[youtube] Fix extraction.

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

diff --git a/youtube_dl/extractor/vier.py b/youtube_dl/extractor/vier.py
index 7fb71834387b2ad46dff0d0b48d89d27976bd5ac..dbd5ba9bad82e3643a462a7db42fce9714dc54bb 100644 (file)
--- a/youtube_dl/extractor/vier.py
+++ b/youtube_dl/extractor/vier.py
@@ -15,7 +15,21 @@ from ..utils import (
 class VierIE(InfoExtractor):
     IE_NAME = 'vier'
     IE_DESC = 'vier.be and vijf.be'
-    _VALID_URL = r'https?://(?:www\.)?(?P<site>vier|vijf)\.be/(?:[^/]+/videos/(?P<display_id>[^/]+)(?:/(?P<id>\d+))?|video/v3/embed/(?P<embed_id>\d+))'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:www\.)?(?P<site>vier|vijf)\.be/
+                        (?:
+                            (?:
+                                [^/]+/videos|
+                                video(?:/[^/]+)*
+                            )/
+                            (?P<display_id>[^/]+)(?:/(?P<id>\d+))?|
+                            (?:
+                                video/v3/embed|
+                                embed/video/public
+                            )/(?P<embed_id>\d+)
+                        )
+                    '''
     _NETRC_MACHINE = 'vier'
     _TESTS = [{
         'url': 'http://www.vier.be/planb/videos/het-wordt-warm-de-moestuin/16129',
@@ -27,6 +41,8 @@ class VierIE(InfoExtractor):
             'title': 'Het wordt warm in De Moestuin',
             'description': 'De vele uren werk eisen hun tol. Wim droomt van assistentie...',
             'upload_date': '20121025',
+            'series': 'Plan B',
+            'tags': ['De Moestuin', 'Moestuin', 'meisjes', 'Tomaat', 'Wim', 'Droom'],
         },
     }, {
         'url': 'http://www.vijf.be/temptationisland/videos/zo-grappig-temptation-island-hosts-moeten-kiezen-tussen-onmogelijke-dilemmas/2561614',
@@ -37,6 +53,8 @@ class VierIE(InfoExtractor):
             'title': 'md5:84f45fe48b8c1fa296a7f6d208d080a7',
             'description': 'md5:0356d4981e58b8cbee19355cbd51a8fe',
             'upload_date': '20170228',
+            'series': 'Temptation Island',
+            'tags': list,
         },
         'params': {
             'skip_download': True,
@@ -50,7 +68,9 @@ class VierIE(InfoExtractor):
             'title': 'Jani gaat naar Tokio - Aflevering 4',
             'description': 'md5:aa8d611541db6ae9e863125704511f88',
             'upload_date': '20170501',
+            'series': 'Jani gaat',
             'episode_number': 4,
+            'tags': ['Jani Gaat', 'Volledige Aflevering'],
         },
         'params': {
             'skip_download': True,
@@ -77,6 +97,15 @@ class VierIE(InfoExtractor):
     }, {
         'url': 'http://www.vier.be/video/v3/embed/16129',
         'only_matching': True,
+    }, {
+        'url': 'https://www.vijf.be/embed/video/public/4093',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.vier.be/video/blockbusters/in-juli-en-augustus-summer-classics',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.vier.be/video/achter-de-rug/2017/achter-de-rug-seizoen-1-aflevering-6',
+        'only_matching': True,
     }]
 
     def _real_initialize(self):
@@ -127,41 +156,50 @@ class VierIE(InfoExtractor):
         video_id = self._search_regex(
             [r'data-nid="(\d+)"', r'"nid"\s*:\s*"(\d+)"'],
             webpage, 'video id', default=video_id or display_id)
-        application = self._search_regex(
-            [r'data-application="([^"]+)"', r'"application"\s*:\s*"([^"]+)"'],
-            webpage, 'application', default=site + '_vod')
-        filename = self._search_regex(
-            [r'data-filename="([^"]+)"', r'"filename"\s*:\s*"([^"]+)"'],
-            webpage, 'filename')
-
-        playlist_url = 'http://vod.streamcloud.be/%s/_definst_/mp4:%s.mp4/playlist.m3u8' % (application, filename)
-        formats = self._extract_wowza_formats(playlist_url, display_id, skip_protocols=['dash'])
+
+        playlist_url = self._search_regex(
+            r'data-file=(["\'])(?P<url>(?:https?:)?//[^/]+/.+?\.m3u8.*?)\1',
+            webpage, 'm3u8 url', default=None, group='url')
+
+        if not playlist_url:
+            application = self._search_regex(
+                [r'data-application="([^"]+)"', r'"application"\s*:\s*"([^"]+)"'],
+                webpage, 'application', default=site + '_vod')
+            filename = self._search_regex(
+                [r'data-filename="([^"]+)"', r'"filename"\s*:\s*"([^"]+)"'],
+                webpage, 'filename')
+            playlist_url = 'http://vod.streamcloud.be/%s/_definst_/mp4:%s.mp4/playlist.m3u8' % (application, filename)
+
+        formats = self._extract_wowza_formats(
+            playlist_url, display_id, skip_protocols=['dash'])
         self._sort_formats(formats)
 
         title = self._og_search_title(webpage, default=display_id)
-        thumbnail = self._og_search_thumbnail(webpage, default=None)
         description = self._html_search_regex(
-            r'''(?x)<div\ class="[^"]*field-type-text-with-summary[^"]*">\s*
-                      (?:<div\ class="[^"]+">\s*)*
-                     <p>\s*(?:<span>)?(.+?)</''',
-            webpage, 'description', default=None)
-        episode_number = int_or_none(self._search_regex(
-            r'(?i)aflevering (\d+)', title, 'episode_number', default=None,
-            fatal=False))
+            r'(?s)<div\b[^>]+\bclass=(["\'])[^>]*?\bfield-type-text-with-summary\b[^>]*?\1[^>]*>.*?<p>(?P<value>.+?)</p>',
+            webpage, 'description', default=None, group='value')
+        thumbnail = self._og_search_thumbnail(webpage, default=None)
         upload_date = unified_strdate(self._html_search_regex(
-            r'''(?x)<div\ class="[^"]*field-name-post-date[^"]*">\s*
-                    (?:<div\ class="[^"]+">\s*)*
-                      (\d{2}/\d{2}/\d{4})''',
-            webpage, 'upload_date', default=None))
+            r'(?s)<div\b[^>]+\bclass=(["\'])[^>]*?\bfield-name-post-date\b[^>]*?\1[^>]*>.*?(?P<value>\d{2}/\d{2}/\d{4})',
+            webpage, 'upload date', default=None, group='value'))
+
+        series = self._search_regex(
+            r'data-program=(["\'])(?P<value>(?:(?!\1).)+)\1', webpage,
+            'series', default=None, group='value')
+        episode_number = int_or_none(self._search_regex(
+            r'(?i)aflevering (\d+)', title, 'episode number', default=None))
+        tags = re.findall(r'<a\b[^>]+\bhref=["\']/tags/[^>]+>([^<]+)<', webpage)
 
         return {
             'id': video_id,
             'display_id': display_id,
             'title': title,
             'description': description,
-            'episode_number': episode_number,
-            'upload_date': upload_date,
             'thumbnail': thumbnail,
+            'upload_date': upload_date,
+            'series': series,
+            'episode_number': episode_number,
+            'tags': tags,
             'formats': formats,
         }