youtube_dl/extractor/afreecatv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import (
   6     compat_urllib_parse_urlparse,
   7     compat_urlparse,
   8 )
   9 from ..utils import (
  10     ExtractorError,
  11     int_or_none,
  12     xpath_text,
  13 )
  14
  15
  16 class AfreecaTVIE(InfoExtractor):
  17     IE_DESC = 'afreecatv.com'
  18     _VALID_URL = r'''(?x)^
  19         https?://(?:(live|afbbs|www)\.)?afreeca(?:tv)?\.com(?::\d+)?
  20         (?:
  21             /app/(?:index|read_ucc_bbs)\.cgi|
  22             /player/[Pp]layer\.(?:swf|html))
  23         \?.*?\bnTitleNo=(?P<id>\d+)'''
  24     _TEST = {
  25         'url': 'http://live.afreecatv.com:8079/app/index.cgi?szType=read_ucc_bbs&szBjId=dailyapril&nStationNo=16711924&nBbsNo=18605867&nTitleNo=36164052&szSkin=',
  26         'md5': 'f72c89fe7ecc14c1b5ce506c4996046e',
  27         'info_dict': {
  28             'id': '36164052',
  29             'ext': 'mp4',
  30             'title': '데일리 에이프릴 요정들의 시상식!',
  31             'thumbnail': 're:^https?://videoimg.afreecatv.com/.*$',
  32             'uploader': 'dailyapril',
  33             'uploader_id': 'dailyapril',
  34         }
  35     }
  36
  37     def _real_extract(self, url):
  38         video_id = self._match_id(url)
  39         parsed_url = compat_urllib_parse_urlparse(url)
  40         info_url = compat_urlparse.urlunparse(parsed_url._replace(
  41             netloc='afbbs.afreecatv.com:8080',
  42             path='/api/video/get_video_info.php'))
  43         video_xml = self._download_xml(info_url, video_id)
  44
  45         if xpath_text(video_xml, './track/flag', default='FAIL') != 'SUCCEED':
  46             raise ExtractorError('Specified AfreecaTV video does not exist',
  47                                  expected=True)
  48         title = xpath_text(video_xml, './track/title', 'title')
  49         uploader = xpath_text(video_xml, './track/nickname', 'uploader')
  50         uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id')
  51         duration = int_or_none(xpath_text(video_xml, './track/duration',
  52                                           'duration'))
  53         thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail')
  54
  55         entries = []
  56         for video_file in video_xml.findall('./track/video/file'):
  57             entries.append({
  58                 'id': video_file.get('key'),
  59                 'title': title,
  60                 'duration': int_or_none(video_file.get('duration')),
  61                 'url': video_file.text,
  62             })
  63
  64         info = {
  65             'id': video_id,
  66             'title': title,
  67             'uploader': uploader,
  68             'uploader_id': uploader_id,
  69             'duration': duration,
  70             'thumbnail': thumbnail,
  71         }
  72
  73         if len(entries) > 1:
  74             info['_type'] = 'multi_video'
  75             info['entries'] = entries
  76         elif len(entries) == 1:
  77             info['url'] = entries[0]['url']
  78         else:
  79             raise ExtractorError(
  80                 'No files found for the specified AfreecaTV video, either'
  81                 ' the URL is incorrect or the video has been made private.',
  82                 expected=True)
  83
  84         return info