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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import (
   6     compat_str,
   7     compat_urlparse,
   8 )
   9 from ..utils import (
  10     ExtractorError,
  11     float_or_none,
  12     parse_duration,
  13     str_to_int,
  14     urlencode_postdata,
  15 )
  16
  17
  18 class PandoraTVIE(InfoExtractor):
  19     IE_NAME = 'pandora.tv'
  20     IE_DESC = '판도라TV'
  21     _VALID_URL = r'https?://(?:.+?\.)?channel\.pandora\.tv/channel/video\.ptv\?'
  22     _TEST = {
  23         'url': 'http://jp.channel.pandora.tv/channel/video.ptv?c1=&prgid=53294230&ch_userid=mikakim&ref=main&lot=cate_01_2',
  24         'info_dict': {
  25             'id': '53294230',
  26             'ext': 'flv',
  27             'title': '頭を撫でてくれる？',
  28             'description': '頭を撫でてくれる？',
  29             'thumbnail': r're:^https?://.*\.jpg$',
  30             'duration': 39,
  31             'upload_date': '20151218',
  32             'uploader': 'カワイイ動物まとめ',
  33             'uploader_id': 'mikakim',
  34             'view_count': int,
  35             'like_count': int,
  36         }
  37     }
  38
  39     def _real_extract(self, url):
  40         qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
  41         video_id = qs.get('prgid', [None])[0]
  42         user_id = qs.get('ch_userid', [None])[0]
  43         if any(not f for f in (video_id, user_id,)):
  44             raise ExtractorError('Invalid URL', expected=True)
  45
  46         data = self._download_json(
  47             'http://m.pandora.tv/?c=view&m=viewJsonApi&ch_userid=%s&prgid=%s'
  48             % (user_id, video_id), video_id)
  49
  50         info = data['data']['rows']['vod_play_info']['result']
  51
  52         formats = []
  53         for format_id, format_url in info.items():
  54             if not format_url:
  55                 continue
  56             height = self._search_regex(
  57                 r'^v(\d+)[Uu]rl$', format_id, 'height', default=None)
  58             if not height:
  59                 continue
  60
  61             play_url = self._download_json(
  62                 'http://m.pandora.tv/?c=api&m=play_url', video_id,
  63                 data=urlencode_postdata({
  64                     'prgid': video_id,
  65                     'runtime': info.get('runtime'),
  66                     'vod_url': format_url,
  67                 }),
  68                 headers={
  69                     'Origin': url,
  70                     'Content-Type': 'application/x-www-form-urlencoded',
  71                 })
  72             format_url = play_url.get('url')
  73             if not format_url:
  74                 continue
  75
  76             formats.append({
  77                 'format_id': '%sp' % height,
  78                 'url': format_url,
  79                 'height': int(height),
  80             })
  81         self._sort_formats(formats)
  82
  83         return {
  84             'id': video_id,
  85             'title': info['subject'],
  86             'description': info.get('body'),
  87             'thumbnail': info.get('thumbnail') or info.get('poster'),
  88             'duration': float_or_none(info.get('runtime'), 1000) or parse_duration(info.get('time')),
  89             'upload_date': info['fid'][:8] if isinstance(info.get('fid'), compat_str) else None,
  90             'uploader': info.get('nickname'),
  91             'uploader_id': info.get('upload_userid'),
  92             'view_count': str_to_int(info.get('hit')),
  93             'like_count': str_to_int(info.get('likecnt')),
  94             'formats': formats,
  95         }