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 ..utils import (
  10     ExtractorError,
  11     float_or_none,
  12     remove_end,
  13     remove_start,
  14     sanitized_Request,
  15     std_headers,
  16     struct_unpack,
  17 )
  18
  19
  20 def _decrypt_url(png):
  21     encrypted_data = base64.b64decode(png.encode('utf-8'))
  22     text_index = encrypted_data.find(b'tEXt')
  23     text_chunk = encrypted_data[text_index - 4:]
  24     length = struct_unpack('!I', text_chunk[:4])[0]
  25     # Use bytearray to get integers when iterating in both python 2.x and 3.x
  26     data = bytearray(text_chunk[8:8 + length])
  27     data = [chr(b) for b in data if b != 0]
  28     hash_index = data.index('#')
  29     alphabet_data = data[:hash_index]
  30     url_data = data[hash_index + 1:]
  31
  32     alphabet = []
  33     e = 0
  34     d = 0
  35     for l in alphabet_data:
  36         if d == 0:
  37             alphabet.append(l)
  38             d = e = (e + 1) % 4
  39         else:
  40             d -= 1
  41     url = ''
  42     f = 0
  43     e = 3
  44     b = 1
  45     for letter in url_data:
  46         if f == 0:
  47             l = int(letter) * 10
  48             f = 1
  49         else:
  50             if e == 0:
  51                 l += int(letter)
  52                 url += alphabet[l]
  53                 e = (b + 3) % 4
  54                 f = 0
  55                 b += 1
  56             else:
  57                 e -= 1
  58
  59     return url
  60
  61
  62 class RTVEALaCartaIE(InfoExtractor):
  63     IE_NAME = 'rtve.es:alacarta'
  64     IE_DESC = 'RTVE a la carta'
  65     _VALID_URL = r'https?://www\.rtve\.es/(m/)?alacarta/videos/[^/]+/[^/]+/(?P<id>\d+)'
  66
  67     _TESTS = [{
  68         'url': 'http://www.rtve.es/alacarta/videos/balonmano/o-swiss-cup-masculina-final-espana-suecia/2491869/',
  69         'md5': '1d49b7e1ca7a7502c56a4bf1b60f1b43',
  70         'info_dict': {
  71             'id': '2491869',
  72             'ext': 'mp4',
  73             'title': 'Balonmano - Swiss Cup masculina. Final: España-Suecia',
  74             'duration': 5024.566,
  75         },
  76     }, {
  77         'note': 'Live stream',
  78         'url': 'http://www.rtve.es/alacarta/videos/television/24h-live/1694255/',
  79         'info_dict': {
  80             'id': '1694255',
  81             'ext': 'flv',
  82             'title': 'TODO',
  83         },
  84         'skip': 'The f4m manifest can\'t be used yet',
  85     }, {
  86         'url': 'http://www.rtve.es/m/alacarta/videos/cuentame-como-paso/cuentame-como-paso-t16-ultimo-minuto-nuestra-vida-capitulo-276/2969138/?media=tve',
  87         'only_matching': True,
  88     }]
  89
  90     def _real_initialize(self):
  91         user_agent_b64 = base64.b64encode(std_headers['User-Agent'].encode('utf-8')).decode('utf-8')
  92         manager_info = self._download_json(
  93             'http://www.rtve.es/odin/loki/' + user_agent_b64,
  94             None, 'Fetching manager info')
  95         self._manager = manager_info['manager']
  96
  97     def _real_extract(self, url):
  98         mobj = re.match(self._VALID_URL, url)
  99         video_id = mobj.group('id')
 100         info = self._download_json(
 101             'http://www.rtve.es/api/videos/%s/config/alacarta_videos.json' % video_id,
 102             video_id)['page']['items'][0]
 103         if info['state'] == 'DESPU':
 104             raise ExtractorError('The video is no longer available', expected=True)
 105         png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/%s/videos/%s.png' % (self._manager, video_id)
 106         png_request = sanitized_Request(png_url)
 107         png_request.add_header('Referer', url)
 108         png = self._download_webpage(png_request, video_id, 'Downloading url information')
 109         video_url = _decrypt_url(png)
 110         if not video_url.endswith('.f4m'):
 111             video_url = video_url.replace(
 112                 'resources/', 'auth/resources/'
 113             ).replace('.net.rtve', '.multimedia.cdn.rtve')
 114
 115         subtitles = None
 116         if info.get('sbtFile') is not None:
