Update changelog.

[youtubedl] / youtube_dl / extractor / ard.py

diff --git a/youtube_dl/extractor/ard.py b/youtube_dl/extractor/ard.py
index 6f465789b497a6625776c383ff699a64b0b5c346..2f47e21c324102d73cf277accc0b30ac41ddfab6 100644 (file)
--- a/youtube_dl/extractor/ard.py
+++ b/youtube_dl/extractor/ard.py
@@ -1,6 +1,7 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
 # coding: utf-8
 from __future__ import unicode_literals
 

+import json

 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor


@@ -8,89 +9,55 @@ from .generic import GenericIE
 from ..utils import (
     determine_ext,
     ExtractorError,
 from ..utils import (
     determine_ext,
     ExtractorError,

-    get_element_by_attribute,
-    qualities,

     int_or_none,
     parse_duration,
     int_or_none,
     parse_duration,

+    qualities,
+    str_or_none,
+    try_get,

     unified_strdate,
     unified_strdate,

+    unified_timestamp,
+    update_url_query,
+    url_or_none,

     xpath_text,
     xpath_text,

-    parse_xml,

 )
 )

+from ..compat import compat_etree_fromstring

 
 
 
 

-class ARDMediathekIE(InfoExtractor):
-    IE_NAME = 'ARD:mediathek'
-    _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
-
-    _TESTS = [{
-        'url': 'http://www.ardmediathek.de/tv/Dokumentation-und-Reportage/Ich-liebe-das-Leben-trotzdem/rbb-Fernsehen/Video?documentId=29582122&bcastId=3822114',
-        'info_dict': {
-            'id': '29582122',
-            'ext': 'mp4',
-            'title': 'Ich liebe das Leben trotzdem',
-            'description': 'md5:45e4c225c72b27993314b31a84a5261c',
-            'duration': 4557,
-        },
-        'params': {
-            # m3u8 download
-            'skip_download': True,
-        },
-    }, {
-        'url': 'http://www.ardmediathek.de/tv/Tatort/Tatort-Scheinwelten-H%C3%B6rfassung-Video/Das-Erste/Video?documentId=29522730&bcastId=602916',
-        'md5': 'f4d98b10759ac06c0072bbcd1f0b9e3e',
-        'info_dict': {
-            'id': '29522730',
-            'ext': 'mp4',
-            'title': 'Tatort: Scheinwelten - Hörfassung (Video tgl. ab 20 Uhr)',
-            'description': 'md5:196392e79876d0ac94c94e8cdb2875f1',
-            'duration': 5252,
-        },
-    }, {
-        # audio
-        'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086',
-        'md5': '219d94d8980b4f538c7fcb0865eb7f2c',
-        'info_dict': {
-            'id': '28488308',
-            'ext': 'mp3',
-            'title': 'Tod eines Fußballers',
-            'description': 'md5:f6e39f3461f0e1f54bfa48c8875c86ef',
-            'duration': 3240,
-        },
-    }, {
-        'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
-        'only_matching': True,
-    }]
+class ARDMediathekBaseIE(InfoExtractor):
+    _GEO_COUNTRIES = ['DE']

 
     def _extract_media_info(self, media_info_url, webpage, video_id):
         media_info = self._download_json(
             media_info_url, video_id, 'Downloading media JSON')
 
     def _extract_media_info(self, media_info_url, webpage, video_id):
         media_info = self._download_json(
             media_info_url, video_id, 'Downloading media JSON')

+        return self._parse_media_info(media_info, video_id, '"fsk"' in webpage)

 
 

+    def _parse_media_info(self, media_info, video_id, fsk):

         formats = self._extract_formats(media_info, video_id)
 
         if not formats:
         formats = self._extract_formats(media_info, video_id)
 
         if not formats:

-            if '"fsk"' in webpage:
+            if fsk:

                 raise ExtractorError(
                     'This video is only available after 20:00', expected=True)
             elif media_info.get('_geoblocked'):
                 raise ExtractorError(
                     'This video is only available after 20:00', expected=True)
             elif media_info.get('_geoblocked'):

-                raise ExtractorError('This video is not available due to geo restriction', expected=True)
+                self.raise_geo_restricted(
+                    'This video is not available due to geoblocking',
+                    countries=self._GEO_COUNTRIES)

 
         self._sort_formats(formats)
 
 
         self._sort_formats(formats)
 

