_ 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 ..utils import (
   6     determine_ext,
   7     int_or_none,
   8     parse_iso8601,
   9     xpath_text,
  10 )
  11
  12
  13 class HeiseIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?heise\.de/(?:[^/]+/)+[^/]+-(?P<id>[0-9]+)\.html'
  15     _TESTS = [{
  16         'url': 'http://www.heise.de/video/artikel/Podcast-c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2404147.html',
  17         'md5': 'ffed432483e922e88545ad9f2f15d30e',
  18         'info_dict': {
  19             'id': '2404147',
  20             'ext': 'mp4',
  21             'title': "Podcast: c't uplink 3.3 – Owncloud / Tastaturen / Peilsender Smartphone",
  22             'format_id': 'mp4_720p',
  23             'timestamp': 1411812600,
  24             'upload_date': '20140927',
  25             'description': 'md5:c934cbfb326c669c2bcabcbe3d3fcd20',
  26             'thumbnail': r're:^https?://.*/gallery/$',
  27         }
  28     }, {
  29         'url': 'http://www.heise.de/ct/artikel/c-t-uplink-3-3-Owncloud-Tastaturen-Peilsender-Smartphone-2403911.html',
  30         'only_matching': True,
  31     }, {
  32         'url': 'http://www.heise.de/newsticker/meldung/c-t-uplink-Owncloud-Tastaturen-Peilsender-Smartphone-2404251.html?wt_mc=rss.ho.beitrag.atom',
  33         'only_matching': True,
  34     }, {
  35         'url': 'http://www.heise.de/ct/ausgabe/2016-12-Spiele-3214137.html',
  36         'only_matching': True,
  37     }]
  38
  39     def _real_extract(self, url):
  40         video_id = self._match_id(url)
  41         webpage = self._download_webpage(url, video_id)
  42
  43         container_id = self._search_regex(
  44             r'<div class="videoplayerjw"[^>]+data-container="([0-9]+)"',
  45             webpage, 'container ID')
  46         sequenz_id = self._search_regex(
  47             r'<div class="videoplayerjw"[^>]+data-sequenz="([0-9]+)"',
  48             webpage, 'sequenz ID')
  49
  50         title = self._html_search_meta('fulltitle', webpage, default=None)
  51         if not title or title == "c't":
  52             title = self._search_regex(
  53                 r'<div[^>]+class="videoplayerjw"[^>]+data-title="([^"]+)"',
  54                 webpage, 'title')
  55
  56         doc = self._download_xml(
  57             'http://www.heise.de/videout/feed', video_id, query={
  58                 'container': container_id,
  59                 'sequenz': sequenz_id,
  60             })
  61
  62         formats = []
  63         for source_node in doc.findall('.//{http://rss.jwpcdn.com/}source'):
  64             label = source_node.attrib['label']
  65             height = int_or_none(self._search_regex(
  66                 r'^(.*?_)?([0-9]+)p$', label, 'height', default=None))
  67             video_url = source_node.attrib['file']
  68             ext = determine_ext(video_url, '')
  69             formats.append({
  70                 'url': video_url,
  71                 'format_note': label,
  72                 'format_id': '%s_%s' % (ext, label),
  73                 'height': height,
  74             })
  75         self._sort_formats(formats)
  76
  77         description = self._og_search_description(
  78             webpage, default=None) or self._html_search_meta(
  79             'description', webpage)
  80
  81         return {
  82             'id': video_id,
  83             'title': title,
  84             'description': description,
  85             'thumbnail': (xpath_text(doc, './/{http://rss.jwpcdn.com/}image') or
  86                           self._og_search_thumbnail(webpage)),
  87             'timestamp': parse_iso8601(
  88                 self._html_search_meta('date', webpage)),
  89             'formats': formats,
  90         }