youtube_dl/extractor/xstream.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     int_or_none,
   9     parse_iso8601,
  10     xpath_with_ns,
  11     xpath_text,
  12     find_xpath_attr,
  13 )
  14
  15
  16 class XstreamIE(InfoExtractor):
  17     _VALID_URL = r'''(?x)
  18                     (?:
  19                         xstream:|
  20                         https?://frontend\.xstream\.(?:dk|net)/
  21                     )
  22                     (?P<partner_id>[^/]+)
  23                     (?:
  24                         :|
  25                         /feed/video/\?.*?\bid=
  26                     )
  27                     (?P<id>\d+)
  28                     '''
  29     _TESTS = [{
  30         'url': 'http://frontend.xstream.dk/btno/feed/video/?platform=web&id=86588',
  31         'md5': 'd7d17e3337dc80de6d3a540aefbe441b',
  32         'info_dict': {
  33             'id': '86588',
  34             'ext': 'mov',
  35             'title': 'Otto Wollertsen',
  36             'description': 'Vestlendingen Otto Fredrik Wollertsen',
  37             'timestamp': 1430473209,
  38             'upload_date': '20150501',
  39         },
  40     }, {
  41         'url': 'http://frontend.xstream.dk/ap/feed/video/?platform=web&id=21039',
  42         'only_matching': True,
  43     }]
  44
  45     def _extract_video_info(self, partner_id, video_id):
  46         data = self._download_xml(
  47             'http://frontend.xstream.dk/%s/feed/video/?platform=web&id=%s'
  48             % (partner_id, video_id),
  49             video_id)
  50
  51         NS_MAP = {
  52             'atom': 'http://www.w3.org/2005/Atom',
  53             'xt': 'http://xstream.dk/',
  54             'media': 'http://search.yahoo.com/mrss/',
  55         }
  56
  57         entry = data.find(xpath_with_ns('./atom:entry', NS_MAP))
  58
  59         title = xpath_text(
  60             entry, xpath_with_ns('./atom:title', NS_MAP), 'title')
  61         description = xpath_text(
  62             entry, xpath_with_ns('./atom:summary', NS_MAP), 'description')
  63         timestamp = parse_iso8601(xpath_text(
  64             entry, xpath_with_ns('./atom:published', NS_MAP), 'upload date'))
  65
  66         formats = []
  67         media_group = entry.find(xpath_with_ns('./media:group', NS_MAP))
  68         for media_content in media_group.findall(xpath_with_ns('./media:content', NS_MAP)):
  69             media_url = media_content.get('url')
  70             if not media_url:
  71                 continue
  72             tbr = int_or_none(media_content.get('bitrate'))
  73             mobj = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', media_url)
  74             if mobj:
  75                 formats.append({
  76                     'url': mobj.group('url'),
  77                     'play_path': 'mp4:%s' % mobj.group('playpath'),
  78                     'app': mobj.group('app'),
  79                     'ext': 'flv',
  80                     'tbr': tbr,
  81                     'format_id': 'rtmp-%d' % tbr,
  82                 })
  83             else:
  84                 formats.append({
  85                     'url': media_url,
  86                     'tbr': tbr,
  87                 })
  88         self._sort_formats(formats)
  89
  90         link = find_xpath_attr(
  91             entry, xpath_with_ns('./atom:link', NS_MAP), 'rel', 'original')
  92         if link is not None:
  93             formats.append({
  94                 'url': link.get('href'),
  95                 'format_id': link.get('rel'),
  96                 'preference': 1,
  97             })
  98
  99         thumbnails = [{
 100             'url': splash.get('url'),
 101             'width': int_or_none(splash.get('width')),
 102             'height': int_or_none(splash.get('height')),
 103         } for splash in media_group.findall(xpath_with_ns('./xt:splash', NS_MAP))]
 104
 105         return {
 106             'id': video_id,
 107             'title': title,
 108             'description': description,
 109             'timestamp': timestamp,
 110             'formats': formats,
 111             'thumbnails': thumbnails,
 112         }
 113
 114     def _real_extract(self, url):
 115         mobj = re.match(self._VALID_URL, url)
 116         partner_id = mobj.group('partner_id')
 117         video_id = mobj.group('id')
 118
 119         return self._extract_video_info(partner_id, video_id)