-        duration = int_or_none(media_info.get('_duration'))
-        thumbnail = media_info.get('_previewImage')
-

         subtitles = {}
         subtitle_url = media_info.get('_subtitleUrl')
         if subtitle_url:
             subtitles['de'] = [{
         subtitles = {}
         subtitle_url = media_info.get('_subtitleUrl')
         if subtitle_url:
             subtitles['de'] = [{

-                'ext': 'srt',
+                'ext': 'ttml',

                 'url': subtitle_url,
             }]
 
         return {
             'id': video_id,
                 'url': subtitle_url,
             }]
 
         return {
             'id': video_id,

-            'duration': duration,
-            'thumbnail': thumbnail,
+            'duration': int_or_none(media_info.get('_duration')),
+            'thumbnail': media_info.get('_previewImage'),
+            'is_live': media_info.get('_isLive') is True,

             'formats': formats,
             'subtitles': subtitles,
         }
             'formats': formats,
             'subtitles': subtitles,
         }


@@ -109,14 +76,21 @@ class ARDMediathekIE(InfoExtractor):
                 quality = stream.get('_quality')
                 server = stream.get('_server')
                 for stream_url in stream_urls:
                 quality = stream.get('_quality')
                 server = stream.get('_server')
                 for stream_url in stream_urls:

+                    if not url_or_none(stream_url):
+                        continue

                     ext = determine_ext(stream_url)
                     ext = determine_ext(stream_url)

+                    if quality != 'auto' and ext in ('f4m', 'm3u8'):
+                        continue

                     if ext == 'f4m':
                         formats.extend(self._extract_f4m_formats(
                     if ext == 'f4m':
                         formats.extend(self._extract_f4m_formats(

-                            stream_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124',
-                            video_id, preference=-1, f4m_id='hds'))
+                            update_url_query(stream_url, {
+                                'hdcore': '3.1.1',
+                                'plugin': 'aasp-3.1.1.69.124'
+                            }), video_id, f4m_id='hds', fatal=False))

                     elif ext == 'm3u8':
                         formats.extend(self._extract_m3u8_formats(
                     elif ext == 'm3u8':
                         formats.extend(self._extract_m3u8_formats(

-                            stream_url, video_id, 'mp4', preference=1, m3u8_id='hls'))
+                            stream_url, video_id, 'mp4', 'm3u8_native',
+                            m3u8_id='hls', fatal=False))

                     else:
                         if server and server.startswith('rtmp'):
                             f = {
                     else:
                         if server and server.startswith('rtmp'):
                             f = {


@@ -124,14 +98,14 @@ class ARDMediathekIE(InfoExtractor):
                                 'play_path': stream_url,
                                 'format_id': 'a%s-rtmp-%s' % (num, quality),
                             }
                                 'play_path': stream_url,
                                 'format_id': 'a%s-rtmp-%s' % (num, quality),
                             }

-                        elif stream_url.startswith('http'):
+                        else:

                             f = {
                                 'url': stream_url,
                                 'format_id': 'a%s-%s-%s' % (num, ext, quality)
                             }
                             f = {
                                 'url': stream_url,
                                 'format_id': 'a%s-%s-%s' % (num, ext, quality)
                             }

-                        else:
-                            continue
-                        m = re.search(r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$', stream_url)
+                        m = re.search(
+                            r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$',
+                            stream_url)

                         if m:
                             f.update({
                                 'width': int(m.group('width')),
                         if m:
                             f.update({
                                 'width': int(m.group('width')),


@@ -142,39 +116,92 @@ class ARDMediathekIE(InfoExtractor):
                         formats.append(f)
         return formats
 
