yt_dlp/extractor/xstream.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     find_xpath_attr,
   6     int_or_none,
   7     parse_iso8601,
   8     xpath_text,
   9     xpath_with_ns,
  10 )
  11
  12
  13 class XstreamIE(InfoExtractor):
  14     _VALID_URL = r'''(?x)
  15                     (?:
  16                         xstream:|
  17                         https?://frontend\.xstream\.(?:dk|net)/
  18                     )
  19                     (?P<partner_id>[^/]+)
  20                     (?:
  21                         :|
  22                         /feed/video/\?.*?\bid=
  23                     )
  24                     (?P<id>\d+)
  25                     '''
  26     _TESTS = [{
  27         'url': 'http://frontend.xstream.dk/btno/feed/video/?platform=web&id=86588',
  28         'md5': 'd7d17e3337dc80de6d3a540aefbe441b',
  29         'info_dict': {
  30             'id': '86588',
  31             'ext': 'mov',
  32             'title': 'Otto Wollertsen',
  33             'description': 'Vestlendingen Otto Fredrik Wollertsen',
  34             'timestamp': 1430473209,
  35             'upload_date': '20150501',
  36         },
  37     }, {
  38         'url': 'http://frontend.xstream.dk/ap/feed/video/?platform=web&id=21039',
  39         'only_matching': True,
  40     }]
  41
  42     def _extract_video_info(self, partner_id, video_id):
  43         data = self._download_xml(
  44             f'http://frontend.xstream.dk/{partner_id}/feed/video/?platform=web&id={video_id}',
  45             video_id)
  46
  47         NS_MAP = {
  48             'atom': 'http://www.w3.org/2005/Atom',
  49             'xt': 'http://xstream.dk/',
  50             'media': 'http://search.yahoo.com/mrss/',
  51         }
  52
  53         entry = data.find(xpath_with_ns('./atom:entry', NS_MAP))
  54
  55         title = xpath_text(
  56             entry, xpath_with_ns('./atom:title', NS_MAP), 'title')
  57         description = xpath_text(
  58             entry, xpath_with_ns('./atom:summary', NS_MAP), 'description')
  59         timestamp = parse_iso8601(xpath_text(
  60             entry, xpath_with_ns('./atom:published', NS_MAP), 'upload date'))
  61
  62         formats = []
  63         media_group = entry.find(xpath_with_ns('./media:group', NS_MAP))
  64         for media_content in media_group.findall(xpath_with_ns('./media:content', NS_MAP)):
  65             media_url = media_content.get('url')
  66             if not media_url:
  67                 continue
  68             tbr = int_or_none(media_content.get('bitrate'))
  69             mobj = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', media_url)
  70             if mobj:
  71                 formats.append({
  72                     'url': mobj.group('url'),
  73                     'play_path': 'mp4:{}'.format(mobj.group('playpath')),
  74                     'app': mobj.group('app'),
  75                     'ext': 'flv',
  76                     'tbr': tbr,
  77                     'format_id': 'rtmp-%d' % tbr,
  78                 })
  79             else:
  80                 formats.append({
  81                     'url': media_url,
  82                     'tbr': tbr,
  83                 })
  84
  85         link = find_xpath_attr(
  86             entry, xpath_with_ns('./atom:link', NS_MAP), 'rel', 'original')
  87         if link is not None:
  88             formats.append({
  89                 'url': link.get('href'),
  90                 'format_id': link.get('rel'),
  91                 'quality': 1,
  92             })
  93
  94         thumbnails = [{
  95             'url': splash.get('url'),
  96             'width': int_or_none(splash.get('width')),
  97             'height': int_or_none(splash.get('height')),
  98         } for splash in media_group.findall(xpath_with_ns('./xt:splash', NS_MAP))]
  99
 100         return {
 101             'id': video_id,
 102             'title': title,
 103             'description': description,
 104             'timestamp': timestamp,
 105             'formats': formats,
 106             'thumbnails': thumbnails,
 107         }
 108
 109     def _real_extract(self, url):
 110         mobj = self._match_valid_url(url)
 111         partner_id = mobj.group('partner_id')
 112         video_id = mobj.group('id')
 113
 114         return self._extract_video_info(partner_id, video_id)