[srgssr] Add generic extractor for SRGSSR Group sites
authorremitamine <remitamine@gmail.com>
Fri, 9 Oct 2015 19:08:37 +0000 (20:08 +0100)
committerremitamine <remitamine@gmail.com>
Fri, 9 Oct 2015 19:08:37 +0000 (20:08 +0100)
youtube_dl/extractor/__init__.py
youtube_dl/extractor/srf.py [deleted file]
youtube_dl/extractor/srgssr.py [new file with mode: 0644]

index 3ace1cc2ca7a375820d423f495d57c4e76aae018..042ad3678907a3ca1fc4a7ec5a2955bef771e169 100644 (file)
@@ -591,7 +591,10 @@ from .sportbox import (
     SportBoxEmbedIE,
 )
 from .sportdeutschland import SportDeutschlandIE
-from .srf import SrfIE
+from .srgssr import (
+    SRGSSRIE,
+    SRGSSRPlayIE,
+)
 from .srmediathek import SRMediathekIE
 from .ssa import SSAIE
 from .stanfordoc import StanfordOpenClassroomIE
diff --git a/youtube_dl/extractor/srf.py b/youtube_dl/extractor/srf.py
deleted file mode 100644 (file)
index 77eec0b..0000000
+++ /dev/null
@@ -1,104 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import re
-from .common import InfoExtractor
-from ..utils import (
-    determine_ext,
-    parse_iso8601,
-    xpath_text,
-)
-
-
-class SrfIE(InfoExtractor):
-    _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})'
-    _TESTS = [{
-        'url': 'http://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5',
-        'md5': '4cd93523723beff51bb4bee974ee238d',
-        'info_dict': {
-            'id': '28e1a57d-5b76-4399-8ab3-9097f071e6c5',
-            'display_id': 'snowden-beantragt-asyl-in-russland',
-            'ext': 'm4v',
-            'upload_date': '20130701',
-            'title': 'Snowden beantragt Asyl in Russland',
-            'timestamp': 1372713995,
-        }
-    }, {
-        # 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': 'flv',
-            'upload_date': '20130710',
-            'title': 'Jaguar XK120, Shadow und Tornado-Dampflokomotive',
-            'timestamp': 1373493600,
-        },
-    }, {
-        '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)
-        display_id = re.match(self._VALID_URL, url).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,
-            display_id)
-
-        title = xpath_text(
-            video_data, './AssetMetadatas/AssetMetadata/title', fatal=True)
-        thumbnails = [{
-            'url': s.text
-        } for s in video_data.findall('.//ImageRepresentation/url')]
-        timestamp = parse_iso8601(xpath_text(video_data, './createdDate'))
-        # The <duration> field in XML is different from the exact duration, skipping
-
-        formats = []
-        for item in video_data.findall('./Playlists/Playlist') + video_data.findall('./Downloads/Download'):
-            for url_node in item.findall('url'):
-                quality = url_node.attrib['quality']
-                full_url = url_node.text
-                original_ext = determine_ext(full_url)
-                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,
-            'formats': formats,
-            'title': title,
-            'thumbnails': thumbnails,
-            'timestamp': timestamp,
-            'subtitles': subtitles,
-        }
diff --git a/youtube_dl/extractor/srgssr.py b/youtube_dl/extractor/srgssr.py
new file mode 100644 (file)
index 0000000..addf4d2
--- /dev/null
@@ -0,0 +1,130 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    parse_iso8601,
+    qualities,
+)
+
+
+class SRGSSRIE(InfoExtractor):
+    _VALID_URL = r'(?:https?://tp\.srgssr\.ch/p(?:/[^/]+)+\?urn=)?urn:(?P<bu>srf|rts|rsi|rtr|swi):(?:[^:]+:)?(?P<type>video|audio):(?P<id>[0-9a-f\-]{36}|\d+)'
+
+    _ERRORS = {
+        'AGERATING12': 'To protect children under the age of 12, this video is only available between 8 p.m. and 6 a.m.',
+        'AGERATING18': 'To protect children under the age of 18, this video is only available between 11 p.m. and 5 a.m.',
+#        'ENDDATE': 'For legal reasons, this video was only available for a specified period of time.',
+        'GEOBLOCK': 'For legal reasons, this video is only available in Switzerland.',
+        'LEGAL': 'The video cannot be transmitted for legal reasons.',
+        'STARTDATE': 'This video is not yet available. Please try again later.',
+    }
+
+    def _real_extract(self, url):
+        bu, media_type, media_id = re.match(self._VALID_URL, url).groups()
+
+        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)
+
+        metadata = media_data['AssetMetadatas']['AssetMetadata'][0]
+        title = metadata['title']
+        description = metadata.get('description')
+        created_date = media_data.get('createdDate') or metadata.get('createdDate')
+        timestamp = parse_iso8601(created_date)
+
+        thumbnails = []
+        for image in media_data['Image']['ImageRepresentations']['ImageRepresentation']:
+            thumbnails.append({
+                'id': image.get('id'),
+                'url': image['url'],
+            })
+
+        preference = qualities(['LQ', 'MQ', 'SD', 'HQ', 'HD'])
+        formats = []
+        for source in media_data['Playlists']['Playlist']:
+            assets = {}
+            protocol = source.get('@protocol')
+            if protocol in ('HTTP-HDS', 'HTTP-HLS'):
+                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'))
+                elif '.m3u8' in asset_url:
+                    formats.extend(self._extract_m3u8_formats(asset_url, media_id, m3u8_id='hls'))
+            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')
+                    formats.append({
+                        'url': asset_url,
+                        'preference': preference(asset['@quality']),
+                        'ext': ext,
+                    })
+
+        downloads = media_data.get('Downloads')
+        if downloads:
+            for source in downloads['Download']:
+                for asset in source['url']:
+                    formats.append({
+                        'url': asset['text'],
+                        'preference': preference(asset['@quality'])
+                    })
+        self._sort_formats(formats)
+
+        return {
+            'id': media_id,
+            'title': title,
+            'description': description,
+            'timestamp': timestamp,
+            'thumbnails': thumbnails,
+            'formats': formats,
+        }
+
+
+class SRGSSRPlayIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:(?:www|play)\.)?(?P<bu>srf|rts|rsi|rtr|swi)\.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',
+        'info_dict': {
+            'id': '28e1a57d-5b76-4399-8ab3-9097f071e6c5',
+            'ext': 'm4v',
+            'upload_date': '20130701',
+            'title': 'Snowden beantragt Asyl in Russland',
+            'timestamp': 1372713995,
+        }
+    }, {
+        # 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': '0a274ce38fda48c53c01890651985bc6',
+        'info_dict': {
+            'id': '677f5829-e473-4823-ac83-a1087fe97faa',
+            'ext': 'flv',
+            'upload_date': '20130710',
+            'title': 'Jaguar XK120, Shadow und Tornado-Dampflokomotive',
+            'description': 'md5:88604432b60d5a38787f152dec89cd56',
+            'timestamp': 1373493600,
+        },
+    }, {
+        'url': 'http://www.srf.ch/player/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        bu, media_type, media_id = re.match(self._VALID_URL, url).groups()
+        return self.url_result('urn:%s:%s:%s' % (bu, media_type, media_id), 'SRGSSR')