                         formats.append(f)
         return formats
 

+
+class ARDMediathekIE(ARDMediathekBaseIE):
+    IE_NAME = 'ARD:mediathek'
+    _VALID_URL = r'^https?://(?:(?:(?:www|classic)\.)?ardmediathek\.de|mediathek\.(?:daserste|rbb-online)\.de|one\.ard\.de)/(?:.*/)(?P<video_id>[0-9]+|[^0-9][^/\?]+)[^/\?]*(?:\?.*)?'
+
+    _TESTS = [{
+        # available till 26.07.2022
+        'url': 'http://www.ardmediathek.de/tv/S%C3%9CDLICHT/Was-ist-die-Kunst-der-Zukunft-liebe-Ann/BR-Fernsehen/Video?bcastId=34633636&documentId=44726822',
+        'info_dict': {
+            'id': '44726822',
+            'ext': 'mp4',
+            'title': 'Was ist die Kunst der Zukunft, liebe Anna McCarthy?',
+            'description': 'md5:4ada28b3e3b5df01647310e41f3a62f5',
+            'duration': 1740,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        }
+    }, {
+        'url': 'https://one.ard.de/tv/Mord-mit-Aussicht/Mord-mit-Aussicht-6-39-T%C3%B6dliche-Nach/ONE/Video?bcastId=46384294&documentId=55586872',
+        'only_matching': True,
+    }, {
+        # audio
+        'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086',
+        'only_matching': True,
+    }, {
+        'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
+        'only_matching': True,
+    }, {
+        # audio
+        'url': 'http://mediathek.rbb-online.de/radio/Hörspiel/Vor-dem-Fest/kulturradio/Audio?documentId=30796318&topRessort=radio&bcastId=9839158',
+        'only_matching': True,
+    }, {
+        'url': 'https://classic.ardmediathek.de/tv/Panda-Gorilla-Co/Panda-Gorilla-Co-Folge-274/Das-Erste/Video?bcastId=16355486&documentId=58234698',
+        'only_matching': True,
+    }]
+
+    @classmethod
+    def suitable(cls, url):
+        return False if ARDBetaMediathekIE.suitable(url) else super(ARDMediathekIE, cls).suitable(url)
+

     def _real_extract(self, url):
         # determine video id from url
         m = re.match(self._VALID_URL, url)
 
     def _real_extract(self, url):
         # determine video id from url
         m = re.match(self._VALID_URL, url)
 

+        document_id = None
+

         numid = re.search(r'documentId=([0-9]+)', url)
         if numid:
         numid = re.search(r'documentId=([0-9]+)', url)
         if numid:

-            video_id = numid.group(1)
+            document_id = video_id = numid.group(1)

         else:
             video_id = m.group('video_id')
 
         webpage = self._download_webpage(url, video_id)
 
         else:
             video_id = m.group('video_id')
 
         webpage = self._download_webpage(url, video_id)
 

-        if '>Der gewünschte Beitrag ist nicht mehr verfügbar.<' in webpage:
-            raise ExtractorError('Video %s is no longer available' % video_id, expected=True)
+        ERRORS = (
+            ('>Leider liegt eine Störung vor.', 'Video %s is unavailable'),
+            ('>Der gewünschte Beitrag ist nicht mehr verfügbar.<',
+             'Video %s is no longer available'),
+        )

 
 

-        if 'Diese Sendung ist für Jugendliche unter 12 Jahren nicht geeignet. Der Clip ist deshalb nur von 20 bis 6 Uhr verfügbar.' in webpage:
-            raise ExtractorError('This program is only suitable for those aged 12 and older. Video %s is therefore only available between 20 pm and 6 am.' % video_id, expected=True)
+        for pattern, message in ERRORS:
+            if pattern in webpage:
+                raise ExtractorError(message % video_id, expected=True)

 
         if re.search(r'[\?&]rss($|[=&])', url):
 
         if re.search(r'[\?&]rss($|[=&])', url):

-            doc = parse_xml(webpage)
+            doc = compat_etree_fromstring(webpage.encode('utf-8'))

             if doc.tag == 'rss':
                 return GenericIE()._extract_rss(url, video_id, doc)
 
         title = self._html_search_regex(
             [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',
             if doc.tag == 'rss':
                 return GenericIE()._extract_rss(url, video_id, doc)
 
         title = self._html_search_regex(
             [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>',

-             r'<meta name="dcterms.title" content="(.*?)"/>',
-             r'<h4 class="headline">(.*?)</h4>'],
+             r'<meta name="dcterms\.title" content="(.*?)"/>',
+             r'<h4 class="headline">(.*?)</h4>',
+             r'<title[^>]*>(.*?)</title>'],

             webpage, 'title')
         description = self._html_search_meta(
             'dcterms.abstract', webpage, 'description', default=None)
         if description is None:
             description = self._html_search_meta(
             webpage, 'title')
         description = self._html_search_meta(
             'dcterms.abstract', webpage, 'description', default=None)
         if description is None:
             description = self._html_search_meta(

