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|radio)/[^/]+/(?P<media_type>video|audio)/(?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/play/radio/hoerspielarchiv-srf-musikwelle/audio/saegel-ohni-wind-von-jakob-stebler?id=415bf3d3-6429-4de7-968d-95866e37cfbc',
  40         'md5': '',
  41         'info_dict': {
  42             'id': '415bf3d3-6429-4de7-968d-95866e37cfbc',
  43             'display_id': 'saegel-ohni-wind-von-jakob-stebler',
  44             'ext': 'mp3',
  45             'upload_date': '20080518',
  46             'title': '«Sägel ohni Wind» von Jakob Stebler',
  47             'timestamp': 1211112000,
  48         },
  49         'params': {
  50             'skip_download': True,  # requires rtmpdump
  51         },
  52     }, {
  53         'url': 'http://www.srf.ch/player/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5',
  54         'only_matching': True,
  55     }, {
  56         'url': 'https://tp.srgssr.ch/p/flash?urn=urn:srf:ais:video:28e1a57d-5b76-4399-8ab3-9097f071e6c5',
  57         'only_matching': True,
  58     }]
  59
  60     def _real_extract(self, url):
  61         mobj = re.match(self._VALID_URL, url)
  62         video_id = mobj.group('id')
  63         media_type = mobj.group('media_type')
  64         display_id = mobj.group('display_id') or video_id
  65
  66         video_data = self._download_xml(
  67             'http://il.srgssr.ch/integrationlayer/1.0/ue/srf/%s/play/%s.xml' % (media_type, video_id),
  68             display_id)
  69
  70         title = xpath_text(
  71             video_data, './AssetMetadatas/AssetMetadata/title', fatal=True)
  72         thumbnails = [{
  73             'url': s.text
  74         } for s in video_data.findall('.//ImageRepresentation/url')]
  75         timestamp = parse_iso8601(xpath_text(video_data, './createdDate'))
  76         # The <duration> field in XML is different from the exact duration, skipping
  77
  78         formats = []
  79         for item in video_data.findall('./Playlists/Playlist') + video_data.findall('./Downloads/Download'):
  80             for url_node in item.findall('url'):
  81                 quality = url_node.attrib['quality']
  82                 full_url = url_node.text
  83                 original_ext = determine_ext(full_url).lower()
  84                 format_id = '%s-%s' % (quality, item.attrib['protocol'])
  85                 if original_ext == 'f4m':
  86                     formats.extend(self._extract_f4m_formats(
  87                         full_url + '?hdcore=3.4.0', display_id, f4m_id=format_id))
  88                 elif original_ext == 'm3u8':
  89                     formats.extend(self._extract_m3u8_formats(
  90                         full_url, display_id, 'mp4', m3u8_id=format_id))
  91                 else:
  92                     formats.append({
  93                         'url': full_url,
  94                         'ext': original_ext,
  95                         'format_id': format_id,
  96                         'quality': 0 if 'HD' in quality else -1,
  97                         'preference': 1,
  98                     })
  99
 100         self._sort_formats(formats)
 101
 102         subtitles = {}
 103         subtitles_data = video_data.find('Subtitles')
 104         if subtitles_data is not None:
 105             subtitles_list = [{
 106                 'url': sub.text,
 107                 'ext': determine_ext(sub.text),
 108             } for sub in subtitles_data]
 109             if subtitles_list:
 110                 subtitles['de'] = subtitles_list
 111
 112         return {
 113             'id': video_id,
 114             'display_id': display_id,
 115             'formats': formats,
 116             'title': title,
 117             'thumbnails': thumbnails,
 118             'timestamp': timestamp,
 119             'subtitles': subtitles,
 120         }