_ Git - youtube-dl/blob - youtube_dl/extractor/kanalplay.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     ExtractorError,
   9     float_or_none,
  10 )
  11
  12
  13 class KanalPlayIE(InfoExtractor):
  14     IE_DESC = 'Kanal 5/9/11 Play'
  15     _VALID_URL = r'https?://(?:www\.)?kanal(?P<channel_id>5|9|11)play\.se/(?:#!/)?(?:play/)?program/\d+/video/(?P<id>\d+)'
  16     _TESTS = [{
  17         'url': 'http://www.kanal5play.se/#!/play/program/3060212363/video/3270012277',
  18         'md5': '',
  19         'info_dict': {
  20             'id': '2609989',
  21             'ext': 'flv',
  22             'title': 'Saknar både dusch och avlopp',
  23             'description': 'md5:',
  24             'duration': 2636.36,
  25         },
  26     }, {
  27         'url': 'http://www.kanal9play.se/#!/play/program/335032/video/246042',
  28         'only_matching': True,
  29     }, {
  30         'url': 'http://www.kanal11play.se/#!/play/program/232835958/video/367135199',
  31         'only_matching': True,
  32     }]
  33
  34     def _fix_subtitles(self, subs):
  35         return '\r\n\r\n'.join(
  36             '%s\r\n%s --> %s\r\n%s'
  37             % (
  38                 num,
  39                 self._subtitles_timecode(item['startMillis'] / 1000.0),
  40                 self._subtitles_timecode(item['endMillis'] / 1000.0),
  41                 item['text'],
  42             ) for num, item in enumerate(subs, 1))
  43
  44     def _get_subtitles(self, channel_id, video_id):
  45         subs = self._download_json(
  46             'http://www.kanal%splay.se/api/subtitles/%s' % (channel_id, video_id),
  47             video_id, 'Downloading subtitles JSON', fatal=False)
  48         return {'se': [{'ext': 'srt', 'data': self._fix_subtitles(subs)}]} if subs else {}
  49
  50     def _real_extract(self, url):
  51         mobj = re.match(self._VALID_URL, url)
  52         video_id = mobj.group('id')
  53         channel_id = mobj.group('channel_id')
  54
  55         video = self._download_json(
  56             'http://www.kanal%splay.se/api/getVideo?format=FLASH&videoId=%s' % (channel_id, video_id),
  57             video_id)
  58
  59         reasons_for_no_streams = video.get('reasonsForNoStreams')
  60         if reasons_for_no_streams:
  61             raise ExtractorError(
  62                 '%s returned error: %s' % (self.IE_NAME, '\n'.join(reasons_for_no_streams)),
  63                 expected=True)
  64
  65         title = video['title']
  66         description = video.get('description')
  67         duration = float_or_none(video.get('length'), 1000)
  68         thumbnail = video.get('posterUrl')
  69
  70         stream_base_url = video['streamBaseUrl']
  71
  72         formats = [{
  73             'url': stream_base_url,
  74             'play_path': stream['source'],
  75             'ext': 'flv',
  76             'tbr': float_or_none(stream.get('bitrate'), 1000),
  77             'rtmp_real_time': True,
  78         } for stream in video['streams']]
  79         self._sort_formats(formats)
  80
  81         subtitles = {}
  82         if video.get('hasSubtitle'):
  83             subtitles = self.extract_subtitles(channel_id, video_id)
  84
  85         return {
  86             'id': video_id,
  87             'title': title,
  88             'description': description,
  89             'thumbnail': thumbnail,
  90             'duration': duration,
  91             'formats': formats,
  92             'subtitles': subtitles,
  93         }