[ard] improve extraction(closes #23761)

- simplify extraction
- extract age limit and series
- bypass geo-restriction

youtube_dl/extractor/ard.py

@@ -1,6 +1,7 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import json
 import re
 
 from .common import InfoExtractor
@@ -22,7 +23,101 @@ from ..utils import (
 from ..compat import compat_etree_fromstring
 
 
-class ARDMediathekIE(InfoExtractor):
+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')
+        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:
+            if fsk:
+                raise ExtractorError(
+                    'This video is only available after 20:00', expected=True)
+            elif media_info.get('_geoblocked'):
+                self.raise_geo_restricted(
+                    'This video is not available due to geoblocking',
+                    countries=self._GEO_COUNTRIES)
+
+        self._sort_formats(formats)
+
+        subtitles = {}
+        subtitle_url = media_info.get('_subtitleUrl')
+        if subtitle_url:
+            subtitles['de'] = [{
+                'ext': 'ttml',
+                'url': subtitle_url,
+            }]
+
+        return {
+            'id': video_id,
+            '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,
+        }
+
+    def _extract_formats(self, media_info, video_id):
+        type_ = media_info.get('_type')
+        media_array = media_info.get('_mediaArray', [])
+        formats = []
+        for num, media in enumerate(media_array):
+            for stream in media.get('_mediaStreamArray', []):
+                stream_urls = stream.get('_stream')
+                if not stream_urls:
+                    continue
+                if not isinstance(stream_urls, list):
+                    stream_urls = [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)
+                    if quality != 'auto' and ext in ('f4m', 'm3u8'):
+                        continue
+                    if ext == 'f4m':
+                        formats.extend(self._extract_f4m_formats(
+                            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(
+                            stream_url, video_id, 'mp4', 'm3u8_native',
+                            m3u8_id='hls', fatal=False))
+                    else:
+                        if server and server.startswith('rtmp'):
+                            f = {
+                                'url': server,
+                                'play_path': stream_url,
+                                'format_id': 'a%s-rtmp-%s' % (num, quality),
+                            }
+                        else:
+                            f = {
+                                'url': stream_url,
+                                'format_id': 'a%s-%s-%s' % (num, ext, quality)
+                            }
+                        m = re.search(
+                            r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$',
+                            stream_url)
+                        if m:
+                            f.update({
+                                'width': int(m.group('width')),
+                                'height': int(m.group('height')),
+                            })
+                        if type_ == 'audio':
+                            f['vcodec'] = 'none'
+                        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][^/\?]+)[^/\?]*(?:\?.*)?'
 
@@ -63,94 +158,6 @@ class ARDMediathekIE(InfoExtractor):
     def suitable(cls, url):
         return False if ARDBetaMediathekIE.suitable(url) else super(ARDMediathekIE, cls).suitable(url)
 
-    def _extract_media_info(self, media_info_url, webpage, video_id):
-        media_info = self._download_json(
-            media_info_url, video_id, 'Downloading media JSON')
-
-        formats = self._extract_formats(media_info, video_id)
-
-        if not formats:
-            if '"fsk"' in webpage:
-                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._sort_formats(formats)
-
-        duration = int_or_none(media_info.get('_duration'))
-        thumbnail = media_info.get('_previewImage')
-        is_live = media_info.get('_isLive') is True
-
-        subtitles = {}
-        subtitle_url = media_info.get('_subtitleUrl')
-        if subtitle_url:
-            subtitles['de'] = [{
-                'ext': 'ttml',
-                'url': subtitle_url,
-            }]
-
-        return {
-            'id': video_id,
-            'duration': duration,
-            'thumbnail': thumbnail,
-            'is_live': is_live,
-            'formats': formats,
-            'subtitles': subtitles,
-        }
-
-    def _extract_formats(self, media_info, video_id):
-        type_ = media_info.get('_type')
-        media_array = media_info.get('_mediaArray', [])
-        formats = []
-        for num, media in enumerate(media_array):
-            for stream in media.get('_mediaStreamArray', []):
-                stream_urls = stream.get('_stream')
-                if not stream_urls:
-                    continue
-                if not isinstance(stream_urls, list):
-                    stream_urls = [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)
-                    if quality != 'auto' and ext in ('f4m', 'm3u8'):
-                        continue
-                    if ext == 'f4m':
-                        formats.extend(self._extract_f4m_formats(
-                            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(
-                            stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
-                    else:
-                        if server and server.startswith('rtmp'):
-                            f = {
-                                'url': server,
-                                'play_path': stream_url,
-                                'format_id': 'a%s-rtmp-%s' % (num, quality),
-                            }
-                        else:
-                            f = {
-                                'url': stream_url,
-                                'format_id': 'a%s-%s-%s' % (num, ext, quality)
-                            }
-                        m = re.search(r'_(?P<width>\d+)x(?P<height>\d+)\.mp4$', stream_url)
-                        if m:
-                            f.update({
-                                'width': int(m.group('width')),
-                                'height': int(m.group('height')),
-                            })
-                        if type_ == 'audio':
-                            f['vcodec'] = 'none'
-                        formats.append(f)
-        return formats
-
     def _real_extract(self, url):
         # determine video id from url
         m = re.match(self._VALID_URL, url)
@@ -302,19 +309,20 @@ class ARDIE(InfoExtractor):
         }
 
