Merge branch 'srgssr' of https://github.com/remitamine/youtube-dl into remitamine-srgssr

youtube_dl/extractor/__init__.py

@@ -643,7 +643,10 @@
     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

youtube_dl/extractor/rts.py

@@ -3,7 +3,7 @@
 
 import re
 
-from .common import InfoExtractor
+from .srgssr import SRGSSRIE
 from ..compat import (
     compat_str,
     compat_urllib_parse_urlparse,
@@ -17,27 +17,18 @@
 )
 
 
-class RTSIE(InfoExtractor):
+class RTSIE(SRGSSRIE):
     IE_DESC = 'RTS.ch'
-    _VALID_URL = r'''(?x)
+    _VALID_URL = r'rts:(?P<rts_id>\d+)|https?://(?:www\.)?rts\.ch/(?:[^/]+/){2,}(?P<id>[0-9]+)-(?P<display_id>.+?)\.html'
-                    (?:
-                        rts:(?P<rts_id>\d+)|
-                        https?://
-                            (?:www\.)?rts\.ch/
-                            (?:
-                                (?:[^/]+/){2,}(?P<id>[0-9]+)-(?P<display_id>.+?)\.html|
-                                play/tv/[^/]+/video/(?P<display_id_new>.+?)\?id=(?P<id_new>[0-9]+)
-                            )
-                    )'''
 
     _TESTS = [
         {
             'url': 'http://www.rts.ch/archives/tv/divers/3449373-les-enfants-terribles.html',
-            'md5': '753b877968ad8afaeddccc374d4256a5',
+            'md5': 'f254c4b26fb1d3c183793d52bc40d3e7',
             'info_dict': {
                 'id': '3449373',
                 'display_id': 'les-enfants-terribles',
-                'ext': 'mp4',
+                'ext': 'flv',
                 'duration': 1488,
                 'title': 'Les Enfants Terribles',
                 'description': 'France Pommier et sa soeur Luce Feral, les deux filles de ce groupe de 5.',
@@ -50,11 +41,11 @@ class RTSIE(InfoExtractor):
         },
         {
             'url': 'http://www.rts.ch/emissions/passe-moi-les-jumelles/5624067-entre-ciel-et-mer.html',
-            'md5': 'c148457a27bdc9e5b1ffe081a7a8337b',
+            'md5': 'f1077ac5af686c76528dc8d7c5df29ba',
             'info_dict': {
-                'id': '5624067',
+                'id': '5742494',
-                'display_id': 'entre-ciel-et-mer',
+                'display_id': '5742494',
-                'ext': 'mp4',
+                'ext': 'flv',
                 'duration': 3720,
                 'title': 'Les yeux dans les cieux - Mon homard au Canada',
                 'description': 'md5:d22ee46f5cc5bac0912e5a0c6d44a9f7',
@@ -85,11 +76,11 @@ class RTSIE(InfoExtractor):
         },
         {
             'url': 'http://www.rts.ch/video/info/journal-continu/5745356-londres-cachee-par-un-epais-smog.html',
-            'md5': '9bb06503773c07ce83d3cbd793cebb91',
+            'md5': '9f713382f15322181bb366cc8c3a4ff0',
             'info_dict': {
                 'id': '5745356',
                 'display_id': 'londres-cachee-par-un-epais-smog',
-                'ext': 'mp4',
+                'ext': 'flv',
                 'duration': 33,
                 'title': 'Londres cachée par un épais smog',
                 'description': 'Un important voile de smog recouvre Londres depuis mercredi, provoqué par la pollution et du sable du Sahara.',
@@ -114,23 +105,6 @@ class RTSIE(InfoExtractor):
                 'timestamp': 1396551600,
             },
         },
-        {
-            'url': 'http://www.rts.ch/play/tv/-/video/le-19h30?id=6348260',
-            'md5': '968777c8779e5aa2434be96c54e19743',
-            'info_dict': {
-                'id': '6348260',
-                'display_id': 'le-19h30',
-                'ext': 'mp4',
-                'duration': 1796,
-                'title': 'Le 19h30',
-                'description': '',
-                'uploader': 'Le 19h30',
-                'upload_date': '20141201',
-                'timestamp': 1417458600,
-                'thumbnail': 're:^https?://.*\.image',
-                'view_count': int,
-            },
-        },
         {
             # article with videos on rhs
             'url': 'http://www.rts.ch/sport/hockey/6693917-hockey-davos-decroche-son-31e-titre-de-champion-de-suisse.html',
@@ -139,42 +113,47 @@ class RTSIE(InfoExtractor):
                 'title': 'Hockey: Davos décroche son 31e titre de champion de Suisse',
             },
             'playlist_mincount': 5,
-        },
-        {
-            'url': 'http://www.rts.ch/play/tv/le-19h30/video/le-chantier-du-nouveau-parlement-vaudois-a-permis-une-trouvaille-historique?id=6348280',
-            'only_matching': True,
         }
     ]
 
