Raphaël G. Git Repositories - youtubedl/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     NO_DEFAULT,
  11     parse_iso8601,
  12     smuggle_url,
  13     xpath_text,
  14 )
  15
  16
  17 class HeiseIE(InfoExtractor):
  18     _VALID_URL = r'https?://(?:www\.)?heise\.de/(?:[^/]+/)+[^/]+-(?P<id>[0-9]+)\.html'
  19     _TESTS = [{
  20         # kaltura embed
  21         'url': 'http://www.heise.de/video/artikel/Podcast-c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2404147.html',
  22         'info_dict': {
  23             'id': '1_kkrq94sm',
  24             'ext': 'mp4',
  25             'title': "Podcast: c't uplink 3.3 – Owncloud / Tastaturen / Peilsender Smartphone",
  26             'timestamp': 1512734959,
  27             'upload_date': '20171208',
  28             'description': 'md5:c934cbfb326c669c2bcabcbe3d3fcd20',
  29         },
  30         'params': {
  31             'skip_download': True,
  32         },
  33     }, {
  34         # YouTube embed
  35         'url': 'http://www.heise.de/newsticker/meldung/Netflix-In-20-Jahren-vom-Videoverleih-zum-TV-Revolutionaer-3814130.html',
  36         'md5': 'e403d2b43fea8e405e88e3f8623909f1',
  37         'info_dict': {
  38             'id': '6kmWbXleKW4',
  39             'ext': 'mp4',
  40             'title': 'NEU IM SEPTEMBER | Netflix',
  41             'description': 'md5:2131f3c7525e540d5fd841de938bd452',
  42             'upload_date': '20170830',
  43             'uploader': 'Netflix Deutschland, Österreich und Schweiz',
  44             'uploader_id': 'netflixdach',
  45         },
  46         'params': {
  47             'skip_download': True,
  48         },
  49     }, {
  50         'url': 'https://www.heise.de/video/artikel/nachgehakt-Wie-sichert-das-c-t-Tool-Restric-tor-Windows-10-ab-3700244.html',
  51         'info_dict': {
  52             'id': '1_ntrmio2s',
  53             'ext': 'mp4',
  54             'title': "nachgehakt: Wie sichert das c't-Tool Restric'tor Windows 10 ab?",
  55             'description': 'md5:47e8ffb6c46d85c92c310a512d6db271',
  56             'timestamp': 1512470717,
  57             'upload_date': '20171205',
  58         },
  59         'params': {
  60             'skip_download': True,
  61         },
  62     }, {
  63         'url': 'https://www.heise.de/ct/artikel/c-t-uplink-20-8-Staubsaugerroboter-Xiaomi-Vacuum-2-AR-Brille-Meta-2-und-Android-rooten-3959893.html',
  64         'info_dict': {
  65             'id': '1_59mk80sf',
  66             'ext': 'mp4',
  67             'title': "c't uplink 20.8: Staubsaugerroboter Xiaomi Vacuum 2, AR-Brille Meta 2 und Android rooten",
  68             'description': 'md5:f50fe044d3371ec73a8f79fcebd74afc',
  69             'timestamp': 1517567237,
  70             'upload_date': '20180202',
  71         },
  72         'params': {
  73             'skip_download': True,
  74         },
  75     }, {
  76         'url': 'http://www.heise.de/ct/artikel/c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2403911.html',
  77         'only_matching': True,
  78     }, {
  79         'url': 'http://www.heise.de/newsticker/meldung/c-t-uplink-Owncloud-Tastaturen-Peilsender-Smartphone-2404251.html?wt_mc=rss.ho.beitrag.atom',
  80         'only_matching': True,
  81     }, {
  82         'url': 'http://www.heise.de/ct/ausgabe/2016-12-Spiele-3214137.html',
  83         'only_matching': True,
  84     }]
  85
  86     def _real_extract(self, url):
  87         video_id = self._match_id(url)
  88         webpage = self._download_webpage(url, video_id)
  89
  90         def extract_title(default=NO_DEFAULT):
  91             title = self._html_search_meta(
  92                 ('fulltitle', 'title'), webpage, default=None)
  93             if not title or title == "c't":
  94                 title = self._search_regex(
  95                     r'<div[^>]+class="videoplayerjw"[^>]+data-title="([^"]+)"',
  96                     webpage, 'title', default=None)
  97             if not title:
  98                 title = self._html_search_regex(
  99                     r'<h1[^>]+\bclass=["\']article_page_title[^>]+>(.+?)<',
 100                     webpage, 'title', default=default)
 101             return title
 102
 103         title = extract_title(default=None)
 104         description = self._og_search_description(
 105             webpage, default=None) or self._html_search_meta(
 106             'description', webpage)
 107
 108         kaltura_url = KalturaIE._extract_url(webpage)
 109         if kaltura_url:
 110             return {
 111                 '_type': 'url_transparent',
 112                 'url': smuggle_url(kaltura_url, {'source_url': url}),
 113                 'ie_key': KalturaIE.ie_key(),
 114                 'title': title,
 115                 'description': description,
 116             }
 117
 118         yt_urls = YoutubeIE._extract_urls(webpage)
 119         if yt_urls:
 120             return self.playlist_from_matches(
 121                 yt_urls, video_id, title, ie=YoutubeIE.ie_key())
 122
 123         title = extract_title()
 124
 125         container_id = self._search_regex(
 126             r'<div class="videoplayerjw"[^>]+data-container="([0-9]+)"',
 127             webpage, 'container ID')
 128
 129         sequenz_id = self._search_regex(
 130             r'<div class="videoplayerjw"[^>]+data-sequenz="([0-9]+)"',
 131             webpage, 'sequenz ID')
 132
 133         doc = self._download_xml(
 134             'http://www.heise.de/videout/feed', video_id, query={
 135                 'container': container_id,
 136                 'sequenz': sequenz_id,
 137             })
 138
 139         formats = []
 140         for source_node in doc.findall('.//{http://rss.jwpcdn.com/}source'):
 141             label = source_node.attrib['label']
 142             height = int_or_none(self._search_regex(
 143                 r'^(.*?_)?([0-9]+)p$', label, 'height', default=None))
 144             video_url = source_node.attrib['file']
 145             ext = determine_ext(video_url, '')
 146             formats.append({
 147                 'url': video_url,
 148                 'format_note': label,
 149                 'format_id': '%s_%s' % (ext, label),
 150                 'height': height,
 151             })
 152         self._sort_formats(formats)
 153
 154         return {
 155             'id': video_id,
 156             'title': title,
 157             'description': description,
 158             'thumbnail': (xpath_text(doc, './/{http://rss.jwpcdn.com/}image') or
 159                           self._og_search_thumbnail(webpage)),
 160             'timestamp': parse_iso8601(
 161                 self._html_search_meta('date', webpage)),
 162             'formats': formats,
 163         }