youtube-dl

commit 2d00be047726086e1deaca1c95dd9367daa85858
parent 0940c5b4c6a068d4919fd29a8db2a85ab3bbf703
Author: Yen Chi Hsuan <yan12125@gmail.com>
Date:   Fri, 25 Sep 2015 19:28:26 +0800

Merge branch 'remitamine-fktv'

Diffstat:
M
youtube_dl/extractor/__init__.py
 | 
5
+----
M
youtube_dl/extractor/fktv.py
 | 
89
++++++++++++++++++++++++++++---------------------------------------------------

2 files changed, 32 insertions(+), 62 deletions(-)
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
@@ -169,10 +169,7 @@ from .firstpost import FirstpostIE
 from .firsttv import FirstTVIE
 from .fivemin import FiveMinIE
 from .fivetv import FiveTVIE
-from .fktv import (
-    FKTVIE,
-    FKTVPosteckeIE,
-)
+from .fktv import FKTVIE
 from .flickr import FlickrIE
 from .folketinget import FolketingetIE
 from .footyroom import FootyRoomIE
diff --git a/youtube_dl/extractor/fktv.py b/youtube_dl/extractor/fktv.py
@@ -1,13 +1,12 @@
 from __future__ import unicode_literals
 
 import re
-import random
-import json
 
 from .common import InfoExtractor
 from ..utils import (
-    get_element_by_id,
     clean_html,
+    determine_ext,
+    ExtractorError,
 )
 
 
@@ -17,66 +16,40 @@ class FKTVIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://fernsehkritik.tv/folge-1',
+        'md5': '21f0b0c99bce7d5b524eb1b17b1c6d79',
         'info_dict': {
-            'id': '00011',
-            'ext': 'flv',
+            'id': '1',
+            'ext': 'mp4',
             'title': 'Folge 1 vom 10. April 2007',
-            'description': 'md5:fb4818139c7cfe6907d4b83412a6864f',
+            'thumbnail': 're:^https?://.*\.jpg$',
         },
     }
 
     def _real_extract(self, url):
-        episode = int(self._match_id(url))
-
-        video_thumbnail = 'http://fernsehkritik.tv/images/magazin/folge%s.jpg' % episode
-        start_webpage = self._download_webpage('http://fernsehkritik.tv/folge-%s/Start' % episode,
-                                               episode)
-        playlist = self._search_regex(r'playlist = (\[.*?\]);', start_webpage,
-                                      'playlist', flags=re.DOTALL)
-        files = json.loads(re.sub('{[^{}]*?}', '{}', playlist))
-
-        videos = []
-        for i, _ in enumerate(files, 1):
-            video_id = '%04d%d' % (episode, i)
-            video_url = 'http://fernsehkritik.tv/js/directme.php?file=%s%s.flv' % (episode, '' if i == 1 else '-%d' % i)
-            videos.append({
-                'ext': 'flv',
-                'id': video_id,
-                'url': video_url,
-                'title': clean_html(get_element_by_id('eptitle', start_webpage)),
-                'description': clean_html(get_element_by_id('contentlist', start_webpage)),
-                'thumbnail': video_thumbnail
-            })
-        return {
-            '_type': 'multi_video',
-            'entries': videos,
-            'id': 'folge-%s' % episode,
-        }
-
-
-class FKTVPosteckeIE(InfoExtractor):
-    IE_NAME = 'fernsehkritik.tv:postecke'
-    _VALID_URL = r'http://(?:www\.)?fernsehkritik\.tv/inline-video/postecke\.php\?(.*&)?ep=(?P<ep>[0-9]+)(&|$)'
-    _TEST = {
-        'url': 'http://fernsehkritik.tv/inline-video/postecke.php?iframe=true&width=625&height=440&ep=120',
-        'md5': '262f0adbac80317412f7e57b4808e5c4',
-        'info_dict': {
-            'id': '0120',
-            'ext': 'flv',
-            'title': 'Postecke 120',
-        }
-    }
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        episode = int(mobj.group('ep'))
-
-        server = random.randint(2, 4)
-        video_id = '%04d' % episode
-        video_url = 'http://dl%d.fernsehkritik.tv/postecke/postecke%d.flv' % (server, episode)
-        video_title = 'Postecke %d' % episode
+        episode = self._match_id(url)
+
+        webpage = self._download_webpage(
+            'http://fernsehkritik.tv/folge-%s/play' % episode, episode)
+        title = clean_html(self._html_search_regex(
+            '<h3>([^<]+)</h3>', webpage, 'title'))
+        matches = re.search(
+            r'(?s)<video(?:(?!poster)[^>])+(?:poster="([^"]+)")?[^>]*>(.*)</video>',
+            webpage)
+        if matches is None:
+            raise ExtractorError('Unable to extract the video')
+
+        poster, sources = matches.groups()
+        if poster is None:
+            self.report_warning('unable to extract thumbnail')
+
+        urls = re.findall(r'<source[^>]+src="([^"]+)"', sources)
+        formats = [{
+            'url': url,
+            'format_id': determine_ext(url),
+        } for url in urls]
         return {
-            'id': video_id,
-            'url': video_url,
-            'title': video_title,
+            'id': episode,
+            'title': title,
+            'formats': formats,
+            'thumbnail': poster,
         }