 
-class ARDBetaMediathekIE(InfoExtractor):
+class ARDBetaMediathekIE(ARDMediathekBaseIE):
-    _VALID_URL = r'https://(?:beta|www)\.ardmediathek\.de/[^/]+/(?:player|live)/(?P<video_id>[a-zA-Z0-9]+)(?:/(?P<display_id>[^/?#]+))?'
+    _VALID_URL = r'https://(?:beta|www)\.ardmediathek\.de/(?P<client>[^/]+)/(?:player|live)/(?P<video_id>[a-zA-Z0-9]+)(?:/(?P<display_id>[^/?#]+))?'
     _TESTS = [{
         'url': 'https://beta.ardmediathek.de/ard/player/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE/die-robuste-roswita',
-        'md5': '2d02d996156ea3c397cfc5036b5d7f8f',
+        'md5': 'dfdc87d2e7e09d073d5a80770a9ce88f',
         'info_dict': {
             'display_id': 'die-robuste-roswita',
-            'id': 'Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE',
+            'id': '70153354',
-            'title': 'Tatort: Die robuste Roswita',
+            'title': 'Die robuste Roswita',
             'description': r're:^Der Mord.*trüber ist als die Ilm.',
             'duration': 5316,
-            'thumbnail': 'https://img.ardmediathek.de/standard/00/55/43/59/34/-1774185891/16x9/960?mandant=ard',
+            'thumbnail': 'https://img.ardmediathek.de/standard/00/70/15/33/90/-1852531467/16x9/960?mandant=ard',
-            'upload_date': '20180826',
+            'timestamp': 1577047500,
+            'upload_date': '20191222',
             'ext': 'mp4',
         },
     }, {
@@ -330,71 +338,68 @@ class ARDBetaMediathekIE(InfoExtractor):
         video_id = mobj.group('video_id')
         display_id = mobj.group('display_id') or video_id
 
-        webpage = self._download_webpage(url, display_id)
+        player_page = self._download_json(
-        data_json = self._search_regex(r'window\.__APOLLO_STATE__\s*=\s*(\{.*);\n', webpage, 'json')
+            'https://api.ardmediathek.de/public-gateway',
-        data = self._parse_json(data_json, display_id)
+            display_id, data=json.dumps({
-
+                'query': '''{
-        res = {
+  playerPage(client:"%s", clipId: "%s") {
-            'id': video_id,
+    blockedByFsk
-            'display_id': display_id,
+    broadcastedOn
+    maturityContentRating
+    mediaCollection {
+      _duration
+      _geoblocked
+      _isLive
+      _mediaArray {
+        _mediaStreamArray {
+          _quality
+          _server
+          _stream
         }
-        formats = []
+      }
-        subtitles = {}
+      _previewImage
-        geoblocked = False
+      _subtitleUrl
-        for widget in data.values():
+      _type
-            if widget.get('_geoblocked') is True:
+    }
-                geoblocked = True
+    show {
-            if '_duration' in widget:
+      title
-                res['duration'] = int_or_none(widget['_duration'])
+    }
-            if 'clipTitle' in widget:
+    synopsis
-                res['title'] = widget['clipTitle']
+    title
-            if '_previewImage' in widget:
+    tracking {
-                res['thumbnail'] = widget['_previewImage']
+      atiCustomVars {
-            if 'broadcastedOn' in widget:
+        contentId
-                res['timestamp'] = unified_timestamp(widget['broadcastedOn'])
+      }
-            if 'synopsis' in widget:
+    }
-                res['description'] = widget['synopsis']
+  }
-            subtitle_url = url_or_none(widget.get('_subtitleUrl'))
+}''' % (mobj.group('client'), video_id),
-            if subtitle_url:
+            }).encode(), headers={
-                subtitles.setdefault('de', []).append({
+                'Content-Type': 'application/json'
-                    'ext': 'ttml',
+            })['data']['playerPage']
-                    'url': subtitle_url,
+        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:
+            age_limit = int_or_none(self._search_regex(r'\(FSK\s*(\d+)\)\s*$', description, 'age limit', default=None))
+        info.update({
+            'age_limit': age_limit,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+            'timestamp': unified_timestamp(player_page.get('broadcastedOn')),
+            'series': try_get(player_page, lambda x: x['show']['title']),
         })
-            if '_quality' in widget:
+        return info
-                format_url = url_or_none(try_get(
-                    widget, lambda x: x['_stream']['json'][0]))
-                if not format_url:
-                    continue
-                ext = determine_ext(format_url)
-                if ext == 'f4m':
-                    formats.extend(self._extract_f4m_formats(
-                        format_url + '?hdcore=3.11.0',
-                        video_id, f4m_id='hds', fatal=False))
-                elif ext == 'm3u8':
-                    formats.extend(self._extract_m3u8_formats(
-                        format_url, video_id, 'mp4', m3u8_id='hls',
-                        fatal=False))
-                else:
-                    # HTTP formats are not available when geoblocked is True,
-                    # other formats are fine though
-                    if geoblocked:
-                        continue
-                    quality = str_or_none(widget.get('_quality'))
-                    formats.append({
-                        'format_id': ('http-' + quality) if quality else 'http',
-                        'url': format_url,
-                        'preference': 10,  # Plain HTTP, that's nice
-                    })
-
-        if not formats and geoblocked:
-            self.raise_geo_restricted(
-                msg='This video is not available due to geoblocking',
-                countries=['DE'])
-
-        self._sort_formats(formats)
-        res.update({
-            'subtitles': subtitles,
-            'formats': formats,
-        })
-
-        return res

youtube_dl/extractor/srmediathek.py

@@ -1,14 +1,14 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-from .ard import ARDMediathekIE
+from .ard import ARDMediathekBaseIE
 from ..utils import (
     ExtractorError,
     get_element_by_attribute,
 )
 
 
-class SRMediathekIE(ARDMediathekIE):
+class SRMediathekIE(ARDMediathekBaseIE):
     IE_NAME = 'sr:mediathek'
     IE_DESC = 'Saarländischer Rundfunk'
     _VALID_URL = r'https?://sr-mediathek(?:\.sr-online)?\.de/index\.php\?.*?&id=(?P<id>[0-9]+)'