Set upstream metadata fields: Bug-Database, Bug-Submit, Repository, Repository-Browse.

[youtubedl] / youtube_dl / extractor / twentymin.py

diff --git a/youtube_dl/extractor/twentymin.py b/youtube_dl/extractor/twentymin.py
index ca7d953b8e2733d404ebe9f7cd90e7e28d083abe..a42977f397e03ab71f70c585e1f3c34113b78eda 100644 (file)
--- a/youtube_dl/extractor/twentymin.py
+++ b/youtube_dl/extractor/twentymin.py
@@ -4,70 +4,88 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import remove_end
+from ..utils import (
+    int_or_none,
+    try_get,
+)
 
 
 class TwentyMinutenIE(InfoExtractor):
     IE_NAME = '20min'
-    _VALID_URL = r'https?://(?:www\.)?20min\.ch/(?:videotv/*\?.*\bvid=(?P<id>\d+)|(?:[^/]+/)*(?P<display_id>[^/#?]+))'
+    _VALID_URL = r'''(?x)
+                    https?://
+                        (?:www\.)?20min\.ch/
+                        (?:
+                            videotv/*\?.*?\bvid=|
+                            videoplayer/videoplayer\.html\?.*?\bvideoId@
+                        )
+                        (?P<id>\d+)
+                    '''
     _TESTS = [{
-        # regular video
         'url': 'http://www.20min.ch/videotv/?vid=469148&cid=2',
-        'md5': 'b52d6bc6ea6398e6a38f12cfd418149c',
+        'md5': 'e7264320db31eed8c38364150c12496e',
         'info_dict': {
             'id': '469148',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': '85 000 Franken für 15 perfekte Minuten',
-            'description': 'Was die Besucher vom Silvesterzauber erwarten können. (Video: Alice Grosjean/Murat Temel)',
-            'thumbnail': 'http://thumbnails.20min-tv.ch/server063/469148/frame-72-469148.jpg'
-        }
+            'thumbnail': r're:https?://.*\.jpg$',
+        },
     }, {
-        # news article with video
-        'url': 'http://www.20min.ch/schweiz/news/story/-Wir-muessen-mutig-nach-vorne-schauen--22050469',
-        'md5': 'cd4cbb99b94130cff423e967cd275e5e',
+        'url': 'http://www.20min.ch/videoplayer/videoplayer.html?params=client@twentyDE|videoId@523629',
         'info_dict': {
-            'id': '469408',
-            'display_id': '-Wir-muessen-mutig-nach-vorne-schauen--22050469',
-            'ext': 'flv',
-            'title': '«Wir müssen mutig nach vorne schauen»',
-            'description': 'Kein Land sei innovativer als die Schweiz, sagte Johann Schneider-Ammann in seiner Neujahrsansprache. Das Land müsse aber seine Hausaufgaben machen.',
-            'thumbnail': 'http://www.20min.ch/images/content/2/2/0/22050469/10/teaserbreit.jpg'
-        }
+            'id': '523629',
+            'ext': 'mp4',
+            'title': 'So kommen Sie bei Eis und Schnee sicher an',
+            'description': 'md5:117c212f64b25e3d95747e5276863f7d',
+            'thumbnail': r're:https?://.*\.jpg$',
+        },
+        'params': {
+            'skip_download': True,
+        },
     }, {
         'url': 'http://www.20min.ch/videotv/?cid=44&vid=468738',
         'only_matching': True,
-    }, {
-        'url': 'http://www.20min.ch/ro/sortir/cinema/story/Grandir-au-bahut--c-est-dur-18927411',
-        'only_matching': True,
     }]
 
+    @staticmethod
+    def _extract_urls(webpage):
+        return [m.group('url') for m in re.finditer(
+            r'<iframe[^>]+src=(["\'])(?P<url>(?:(?:https?:)?//)?(?:www\.)?20min\.ch/videoplayer/videoplayer.html\?.*?\bvideoId@\d+.*?)\1',
+            webpage)]
+
     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
+        video_id = self._match_id(url)
+
+        video = self._download_json(
+            'http://api.20min.ch/video/%s/show' % video_id,
+            video_id)['content']
+
+        title = video['title']
 
-        webpage = self._download_webpage(url, display_id)
+        formats = [{
+            'format_id': format_id,
+            'url': 'http://podcast.20min-tv.ch/podcast/20min/%s%s.mp4' % (video_id, p),
+            'quality': quality,
+        } for quality, (format_id, p) in enumerate([('sd', ''), ('hd', 'h')])]
+        self._sort_formats(formats)
 
-        title = self._html_search_regex(
-            r'<h1>.*?<span>(.+?)</span></h1>',
-            webpage, 'title', default=None)
-        if not title:
-            title = remove_end(re.sub(
-                r'^20 [Mm]inuten.*? -', '', self._og_search_title(webpage)), ' - News')
+        description = video.get('lead')
+        thumbnail = video.get('thumbnail')
 
-        if not video_id:
-            video_id = self._search_regex(
-                r'"file\d?"\s*,\s*\"(\d+)', webpage, 'video id')
+        def extract_count(kind):
+            return try_get(
+                video,
+                lambda x: int_or_none(x['communityobject']['thumbs_%s' % kind]))
 
-        description = self._html_search_meta(
-            'description', webpage, 'description')
-        thumbnail = self._og_search_thumbnail(webpage)
+        like_count = extract_count('up')
+        dislike_count = extract_count('down')
 
         return {
             'id': video_id,
-            'display_id': display_id,
-            'url': 'http://speed.20min-tv.ch/%sm.flv' % video_id,
             'title': title,
             'description': description,
             'thumbnail': thumbnail,
+            'like_count': like_count,
+            'dislike_count': dislike_count,
+            'formats': formats,
         }