]> git.bitcoin.ninja Git - youtube-dl/blob - youtube_dl/extractor/heise.py
[pornhub] Extract categories and tags (closes #10499)
[youtube-dl] / youtube_dl / extractor / heise.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 from .common import InfoExtractor
5 from ..utils import (
6     determine_ext,
7     int_or_none,
8     parse_iso8601,
9 )
10
11
12 class HeiseIE(InfoExtractor):
13     _VALID_URL = r'''(?x)
14         https?://(?:www\.)?heise\.de/video/artikel/
15         .+?(?P<id>[0-9]+)\.html(?:$|[?#])
16     '''
17     _TEST = {
18         'url': (
19             'http://www.heise.de/video/artikel/Podcast-c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2404147.html'
20         ),
21         'md5': 'ffed432483e922e88545ad9f2f15d30e',
22         'info_dict': {
23             'id': '2404147',
24             'ext': 'mp4',
25             'title': (
26                 "Podcast: c't uplink 3.3 – Owncloud / Tastaturen / Peilsender Smartphone"
27             ),
28             'format_id': 'mp4_720p',
29             'timestamp': 1411812600,
30             'upload_date': '20140927',
31             'description': 'In uplink-Episode 3.3 geht es darum, wie man sich von Cloud-Anbietern emanzipieren kann, worauf man beim Kauf einer Tastatur achten sollte und was Smartphones über uns verraten.',
32             'thumbnail': 're:^https?://.*\.jpe?g$',
33         }
34     }
35
36     def _real_extract(self, url):
37         video_id = self._match_id(url)
38         webpage = self._download_webpage(url, video_id)
39
40         container_id = self._search_regex(
41             r'<div class="videoplayerjw".*?data-container="([0-9]+)"',
42             webpage, 'container ID')
43         sequenz_id = self._search_regex(
44             r'<div class="videoplayerjw".*?data-sequenz="([0-9]+)"',
45             webpage, 'sequenz ID')
46         data_url = 'http://www.heise.de/videout/feed?container=%s&sequenz=%s' % (container_id, sequenz_id)
47         doc = self._download_xml(data_url, video_id)
48
49         info = {
50             'id': video_id,
51             'thumbnail': self._og_search_thumbnail(webpage),
52             'timestamp': parse_iso8601(
53                 self._html_search_meta('date', webpage)),
54             'description': self._og_search_description(webpage),
55         }
56
57         title = self._html_search_meta('fulltitle', webpage)
58         if title:
59             info['title'] = title
60         else:
61             info['title'] = self._og_search_title(webpage)
62
63         formats = []
64         for source_node in doc.findall('.//{http://rss.jwpcdn.com/}source'):
65             label = source_node.attrib['label']
66             height = int_or_none(self._search_regex(
67                 r'^(.*?_)?([0-9]+)p$', label, 'height', default=None))
68             video_url = source_node.attrib['file']
69             ext = determine_ext(video_url, '')
70             formats.append({
71                 'url': video_url,
72                 'format_note': label,
73                 'format_id': '%s_%s' % (ext, label),
74                 'height': height,
75             })
76         self._sort_formats(formats)
77         info['formats'] = formats
78
79         return info