yt_dlp/extractor/heise.py

   1 import urllib.parse
   2
   3 from .common import InfoExtractor
   4 from .kaltura import KalturaIE
   5 from .youtube import YoutubeIE
   6 from ..utils import (
   7     NO_DEFAULT,
   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         # kaltura embed
  20         'url': 'http://www.heise.de/video/artikel/Podcast-c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2404147.html',
  21         'info_dict': {
  22             'id': '1_kkrq94sm',
  23             'ext': 'mp4',
  24             'title': "Podcast: c't uplink 3.3 – Owncloud / Tastaturen / Peilsender Smartphone",
  25             'timestamp': 1512734959,
  26             'upload_date': '20171208',
  27             'description': 'md5:c934cbfb326c669c2bcabcbe3d3fcd20',
  28             'thumbnail': 're:^https?://.*/thumbnail/.*',
  29             'duration': 2845,
  30             'view_count': int,
  31         },
  32         'params': {
  33             'skip_download': True,
  34         },
  35     }, {
  36         # YouTube embed
  37         'url': 'http://www.heise.de/newsticker/meldung/Netflix-In-20-Jahren-vom-Videoverleih-zum-TV-Revolutionaer-3814130.html',
  38         'md5': 'e403d2b43fea8e405e88e3f8623909f1',
  39         'info_dict': {
  40             'id': '6kmWbXleKW4',
  41             'ext': 'mp4',
  42             'title': 'Neu im September 2017 | Netflix',
  43             'description': 'md5:d6852d1f96bb80760608eed3b907437c',
  44             'upload_date': '20170830',
  45             'uploader': 'Netflix Deutschland, Österreich und Schweiz',
  46             'uploader_id': 'netflixdach',
  47             'categories': ['Entertainment'],
  48             'tags': 'count:27',
  49             'age_limit': 0,
  50             'availability': 'public',
  51             'comment_count': int,
  52             'channel_id': 'UCZqgRlLcvO3Fnx_npQJygcQ',
  53             'thumbnail': 'https://i.ytimg.com/vi_webp/6kmWbXleKW4/maxresdefault.webp',
  54             'uploader_url': 'http://www.youtube.com/user/netflixdach',
  55             'playable_in_embed': True,
  56             'live_status': 'not_live',
  57             'channel_url': 'https://www.youtube.com/channel/UCZqgRlLcvO3Fnx_npQJygcQ',
  58             'view_count': int,
  59             'channel': 'Netflix Deutschland, Österreich und Schweiz',
  60             'channel_follower_count': int,
  61             'like_count': int,
  62             'duration': 67,
  63         },
  64         'params': {
  65             'skip_download': True,
  66         },
  67     }, {
  68         'url': 'https://www.heise.de/video/artikel/nachgehakt-Wie-sichert-das-c-t-Tool-Restric-tor-Windows-10-ab-3700244.html',
  69         'info_dict': {
  70             'id': '1_ntrmio2s',
  71             'ext': 'mp4',
  72             'title': "nachgehakt: Wie sichert das c't-Tool Restric'tor Windows 10 ab?",
  73             'description': 'md5:47e8ffb6c46d85c92c310a512d6db271',
  74             'timestamp': 1512470717,
  75             'upload_date': '20171205',
  76             'duration': 786,
  77             'view_count': int,
  78             'thumbnail': 're:^https?://.*/thumbnail/.*',
  79         },
  80         'params': {
  81             'skip_download': True,
  82         },
  83     }, {
  84         # FIXME: Video m3u8 fails to download; issue with Kaltura extractor
  85         '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',
  86         'info_dict': {
  87             'id': '1_59mk80sf',
  88             'ext': 'mp4',
  89             'title': "c't uplink 20.8: Staubsaugerroboter Xiaomi Vacuum 2, AR-Brille Meta 2 und Android rooten",
  90             'description': 'md5:f50fe044d3371ec73a8f79fcebd74afc',
  91             'timestamp': 1517567237,
  92             'upload_date': '20180202',
  93         },
  94         'params': {
  95             'skip_download': True,
  96         },
  97     }, {
  98         # videout
  99         'url': 'https://www.heise.de/ct/artikel/c-t-uplink-3-8-Anonyme-SIM-Karten-G-Sync-Monitore-Citizenfour-2440327.html',
 100         'info_dict': {
 101             'id': '2440327',
 102             'ext': 'mp4',
 103             'title': 'c\'t uplink 3.8: Anonyme SIM-Karten, G-Sync-Monitore, Citizenfour',
 104             'thumbnail': 'http://www.heise.de/imagine/yxM2qmol0xV3iFB7qFb70dGvXjc/gallery/',
 105             'description': 'md5:fa164d8c8707dff124a9626d39205f5d',
 106             'timestamp': 1414825200,
 107             'upload_date': '20141101',
 108         },
 109     }, {