 117             subtitles = self.extract_subtitles(video_id, info['sbtFile'])
 118
 119         return {
 120             'id': video_id,
 121             'title': info['title'],
 122             'url': video_url,
 123             'thumbnail': info.get('image'),
 124             'page_url': url,
 125             'subtitles': subtitles,
 126             'duration': float_or_none(info.get('duration'), scale=1000),
 127         }
 128
 129     def _get_subtitles(self, video_id, sub_file):
 130         subs = self._download_json(
 131             sub_file + '.json', video_id,
 132             'Downloading subtitles info')['page']['items']
 133         return dict(
 134             (s['lang'], [{'ext': 'vtt', 'url': s['src']}])
 135             for s in subs)
 136
 137
 138 class RTVEInfantilIE(InfoExtractor):
 139     IE_NAME = 'rtve.es:infantil'
 140     IE_DESC = 'RTVE infantil'
 141     _VALID_URL = r'https?://(?:www\.)?rtve\.es/infantil/serie/(?P<show>[^/]*)/video/(?P<short_title>[^/]*)/(?P<id>[0-9]+)/'
 142
 143     _TESTS = [{
 144         'url': 'http://www.rtve.es/infantil/serie/cleo/video/maneras-vivir/3040283/',
 145         'md5': '915319587b33720b8e0357caaa6617e6',
 146         'info_dict': {
 147             'id': '3040283',
 148             'ext': 'mp4',
 149             'title': 'Maneras de vivir',
 150             'thumbnail': 'http://www.rtve.es/resources/jpg/6/5/1426182947956.JPG',
 151             'duration': 357.958,
 152         },
 153     }]
 154
 155     def _real_extract(self, url):
 156         video_id = self._match_id(url)
 157         info = self._download_json(
 158             'http://www.rtve.es/api/videos/%s/config/alacarta_videos.json' % video_id,
 159             video_id)['page']['items'][0]
 160
 161         webpage = self._download_webpage(url, video_id)
 162         vidplayer_id = self._search_regex(
 163             r' id="vidplayer([0-9]+)"', webpage, 'internal video ID')
 164
 165         png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/default/videos/%s.png' % vidplayer_id
 166         png = self._download_webpage(png_url, video_id, 'Downloading url information')
 167         video_url = _decrypt_url(png)
 168
 169         return {
 170             'id': video_id,
 171             'ext': 'mp4',
 172             'title': info['title'],
 173             'url': video_url,
 174             'thumbnail': info.get('image'),
 175             'duration': float_or_none(info.get('duration'), scale=1000),
 176         }
 177
 178
 179 class RTVELiveIE(InfoExtractor):
 180     IE_NAME = 'rtve.es:live'
 181     IE_DESC = 'RTVE.es live streams'
 182     _VALID_URL = r'https?://www\.rtve\.es/directo/(?P<id>[a-zA-Z0-9-]+)'
 183
 184     _TESTS = [{
 185         'url': 'http://www.rtve.es/directo/la-1/',
 186         'info_dict': {
 187             'id': 'la-1',
 188             'ext': 'mp4',
 189             'title': 're:^La 1 [0-9]{4}-[0-9]{2}-[0-9]{2}Z[0-9]{6}$',
 190         },
 191         'params': {
 192             'skip_download': 'live stream',
 193         }
 194     }]
 195
 196     def _real_extract(self, url):
 197         mobj = re.match(self._VALID_URL, url)
 198         start_time = time.gmtime()
 199         video_id = mobj.group('id')
 200
 201         webpage = self._download_webpage(url, video_id)
 202         title = remove_end(self._og_search_title(webpage), ' en directo en RTVE.es')
 203         title = remove_start(title, 'Estoy viendo ')
 204         title += ' ' + time.strftime('%Y-%m-%dZ%H%M%S', start_time)
 205
 206         vidplayer_id = self._search_regex(
 207             r'playerId=player([0-9]+)', webpage, 'internal video ID')
 208         png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/amonet/videos/%s.png' % vidplayer_id
 209         png = self._download_webpage(png_url, video_id, 'Downloading url information')
 210         m3u8_url = _decrypt_url(png)
 211         formats = self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4')
 212
 213         return {
 214             'id': video_id,
 215             'title': title,
 216             'formats': formats,
 217             'is_live': True,
 218         }