     def _real_extract(self, url):
         m = re.match(self._VALID_URL, url)
-        video_id = m.group('rts_id') or m.group('id') or m.group('id_new')
+        media_id = m.group('rts_id') or m.group('id')
-        display_id = m.group('display_id') or m.group('display_id_new')
+        display_id = m.group('display_id') or media_id
 
         def download_json(internal_id):
             return self._download_json(
                 'http://www.rts.ch/a/%s.html?f=json/article' % internal_id,
                 display_id)
 
-        all_info = download_json(video_id)
+        all_info = download_json(media_id)
 
-        # video_id extracted out of URL is not always a real id
+        # media_id extracted out of URL is not always a real id
         if 'video' not in all_info and 'audio' not in all_info:
             page = self._download_webpage(url, display_id)
 
             # article with videos on rhs
             videos = re.findall(
-                r'<article[^>]+class="content-item"[^>]*>\s*<a[^>]+data-video-urn="urn:rts:video:(\d+)"',
+                r'<article[^>]+class="content-item"[^>]*>\s*<a[^>]+data-video-urn="urn:([^"]+)"',
+                page)
+            if not videos:
+                videos = re.findall(
+                    r'(?s)<iframe[^>]+class="srg-player"[^>]+src="[^"]+urn:([^"]+)"',
                     page)
             if videos:
-                entries = [self.url_result('rts:%s' % video_urn, 'RTS') for video_urn in videos]
+                entries = [self.url_result('srgssr:%s' % video_urn, 'SRGSSR') for video_urn in videos]
-                return self.playlist_result(entries, video_id, self._og_search_title(page))
+                return self.playlist_result(entries, media_id, self._og_search_title(page))
 
             internal_id = self._html_search_regex(
                 r'<(?:video|audio) data-id="([0-9]+)"', page,
                 'internal video id')
             all_info = download_json(internal_id)
 
+        media_type = 'video' if 'video' in all_info else 'audio'
+
+        # check for errors
+        self.get_media_data('rts', media_type, media_id)
+
         info = all_info['video']['JSONinfo'] if 'video' in all_info else all_info['audio']
 
         upload_timestamp = parse_iso8601(info.get('broadcast_date'))
@@ -190,19 +169,27 @@ def extract_bitrate(url):
 
         formats = []
         for format_id, format_url in info['streams'].items():
+            if format_id == 'hds_sd' and 'hds' in info['streams']:
+                continue
+            if format_id == 'hls_sd' and 'hls' in info['streams']:
+                continue
             if format_url.endswith('.f4m'):
                 token = self._download_xml(
                     'http://tp.srgssr.ch/token/akahd.xml?stream=%s/*' % compat_urllib_parse_urlparse(format_url).path,
-                    video_id, 'Downloading %s token' % format_id)
+                    media_id, 'Downloading %s token' % format_id)
                 auth_params = xpath_text(token, './/authparams', 'auth params')
                 if not auth_params:
                     continue
-                formats.extend(self._extract_f4m_formats(
+                f4m_formats = self._extract_f4m_formats(
                     '%s?%s&hdcore=3.4.0&plugin=aasp-3.4.0.132.66' % (format_url, auth_params),
-                    video_id, f4m_id=format_id))
+                    media_id, f4m_id=format_id, fatal=False)
+                if f4m_formats:
+                    formats.extend(f4m_formats)
             elif format_url.endswith('.m3u8'):
-                formats.extend(self._extract_m3u8_formats(
+                m3u8_formats = self._extract_m3u8_formats(
-                    format_url, video_id, 'mp4', m3u8_id=format_id))
+                    format_url, media_id, 'mp4', m3u8_id=format_id, fatal=False)
+                if m3u8_formats:
+                    formats.extend(m3u8_formats)
             else:
                 formats.append({
                     'format_id': format_id,
@@ -217,11 +204,11 @@ def extract_bitrate(url):
                 'tbr': media['rate'] or extract_bitrate(media['url']),
             } for media in info['media'] if media.get('rate')])
 
