_ Git - youtube-dl/blob - youtube_dl/extractor/heise.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from .kaltura import KalturaIE
   6 from .youtube import YoutubeIE
   7 from ..utils import (
   8     determine_ext,
   9     int_or_none,
  10     parse_iso8601,
  11     smuggle_url,
  12     xpath_text,
  13 )
  14
  15
  16 class HeiseIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?heise\.de/(?:[^/]+/)+[^/]+-(?P<id>[0-9]+)\.html'
  18     _TESTS = [{
  19         'url': 'http://www.heise.de/video/artikel/Podcast-c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2404147.html',
  20         'md5': 'ffed432483e922e88545ad9f2f15d30e',
  21         'info_dict': {
  22             'id': '2404147',
  23             'ext': 'mp4',
  24             'title': "Podcast: c't uplink 3.3 – Owncloud / Tastaturen / Peilsender Smartphone",
  25             'format_id': 'mp4_720p',
  26             'timestamp': 1411812600,
  27             'upload_date': '20140927',
  28             'description': 'md5:c934cbfb326c669c2bcabcbe3d3fcd20',
  29             'thumbnail': r're:^https?://.*/gallery/$',
  30         }
  31     }, {
  32         # YouTube embed
  33         'url': 'http://www.heise.de/newsticker/meldung/Netflix-In-20-Jahren-vom-Videoverleih-zum-TV-Revolutionaer-3814130.html',
  34         'md5': 'e403d2b43fea8e405e88e3f8623909f1',
  35         'info_dict': {
  36             'id': '6kmWbXleKW4',
  37             'ext': 'mp4',
  38             'title': 'NEU IM SEPTEMBER | Netflix',
  39             'description': 'md5:2131f3c7525e540d5fd841de938bd452',
  40             'upload_date': '20170830',
  41             'uploader': 'Netflix Deutschland, Österreich und Schweiz',
  42             'uploader_id': 'netflixdach',
  43         },
  44         'params': {
  45             'skip_download': True,
  46         },
  47     }, {
  48         'url': 'https://www.heise.de/video/artikel/nachgehakt-Wie-sichert-das-c-t-Tool-Restric-tor-Windows-10-ab-3700244.html',
  49         'md5': '4b58058b46625bdbd841fc2804df95fc',
  50         'info_dict': {
  51             'id': '1_ntrmio2s',
  52             'timestamp': 1512470717,
  53             'upload_date': '20171205',
  54             'ext': 'mp4',
  55             'title': 'ct10 nachgehakt hos restrictor',
  56         },
  57         'params': {
  58             'skip_download': True,
  59         },
  60     }, {
  61         'url': 'http://www.heise.de/ct/artikel/c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2403911.html',
  62         'only_matching': True,
  63     }, {
  64         'url': 'http://www.heise.de/newsticker/meldung/c-t-uplink-Owncloud-Tastaturen-Peilsender-Smartphone-2404251.html?wt_mc=rss.ho.beitrag.atom',
  65         'only_matching': True,
  66     }, {
  67         'url': 'http://www.heise.de/ct/ausgabe/2016-12-Spiele-3214137.html',
  68         'only_matching': True,
  69     }]
  70
  71     def _real_extract(self, url):
  72         video_id = self._match_id(url)
  73         webpage = self._download_webpage(url, video_id)
  74
  75         title = self._html_search_meta('fulltitle', webpage, default=None)
  76         if not title or title == "c't":
  77             title = self._search_regex(
  78                 r'<div[^>]+class="videoplayerjw"[^>]+data-title="([^"]+)"',
  79                 webpage, 'title')
  80
  81         yt_urls = YoutubeIE._extract_urls(webpage)
  82         if yt_urls:
  83             return self.playlist_from_matches(yt_urls, video_id, title, ie=YoutubeIE.ie_key())
  84
  85         kaltura_url = KalturaIE._extract_url(webpage)
  86         if kaltura_url:
  87             return self.url_result(smuggle_url(kaltura_url, {'source_url': url}), KalturaIE.ie_key())
  88
  89         container_id = self._search_regex(
  90             r'<div class="videoplayerjw"[^>]+data-container="([0-9]+)"',
  91             webpage, 'container ID')
  92
  93         sequenz_id = self._search_regex(
  94             r'<div class="videoplayerjw"[^>]+data-sequenz="([0-9]+)"',
  95             webpage, 'sequenz ID')
  96
  97         doc = self._download_xml(
  98             'http://www.heise.de/videout/feed', video_id, query={
  99                 'container': container_id,
 100                 'sequenz': sequenz_id,
 101             })
 102
 103         formats = []
 104         for source_node in doc.findall('.//{http://rss.jwpcdn.com/}source'):
 105             label = source_node.attrib['label']
 106             height = int_or_none(self._search_regex(
 107                 r'^(.*?_)?([0-9]+)p$', label, 'height', default=None))
 108             video_url = source_node.attrib['file']
 109             ext = determine_ext(video_url, '')
 110             formats.append({
 111                 'url': video_url,
 112                 'format_note': label,
 113                 'format_id': '%s_%s' % (ext, label),
 114                 'height': height,
 115             })
 116         self._sort_formats(formats)
 117
 118         description = self._og_search_description(
 119             webpage, default=None) or self._html_search_meta(
 120             'description', webpage)
 121
 122         return {
 123             'id': video_id,
 124             'title': title,
 125             'description': description,
 126             'thumbnail': (xpath_text(doc, './/{http://rss.jwpcdn.com/}image') or
 127                           self._og_search_thumbnail(webpage)),
 128             'timestamp': parse_iso8601(
 129                 self._html_search_meta('date', webpage)),
 130             'formats': formats,
 131         }