_ Git - youtube-dl/blob - youtube_dl/extractor/srf.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     determine_ext,
   8     parse_iso8601,
   9     xpath_text,
  10 )
  11
  12
  13 class SrfIE(InfoExtractor):
  14     _VALID_URL = r'http://www\.srf\.ch/play(?:er)?/tv/[^/]+/video/(?P<display_id>[^?]+)\?id=(?P<id>[0-9a-f\-]{36})'
  15     _TESTS = [{
  16         'url': 'http://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5',
  17         'md5': '4cd93523723beff51bb4bee974ee238d',
  18         'info_dict': {
  19             'id': '28e1a57d-5b76-4399-8ab3-9097f071e6c5',
  20             'display_id': 'snowden-beantragt-asyl-in-russland',
  21             'ext': 'm4v',
  22             'upload_date': '20130701',
  23             'title': 'Snowden beantragt Asyl in Russland',
  24             'timestamp': 1372713995,
  25         }
  26     }, {
  27         # No Speichern (Save) button
  28         'url': 'http://www.srf.ch/play/tv/top-gear/video/jaguar-xk120-shadow-und-tornado-dampflokomotive?id=677f5829-e473-4823-ac83-a1087fe97faa',
  29         'info_dict': {
  30             'id': '677f5829-e473-4823-ac83-a1087fe97faa',
  31             'display_id': 'jaguar-xk120-shadow-und-tornado-dampflokomotive',
  32             'ext': 'mp4',
  33             'upload_date': '20130710',
  34             'title': 'Jaguar XK120, Shadow und Tornado-Dampflokomotive',
  35             'timestamp': 1373493600,
  36         },
  37         'params': {
  38             # Require ffmpeg/avconv
  39             'skip_download': True,
  40         }
  41     }, {
  42         'url': 'http://www.srf.ch/player/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5',
  43         'only_matching': True,
  44     }]
  45
  46     def _real_extract(self, url):
  47         video_id = self._match_id(url)
  48
  49         video_data = self._download_xml(
  50             'http://il.srgssr.ch/integrationlayer/1.0/ue/srf/video/play/%s.xml' % video_id,
  51             video_id)
  52
  53         display_id = re.match(self._VALID_URL, url).group('display_id')
  54         title = xpath_text(
  55             video_data, './AssetMetadatas/AssetMetadata/title', fatal=True)
  56         thumbnails = [{
  57             'url': s.text
  58         } for s in video_data.findall('.//ImageRepresentation/url')]
  59         timestamp = parse_iso8601(xpath_text(video_data, './createdDate'))
  60         # The <duration> field in XML is different from the exact duration, skipping
  61
  62         formats = []
  63         for item in video_data.findall('./Playlists/Playlist') + video_data.findall('./Downloads/Download'):
  64             url_node = item.find('url')
  65             quality = url_node.attrib['quality']
  66             full_url = url_node.text
  67             original_ext = determine_ext(full_url)
  68             if original_ext == 'f4m':
  69                 full_url += '?hdcore=3.4.0'  # Without this, you get a 403 error
  70             formats.append({
  71                 'url': full_url,
  72                 'ext': 'mp4' if original_ext == 'm3u8' else original_ext,
  73                 'format_id': '%s-%s' % (quality, item.attrib['protocol']),
  74                 'preference': 0 if 'HD' in quality else -1,
  75             })
  76
  77         self._sort_formats(formats)
  78
  79         return {
  80             'id': video_id,
  81             'display_id': display_id,
  82             'formats': formats,
  83             'title': title,
  84             'thumbnails': thumbnails,
  85             'timestamp': timestamp,
  86         }