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

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import base64
   5 import re
   6 import time
   7
   8 from .common import InfoExtractor
   9 from ..compat import compat_urlparse
  10 from ..utils import (
  11     float_or_none,
  12     remove_end,
  13     struct_unpack,
  14 )
  15
  16
  17 def _decrypt_url(png):
  18     encrypted_data = base64.b64decode(png)
  19     text_index = encrypted_data.find(b'tEXt')
  20     text_chunk = encrypted_data[text_index - 4:]
  21     length = struct_unpack('!I', text_chunk[:4])[0]
  22     # Use bytearray to get integers when iterating in both python 2.x and 3.x
  23     data = bytearray(text_chunk[8:8 + length])
  24     data = [chr(b) for b in data if b != 0]
  25     hash_index = data.index('#')
  26     alphabet_data = data[:hash_index]
  27     url_data = data[hash_index + 1:]
  28
  29     alphabet = []
  30     e = 0
  31     d = 0
  32     for l in alphabet_data:
  33         if d == 0:
  34             alphabet.append(l)
  35             d = e = (e + 1) % 4
  36         else:
  37             d -= 1
  38     url = ''
  39     f = 0
  40     e = 3
  41     b = 1
  42     for letter in url_data:
  43         if f == 0:
  44             l = int(letter) * 10
  45             f = 1
  46         else:
  47             if e == 0:
  48                 l += int(letter)
  49                 url += alphabet[l]
  50                 e = (b + 3) % 4
  51                 f = 0
  52                 b += 1
  53             else:
  54                 e -= 1
  55
  56     return url
  57
  58
  59 class RTVEALaCartaIE(InfoExtractor):
  60     IE_NAME = 'rtve.es:alacarta'
  61     IE_DESC = 'RTVE a la carta'
  62     _VALID_URL = r'http://www\.rtve\.es/(m/)?alacarta/videos/[^/]+/[^/]+/(?P<id>\d+)'
  63
  64     _TESTS = [{
  65         'url': 'http://www.rtve.es/alacarta/videos/balonmano/o-swiss-cup-masculina-final-espana-suecia/2491869/',
  66         'md5': '1d49b7e1ca7a7502c56a4bf1b60f1b43',
  67         'info_dict': {
  68             'id': '2491869',
  69             'ext': 'mp4',
  70             'title': 'Balonmano - Swiss Cup masculina. Final: España-Suecia',
  71             'duration': 5024.566,
  72         },
  73     }, {
  74         'note': 'Live stream',
  75         'url': 'http://www.rtve.es/alacarta/videos/television/24h-live/1694255/',
  76         'info_dict': {
  77             'id': '1694255',
  78             'ext': 'flv',
  79             'title': 'TODO',
  80         },
  81         'skip': 'The f4m manifest can\'t be used yet',
  82     }, {
  83         'url': 'http://www.rtve.es/m/alacarta/videos/cuentame-como-paso/cuentame-como-paso-t16-ultimo-minuto-nuestra-vida-capitulo-276/2969138/?media=tve',
  84         'only_matching': True,
  85     }]
  86
  87     def _real_extract(self, url):
  88         mobj = re.match(self._VALID_URL, url)
  89         video_id = mobj.group('id')
  90         info = self._download_json(
  91             'http://www.rtve.es/api/videos/%s/config/alacarta_videos.json' % video_id,
  92             video_id)['page']['items'][0]
  93         png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/default/videos/%s.png' % video_id
  94         png = self._download_webpage(png_url, video_id, 'Downloading url information')
  95         video_url = _decrypt_url(png)
  96         if not video_url.endswith('.f4m'):
  97             auth_url = video_url.replace(
  98                 'resources/', 'auth/resources/'
  99             ).replace('.net.rtve', '.multimedia.cdn.rtve')
 100             video_path = self._download_webpage(
 101                 auth_url, video_id, 'Getting video url')
 102             # Use mvod1.akcdn instead of flash.akamaihd.multimedia.cdn to get
 103             # the right Content-Length header and the mp4 format
 104             video_url = compat_urlparse.urljoin(
 105                 'http://mvod1.akcdn.rtve.es/', video_path)
 106
 107         subtitles = None
 108         if info.get('sbtFile') is not None:
 109             subtitles = self.extract_subtitles(video_id, info['sbtFile'])
 110
 111         return {
 112             'id': video_id,
 113             'title': info['title'],
 114             'url': video_url,
 115             'thumbnail': info.get('image'),
 116             'page_url': url,
 117             'subtitles': subtitles,
 118             'duration': float_or_none(info.get('duration'), scale=1000),
 119         }
 120
 121     def _get_subtitles(self, video_id, sub_file):
 122         subs = self._download_json(
 123             sub_file + '.json', video_id,
 124             'Downloading subtitles info')['page']['items']
 125         return dict(
 126             (s['lang'], [{'ext': 'vtt', 'url': s['src']}])
 127             for s in subs)
 128
 129
 130 class RTVELiveIE(InfoExtractor):
 131     IE_NAME = 'rtve.es:live'
 132     IE_DESC = 'RTVE.es live streams'
 133     _VALID_URL = r'http://www\.rtve\.es/(?:deportes/directo|noticias|television)/(?P<id>[a-zA-Z0-9-]+)'
 134
 135     _TESTS = [{
 136         'url': 'http://www.rtve.es/noticias/directo-la-1/',
 137         'info_dict': {
 138             'id': 'directo-la-1',
 139             'ext': 'flv',
 140             'title': 're:^La 1 de TVE [0-9]{4}-[0-9]{2}-[0-9]{2}Z[0-9]{6}$',
 141         },
 142         'params': {
 143             'skip_download': 'live stream',
 144         }
 145     }]
 146
 147     def _real_extract(self, url):
 148         mobj = re.match(self._VALID_URL, url)
 149         start_time = time.gmtime()
 150         video_id = mobj.group('id')
 151
 152         webpage = self._download_webpage(url, video_id)
 153         player_url = self._search_regex(
 154             r'<param name="movie" value="([^"]+)"/>', webpage, 'player URL')
 155         title = remove_end(self._og_search_title(webpage), ' en directo')
 156         title += ' ' + time.strftime('%Y-%m-%dZ%H%M%S', start_time)
 157
 158         vidplayer_id = self._search_regex(
 159             r' id="vidplayer([0-9]+)"', webpage, 'internal video ID')
 160         png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/default/videos/%s.png' % vidplayer_id
 161         png = self._download_webpage(png_url, video_id, 'Downloading url information')
 162         video_url = _decrypt_url(png)
 163
 164         return {
 165             'id': video_id,
 166             'ext': 'flv',
 167             'title': title,
 168             'url': video_url,
 169             'app': 'rtve-live-live?ovpfv=2.1.2',
 170             'player_url': player_url,
 171             'rtmp_live': True,
 172         }