Raphaël G. Git Repositories - youtubedl/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'https?://(?:www\.srf\.ch/play(?:er)?/tv/[^/]+/video/(?P<display_id>[^?]+)\?id=|tp\.srgssr\.ch/p/flash\?urn=urn:srf:ais:video:)(?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         'md5': 'd97e236e80d1d24729e5d0953d276a4f',
  30         'info_dict': {
  31             'id': '677f5829-e473-4823-ac83-a1087fe97faa',
  32             'display_id': 'jaguar-xk120-shadow-und-tornado-dampflokomotive',
  33             'ext': 'flv',
  34             'upload_date': '20130710',
  35             'title': 'Jaguar XK120, Shadow und Tornado-Dampflokomotive',
  36             'timestamp': 1373493600,
  37         },
  38     }, {
  39         'url': 'http://www.srf.ch/player/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5',
  40         'only_matching': True,
  41     }, {
  42         'url': 'https://tp.srgssr.ch/p/flash?urn=urn:srf:ais:video: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         display_id = re.match(self._VALID_URL, url).group('display_id') or video_id
  49
  50         video_data = self._download_xml(
  51             'http://il.srgssr.ch/integrationlayer/1.0/ue/srf/video/play/%s.xml' % video_id,
  52             display_id)
  53
  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             for url_node in item.findall('url'):
  65                 quality = url_node.attrib['quality']
  66                 full_url = url_node.text
  67                 original_ext = determine_ext(full_url)
  68                 format_id = '%s-%s' % (quality, item.attrib['protocol'])
  69                 if original_ext == 'f4m':
  70                     formats.extend(self._extract_f4m_formats(
  71                         full_url + '?hdcore=3.4.0', display_id, f4m_id=format_id))
  72                 elif original_ext == 'm3u8':
  73                     formats.extend(self._extract_m3u8_formats(
  74                         full_url, display_id, 'mp4', m3u8_id=format_id))
  75                 else:
  76                     formats.append({
  77                         'url': full_url,
  78                         'ext': original_ext,
  79                         'format_id': format_id,
  80                         'quality': 0 if 'HD' in quality else -1,
  81                         'preference': 1,
  82                     })
  83
  84         self._sort_formats(formats)
  85
  86         subtitles = {}
  87         subtitles_data = video_data.find('Subtitles')
  88         if subtitles_data is not None:
  89             subtitles_list = [{
  90                 'url': sub.text,
  91                 'ext': determine_ext(sub.text),
  92             } for sub in subtitles_data]
  93             if subtitles_list:
  94                 subtitles['de'] = subtitles_list
  95
  96         return {
  97             'id': video_id,
  98             'display_id': display_id,
  99             'formats': formats,
 100             'title': title,
 101             'thumbnails': thumbnails,
 102             'timestamp': timestamp,
 103             'subtitles': subtitles,
 104         }