/',
  33         'only_matching': True,
  34     }, {
  35         'url': 'https://bx1.be/berchem-sainte-agathe/personnel-carrefour-de-berchem-sainte-agathe-inquiet/',
  36         'only_matching': True,
  37     }, {
  38         'url': 'https://bx1.be/dernier-jt/',
  39         'only_matching': True,
  40     }, {
  41         # live stream
  42         'url': 'https://bx1.be/lives/direct-tv/',
  43         'only_matching': True,
  44     }]
  45
  46     def _real_extract(self, url):
  47         display_id = self._match_id(url)
  48         webpage = self._download_webpage(url, display_id)
  49
  50         article_id = self._html_search_regex(
  51             r'<article[^>]+\bid=["\']post-(\d+)', webpage, 'article ID', default=None)
  52         title = self._html_search_regex(
  53             r'<h1[^>]*>(.+?)</h1>', webpage, 'title',
  54             default=None) or self._og_search_title(webpage)
  55         description = self._og_search_description(webpage, default=None)
  56
  57         rtmp_url = self._html_search_regex(
  58             r'file["\']?\s*:\s*"(r(?:tm|mt)ps?://[^/]+/(?:vod/mp4:"\s*\+\s*"[^"]+"\s*\+\s*"\.mp4|stream/live))"',
  59             webpage, 'RTMP url')
  60         # Yes, they have a typo in scheme name for live stream URLs (e.g.
  61         # https://bx1.be/lives/direct-tv/)
  62         rtmp_url = re.sub(r'^rmtp', 'rtmp', rtmp_url)
  63         rtmp_url = re.sub(r'"\s*\+\s*"', '', rtmp_url)
  64         formats = self._extract_wowza_formats(rtmp_url, article_id or display_id)
  65         self._sort_formats(formats)
  66
  67         is_live = 'stream/live' in rtmp_url
  68
  69         return {
  70             'id': article_id or display_id,
  71             'display_id': display_id,
  72             'title': self._live_title(title) if is_live else title,
  73             'description': description,
  74             'formats': formats,
  75             'is_live': is_live,
  76         }