Fix "invalid escape sequences" error on Python 3.6

[youtube-dl] / youtube_dl / extractor / srgssr.py

diff --git a/youtube_dl/extractor/srgssr.py b/youtube_dl/extractor/srgssr.py
index 4962fe885d3a6620c85b5d1efade05b12c24e879..47aa887cc25f5254d8bbb0d6465a9ad499b4afaa 100644 (file)
--- a/youtube_dl/extractor/srgssr.py
+++ b/youtube_dl/extractor/srgssr.py
@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_urllib_parse_urlparse
 from ..utils import (
     ExtractorError,
     parse_iso8601,
@@ -23,13 +24,24 @@ class SRGSSRIE(InfoExtractor):
         'STARTDATE': 'This video is not yet available. Please try again later.',
     }
 
+    def _get_tokenized_src(self, url, video_id, format_id):
+        sp = compat_urllib_parse_urlparse(url).path.split('/')
+        token = self._download_json(
+            'http://tp.srgssr.ch/akahd/token?acl=/%s/%s/*' % (sp[1], sp[2]),
+            video_id, 'Downloading %s token' % format_id, fatal=False) or {}
+        auth_params = token.get('token', {}).get('authparams')
+        if auth_params:
+            url += '?' + auth_params
+        return url
+
     def get_media_data(self, bu, media_type, media_id):
         media_data = self._download_json(
             'http://il.srgssr.ch/integrationlayer/1.0/ue/%s/%s/play/%s.json' % (bu, media_type, media_id),
             media_id)[media_type.capitalize()]
 
         if media_data.get('block') and media_data['block'] in self._ERRORS:
-            raise ExtractorError('%s said: %s' % (self.IE_NAME, self._ERRORS[media_data['block']]), expected=True)
+            raise ExtractorError('%s said: %s' % (
+                self.IE_NAME, self._ERRORS[media_data['block']]), expected=True)
 
         return media_data
 
@@ -56,26 +68,26 @@ class SRGSSRIE(InfoExtractor):
         formats = []
         for source in media_data.get('Playlists', {}).get('Playlist', []) + media_data.get('Downloads', {}).get('Download', []):
             protocol = source.get('@protocol')
-            if protocol in ('HTTP-HDS', 'HTTP-HLS'):
-                assets = {}
-                for quality in source['url']:
-                    assets[quality['@quality']] = quality['text']
-                asset_url = assets.get('HD') or assets.get('HQ') or assets.get('SD') or assets.get('MQ') or assets.get('LQ')
-                if '.f4m' in asset_url:
-                    formats.extend(self._extract_f4m_formats(asset_url + '?hdcore=3.4.0', media_id, f4m_id='hds', fatal=False))
-                elif '.m3u8' in asset_url:
-                    formats.extend(self._extract_m3u8_formats(asset_url, media_id, m3u8_id='hls', fatal=False))
-            else:
-                for asset in source['url']:
-                    asset_url = asset['text']
-                    ext = None
-                    if asset_url.startswith('rtmp'):
-                        ext = self._search_regex(r'([a-z0-9]+):[^/]+', asset_url, 'ext')
+            for asset in source['url']:
+                asset_url = asset['text']
+                quality = asset['@quality']
+                format_id = '%s-%s' % (protocol, quality)
+                if protocol.startswith('HTTP-HDS') or protocol.startswith('HTTP-HLS'):
+                    asset_url = self._get_tokenized_src(asset_url, media_id, format_id)
+                    if protocol.startswith('HTTP-HDS'):
+                        formats.extend(self._extract_f4m_formats(
+                            asset_url + ('?' if '?' not in asset_url else '&') + 'hdcore=3.4.0',
+                            media_id, f4m_id=format_id, fatal=False))
+                    elif protocol.startswith('HTTP-HLS'):
+                        formats.extend(self._extract_m3u8_formats(
+                            asset_url, media_id, 'mp4', 'm3u8_native',
+                            m3u8_id=format_id, fatal=False))
+                else:
                     formats.append({
-                        'format_id': asset['@quality'],
+                        'format_id': format_id,
                         'url': asset_url,
-                        'preference': preference(asset['@quality']),
-                        'ext': ext,
+                        'preference': preference(quality),
+                        'ext': 'flv' if protocol == 'RTMP' else None,
                     })
         self._sort_formats(formats)
 
@@ -90,14 +102,15 @@ class SRGSSRIE(InfoExtractor):
 
 
 class SRGSSRPlayIE(InfoExtractor):
+    IE_DESC = 'srf.ch, rts.ch, rsi.ch, rtr.ch and swissinfo.ch play sites'
     _VALID_URL = r'https?://(?:(?:www|play)\.)?(?P<bu>srf|rts|rsi|rtr|swissinfo)\.ch/play/(?:tv|radio)/[^/]+/(?P<type>video|audio)/[^?]+\?id=(?P<id>[0-9a-f\-]{36}|\d+)'
 
     _TESTS = [{
         'url': 'http://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5',
-        'md5': '4cd93523723beff51bb4bee974ee238d',
+        'md5': 'da6b5b3ac9fa4761a942331cef20fcb3',
         'info_dict': {
             'id': '28e1a57d-5b76-4399-8ab3-9097f071e6c5',
-            'ext': 'm4v',
+            'ext': 'mp4',
             'upload_date': '20130701',
             'title': 'Snowden beantragt Asyl in Russland',
             'timestamp': 1372713995,
@@ -140,7 +153,7 @@ class SRGSSRPlayIE(InfoExtractor):
             'uploader': '19h30',
             'upload_date': '20141201',
             'timestamp': 1417458600,
-            'thumbnail': 're:^https?://.*\.image',
+            'thumbnail': r're:^https?://.*\.image',
             'view_count': int,
         },
         'params': {