]> gitweb @ CieloNegro.org - youtube-dl.git/blob - youtube_dl/extractor/fktv.py
[generic] Prevent <video> search from skipping over empty sources (#3546)
[youtube-dl.git] / youtube_dl / extractor / fktv.py
1 import re
2 import random
3 import json
4
5 from .common import InfoExtractor
6 from ..utils import (
7     determine_ext,
8     get_element_by_id,
9     clean_html,
10 )
11
12
13 class FKTVIE(InfoExtractor):
14     IE_NAME = u'fernsehkritik.tv'
15     _VALID_URL = r'(?:http://)?(?:www\.)?fernsehkritik\.tv/folge-(?P<ep>[0-9]+)(?:/.*)?'
16
17     _TEST = {
18         u'url': u'http://fernsehkritik.tv/folge-1',
19         u'file': u'00011.flv',
20         u'info_dict': {
21             u'title': u'Folge 1 vom 10. April 2007',
22             u'description': u'md5:fb4818139c7cfe6907d4b83412a6864f',
23         },
24     }
25
26     def _real_extract(self, url):
27         mobj = re.match(self._VALID_URL, url)
28         episode = int(mobj.group('ep'))
29
30         server = random.randint(2, 4)
31         video_thumbnail = 'http://fernsehkritik.tv/images/magazin/folge%d.jpg' % episode
32         start_webpage = self._download_webpage('http://fernsehkritik.tv/folge-%d/Start' % episode,
33             episode)
34         playlist = self._search_regex(r'playlist = (\[.*?\]);', start_webpage,
35             u'playlist', flags=re.DOTALL)
36         files = json.loads(re.sub('{[^{}]*?}', '{}', playlist))
37         # TODO: return a single multipart video
38         videos = []
39         for i, _ in enumerate(files, 1):
40             video_id = '%04d%d' % (episode, i)
41             video_url = 'http://dl%d.fernsehkritik.tv/fernsehkritik%d%s.flv' % (server, episode, '' if i == 1 else '-%d' % i)
42             videos.append({
43                 'id': video_id,
44                 'url': video_url,
45                 'ext': determine_ext(video_url),
46                 'title': clean_html(get_element_by_id('eptitle', start_webpage)),
47                 'description': clean_html(get_element_by_id('contentlist', start_webpage)),
48                 'thumbnail': video_thumbnail
49             })
50         return videos
51
52
53 class FKTVPosteckeIE(InfoExtractor):
54     IE_NAME = u'fernsehkritik.tv:postecke'
55     _VALID_URL = r'(?:http://)?(?:www\.)?fernsehkritik\.tv/inline-video/postecke\.php\?(.*&)?ep=(?P<ep>[0-9]+)(&|$)'
56     _TEST = {
57         u'url': u'http://fernsehkritik.tv/inline-video/postecke.php?iframe=true&width=625&height=440&ep=120',
58         u'file': u'0120.flv',
59         u'md5': u'262f0adbac80317412f7e57b4808e5c4',
60         u'info_dict': {
61             u"title": u"Postecke 120"
62         }
63     }
64
65     def _real_extract(self, url):
66         mobj = re.match(self._VALID_URL, url)
67         episode = int(mobj.group('ep'))
68
69         server = random.randint(2, 4)
70         video_id = '%04d' % episode
71         video_url = 'http://dl%d.fernsehkritik.tv/postecke/postecke%d.flv' % (server, episode)
72         video_title = 'Postecke %d' % episode
73         return {
74             'id':       video_id,
75             'url':      video_url,
76             'ext':      determine_ext(video_url),
77             'title':    video_title,
78         }