[nhk:vod] Improve extraction (Closes #10424)

1 files changed, 32 insertions, 11 deletions

diff --git a/youtube_dl/extractor/nhk.py b/youtube_dl/extractor/nhk.py
index 90e935351..691bdfa4e 100644
--- a/youtube_dl/extractor/nhk.py
+++ b/youtube_dl/extractor/nhk.py
@@ -4,26 +4,47 @@ from .common import InfoExtractor
 
 
 class NhkVodIE(InfoExtractor):
-    _VALID_URL = r'http://www3\.nhk\.or\.jp/nhkworld/en/vod/(?P<id>.+)\.html'
-    _TESTS = [{
+    _VALID_URL = r'https?://www3\.nhk\.or\.jp/nhkworld/en/vod/(?P<id>.+?)\.html'
+    _TEST = {
+        # Videos available only for a limited period of time. Visit
+        # http://www3.nhk.or.jp/nhkworld/en/vod/ for working samples.
         'url': 'http://www3.nhk.or.jp/nhkworld/en/vod/tokyofashion/20160815.html',
         'info_dict': {
             'id': 'A1bnNiNTE6nY3jLllS-BIISfcC_PpvF5',
             'ext': 'flv',
-            'title': '[nhkworld]VOD;2009-251-2016;TOKYO FASHION EXPRESS;The Kimono as Global Fashion;en',
+            'title': 'TOKYO FASHION EXPRESS - The Kimono as Global Fashion',
+            'description': 'md5:db338ee6ce8204f415b754782f819824',
+            'series': 'TOKYO FASHION EXPRESS',
+            'episode': 'The Kimono as Global Fashion',
         },
-        'params': {
-            'skip_download': True  # Videos available only for a limited period of time.
-        },
-    }]
+        'skip': 'Videos available only for a limited period of time',
+    }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
+
         webpage = self._download_webpage(url, video_id)
 
         embed_code = self._search_regex(
-            r'''nw_vod_ooplayer\('movie-area', '([^']+)'\);''',
-            webpage,
-            'ooyala embed code')
+            r'nw_vod_ooplayer\([^,]+,\s*(["\'])(?P<id>(?:(?!\1).)+)\1',
+            webpage, 'ooyala embed code', group='id')
+
+        title = self._search_regex(
+            r'<div[^>]+class=["\']episode-detail["\']>\s*<h\d+>([^<]+)',
+            webpage, 'title', default=None)
+        description = self._html_search_regex(
+            r'(?s)<p[^>]+class=["\']description["\'][^>]*>(.+?)</p>',
+            webpage, 'description', default=None)
+        series = self._search_regex(
+            r'<h2[^>]+class=["\']detail-top-player-title[^>]+><a[^>]+>([^<]+)',
+            webpage, 'series', default=None)
 
-        return self.url_result('ooyala:' + embed_code, 'Ooyala')
+        return {
+            '_type': 'url_transparent',
+            'ie_key': 'Ooyala',
+            'url': 'ooyala:%s' % embed_code,
+            'title': '%s - %s' % (series, title) if series and title else title,
+            'description': description,
+            'series': series,
+            'episode': title,
+        }