[youtube] Fix extraction.

[youtube-dl] / youtube_dl / extractor / filmweb.py

diff --git a/youtube_dl/extractor/filmweb.py b/youtube_dl/extractor/filmweb.py
index a3d9f872ef06b42616ede3d311338ba1ea3e2c72..56000bc5ba941dfe444fd4c29cbb3b6dc8eff0a4 100644 (file)
--- a/youtube_dl/extractor/filmweb.py
+++ b/youtube_dl/extractor/filmweb.py
@@ -1,45 +1,42 @@
 from __future__ import unicode_literals
 
-from .twentythreevideo import TwentyThreeVideoIE
+import re
 
+from .common import InfoExtractor
 
-class FilmwebIE(TwentyThreeVideoIE):
-    IE_NAME = 'Filmweb'
-    _VALID_URL = r'https?://(?:www\.)?filmweb\.no/trailere/article(?P<id>\d+).ece'
+
+class FilmwebIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?filmweb\.no/(?P<type>trailere|filmnytt)/article(?P<id>\d+)\.ece'
     _TEST = {
         'url': 'http://www.filmweb.no/trailere/article1264921.ece',
         'md5': 'e353f47df98e557d67edaceda9dece89',
         'info_dict': {
-            'id': '1264921',
-            'title': 'Det som en gang var',
+            'id': '13033574',
             'ext': 'mp4',
-            'description': 'Trailer: Scener fra et vennskap',
+            'title': 'Det som en gang var',
+            'upload_date': '20160316',
+            'timestamp': 1458140101,
+            'uploader_id': '12639966',
+            'uploader': 'Live Roaldset',
         }
     }
 
-    _CLIENT_NAME = 'filmweb'
-    _CLIENT_ID = '12732917'
-    _EMBED_BASE_URL = 'http://www.filmweb.no/template/ajax/json_trailerEmbed.jsp?articleId=%s&autoplay=true'
-
     def _real_extract(self, url):
-        article_id = self._match_id(url)
-        webpage = self._download_webpage(url, article_id)
-
-        title = self._search_regex(r'var\s+jsTitle\s*=\s*escape\("([^"]+)"\);',
-            webpage, 'title', fatal=True)
-
-        format_url = self._proto_relative_url(
-            self._html_search_regex(r'"(//filmweb\.23video\.com/[^"]+)"',
-                self._download_json(self._EMBED_BASE_URL % article_id,
-                    article_id)['embedCode'], 'format url'))
-
-        formats = self._extract_formats(format_url, self._CLIENT_ID)
-        self._sort_formats(formats)
+        article_type, article_id = re.match(self._VALID_URL, url).groups()
+        if article_type == 'filmnytt':
+            webpage = self._download_webpage(url, article_id)
+            article_id = self._search_regex(r'data-videoid="(\d+)"', webpage, 'article id')
+        embed_code = self._download_json(
+            'https://www.filmweb.no/template_v2/ajax/json_trailerEmbed.jsp',
+            article_id, query={
+                'articleId': article_id,
+            })['embedCode']
+        iframe_url = self._proto_relative_url(self._search_regex(
+            r'<iframe[^>]+src="([^"]+)', embed_code, 'iframe url'))
 
         return {
+            '_type': 'url_transparent',
             'id': article_id,
-            'title': title,
-            'alt_title': self._og_search_title(webpage),
-            'formats': formats,
-            'description': self._og_search_description(webpage),
+            'url': iframe_url,
+            'ie_key': 'TwentyThreeVideo',
         }