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     _TESTS = [{
  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$',
  30             'duration': 39,
  31             'upload_date': '20151218',
  32             'uploader': 'カワイイ動物まとめ',
  33             'uploader_id': 'mikakim',
  34             'view_count': int,
  35             'like_count': int,
  36         }
  37     }, {
  38         'url': 'http://channel.pandora.tv/channel/video.ptv?ch_userid=gogoucc&prgid=54721744',
  39         'info_dict': {
  40             'id': '54721744',
  41             'ext': 'flv',
  42             'title': '[HD] JAPAN COUNTDOWN 170423',
  43             'description': '[HD] JAPAN COUNTDOWN 170423',
  44             'thumbnail': r're:^https?://.*\.jpg$',
  45             'duration': 1704.9,
  46             'upload_date': '20170423',
  47             'uploader': 'GOGO_UCC',
  48             'uploader_id': 'gogoucc',
  49             'view_count': int,
  50             'like_count': int,
  51         },
  52         'params': {
  53             # Test metadata only
  54             'skip_download': True,
  55         },
  56     }]
  57
  58     def _real_extract(self, url):
  59         qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
  60         video_id = qs.get('prgid', [None])[0]
  61         user_id = qs.get('ch_userid', [None])[0]
  62         if any(not f for f in (video_id, user_id,)):
  63             raise ExtractorError('Invalid URL', expected=True)
  64
  65         data = self._download_json(
  66             'http://m.pandora.tv/?c=view&m=viewJsonApi&ch_userid=%s&prgid=%s'
  67             % (user_id, video_id), video_id)
  68
  69         info = data['data']['rows']['vod_play_info']['result']
  70
  71         formats = []
  72         for format_id, format_url in info.items():
  73             if not format_url:
  74                 continue
  75             height = self._search_regex(
  76                 r'^v(\d+)[Uu]rl$', format_id, 'height', default=None)
  77             if not height:
  78                 continue
  79
  80             play_url = self._download_json(
  81                 'http://m.pandora.tv/?c=api&m=play_url', video_id,
  82                 data=urlencode_postdata({
  83                     'prgid': video_id,
  84                     'runtime': info.get('runtime'),
  85                     'vod_url': format_url,
  86                 }),
  87                 headers={
  88                     'Origin': url,
  89                     'Content-Type': 'application/x-www-form-urlencoded',
  90                 })
  91             format_url = play_url.get('url')
  92             if not format_url:
  93                 continue
  94
  95             formats.append({
  96                 'format_id': '%sp' % height,
  97                 'url': format_url,
  98                 'height': int(height),
  99             })
 100         self._sort_formats(formats)
 101
 102         return {
 103             'id': video_id,
 104             'title': info['subject'],
 105             'description': info.get('body'),
 106             'thumbnail': info.get('thumbnail') or info.get('poster'),
 107             'duration': float_or_none(info.get('runtime'), 1000) or parse_duration(info.get('time')),
 108             'upload_date': info['fid'].split('/')[-1][:8] if isinstance(info.get('fid'), compat_str) else None,
 109             'uploader': info.get('nickname'),
 110             'uploader_id': info.get('upload_userid'),
 111             'view_count': str_to_int(info.get('hit')),
 112             'like_count': str_to_int(info.get('likecnt')),
 113             'formats': formats,
 114         }