Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vrt.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     float_or_none,
   9 )
  10
  11
  12 class VRTIE(InfoExtractor):
  13     IE_DESC = 'deredactie.be, sporza.be, cobra.be and cobra.canvas.be'
  14     _VALID_URL = r'https?://(?:deredactie|sporza|cobra(?:\.canvas)?)\.be/cm/(?:[^/]+/)+(?P<id>[^/]+)/*'
  15     _TESTS = [
  16         # deredactie.be
  17         {
  18             'url': 'http://deredactie.be/cm/vrtnieuws/videozone/programmas/journaal/EP_141025_JOL',
  19             'md5': '4cebde1eb60a53782d4f3992cbd46ec8',
  20             'info_dict': {
  21                 'id': '2129880',
  22                 'ext': 'flv',
  23                 'title': 'Het journaal L - 25/10/14',
  24                 'description': None,
  25                 'timestamp': 1414271750.949,
  26                 'upload_date': '20141025',
  27                 'duration': 929,
  28             },
  29             'skip': 'HTTP Error 404: Not Found',
  30         },
  31         # sporza.be
  32         {
  33             'url': 'http://sporza.be/cm/sporza/videozone/programmas/extratime/EP_141020_Extra_time',
  34             'md5': '11f53088da9bf8e7cfc42456697953ff',
  35             'info_dict': {
  36                 'id': '2124639',
  37                 'ext': 'flv',
  38                 'title': 'Bekijk Extra Time van 20 oktober',
  39                 'description': 'md5:83ac5415a4f1816c6a93f8138aef2426',
  40                 'timestamp': 1413835980.560,
  41                 'upload_date': '20141020',
  42                 'duration': 3238,
  43             },
  44             'skip': 'HTTP Error 404: Not Found',
  45         },
  46         # cobra.be
  47         {
  48             'url': 'http://cobra.be/cm/cobra/videozone/rubriek/film-videozone/141022-mv-ellis-cafecorsari',
  49             'md5': '78a2b060a5083c4f055449a72477409d',
  50             'info_dict': {
  51                 'id': '2126050',
  52                 'ext': 'flv',
  53                 'title': 'Bret Easton Ellis in Café Corsari',
  54                 'description': 'md5:f699986e823f32fd6036c1855a724ee9',
  55                 'timestamp': 1413967500.494,
  56                 'upload_date': '20141022',
  57                 'duration': 661,
  58             },
  59             'skip': 'HTTP Error 404: Not Found',
  60         },
  61         {
  62             # YouTube video
  63             'url': 'http://deredactie.be/cm/vrtnieuws/videozone/nieuws/cultuurenmedia/1.2622957',
  64             'md5': 'b8b93da1df1cea6c8556255a796b7d61',
  65             'info_dict': {
  66                 'id': 'Wji-BZ0oCwg',
  67                 'ext': 'mp4',
  68                 'title': 'ROGUE ONE: A STAR WARS STORY Official Teaser Trailer',
  69                 'description': 'md5:8e468944dce15567a786a67f74262583',
  70                 'uploader': 'Star Wars',
  71                 'uploader_id': 'starwars',
  72                 'upload_date': '20160407',
  73             },
  74             'add_ie': ['Youtube'],
  75         },
  76         {
  77             'url': 'http://cobra.canvas.be/cm/cobra/videozone/rubriek/film-videozone/1.2377055',
  78             'info_dict': {
  79                 'id': '2377055',
  80                 'ext': 'mp4',
  81                 'title': 'Cafe Derby',
  82                 'description': 'Lenny Van Wesemael debuteert met de langspeelfilm Café Derby. Een waar gebeurd maar ook verzonnen verhaal.',
  83                 'upload_date': '20150626',
  84                 'timestamp': 1435305240.769,
  85             },
  86             'params': {
  87                 # m3u8 download
  88                 'skip_download': True,
  89             }
  90         }
  91     ]
  92
  93     def _real_extract(self, url):
  94         video_id = self._match_id(url)
  95
  96         webpage = self._download_webpage(url, video_id)
  97
  98         video_id = self._search_regex(
  99             r'data-video-id="([^"]+)_[^"]+"', webpage, 'video id', fatal=False)
 100
 101         src = self._search_regex(
 102             r'data-video-src="([^"]+)"', webpage, 'video src', default=None)
 103
 104         video_type = self._search_regex(
 105             r'data-video-type="([^"]+)"', webpage, 'video type', default=None)
 106
 107         if video_type == 'YouTubeVideo':
 108             return self.url_result(src, 'Youtube')
 109
 110         formats = []
 111
 112         mobj = re.search(
 113             r'data-video-iphone-server="(?P<server>[^"]+)"\s+data-video-iphone-path="(?P<path>[^"]+)"',
 114             webpage)
 115         if mobj:
 116             formats.extend(self._extract_m3u8_formats(
 117                 '%s/%s' % (mobj.group('server'), mobj.group('path')),
 118                 video_id, 'mp4', m3u8_id='hls', fatal=False))
 119
 120         if src:
 121             formats = self._extract_wowza_formats(src, video_id)
 122             if 'data-video-geoblocking="true"' not in webpage:
 123                 for f in formats:
 124                     if f['url'].startswith('rtsp://'):
 125                         http_format = f.copy()
 126                         http_format.update({
 127                             'url': f['url'].replace('rtsp://', 'http://').replace('vod.', 'download.').replace('/_definst_/', '/').replace('mp4:', ''),
 128                             'format_id': f['format_id'].replace('rtsp', 'http'),
 129                             'protocol': 'http',
 130                         })
 131                         formats.append(http_format)
 132
 133         if not formats and 'data-video-geoblocking="true"' in webpage:
 134             self.raise_geo_restricted('This video is only available in Belgium')
 135
 136         self._sort_formats(formats)
 137
 138         title = self._og_search_title(webpage)
 139         description = self._og_search_description(webpage, default=None)
 140         thumbnail = self._og_search_thumbnail(webpage)
 141         timestamp = float_or_none(self._search_regex(
 142             r'data-video-sitestat-pubdate="(\d+)"', webpage, 'timestamp', fatal=False), 1000)
 143         duration = float_or_none(self._search_regex(
 144             r'data-video-duration="(\d+)"', webpage, 'duration', fatal=False), 1000)
 145
 146         return {
 147             'id': video_id,
 148             'title': title,
 149             'description': description,
 150             'thumbnail': thumbnail,
 151             'timestamp': timestamp,
 152             'duration': duration,
 153             'formats': formats,
 154         }