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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     compat_urllib_parse,
  10     get_meta_content,
  11     parse_iso8601,
  12 )
  13
  14
  15 class HeiseIE(InfoExtractor):
  16     _VALID_URL = (
  17         r'^https?://(?:www\.)?heise\.de/video/artikel/' +
  18         r'.+?(?P<id>[0-9]+)\.html$'
  19     )
  20     _TEST = {
  21         'url': (
  22             'http://www.heise.de/video/artikel/Podcast-c-t-uplink-3-3-' +
  23             'Owncloud-Tastaturen-Peilsender-Smartphone-2404147.html'
  24         ),
  25         'md5': 'ffed432483e922e88545ad9f2f15d30e',
  26         'info_dict': {
  27             'id': '2404147',
  28             'ext': 'mp4',
  29             'title': (
  30                 "Podcast: c't uplink 3.3 – Owncloud / Tastaturen / " +
  31                 "Peilsender Smartphone"
  32             ),
  33             'format_id': 'mp4_720',
  34             'timestamp': 1411812600,
  35             'upload_date': '20140927',
  36         }
  37     }
  38
  39     _CONFIG = (
  40         r'".+?\?sequenz=(?P<sequenz>.+?)&container=(?P<container>.+?)' +
  41         r'(?:&hd=(?P<hd>.+?))?(?:&signature=(?P<signature>.+?))?&callback=\?"'
  42     )
  43     _PREFIX = 'http://www.heise.de/videout/info?'
  44
  45     def _warn(self, fmt, *args):
  46         self.report_warning(fmt.format(*args), self._id)
  47
  48     def _parse_config_url(self, html):
  49         m = re.search(self._CONFIG, html)
  50         if not m:
  51             raise ExtractorError('No config found')
  52
  53         qs = compat_urllib_parse.urlencode(dict((k, v) for k, v
  54                                                 in m.groupdict().items()
  55                                                 if v is not None))
  56         return self._PREFIX + qs
  57
  58     def _real_extract(self, url):
  59         mobj = re.match(self._VALID_URL, url)
  60         self._id = mobj.group('id')
  61
  62         html = self._download_webpage(url, self._id)
  63         config = self._download_json(self._parse_config_url(html), self._id)
  64
  65         info = {
  66             'id': self._id
  67         }
  68
  69         title = get_meta_content('fulltitle', html)
  70         if title:
  71             info['title'] = title
  72         elif config.get('title'):
  73             info['title'] = config['title']
  74         else:
  75             self._warn('title: not found')
  76             info['title'] = 'heise'
  77
  78         if (not config.get('formats') or
  79                 not hasattr(config['formats'], 'items')):
  80             raise ExtractorError('No formats found')
  81
  82         formats = []
  83         for t, rs in config['formats'].items():
  84             if not rs or not hasattr(rs, 'items'):
  85                 self._warn('formats: {0}: no resolutions', t)
  86                 continue
  87
  88             for res, obj in rs.items():
  89                 format_id = '{0}_{1}'.format(t, res)
  90
  91                 if (not obj or not obj.get('url') or
  92                         not isinstance(obj['url'], str)):
  93                     self._warn('formats: {0}: no url', format_id)
  94                     continue
  95
  96                 fmt = {
  97                     'url': obj['url'],
  98                     'format_id': format_id
  99                 }
 100                 try:
 101                     fmt['height'] = int(res)
 102                 except ValueError as e:
 103                     self._warn('formats: {0}: height: {1}', t, e)
 104
 105                 formats.append(fmt)
 106
 107         self._sort_formats(formats)
 108         info['formats'] = formats
 109
 110         if config.get('poster') and isinstance(config['poster'], str):
 111             info['thumbnail'] = config['poster']
 112
 113         date = get_meta_content('date', html)
 114         if date and isinstance(date, str):
 115             try:
 116                 info['timestamp'] = parse_iso8601(date)
 117             except ValueError as e:
 118                 self._warn('timestamp: {0}', e)
 119
 120         return info