_ Git - youtube-dl/blob - 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 )
  13
  14
  15 class AfreecaTVIE(InfoExtractor):
  16     IE_DESC = 'afreecatv.com'
  17     _VALID_URL = r'''(?x)^
  18         https?://(?:(live|afbbs|www)\.)?afreeca(?:tv)?\.com(?::\d+)?
  19         (?:
  20             /app/(?:index|read_ucc_bbs)\.cgi|
  21             /player/[Pp]layer\.(?:swf|html))
  22         \?.*?\bnTitleNo=(?P<id>\d+)'''
  23     _TEST = {
  24         'url': 'http://live.afreecatv.com:8079/app/index.cgi?szType=read_ucc_bbs&szBjId=dailyapril&nStationNo=16711924&nBbsNo=18605867&nTitleNo=36164052&szSkin=',
  25         'md5': 'f72c89fe7ecc14c1b5ce506c4996046e',
  26         'info_dict': {
  27             'id': '36164052',
  28             'ext': 'mp4',
  29             'title': '데일리 에이프릴 요정들의 시상식!',
  30             'thumbnail': 're:^https?://videoimg.afreecatv.com/.*$',
  31             'uploader': 'dailyapril',
  32             'uploader_id': 'dailyapril',
  33         }
  34     }
  35
  36     def _real_extract(self, url):
  37         video_id = self._match_id(url)
  38         parsed_url = compat_urllib_parse_urlparse(url)
  39         info_url = compat_urlparse.urlunparse(parsed_url._replace(
  40             netloc='afbbs.afreecatv.com:8080',
  41             path='/api/video/get_video_info.php'))
  42         video_xml = self._download_xml(info_url, video_id)
  43
  44         track = video_xml.find('track')
  45         if track.find('flag').text != 'SUCCEED':
  46             raise ExtractorError('Specified AfreecaTV video does not exist',
  47                                  expected=True)
  48         title = track.find('title').text
  49         uploader = track.find('nickname').text
  50         uploader_id = track.find('bj_id').text
  51         duration = int_or_none(track.find('duration').text)
  52         thumbnail = track.find('titleImage').text
  53
  54         entries = []
  55         for video in track.findall('video'):
  56             for video_file in video.findall('file'):
  57                 entries.append({
  58                     'id': video_file.get('key'),
  59                     'title': title,
  60                     'duration': int_or_none(video_file.get('duration')),
  61                     'formats': [{'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['formats'] = entries[0]['formats']
  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