[nowness] fix video extraction and add support serie and playlist extraction (fixes #6720)

youtube_dl/extractor/__init__.py

@@ -403,7 +403,11 @@ from .normalboots import NormalbootsIE
 from .nosvideo import NosVideoIE
 from .nova import NovaIE
 from .novamov import NovaMovIE
-from .nowness import NownessIE
+from .nowness import (
+    NownessIE,
+    NownessPlaylistIE,
+    NownessSerieIE,
+)
 from .nowtv import NowTVIE
 from .nowvideo import NowVideoIE
 from .npo import (

youtube_dl/extractor/nowness.py

@@ -1,19 +1,42 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .brightcove import BrightcoveIE
 from .common import InfoExtractor
 from ..utils import ExtractorError
+from ..compat import compat_urllib_request
+
 
+class NownessBaseIE(InfoExtractor):
+    def extract_url_result(self, post):
+        if post['type'] == 'video':
+            for media in post['media']:
+                if media['type'] == 'video':
+                    video_id = media['content']
+                    source = media['source']
+                    if source == 'brightcove':
+                        player_code = self._download_webpage(
+                            'http://www.nowness.com/iframe?id=%s' % video_id, video_id,
+                            note='Downloading player JavaScript',
+                            errnote='Player download failed')
+                        bc_url = BrightcoveIE._extract_brightcove_url(player_code)
+                        if bc_url is None:
+                            raise ExtractorError('Could not find player definition')
+                        return self.url_result(bc_url, 'Brightcove')
+                    elif source == 'vimeo':
+                        return self.url_result('http://vimeo.com/%s' % video_id, 'Vimeo')
+                    elif source == 'youtube':
+                        return self.url_result(video_id, 'Youtube')
+                    elif source == 'cinematique':
+                        return self.url_result('http://cinematique.com/embed/%s' % video_id, 'Cinematique')
 
-class NownessIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/[^?#]*?/(?P<id>[0-9]+)/(?P<slug>[^/]+?)(?:$|[?#])'
 
+class NownessIE(NownessBaseIE):
+    IE_NAME = 'nowness'
+    _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/(story|series/[^/])/(?P<id>[0-9a-z-]+)'
     _TESTS = [
         {
-            'url': 'http://www.nowness.com/day/2013/6/27/3131/candor--the-art-of-gesticulation',
+            'url': 'https://www.nowness.com/story/candor-the-art-of-gesticulation',
             'md5': '068bc0202558c2e391924cb8cc470676',
             'info_dict': {
                 'id': '2520295746001',
@@ -25,7 +48,7 @@ class NownessIE(InfoExtractor):
             }
         },
         {
-            'url': 'http://cn.nowness.com/day/2014/8/7/4069/kasper-bj-rke-ft-jaakko-eino-kalevi--tnr',
+            'url': 'https://cn.nowness.com/story/kasper-bjorke-ft-jaakko-eino-kalevi-tnr',
             'md5': 'e79cf125e387216f86b2e0a5b5c63aa3',
             'info_dict': {
                 'id': '3716354522001',
@@ -39,26 +62,60 @@ class NownessIE(InfoExtractor):
     ]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        display_id = self._match_id(url)
-        video_id = mobj.group('slug')
+
+        lang = 'zh-cn' if 'cn.nowness.com' in url else 'en-us'
+        request = compat_urllib_request.Request('http://api.nowness.com/api/post/getBySlug/%s' % display_id, headers={
+            'X-Nowness-Language': lang,
+        })
+        post = self._download_json(request, display_id)
+        return self.extract_url_result(post)
+
+
+class NownessPlaylistIE(NownessBaseIE):
+    IE_NAME = 'nowness:playlist'
+    _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/playlist/(?P<id>\d+)/[0-9a-z-]+'
+    _TEST = {
+        'url': 'https://www.nowness.com/playlist/3286/i-guess-thats-why-they-call-it-the-blues',
+        'info_dict':
+        {
+            'id': '3286',
+        },
+        'playlist_mincount': 8,
+    }
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
+        lang = 'zh-cn' if 'cn.nowness.com' in url else 'en-us'
-        player_url = self._search_regex(
+        request = compat_urllib_request.Request('http://api.nowness.com/api/post?PlaylistId=%s' % playlist_id, headers={
-            r'"([^"]+/content/issue-[0-9.]+.js)"', webpage, 'player URL')
+            'X-Nowness-Language': lang,
-        real_id = self._search_regex(
+        })
-            r'\sdata-videoId="([0-9]+)"', webpage, 'internal video ID')
+        playlist = self._download_json(request, playlist_id)
+        entries = [self.extract_url_result(item) for item in playlist['items']]
+        return self.playlist_result(entries, playlist_id)
 
-        player_code = self._download_webpage(
-            player_url, video_id,
-            note='Downloading player JavaScript',
-            errnote='Player download failed')
-        player_code = player_code.replace("'+d+'", real_id)
 
-        bc_url = BrightcoveIE._extract_brightcove_url(player_code)
+class NownessSerieIE(NownessBaseIE):
-        if bc_url is None:
+    IE_NAME = 'nowness:serie'
-            raise ExtractorError('Could not find player definition')
+    _VALID_URL = r'https?://(?:(?:www|cn)\.)?nowness\.com/series/(?P<id>[0-9a-z-]+)'
-        return {
+    _TEST = {
-            '_type': 'url',
+        'url': 'https://www.nowness.com/series/60-seconds',
-            'url': bc_url,
+        'info_dict':
-            'ie_key': 'Brightcove',
+        {
-        }
+            'id': '60',
+        },
+        'playlist_mincount': 4,
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        lang = 'zh-cn' if 'cn.nowness.com' in url else 'en-us'
+        request = compat_urllib_request.Request('https://api.nowness.com/api/series/getBySlug/%s' % display_id, headers={
+            'X-Nowness-Language': lang,
+        })
+        serie = self._download_json(request, display_id)
+        serie_id = str(serie['id'])
+        entries = [self.extract_url_result(post) for post in serie['posts']]
+        return self.playlist_result(entries, serie_id)