[20min] Improve (Closes #8110)

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index 64e1fd334af08d05dd33cf10bf91f3d9782fc30b..625b0bf1667c07aaafe17f67d3425fa4f9082f65 100644 (file)
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -364,7 +364,6 @@ from .mdr import MDRIE
 from .metacafe import MetacafeIE
 from .metacritic import MetacriticIE
 from .mgoon import MgoonIE
-from .min20 import Min20IE
 from .minhateca import MinhatecaIE
 from .ministrygrid import MinistryGridIE
 from .miomio import MioMioIE
@@ -747,6 +746,7 @@ from .tvp import TvpIE, TvpSeriesIE
 from .tvplay import TVPlayIE
 from .tweakers import TweakersIE
 from .twentyfourvideo import TwentyFourVideoIE
+from .twentymin import TwentyMinutenIE
 from .twentytwotracks import (
     TwentyTwoTracksIE,
     TwentyTwoTracksGenreIE

diff --git a/youtube_dl/extractor/min20.py b/youtube_dl/extractor/min20.py
deleted file mode 100644 (file)
index 23aead1..0000000
--- a/youtube_dl/extractor/min20.py
+++ /dev/null
@@ -1,56 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-
-
-class Min20IE(InfoExtractor):
-    _VALID_URL = r'http://www\.20min\.ch/(videotv/\?vid=(?P<video_id>[0-9]+)|.+?-(?P<page_id>[0-9]+)$)'
-    _TESTS = [{
-        'url': 'http://www.20min.ch/schweiz/news/story/-Wir-muessen-mutig-nach-vorne-schauen--22050469',
-        'md5': 'cd4cbb99b94130cff423e967cd275e5e',
-        'info_dict': {
-            'id': '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'
-        }
-    }, {
-        'url': 'http://www.20min.ch/videotv/?vid=469148&cid=2',
-        'md5': 'b52d6bc6ea6398e6a38f12cfd418149c',
-        'info_dict': {
-            'id': '469148',
-            'ext': 'flv',
-            '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'
-        }
-    }]
-
-    # location of the flv videos, can't be extracted from the web page
-    _BASE_URL = "http://flv-rr.20min-tv.ch/videos/"
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('page_id')
-        if video_id is None:
-            # URL from the videoportal
-            video_id = mobj.group('video_id')
-        webpage = self._download_webpage(url, video_id)
-        title = self._html_search_regex(r'<h1>.*<span>(.+?)</span></h1>', webpage, 'title')
-        flash_id = self._search_regex(r"so\.addVariable\(\"file1\",\"([0-9]+)\"\)", webpage, 'flash_id')
-
-        description = self._html_search_regex(r'<meta name="description" content="(.+?)" />', webpage, 'description')
-        thumbnail = self._html_search_regex(r'<meta property="og:image" content="(.+?)" />', webpage, 'thumbnail')
-        url = self._BASE_URL + flash_id + "m.flv"
-
-        return {
-            'id': video_id,
-            'url': url,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail
-        }

diff --git a/youtube_dl/extractor/twentymin.py b/youtube_dl/extractor/twentymin.py
new file mode 100644 (file)
index 0000000..ca7d953
--- /dev/null
+++ b/youtube_dl/extractor/twentymin.py
@@ -0,0 +1,73 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import remove_end
+
+
+class TwentyMinutenIE(InfoExtractor):
+    IE_NAME = '20min'
+    _VALID_URL = r'https?://(?:www\.)?20min\.ch/(?:videotv/*\?.*\bvid=(?P<id>\d+)|(?:[^/]+/)*(?P<display_id>[^/#?]+))'
+    _TESTS = [{
+        # regular video
+        'url': 'http://www.20min.ch/videotv/?vid=469148&cid=2',
+        'md5': 'b52d6bc6ea6398e6a38f12cfd418149c',
+        'info_dict': {
+            'id': '469148',
+            'ext': 'flv',
+            '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'
+        }
+    }, {
+        # news article with video
+        'url': 'http://www.20min.ch/schweiz/news/story/-Wir-muessen-mutig-nach-vorne-schauen--22050469',
+        'md5': 'cd4cbb99b94130cff423e967cd275e5e',
+        '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'
+        }
+    }, {
+        '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,
+    }]
+
+    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
+
+        webpage = self._download_webpage(url, display_id)
+
+        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')
+
+        if not video_id:
+            video_id = self._search_regex(
+                r'"file\d?"\s*,\s*\"(\d+)', webpage, 'video id')
+
+        description = self._html_search_meta(
+            'description', webpage, 'description')
+        thumbnail = self._og_search_thumbnail(webpage)
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'url': 'http://speed.20min-tv.ch/%sm.flv' % video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+        }