Update upstream source from tag 'upstream/2020.06.16.1'

[youtubedl] / youtube_dl / extractor / npo.py

diff --git a/youtube_dl/extractor/npo.py b/youtube_dl/extractor/npo.py
index ff215338744893a29b93202e6f1f72dd0186546f..e525ad928433bc64521cce3e9c2bdd2a7ead65b1 100644 (file)
--- a/youtube_dl/extractor/npo.py
+++ b/youtube_dl/extractor/npo.py
@@ -12,11 +12,16 @@ from ..utils import (
     ExtractorError,
     fix_xml_ampersands,
     int_or_none,
     ExtractorError,
     fix_xml_ampersands,
     int_or_none,

+    merge_dicts,

     orderedSet,
     parse_duration,
     qualities,
     orderedSet,
     parse_duration,
     qualities,

+    str_or_none,

     strip_jsonp,
     unified_strdate,
     strip_jsonp,
     unified_strdate,

+    unified_timestamp,
+    url_or_none,
+    urlencode_postdata,

 )
 
 
 )
 
 


@@ -36,8 +41,8 @@ class NPOIE(NPOBaseIE):
                         https?://
                             (?:www\.)?
                             (?:
                         https?://
                             (?:www\.)?
                             (?:

-                                npo\.nl/(?!(?:live|radio)/)(?:[^/]+/){2}|
-                                ntr\.nl/(?:[^/]+/){2,}|
+                                npo\.nl/(?:[^/]+/)*|
+                                (?:ntr|npostart)\.nl/(?:[^/]+/){2,}|

                                 omroepwnl\.nl/video/fragment/[^/]+__|
                                 (?:zapp|npo3)\.nl/(?:[^/]+/){2,}
                             )
                                 omroepwnl\.nl/video/fragment/[^/]+__|
                                 (?:zapp|npo3)\.nl/(?:[^/]+/){2,}
                             )


@@ -160,13 +165,138 @@ class NPOIE(NPOBaseIE):
     }, {
         'url': 'https://www.zapp.nl/1803-skelterlab/instructie-video-s/740-instructievideo-s/POMS_AT_11736927',
         'only_matching': True,
     }, {
         'url': 'https://www.zapp.nl/1803-skelterlab/instructie-video-s/740-instructievideo-s/POMS_AT_11736927',
         'only_matching': True,

+    }, {
+        'url': 'https://www.npostart.nl/broodje-gezond-ei/28-05-2018/KN_1698996',
+        'only_matching': True,
+    }, {
+        'url': 'https://npo.nl/KN_1698996',
+        'only_matching': True,

     }]
 
     }]
 

+    @classmethod
+    def suitable(cls, url):
+        return (False if any(ie.suitable(url)
+                for ie in (NPOLiveIE, NPORadioIE, NPORadioFragmentIE))
+                else super(NPOIE, cls).suitable(url))
+

     def _real_extract(self, url):
         video_id = self._match_id(url)
     def _real_extract(self, url):
         video_id = self._match_id(url)

-        return self._get_info(video_id)
+        return self._get_info(url, video_id) or self._get_old_info(video_id)
+
+    def _get_info(self, url, video_id):
+        token = self._download_json(
+            'https://www.npostart.nl/api/token', video_id,
+            'Downloading token', headers={
+                'Referer': url,
+                'X-Requested-With': 'XMLHttpRequest',
+            })['token']
+
+        player = self._download_json(
+            'https://www.npostart.nl/player/%s' % video_id, video_id,
+            'Downloading player JSON', data=urlencode_postdata({
+                'autoplay': 0,
+                'share': 1,
+                'pageUrl': url,
+                'hasAdConsent': 0,
+                '_token': token,
+            }))
+
+        player_token = player['token']
+
+        drm = False
+        format_urls = set()
+        formats = []
+        for profile in ('hls', 'dash-widevine', 'dash-playready', 'smooth'):
+            streams = self._download_json(
+                'https://start-player.npo.nl/video/%s/streams' % video_id,
+                video_id, 'Downloading %s profile JSON' % profile, fatal=False,
+                query={
+                    'profile': profile,
+                    'quality': 'npo',
+                    'tokenId': player_token,
+                    'streamType': 'broadcast',
+                })
+            if not streams:
+                continue
+            stream = streams.get('stream')
+            if not isinstance(stream, dict):
+                continue
+            stream_url = url_or_none(stream.get('src'))
+            if not stream_url or stream_url in format_urls:
+                continue
+            format_urls.add(stream_url)
+            if stream.get('protection') is not None or stream.get('keySystemOptions') is not None:
+                drm = True
+                continue
+            stream_type = stream.get('type')
+            stream_ext = determine_ext(stream_url)
+            if stream_type == 'application/dash+xml' or stream_ext == 'mpd':
+                formats.extend(self._extract_mpd_formats(
+                    stream_url, video_id, mpd_id='dash', fatal=False))
+            elif stream_type == 'application/vnd.apple.mpegurl' or stream_ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    stream_url, video_id, ext='mp4',
+                    entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
+            elif re.search(r'\.isml?/Manifest', stream_url):
+                formats.extend(self._extract_ism_formats(
+                    stream_url, video_id, ism_id='mss', fatal=False))
+            else:
+                formats.append({
+                    'url': stream_url,
+                })

