youtube-dl

commit c5e743f66f5637fe02fe0b5167fab99a06b903e6
parent 6c36d8d6fb39b686ae5b475868de3523b6c76f76
Author: Jaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>
Date:   Wed, 18 Sep 2013 23:32:37 +0200

[fktv] support videos splitted in any number of parts and some style changes

Diffstat:
M
youtube_dl/extractor/fktv.py
 | 
57
+++++++++++++++++++++++++++++++++++++++------------------

1 file changed, 39 insertions(+), 18 deletions(-)
diff --git a/youtube_dl/extractor/fktv.py b/youtube_dl/extractor/fktv.py
@@ -1,37 +1,58 @@
-import re,random
+import re
+import random
+import json
 
 from .common import InfoExtractor
 from ..utils import (
     determine_ext,
+    get_element_by_id,
+    clean_html,
 )
 
+
 class FKTVIE(InfoExtractor):
-    """Information Extractor for Fernsehkritik-TV"""
+    IE_NAME = u'fernsehkritik.tv'
     _VALID_URL = r'(?:http://)?(?:www\.)?fernsehkritik.tv/folge-(?P<ep>[0-9]+)(?:/.*)?'
 
-    def _real_extract(self,url):
+    _TEST = {
+        u'url': u'http://fernsehkritik.tv/folge-1',
+        u'file': u'00011.flv',
+        u'info_dict': {
+            u'title': u'Folge 1 vom 10. April 2007',
+            u'description': u'md5:fb4818139c7cfe6907d4b83412a6864f',
+        },
+    }
+
+    def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         episode = int(mobj.group('ep'))
-        
-        server = random.randint(2,4)
+
+        server = random.randint(2, 4)
         video_thumbnail = 'http://fernsehkritik.tv/images/magazin/folge%d.jpg' % episode
+        start_webpage = self._download_webpage('http://fernsehkritik.tv/folge-%d/Start' % episode,
+            episode)
+        playlist = self._search_regex(r'playlist = (\[.*?\]);', start_webpage,
+            u'playlist', flags=re.DOTALL)
+        files = json.loads(re.sub('{[^{}]*?}', '{}', playlist))
+        # TODO: return a single multipart video
         videos = []
-        # Download all three parts
-        for i in range(1,4):
+        for i, _ in enumerate(files, 1):
             video_id = '%04d%d' % (episode, i)
-            video_url = 'http://dl%d.fernsehkritik.tv/fernsehkritik%d%s.flv' % (server, episode, '' if i==1 else '-%d'%i)
+            video_url = 'http://dl%d.fernsehkritik.tv/fernsehkritik%d%s.flv' % (server, episode, '' if i == 1 else '-%d' % i)
             video_title = 'Fernsehkritik %d.%d' % (episode, i)
             videos.append({
-                'id':       video_id,
-                'url':      video_url,
-                'ext':      determine_ext(video_url),
-                'title':    video_title,
+                'id': video_id,
+                'url': video_url,
+                'ext': determine_ext(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 videos
 
+
 class FKTVPosteckeIE(InfoExtractor):
-    """Information Extractor for Fernsehkritik-TV Postecke"""
+    IE_NAME = u'fernsehkritik.tv:postecke'
     _VALID_URL = r'(?:http://)?(?:www\.)?fernsehkritik.tv/inline-video/postecke.php\?(.*&)?ep=(?P<ep>[0-9]+)(&|$)'
     _TEST = {
         u'url': u'http://fernsehkritik.tv/inline-video/postecke.php?iframe=true&width=625&height=440&ep=120',
@@ -42,17 +63,17 @@ class FKTVPosteckeIE(InfoExtractor):
         }
     }
 
-    def _real_extract(self,url):
+    def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         episode = int(mobj.group('ep'))
-        
-        server = random.randint(2,4)
+
+        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
-        return[{
+        return {
             'id':       video_id,
             'url':      video_url,
             'ext':      determine_ext(video_url),
             'title':    video_title,
-        }]
+        }