 110         'url': 'http://www.heise.de/ct/artikel/c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2403911.html',
 111         'only_matching': True,
 112     }, {
 113         'url': 'http://www.heise.de/newsticker/meldung/c-t-uplink-Owncloud-Tastaturen-Peilsender-Smartphone-2404251.html?wt_mc=rss.ho.beitrag.atom',
 114         'only_matching': True,
 115     }, {
 116         'url': 'http://www.heise.de/ct/ausgabe/2016-12-Spiele-3214137.html',
 117         'only_matching': True,
 118     }]
 119
 120     def _real_extract(self, url):
 121         video_id = self._match_id(url)
 122         webpage = self._download_webpage(url, video_id)
 123
 124         def extract_title(default=NO_DEFAULT):
 125             title = self._html_search_meta(
 126                 ('fulltitle', 'title'), webpage, default=None)
 127             if not title or title == "c't":
 128                 title = self._search_regex(
 129                     r'<div[^>]+class="videoplayerjw"[^>]+data-title="([^"]+)"',
 130                     webpage, 'title', default=None)
 131             if not title:
 132                 title = self._html_search_regex(
 133                     r'<h1[^>]+\bclass=["\']article_page_title[^>]+>(.+?)<',
 134                     webpage, 'title', default=default)
 135             return title
 136
 137         title = extract_title(default=None)
 138         description = self._og_search_description(
 139             webpage, default=None) or self._html_search_meta(
 140             'description', webpage)
 141
 142         def _make_kaltura_result(kaltura_url):
 143             return {
 144                 '_type': 'url_transparent',
 145                 'url': smuggle_url(kaltura_url, {'source_url': url}),
 146                 'ie_key': KalturaIE.ie_key(),
 147                 'title': title,
 148                 'description': description,
 149             }
 150
 151         kaltura_url = KalturaIE._extract_url(webpage)
 152         if kaltura_url:
 153             return _make_kaltura_result(kaltura_url)
 154
 155         kaltura_id = self._search_regex(
 156             r'entry-id=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'kaltura id',
 157             default=None, group='id')
 158         if kaltura_id:
 159             return _make_kaltura_result(f'kaltura:2238431:{kaltura_id}')
 160
 161         yt_urls = tuple(YoutubeIE._extract_embed_urls(url, webpage))
 162         if yt_urls:
 163             return self.playlist_from_matches(
 164                 yt_urls, video_id, title, ie=YoutubeIE.ie_key())
 165
 166         title = extract_title()
 167         api_params = urllib.parse.parse_qs(
 168             self._search_regex(r'/videout/feed\.json\?([^\']+)', webpage, 'feed params', default=None) or '')
 169         if not api_params or 'container' not in api_params or 'sequenz' not in api_params:
 170             container_id = self._search_regex(
 171                 r'<div class="videoplayerjw"[^>]+data-container="([0-9]+)"',
 172                 webpage, 'container ID')
 173
 174             sequenz_id = self._search_regex(
 175                 r'<div class="videoplayerjw"[^>]+data-sequenz="([0-9]+)"',
 176                 webpage, 'sequenz ID')
 177             api_params = {
 178                 'container': container_id,
 179                 'sequenz': sequenz_id,
 180             }
 181         doc = self._download_xml(
 182             'http://www.heise.de/videout/feed', video_id, query=api_params)
 183
 184         formats = []
 185         for source_node in doc.findall('.//{http://rss.jwpcdn.com/}source'):
 186             label = source_node.attrib['label']
 187             height = int_or_none(self._search_regex(
 188                 r'^(.*?_)?([0-9]+)p$', label, 'height', default=None))
 189             video_url = source_node.attrib['file']
 190             ext = determine_ext(video_url, '')
 191             formats.append({
 192                 'url': video_url,
 193                 'format_note': label,
 194                 'format_id': f'{ext}_{label}',
 195                 'height': height,
 196             })
 197
 198         return {
 199             'id': video_id,
 200             'title': title,
 201             'description': description,
 202             'thumbnail': (xpath_text(doc, './/{http://rss.jwpcdn.com/}image')
 203                           or self._og_search_thumbnail(webpage)),
 204             'timestamp': parse_iso8601(
 205                 self._html_search_meta('date', webpage)),
 206             'formats': formats,
 207         }