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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     determine_ext,
   7     float_or_none,
   8     int_or_none,
   9 )
  10
  11
  12 class KonserthusetPlayIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?konserthusetplay\.se/\?.*\bm=(?P<id>[^&]+)'
  14     _TEST = {
  15         'url': 'http://www.konserthusetplay.se/?m=CKDDnlCY-dhWAAqiMERd-A',
  16         'info_dict': {
  17             'id': 'CKDDnlCY-dhWAAqiMERd-A',
  18             'ext': 'flv',
  19             'title': 'Orkesterns instrument: Valthornen',
  20             'description': 'md5:f10e1f0030202020396a4d712d2fa827',
  21             'thumbnail': 're:^https?://.*$',
  22             'duration': 398.8,
  23         },
  24         'params': {
  25             # rtmp download
  26             'skip_download': True,
  27         },
  28     }
  29
  30     def _real_extract(self, url):
  31         video_id = self._match_id(url)
  32
  33         webpage = self._download_webpage(url, video_id)
  34
  35         e = self._search_regex(
  36             r'https?://csp\.picsearch\.com/rest\?.*\be=(.+?)[&"\']', webpage, 'e')
  37
  38         rest = self._download_json(
  39             'http://csp.picsearch.com/rest?e=%s&containerId=mediaplayer&i=object' % e,
  40             video_id, transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1])
  41
  42         media = rest['media']
  43         player_config = media['playerconfig']
  44         playlist = player_config['playlist']
  45
  46         source = next(f for f in playlist if f.get('bitrates') or f.get('provider'))
  47
  48         FORMAT_ID_REGEX = r'_([^_]+)_h264m\.mp4'
  49
  50         formats = []
  51
  52         m3u8_url = source.get('url')
  53         if m3u8_url and determine_ext(m3u8_url) == 'm3u8':
  54             formats.extend(self._extract_m3u8_formats(
  55                 m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
  56                 m3u8_id='hls', fatal=False))
  57
  58         fallback_url = source.get('fallbackUrl')
  59         fallback_format_id = None
  60         if fallback_url:
  61             fallback_format_id = self._search_regex(
  62                 FORMAT_ID_REGEX, fallback_url, 'format id', default=None)
  63
  64         connection_url = (player_config.get('rtmp', {}).get(
  65             'netConnectionUrl') or player_config.get(
  66             'plugins', {}).get('bwcheck', {}).get('netConnectionUrl'))
  67         if connection_url:
  68             for f in source['bitrates']:
  69                 video_url = f.get('url')
  70                 if not video_url:
  71                     continue
  72                 format_id = self._search_regex(
  73                     FORMAT_ID_REGEX, video_url, 'format id', default=None)
  74                 f_common = {
  75                     'vbr': int_or_none(f.get('bitrate')),
  76                     'width': int_or_none(f.get('width')),
  77                     'height': int_or_none(f.get('height')),
  78                 }
  79                 f = f_common.copy()
  80                 f.update({
  81                     'url': connection_url,
  82                     'play_path': video_url,
  83                     'format_id': 'rtmp-%s' % format_id if format_id else 'rtmp',
  84                     'ext': 'flv',
  85                 })
  86                 formats.append(f)
  87                 if format_id and format_id == fallback_format_id:
  88                     f = f_common.copy()
  89                     f.update({
  90                         'url': fallback_url,
  91                         'format_id': 'http-%s' % format_id if format_id else 'http',
  92                     })
  93                     formats.append(f)
  94
  95         if not formats and fallback_url:
  96             formats.append({
  97                 'url': fallback_url,
  98             })
  99
 100         self._sort_formats(formats)
 101
 102         title = player_config.get('title') or media['title']
 103         description = player_config.get('mediaInfo', {}).get('description')
 104         thumbnail = media.get('image')
 105         duration = float_or_none(media.get('duration'), 1000)
 106
 107         return {
 108             'id': video_id,
 109             'title': title,
 110             'description': description,
 111             'thumbnail': thumbnail,
 112             'duration': duration,
 113             'formats': formats,
 114         }