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

   1 from __future__ import unicode_literals
   2
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import (
   7     compat_urllib_parse,
   8 )
   9 from ..utils import (
  10     parse_duration,
  11     unified_strdate,
  12 )
  13
  14
  15 class RaiIE(InfoExtractor):
  16     _VALID_URL = r'(?P<url>(?P<host>http://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it))/dl/.+?-(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html)'
  17     _TESTS = [
  18         {
  19             'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-cb27157f-9dd0-4aee-b788-b1f67643a391.html',
  20             'md5': 'c064c0b2d09c278fb293116ef5d0a32d',
  21             'info_dict': {
  22                 'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
  23                 'ext': 'mp4',
  24                 'title': 'Report del 07/04/2014',
  25                 'description': 'md5:f27c544694cacb46a078db84ec35d2d9',
  26                 'upload_date': '20140407',
  27                 'duration': 6160,
  28             }
  29         },
  30         {
  31             'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
  32             'md5': '8bb9c151924ce241b74dd52ef29ceafa',
  33             'info_dict': {
  34                 'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
  35                 'ext': 'mp4',
  36                 'title': 'TG PRIMO TEMPO',
  37                 'description': '',
  38                 'upload_date': '20140612',
  39                 'duration': 1758,
  40             },
  41             'skip': 'Error 404',
  42         },
  43         {
  44             'url': 'http://www.rainews.it/dl/rainews/media/state-of-the-net-Antonella-La-Carpia-regole-virali-7aafdea9-0e5d-49d5-88a6-7e65da67ae13.html',
  45             'md5': '35cf7c229f22eeef43e48b5cf923bef0',
  46             'info_dict': {
  47                 'id': '7aafdea9-0e5d-49d5-88a6-7e65da67ae13',
  48                 'ext': 'mp4',
  49                 'title': 'State of the Net, Antonella La Carpia: regole virali',
  50                 'description': 'md5:b0ba04a324126903e3da7763272ae63c',
  51                 'upload_date': '20140613',
  52             },
  53             'skip': 'Error 404',
  54         },
  55         {
  56             'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-b4a49761-e0cc-4b14-8736-2729f6f73132-tg2.html',
  57             'md5': '35694f062977fe6619943f08ed935730',
  58             'info_dict': {
  59                 'id': 'b4a49761-e0cc-4b14-8736-2729f6f73132',
  60                 'ext': 'mp4',
  61                 'title': 'Alluvione in Sardegna e dissesto idrogeologico',
  62                 'description': 'Edizione delle ore 20:30 ',
  63             }
  64         },
  65         {
  66             'url': 'http://www.ilcandidato.rai.it/dl/ray/media/Il-Candidato---Primo-episodio-Le-Primarie-28e5525a-b495-45e8-a7c3-bc48ba45d2b6.html',
  67             'md5': '02b64456f7cc09f96ff14e7dd489017e',
  68             'info_dict': {
  69                 'id': '28e5525a-b495-45e8-a7c3-bc48ba45d2b6',
  70                 'ext': 'flv',
  71                 'title': 'Il Candidato - Primo episodio: "Le Primarie"',
  72                 'description': 'Primo appuntamento con "Il candidato" con Filippo Timi, alias Piero Zucca presidente!',
  73                 'uploader': 'RaiTre',
  74             }
  75         }
  76     ]
  77
  78     def _extract_relinker_url(self, webpage):
  79         return self._proto_relative_url(self._search_regex(
  80             [r'name="videourl" content="([^"]+)"', r'var\s+videoURL(?:_MP4)?\s*=\s*"([^"]+)"'],
  81             webpage, 'relinker url', default=None))
  82
  83     def _real_extract(self, url):
  84         mobj = re.match(self._VALID_URL, url)
  85         video_id = mobj.group('id')
  86         host = mobj.group('host')
  87
  88         webpage = self._download_webpage(url, video_id)
  89
  90         relinker_url = self._extract_relinker_url(webpage)
  91
  92         if not relinker_url:
  93             iframe_path = self._search_regex(
  94                 r'<iframe[^>]+src="/?(dl/[^"]+\?iframe\b[^"]*)"',
  95                 webpage, 'iframe')
  96             webpage = self._download_webpage(
  97                 '%s/%s' % (host, iframe_path), video_id)
  98             relinker_url = self._extract_relinker_url(webpage)
  99
 100         relinker = self._download_json(
 101             '%s&output=47' % relinker_url, video_id)
 102
 103         media_url = relinker['video'][0]
 104         ct = relinker.get('ct')
 105         if ct == 'f4m':
 106             formats = self._extract_f4m_formats(
 107                 media_url + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44', video_id)
 108         else:
 109             formats = [{
 110                 'url': media_url,
 111                 'format_id': ct,
 112             }]
 113
 114         json_link = self._html_search_meta(
 115             'jsonlink', webpage, 'JSON link', default=None)
 116         if json_link:
 117             media = self._download_json(
 118                 host + json_link, video_id, 'Downloading video JSON')
 119             title = media.get('name')
 120             description = media.get('desc')
 121             thumbnail = media.get('image_300') or media.get('image_medium') or media.get('image')
 122             duration = parse_duration(media.get('length'))
 123             uploader = media.get('author')
 124             upload_date = unified_strdate(media.get('date'))
 125         else:
 126             title = (self._search_regex(
 127                 r'var\s+videoTitolo\s*=\s*"(.+?)";',
 128                 webpage, 'title', default=None) or self._og_search_title(webpage)).replace('\\"', '"')
 129             description = self._og_search_description(webpage)
 130             thumbnail = self._og_search_thumbnail(webpage)
 131             duration = None
 132             uploader = self._html_search_meta('Editore', webpage, 'uploader')
 133             upload_date = unified_strdate(self._html_search_meta(
 134                 'item-date', webpage, 'upload date', default=None))
 135
 136         subtitles = self.extract_subtitles(video_id, webpage)
 137
 138         return {
 139             'id': video_id,
 140             'title': title,
 141             'description': description,
 142             'thumbnail': thumbnail,
 143             'uploader': uploader,
 144             'upload_date': upload_date,
 145             'duration': duration,
 146             'formats': formats,
 147             'subtitles': subtitles,
 148         }
 149
 150     def _get_subtitles(self, video_id, webpage):
 151         subtitles = {}
 152         m = re.search(r'<meta name="closedcaption" content="(?P<captions>[^"]+)"', webpage)
 153         if m:
 154             captions = m.group('captions')
 155             STL_EXT = '.stl'
 156             SRT_EXT = '.srt'
 157             if captions.endswith(STL_EXT):
 158                 captions = captions[:-len(STL_EXT)] + SRT_EXT
 159             subtitles['it'] = [{
 160                 'ext': 'srt',
 161                 'url': 'http://www.rai.tv%s' % compat_urllib_parse.quote(captions),
 162             }]
 163         return subtitles