Merge branch 'pr-twitter' of https://github.com/atomicdryad/youtube-dl into atomicdry...

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

diff --git a/youtube_dl/extractor/fktv.py b/youtube_dl/extractor/fktv.py
index 239d9df38190536633f214795f2a8ad12ce89eb6..40ea2789525ab720beb9e6e42a096682a9581027 100644 (file)
--- a/youtube_dl/extractor/fktv.py
+++ b/youtube_dl/extractor/fktv.py
@@ -1,58 +1,55 @@
-import re,random
+from __future__ import unicode_literals
+
+import re
 
 from .common import InfoExtractor
 from ..utils import (
+    clean_html,
     determine_ext,
+    ExtractorError,
 )
 
+
 class FKTVIE(InfoExtractor):
-    """Information Extractor for Fernsehkritik-TV"""
-    _VALID_URL = r'(?:http://)?(?:www\.)?fernsehkritik.tv/folge-(?P<ep>[0-9]+)(?:/.*)?'
-
-    def _real_extract(self,url):
-        mobj = re.match(self._VALID_URL, url)
-        episode = int(mobj.group('ep'))
-        
-        server = random.randint(2,4)
-        video_thumbnail = 'http://fernsehkritik.tv/images/magazin/folge%d.jpg' % episode
-        videos = []
-        # Download all three parts
-        for i in range(1,4):
-            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_title = 'Fernsehkritik %d.%d' % (episode, i)
-            videos.append({
-                'id':       video_id,
-                'url':      video_url,
-                'ext':      determine_ext(video_url),
-                'title':    video_title,
-                'thumbnail': video_thumbnail
-            })
-        return videos
-
-class FKTVPosteckeIE(InfoExtractor):
-    """Information Extractor for Fernsehkritik-TV Postecke"""
-    _VALID_URL = r'(?:http://)?(?:www\.)?fernsehkritik.tv/inline-video/postecke.php\?(.*&)?ep=(?P<ep>[0-9]+)(&|$)'
+    IE_NAME = 'fernsehkritik.tv'
+    _VALID_URL = r'http://(?:www\.)?fernsehkritik\.tv/folge-(?P<id>[0-9]+)(?:/.*)?'
+
     _TEST = {
-        u'url': u'http://fernsehkritik.tv/inline-video/postecke.php?iframe=true&width=625&height=440&ep=120',
-        u'file': u'0120.flv',
-        u'md5': u'262f0adbac80317412f7e57b4808e5c4',
-        u'info_dict': {
-            u"title": u"Postecke 120"
-        }
+        'url': 'http://fernsehkritik.tv/folge-1',
+        'md5': '21f0b0c99bce7d5b524eb1b17b1c6d79',
+        'info_dict': {
+            'id': '1',
+            'ext': 'mp4',
+            'title': 'Folge 1 vom 10. April 2007',
+            'thumbnail': 're:^https?://.*\.jpg$',
+        },
     }
 
-    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
-        return[{
-            'id':       video_id,
-            'url':      video_url,
-            'ext':      determine_ext(video_url),
-            'title':    video_title,
-        }]
+    def _real_extract(self, url):
+        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': furl,
+            'format_id': determine_ext(furl),
+        } for furl in urls]
+        return {
+            'id': episode,
+            'title': title,
+            'formats': formats,
+            'thumbnail': poster,
+        }