Merge pull request #7045 from remitamine/ign

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

diff --git a/youtube_dl/extractor/srf.py b/youtube_dl/extractor/srf.py
index 87b4a676ac72522f5b68eac6169b90105ba8c8a1..16e1bf2d6d27589bdc05a1d5e8f0f903e8aff4c5 100644 (file)
--- a/youtube_dl/extractor/srf.py
+++ b/youtube_dl/extractor/srf.py
@@ -11,7 +11,7 @@ from ..utils import (
 
 
 class SrfIE(InfoExtractor):
-    _VALID_URL = r'http://www\.srf\.ch/play(?:er)?/tv/[^/]+/video/(?P<display_id>[^?]+)\?id=(?P<id>[0-9a-f\-]{36})'
+    _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})'
     _TESTS = [{
         'url': 'http://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5',
         'md5': '4cd93523723beff51bb4bee974ee238d',
@@ -26,31 +26,47 @@ class SrfIE(InfoExtractor):
     }, {
         # No Speichern (Save) button
         'url': 'http://www.srf.ch/play/tv/top-gear/video/jaguar-xk120-shadow-und-tornado-dampflokomotive?id=677f5829-e473-4823-ac83-a1087fe97faa',
+        'md5': 'd97e236e80d1d24729e5d0953d276a4f',
         'info_dict': {
             'id': '677f5829-e473-4823-ac83-a1087fe97faa',
             'display_id': 'jaguar-xk120-shadow-und-tornado-dampflokomotive',
-            'ext': 'mp4',
+            'ext': 'flv',
             'upload_date': '20130710',
             'title': 'Jaguar XK120, Shadow und Tornado-Dampflokomotive',
             'timestamp': 1373493600,
         },
+    }, {
+        'url': 'http://www.srf.ch/play/radio/hoerspielarchiv-srf-musikwelle/audio/saegel-ohni-wind-von-jakob-stebler?id=415bf3d3-6429-4de7-968d-95866e37cfbc',
+        'md5': '',
+        'info_dict': {
+            'id': '415bf3d3-6429-4de7-968d-95866e37cfbc',
+            'display_id': 'saegel-ohni-wind-von-jakob-stebler',
+            'ext': 'mp3',
+            'upload_date': '20080518',
+            'title': '«Sägel ohni Wind» von Jakob Stebler',
+            'timestamp': 1211112000,
+        },
         'params': {
-            # Require ffmpeg/avconv
-            'skip_download': True,
-        }
+            'skip_download': True,  # requires rtmpdump
+        },
     }, {
         'url': 'http://www.srf.ch/player/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5',
         'only_matching': True,
+    }, {
+        'url': 'https://tp.srgssr.ch/p/flash?urn=urn:srf:ais:video:28e1a57d-5b76-4399-8ab3-9097f071e6c5',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        media_type = mobj.group('media_type')
+        display_id = mobj.group('display_id') or video_id
 
         video_data = self._download_xml(
-            'http://il.srgssr.ch/integrationlayer/1.0/ue/srf/video/play/%s.xml' % video_id,
-            video_id)
+            'http://il.srgssr.ch/integrationlayer/1.0/ue/srf/%s/play/%s.xml' % (media_type, video_id),
+            display_id)
 
-        display_id = re.match(self._VALID_URL, url).group('display_id')
         title = xpath_text(
             video_data, './AssetMetadatas/AssetMetadata/title', fatal=True)
         thumbnails = [{
@@ -61,21 +77,38 @@ class SrfIE(InfoExtractor):
 
         formats = []
         for item in video_data.findall('./Playlists/Playlist') + video_data.findall('./Downloads/Download'):
-            url_node = item.find('url')
-            quality = url_node.attrib['quality']
-            full_url = url_node.text
-            original_ext = determine_ext(full_url)
-            if original_ext == 'f4m':
-                full_url += '?hdcore=3.4.0'  # Without this, you get a 403 error
-            formats.append({
-                'url': full_url,
-                'ext': 'mp4' if original_ext == 'm3u8' else original_ext,
-                'format_id': '%s-%s' % (quality, item.attrib['protocol']),
-                'preference': 0 if 'HD' in quality else -1,
-            })
+            for url_node in item.findall('url'):
+                quality = url_node.attrib['quality']
+                full_url = url_node.text
+                original_ext = determine_ext(full_url).lower()
+                format_id = '%s-%s' % (quality, item.attrib['protocol'])
+                if original_ext == 'f4m':
+                    formats.extend(self._extract_f4m_formats(
+                        full_url + '?hdcore=3.4.0', display_id, f4m_id=format_id))
+                elif original_ext == 'm3u8':
+                    formats.extend(self._extract_m3u8_formats(
+                        full_url, display_id, 'mp4', m3u8_id=format_id))
+                else:
+                    formats.append({
+                        'url': full_url,
+                        'ext': original_ext,
+                        'format_id': format_id,
+                        'quality': 0 if 'HD' in quality else -1,
+                        'preference': 1,
+                    })
 
         self._sort_formats(formats)
 
+        subtitles = {}
+        subtitles_data = video_data.find('Subtitles')
+        if subtitles_data is not None:
+            subtitles_list = [{
+                'url': sub.text,
+                'ext': determine_ext(sub.text),
+            } for sub in subtitles_data]
+            if subtitles_list:
+                subtitles['de'] = subtitles_list
+
         return {
             'id': video_id,
             'display_id': display_id,
@@ -83,4 +116,5 @@ class SrfIE(InfoExtractor):
             'title': title,
             'thumbnails': thumbnails,
             'timestamp': timestamp,
+            'subtitles': subtitles,
         }