 
 

-    def _get_info(self, video_id):
+        if not formats:
+            if drm:
+                raise ExtractorError('This video is DRM protected.', expected=True)
+            return
+
+        self._sort_formats(formats)
+
+        info = {
+            'id': video_id,
+            'title': video_id,
+            'formats': formats,
+        }
+
+        embed_url = url_or_none(player.get('embedUrl'))
+        if embed_url:
+            webpage = self._download_webpage(
+                embed_url, video_id, 'Downloading embed page', fatal=False)
+            if webpage:
+                video = self._parse_json(
+                    self._search_regex(
+                        r'\bvideo\s*=\s*({.+?})\s*;', webpage, 'video',
+                        default='{}'), video_id)
+                if video:
+                    title = video.get('episodeTitle')
+                    subtitles = {}
+                    subtitles_list = video.get('subtitles')
+                    if isinstance(subtitles_list, list):
+                        for cc in subtitles_list:
+                            cc_url = url_or_none(cc.get('src'))
+                            if not cc_url:
+                                continue
+                            lang = str_or_none(cc.get('language')) or 'nl'
+                            subtitles.setdefault(lang, []).append({
+                                'url': cc_url,
+                            })
+                    return merge_dicts({
+                        'title': title,
+                        'description': video.get('description'),
+                        'thumbnail': url_or_none(
+                            video.get('still_image_url') or video.get('orig_image_url')),
+                        'duration': int_or_none(video.get('duration')),
+                        'timestamp': unified_timestamp(video.get('broadcastDate')),
+                        'creator': video.get('channel'),
+                        'series': video.get('title'),
+                        'episode': title,
+                        'episode_number': int_or_none(video.get('episodeNumber')),
+                        'subtitles': subtitles,
+                    }, info)
+
+        return info
+
+    def _get_old_info(self, video_id):

         metadata = self._download_json(
             'http://e.omroep.nl/metadata/%s' % video_id,
             video_id,
         metadata = self._download_json(
             'http://e.omroep.nl/metadata/%s' % video_id,
             video_id,


@@ -268,9 +398,9 @@ class NPOIE(NPOBaseIE):
             # JSON
             else:
                 video_url = stream_info.get('url')
             # JSON
             else:
                 video_url = stream_info.get('url')

-            if not video_url or video_url in urls:
+            if not video_url or 'vodnotavailable.' in video_url or video_url in urls:

                 continue
                 continue

-            urls.add(item_url)
+            urls.add(video_url)

             if determine_ext(video_url) == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
                     video_url, video_id, ext='mp4',
             if determine_ext(video_url) == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
                     video_url, video_id, ext='mp4',


@@ -351,7 +481,7 @@ class NPOIE(NPOBaseIE):
 
 class NPOLiveIE(NPOBaseIE):
     IE_NAME = 'npo.nl:live'
 
 class NPOLiveIE(NPOBaseIE):
     IE_NAME = 'npo.nl:live'

-    _VALID_URL = r'https?://(?:www\.)?npo\.nl/live(?:/(?P<id>[^/?#&]+))?'
+    _VALID_URL = r'https?://(?:www\.)?npo(?:start)?\.nl/live(?:/(?P<id>[^/?#&]+))?'

 
     _TESTS = [{
         'url': 'http://www.npo.nl/live/npo-1',
 
     _TESTS = [{
         'url': 'http://www.npo.nl/live/npo-1',


@@ -368,6 +498,9 @@ class NPOLiveIE(NPOBaseIE):
     }, {
         'url': 'http://www.npo.nl/live',
         'only_matching': True,
     }, {
         'url': 'http://www.npo.nl/live',
         'only_matching': True,

+    }, {
+        'url': 'https://www.npostart.nl/live/npo-1',
+        'only_matching': True,

     }]
 
     def _real_extract(self, url):
     }]
 
     def _real_extract(self, url):


@@ -389,7 +522,7 @@ class NPOLiveIE(NPOBaseIE):
 
 class NPORadioIE(InfoExtractor):
     IE_NAME = 'npo.nl:radio'
 
 class NPORadioIE(InfoExtractor):
     IE_NAME = 'npo.nl:radio'

-    _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P<id>[^/]+)/?$'
+    _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P<id>[^/]+)'

 
     _TEST = {
         'url': 'http://www.npo.nl/radio/radio-1',
 
     _TEST = {
         'url': 'http://www.npo.nl/radio/radio-1',


@@ -404,6 +537,10 @@ class NPORadioIE(InfoExtractor):
         }
     }
 
         }
     }
 

+    @classmethod
+    def suitable(cls, url):
+        return False if NPORadioFragmentIE.suitable(url) else super(NPORadioIE, cls).suitable(url)
+

     @staticmethod
     def _html_get_attribute_regex(attribute):
         return r'{0}\s*=\s*\'([^\']+)\''.format(attribute)
     @staticmethod
     def _html_get_attribute_regex(attribute):
         return r'{0}\s*=\s*\'([^\']+)\''.format(attribute)