Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/ndr.py

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import ExtractorError
   8
   9
  10 class NDRIE(InfoExtractor):
  11     IE_NAME = 'ndr'
  12     IE_DESC = 'NDR.de - Mediathek'
  13     _VALID_URL = r'https?://www\.ndr\.de/.+?(?P<id>\d+)\.html'
  14
  15     _TESTS = [
  16         {
  17             'url': 'http://www.ndr.de/fernsehen/sendungen/markt/markt7959.html',
  18             'md5': 'e7a6079ca39d3568f4996cb858dd6708',
  19             'note': 'Video file',
  20             'info_dict': {
  21                 'id': '7959',
  22                 'ext': 'mp4',
  23                 'title': 'Markt - die ganze Sendung',
  24                 'description': 'md5:af9179cf07f67c5c12dc6d9997e05725',
  25                 'duration': 2655,
  26             },
  27         },
  28         {
  29             'url': 'http://www.ndr.de/info/audio51535.html',
  30             'md5': 'bb3cd38e24fbcc866d13b50ca59307b8',
  31             'note': 'Audio file',
  32             'info_dict': {
  33                 'id': '51535',
  34                 'ext': 'mp3',
  35                 'title': 'La Valette entgeht der Hinrichtung',
  36                 'description': 'md5:22f9541913a40fe50091d5cdd7c9f536',
  37                 'duration': 884,
  38             }
  39         }
  40     ]
  41
  42     def _real_extract(self, url):
  43         mobj = re.match(self._VALID_URL, url)
  44         video_id = mobj.group('id')
  45
  46         page = self._download_webpage(url, video_id, 'Downloading page')
  47
  48         title = self._og_search_title(page)
  49         description = self._og_search_description(page)
  50
  51         mobj = re.search(
  52             r'<div class="duration"><span class="min">(?P<minutes>\d+)</span>:<span class="sec">(?P<seconds>\d+)</span></div>',
  53             page)
  54         duration = int(mobj.group('minutes')) * 60 + int(mobj.group('seconds')) if mobj else None
  55
  56         formats = []
  57
  58         mp3_url = re.search(r'''{src:'(?P<audio>[^']+)', type:"audio/mp3"},''', page)
  59         if mp3_url:
  60             formats.append({
  61                 'url': mp3_url.group('audio'),
  62                 'format_id': 'mp3',
  63             })
  64
  65         thumbnail = None
  66
  67         video_url = re.search(r'''3: {src:'(?P<video>.+?)\.hi\.mp4', type:"video/mp4"},''', page)
  68         if video_url:
  69             thumbnail = self._html_search_regex(r'(?m)title: "NDR PLAYER",\s*poster: "([^"]+)",',
  70                 page, 'thumbnail', fatal=False)
  71             if thumbnail:
  72                 thumbnail = 'http://www.ndr.de' + thumbnail
  73             for format_id in ['lo', 'hi', 'hq']:
  74                 formats.append({
  75                     'url': '%s.%s.mp4' % (video_url.group('video'), format_id),
  76                     'format_id': format_id,
  77                 })
  78
  79         if not formats:
  80             raise ExtractorError('No media links available for %s' % video_id)
  81
  82         return {
  83             'id': video_id,
  84             'title': title,
  85             'description': description,
  86             'thumbnail': thumbnail,
  87             'duration': duration,
  88             'formats': formats,
  89         }