Update upstream source from tag 'upstream/2019.09.28'

[youtubedl] / youtube_dl / extractor / dbtv.py

diff --git a/youtube_dl/extractor/dbtv.py b/youtube_dl/extractor/dbtv.py
index 133cdc50b8c8379021d6d7fffb7c7c28dd2ba30d..aaedf2e3d36084d90699277a52aebb6c3a71a335 100644 (file)
--- a/youtube_dl/extractor/dbtv.py
+++ b/youtube_dl/extractor/dbtv.py
@@ -4,78 +4,54 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

-from ..compat import compat_str
-from ..utils import (
-    float_or_none,
-    int_or_none,
-    clean_html,
-)

 
 
 class DBTVIE(InfoExtractor):
 
 
 class DBTVIE(InfoExtractor):

-    _VALID_URL = r'https?://(?:www\.)?dbtv\.no/(?:(?:lazyplayer|player)/)?(?P<id>[0-9]+)(?:#(?P<display_id>.+))?'
+    _VALID_URL = r'https?://(?:www\.)?dagbladet\.no/video/(?:(?:embed|(?P<display_id>[^/]+))/)?(?P<id>[0-9A-Za-z_-]{11}|[a-zA-Z0-9]{8})'

     _TESTS = [{
     _TESTS = [{

-        'url': 'http://dbtv.no/3649835190001#Skulle_teste_ut_fornøyelsespark,_men_kollegaen_var_bare_opptatt_av_bikinikroppen',
-        'md5': 'b89953ed25dacb6edb3ef6c6f430f8bc',
+        'url': 'https://www.dagbladet.no/video/PynxJnNWChE/',
+        'md5': 'b8f850ba1860adbda668d367f9b77699',

         'info_dict': {
         'info_dict': {

-            'id': '33100',
-            'display_id': 'Skulle_teste_ut_fornøyelsespark,_men_kollegaen_var_bare_opptatt_av_bikinikroppen',
+            'id': 'PynxJnNWChE',

             'ext': 'mp4',
             'title': 'Skulle teste ut fornøyelsespark, men kollegaen var bare opptatt av bikinikroppen',
             'ext': 'mp4',
             'title': 'Skulle teste ut fornøyelsespark, men kollegaen var bare opptatt av bikinikroppen',

-            'description': 'md5:1504a54606c4dde3e4e61fc97aa857e0',
-            'thumbnail': 're:https?://.*\.jpg$',
-            'timestamp': 1404039863.438,
-            'upload_date': '20140629',
-            'duration': 69.544,
-            'view_count': int,
-            'categories': list,
-        }
+            'description': 'md5:49cc8370e7d66e8a2ef15c3b4631fd3f',
+            'thumbnail': r're:https?://.*\.jpg',
+            'upload_date': '20160916',
+            'duration': 69,
+            'uploader_id': 'UCk5pvsyZJoYJBd7_oFPTlRQ',
+            'uploader': 'Dagbladet',
+        },
+        'add_ie': ['Youtube']

     }, {
     }, {

-        'url': 'http://dbtv.no/3649835190001',
+        'url': 'https://www.dagbladet.no/video/embed/xlGmyIeN9Jo/?autoplay=false',

         'only_matching': True,
     }, {
         'only_matching': True,
     }, {

-        'url': 'http://www.dbtv.no/lazyplayer/4631135248001',
+        'url': 'https://www.dagbladet.no/video/truer-iran-bor-passe-dere/PalfB2Cw',

         'only_matching': True,
     }]
 
         'only_matching': True,
     }]
 

-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        display_id = mobj.group('display_id') or video_id
-
-        data = self._download_json(
-            'http://api.dbtv.no/discovery/%s' % video_id, display_id)
-
-        video = data['playlist'][0]
+    @staticmethod
+    def _extract_urls(webpage):
+        return [url for _, url in re.findall(
+            r'<iframe[^>]+src=(["\'])((?:https?:)?//(?:www\.)?dagbladet\.no/video/embed/(?:[0-9A-Za-z_-]{11}|[a-zA-Z0-9]{8}).*?)\1',
+            webpage)]

 
 

-        formats = [{
-            'url': f['URL'],
-            'vcodec': f.get('container'),
-            'width': int_or_none(f.get('width')),
-            'height': int_or_none(f.get('height')),
-            'vbr': float_or_none(f.get('rate'), 1000),
-            'filesize': int_or_none(f.get('size')),
-        } for f in video['renditions'] if 'URL' in f]
-
-        if not formats:
-            for url_key, format_id in [('URL', 'mp4'), ('HLSURL', 'hls')]:
-                if url_key in video:
-                    formats.append({
-                        'url': video[url_key],
-                        'format_id': format_id,
-                    })
-
-        self._sort_formats(formats)
-
-        return {
-            'id': compat_str(video['id']),
+    def _real_extract(self, url):
+        display_id, video_id = re.match(self._VALID_URL, url).groups()
+        info = {
+            '_type': 'url_transparent',
+            'id': video_id,

             'display_id': display_id,
             'display_id': display_id,

-            'title': video['title'],
-            'description': clean_html(video['desc']),
-            'thumbnail': video.get('splash') or video.get('thumb'),
-            'timestamp': float_or_none(video.get('publishedAt'), 1000),
-            'duration': float_or_none(video.get('length'), 1000),
-            'view_count': int_or_none(video.get('views')),
-            'categories': video.get('tags'),
-            'formats': formats,

         }
         }

+        if len(video_id) == 11:
+            info.update({
+                'url': video_id,
+                'ie_key': 'Youtube',
+            })
+        else:
+            info.update({
+                'url': 'jwplatform:' + video_id,
+                'ie_key': 'JWPlatform',
+            })
+        return info