-        self._check_formats(formats, video_id)
+        self._check_formats(formats, media_id)
         self._sort_formats(formats)
 
         return {
-            'id': video_id,
+            'id': media_id,
             'display_id': display_id,
             'formats': formats,
             'title': info['title'],

youtube_dl/extractor/srf.py

@@ -1,120 +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|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',
-        '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/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': {
-            '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):
-        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/%s/play/%s.xml' % (media_type, 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).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,
-            'formats': formats,
-            'title': title,
-            'thumbnails': thumbnails,
-            'timestamp': timestamp,
-            'subtitles': subtitles,
-        }

youtube_dl/extractor/srgssr.py

@@ -0,0 +1,167 @@
+# 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|srgssr):(?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 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)
+
+        return media_data
+
+    def _real_extract(self, url):
+        bu, media_type, media_id = re.match(self._VALID_URL, url).groups()
+
+        if bu == 'rts':
+            return self.url_result('rts:%s' % media_id, 'RTS')
+
+        media_data = self.get_media_data(bu, media_type, media_id)
+
+        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 = []
+        if 'Image' in media_data:
+            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:
+                    f4m_formats = formats.extend(self._extract_f4m_formats(asset_url + '?hdcore=3.4.0', media_id, f4m_id='hds', fatal=False))
+                    if f4m_formats:
+                        formats.extend(f4m_formats)
+                elif '.m3u8' in asset_url:
+                    m3u8_formats = formats.extend(self._extract_m3u8_formats(asset_url, media_id, m3u8_id='hls', fatal=False))
+                    if m3u8_formats:
+                        formats.extend(m3u8_formats)
+            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({
+                        'format_id': asset['@quality'],
+                        'url': asset_url,
+                        'preference': preference(asset['@quality']),
+                        'ext': ext,
+                    })
+
+        if 'Downloads' in media_data:
+            for source in media_data['Downloads']['Download']:
+                for asset in source['url']:
+                    formats.append({
+                        'format_id': asset['@quality'],
+                        '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|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',
+        '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.rtr.ch/play/radio/actualitad/audio/saira-tujetsch-tuttina-cuntinuar-cun-sedrun-muster-turissem?id=63cb0778-27f8-49af-9284-8c7a8c6d15fc',
+        'info_dict': {
+            'id': '63cb0778-27f8-49af-9284-8c7a8c6d15fc',
+            'ext': 'mp3',
+            'upload_date': '20151013',
+            'title': 'Saira: Tujetsch - tuttina cuntinuar cun Sedrun Mustér Turissem',
+            'timestamp': 1444750398,
+        },
+        'params': {
+            # rtmp download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://www.rts.ch/play/tv/-/video/le-19h30?id=6348260',
+        'md5': '67a2a9ae4e8e62a68d0e9820cc9782df',
+        'info_dict': {
+            'id': '6348260',
+            'display_id': '6348260',
+            'ext': 'flv',
+            'duration': 1796,
+            'title': 'Le 19h30',
+            'description': '',
+            'uploader': '19h30',
+            'upload_date': '20141201',
+            'timestamp': 1417458600,
+            'thumbnail': 're:^https?://.*\.image',
+            'view_count': int,
+        },
+    }]
+
+    def _real_extract(self, url):
+        bu, media_type, media_id = re.match(self._VALID_URL, url).groups()
+        # other info can be extracted from url + '&layout=json'
+        return self.url_result('srgssr:%s:%s:%s' % (bu[:3], media_type, media_id), 'SRGSSR')