-                'description', webpage, 'meta description')
+                'description', webpage, 'meta description', default=None)
+        if description is None:
+            description = self._html_search_regex(
+                r'<p\s+class="teasertext">(.+?)</p>',
+                webpage, 'teaser text', default=None)

 
         # Thumbnail is sometimes not present.
         # It is in the mobile version, but that seems to use a different URL
 
         # Thumbnail is sometimes not present.
         # It is in the mobile version, but that seems to use a different URL


@@ -204,12 +231,16 @@ class ARDMediathekIE(InfoExtractor):
                 'formats': formats,
             }
         else:  # request JSON file
                 'formats': formats,
             }
         else:  # request JSON file

+            if not document_id:
+                video_id = self._search_regex(
+                    r'/play/(?:config|media)/(\d+)', webpage, 'media id')

             info = self._extract_media_info(
             info = self._extract_media_info(

-                'http://www.ardmediathek.de/play/media/%s' % video_id, webpage, video_id)
+                'http://www.ardmediathek.de/play/media/%s' % video_id,
+                webpage, video_id)

 
         info.update({
             'id': video_id,
 
         info.update({
             'id': video_id,

-            'title': title,
+            'title': self._live_title(title) if info.get('is_live') else title,

             'description': description,
             'thumbnail': thumbnail,
         })
             'description': description,
             'thumbnail': thumbnail,
         })


@@ -218,20 +249,24 @@ class ARDMediathekIE(InfoExtractor):
 
 
 class ARDIE(InfoExtractor):
 
 
 class ARDIE(InfoExtractor):

-    _VALID_URL = '(?P<mainurl>https?://(www\.)?daserste\.de/[^?#]+/videos/(?P<display_id>[^/?#]+)-(?P<id>[0-9]+))\.html'
-    _TEST = {
-        'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html',
-        'md5': 'd216c3a86493f9322545e045ddc3eb35',
+    _VALID_URL = r'(?P<mainurl>https?://(www\.)?daserste\.de/[^?#]+/videos/(?P<display_id>[^/?#]+)-(?P<id>[0-9]+))\.html'
+    _TESTS = [{
+        # available till 14.02.2019
+        'url': 'http://www.daserste.de/information/talk/maischberger/videos/das-groko-drama-zerlegen-sich-die-volksparteien-video-102.html',
+        'md5': '8e4ec85f31be7c7fc08a26cdbc5a1f49',

         'info_dict': {
         'info_dict': {

-            'display_id': 'die-story-im-ersten-mission-unter-falscher-flagge',
-            'id': '100',
+            'display_id': 'das-groko-drama-zerlegen-sich-die-volksparteien-video',
+            'id': '102',

             'ext': 'mp4',
             'ext': 'mp4',

-            'duration': 2600,
-            'title': 'Die Story im Ersten: Mission unter falscher Flagge',
-            'upload_date': '20140804',
-            'thumbnail': 're:^https?://.*\.jpg$',
-        }
-    }
+            'duration': 4435.0,
+            'title': 'Das GroKo-Drama: Zerlegen sich die Volksparteien?',
+            'upload_date': '20180214',
+            'thumbnail': r're:^https?://.*\.jpg$',
+        },
+    }, {
+        'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html',
+        'only_matching': True,
+    }]

 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)


@@ -274,39 +309,98 @@ class ARDIE(InfoExtractor):
         }
 
 
         }
 
 

-class SportschauIE(ARDMediathekIE):
-    IE_NAME = 'Sportschau'
-    _VALID_URL = r'(?P<baseurl>https?://(?:www\.)?sportschau\.de/(?:[^/]+/)+video(?P<id>[^/#?]+))\.html'
+class ARDBetaMediathekIE(ARDMediathekBaseIE):
+    _VALID_URL = r'https://(?:beta|www)\.ardmediathek\.de/(?P<client>[^/]+)/(?:player|live)/(?P<video_id>[a-zA-Z0-9]+)(?:/(?P<display_id>[^/?#]+))?'

