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