]> gitweb @ CieloNegro.org - youtube-dl.git/blob - youtube_dl/extractor/srf.py
e303516027cf9f941a2510d84cd7620cd9bf4231
[youtube-dl.git] / 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         '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
43     def _real_extract(self, url):
44         video_id = self._match_id(url)
45
46         video_data = self._download_xml(
47             'http://il.srgssr.ch/integrationlayer/1.0/ue/srf/video/play/%s.xml' % video_id,
48             video_id)
49
50         display_id = re.match(self._VALID_URL, url).group('display_id')
51         title = xpath_text(
52             video_data, './AssetMetadatas/AssetMetadata/title', fatal=True)
53         thumbnails = [{
54             'url': s.text
55         } for s in video_data.findall('.//ImageRepresentation/url')]
56         timestamp = parse_iso8601(xpath_text(video_data, './createdDate'))
57         # The <duration> field in XML is different from the exact duration, skipping
58
59         formats = []
60         for item in video_data.findall('./Playlists/Playlist') + video_data.findall('./Downloads/Download'):
61             for url_node in item.findall('url'):
62                 quality = url_node.attrib['quality']
63                 full_url = url_node.text
64                 original_ext = determine_ext(full_url)
65                 format_id = '%s-%s' % (quality, item.attrib['protocol'])
66                 if original_ext == 'f4m':
67                     formats.extend(self._extract_f4m_formats(
68                         full_url + '?hdcore=3.4.0', video_id, f4m_id=format_id))
69                 elif original_ext == 'm3u8':
70                     formats.extend(self._extract_m3u8_formats(
71                         full_url, video_id, 'mp4', m3u8_id=format_id))
72                 else:
73                     formats.append({
74                         'url': full_url,
75                         'ext': original_ext,
76                         'format_id': format_id,
77                         'quality': 0 if 'HD' in quality else -1,
78                         'preference': 1,
79                     })
80
81         self._sort_formats(formats)
82
83         subtitles = {}
84         subtitles_data = video_data.find('Subtitles')
85         if subtitles_data is not None:
86             subtitles_list = [{
87                 'url': sub.text,
88                 'ext': determine_ext(sub.text),
89             } for sub in subtitles_data]
90             if subtitles_list:
91                 subtitles['de'] = subtitles_list
92
93         return {
94             'id': video_id,
95             'display_id': display_id,
96             'formats': formats,
97             'title': title,
98             'thumbnails': thumbnails,
99             'timestamp': timestamp,
100             'subtitles': subtitles,
101         }