Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/aftenposten.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 AftenpostenIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?aftenposten\.no/webtv/([^/]+/)*(?P<id>[^/]+)-\d+\.html'
  18
  19     _TEST = {
  20         'url': 'http://www.aftenposten.no/webtv/serier-og-programmer/sweatshopenglish/TRAILER-SWEATSHOP---I-cant-take-any-more-7800835.html?paging=&section=webtv_serierogprogrammer_sweatshop_sweatshopenglish',
  21         'md5': 'fd828cd29774a729bf4d4425fe192972',
  22         'info_dict': {
  23             'id': '21039',
  24             'ext': 'mov',
  25             'title': 'TRAILER: "Sweatshop" - I can´t take any more',
  26             'description': 'md5:21891f2b0dd7ec2f78d84a50e54f8238',
  27             'timestamp': 1416927969,
  28             'upload_date': '20141125',
  29         }
  30     }
  31
  32     def _real_extract(self, url):
  33         display_id = self._match_id(url)
  34
  35         webpage = self._download_webpage(url, display_id)
  36
  37         video_id = self._html_search_regex(
  38             r'data-xs-id="(\d+)"', webpage, 'video id')
  39
  40         data = self._download_xml(
  41             'http://frontend.xstream.dk/ap/feed/video/?platform=web&id=%s' % video_id, video_id)
  42
  43         NS_MAP = {
  44             'atom': 'http://www.w3.org/2005/Atom',
  45             'xt': 'http://xstream.dk/',
  46             'media': 'http://search.yahoo.com/mrss/',
  47         }
  48
  49         entry = data.find(xpath_with_ns('./atom:entry', NS_MAP))
  50
  51         title = xpath_text(
  52             entry, xpath_with_ns('./atom:title', NS_MAP), 'title')
  53         description = xpath_text(
  54             entry, xpath_with_ns('./atom:summary', NS_MAP), 'description')
  55         timestamp = parse_iso8601(xpath_text(
  56             entry, xpath_with_ns('./atom:published', NS_MAP), 'upload date'))
  57
  58         formats = []
  59         media_group = entry.find(xpath_with_ns('./media:group', NS_MAP))
  60         for media_content in media_group.findall(xpath_with_ns('./media:content', NS_MAP)):
  61             media_url = media_content.get('url')
  62             if not media_url:
  63                 continue
  64             tbr = int_or_none(media_content.get('bitrate'))
  65             mobj = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', media_url)
  66             if mobj:
  67                 formats.append({
  68                     'url': mobj.group('url'),
  69                     'play_path': 'mp4:%s' % mobj.group('playpath'),
  70                     'app': mobj.group('app'),
  71                     'ext': 'flv',
  72                     'tbr': tbr,
  73                     'format_id': 'rtmp-%d' % tbr,
  74                 })
  75             else:
  76                 formats.append({
  77                     'url': media_url,
  78                     'tbr': tbr,
  79                 })
  80         self._sort_formats(formats)
  81
  82         link = find_xpath_attr(
  83             entry, xpath_with_ns('./atom:link', NS_MAP), 'rel', 'original')
  84         if link is not None:
  85             formats.append({
  86                 'url': link.get('href'),
  87                 'format_id': link.get('rel'),
  88             })
  89
  90         thumbnails = [{
  91             'url': splash.get('url'),
  92             'width': int_or_none(splash.get('width')),
  93             'height': int_or_none(splash.get('height')),
  94         } for splash in media_group.findall(xpath_with_ns('./xt:splash', NS_MAP))]
  95
  96         return {
  97             'id': video_id,
  98             'title': title,
  99             'description': description,
 100             'timestamp': timestamp,
 101             'formats': formats,
 102             'thumbnails': thumbnails,
 103         }