     _TESTS = [{
     _TESTS = [{

-        'url': 'http://www.sportschau.de/tourdefrance/videoseppeltkokainhatnichtsmitklassischemdopingzutun100.html',
+        'url': 'https://beta.ardmediathek.de/ard/player/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE/die-robuste-roswita',
+        'md5': 'dfdc87d2e7e09d073d5a80770a9ce88f',

         'info_dict': {
         'info_dict': {

-            'id': 'seppeltkokainhatnichtsmitklassischemdopingzutun100',
+            'display_id': 'die-robuste-roswita',
+            'id': '70153354',
+            'title': 'Die robuste Roswita',
+            'description': r're:^Der Mord.*trüber ist als die Ilm.',
+            'duration': 5316,
+            'thumbnail': 'https://img.ardmediathek.de/standard/00/70/15/33/90/-1852531467/16x9/960?mandant=ard',
+            'timestamp': 1577047500,
+            'upload_date': '20191222',

             'ext': 'mp4',
             'ext': 'mp4',

-            'title': 'Seppelt: "Kokain hat nichts mit klassischem Doping zu tun"',
-            'thumbnail': 're:^https?://.*\.jpg$',
-            'description': 'Der ARD-Doping Experte Hajo Seppelt gibt seine Einschätzung zum ersten Dopingfall der diesjährigen Tour de France um den Italiener Luca Paolini ab.',
-        },
-        'params': {
-            # m3u8 download
-            'skip_download': True,

         },
         },

+    }, {
+        'url': 'https://www.ardmediathek.de/ard/player/Y3JpZDovL3N3ci5kZS9hZXgvbzEwNzE5MTU/',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.ardmediathek.de/swr/live/Y3JpZDovL3N3ci5kZS8xMzQ4MTA0Mg',
+        'only_matching': True,

     }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
     }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

-        video_id = mobj.group('id')
-        base_url = mobj.group('baseurl')
-
-        webpage = self._download_webpage(url, video_id)
-        title = get_element_by_attribute('class', 'headline', webpage)
-        description = self._html_search_meta('description', webpage, 'description')
-
-        info = self._extract_media_info(
-            base_url + '-mc_defaultQuality-h.json', webpage, video_id)
-
+        video_id = mobj.group('video_id')
+        display_id = mobj.group('display_id') or video_id
+
+        player_page = self._download_json(
+            'https://api.ardmediathek.de/public-gateway',
+            display_id, data=json.dumps({
+                'query': '''{
+  playerPage(client:"%s", clipId: "%s") {
+    blockedByFsk
+    broadcastedOn
+    maturityContentRating
+    mediaCollection {
+      _duration
+      _geoblocked
+      _isLive
+      _mediaArray {
+        _mediaStreamArray {
+          _quality
+          _server
+          _stream
+        }
+      }
+      _previewImage
+      _subtitleUrl
+      _type
+    }
+    show {
+      title
+    }
+    synopsis
+    title
+    tracking {
+      atiCustomVars {
+        contentId
+      }
+    }
+  }
+}''' % (mobj.group('client'), video_id),
+            }).encode(), headers={
+                'Content-Type': 'application/json'
+            })['data']['playerPage']
+        title = player_page['title']
+        content_id = str_or_none(try_get(
+            player_page, lambda x: x['tracking']['atiCustomVars']['contentId']))
+        media_collection = player_page.get('mediaCollection') or {}
+        if not media_collection and content_id:
+            media_collection = self._download_json(
+                'https://www.ardmediathek.de/play/media/' + content_id,
+                content_id, fatal=False) or {}
+        info = self._parse_media_info(
+            media_collection, content_id or video_id,
+            player_page.get('blockedByFsk'))
+        age_limit = None
+        description = player_page.get('synopsis')
+        maturity_content_rating = player_page.get('maturityContentRating')
+        if maturity_content_rating:
+            age_limit = int_or_none(maturity_content_rating.lstrip('FSK'))
+        if not age_limit and description:
+            age_limit = int_or_none(self._search_regex(
+                r'\(FSK\s*(\d+)\)\s*$', description, 'age limit', default=None))

         info.update({
         info.update({

+            'age_limit': age_limit,
+            'display_id': display_id,

             'title': title,
             'description': description,
             'title': title,
             'description': description,

+            'timestamp': unified_timestamp(player_page.get('broadcastedOn')),
+            'series': try_get(player_page, lambda x: x['show']['title']),

         })
         })

-

         return info
         return info