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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_urllib_parse_urlparse,
   9     compat_urlparse,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     int_or_none,
  14     update_url_query,
  15     xpath_element,
  16     xpath_text,
  17 )
  18
  19
  20 class AfreecaTVIE(InfoExtractor):
  21     IE_NAME = 'afreecatv'
  22     IE_DESC = 'afreecatv.com'
  23     _VALID_URL = r'''(?x)
  24                     https?://
  25                         (?:
  26                             (?:(?:live|afbbs|www)\.)?afreeca(?:tv)?\.com(?::\d+)?
  27                             (?:
  28                                 /app/(?:index|read_ucc_bbs)\.cgi|
  29                                 /player/[Pp]layer\.(?:swf|html)
  30                             )\?.*?\bnTitleNo=|
  31                             vod\.afreecatv\.com/PLAYER/STATION/
  32                         )
  33                         (?P<id>\d+)
  34                     '''
  35     _TESTS = [{
  36         'url': 'http://live.afreecatv.com:8079/app/index.cgi?szType=read_ucc_bbs&szBjId=dailyapril&nStationNo=16711924&nBbsNo=18605867&nTitleNo=36164052&szSkin=',
  37         'md5': 'f72c89fe7ecc14c1b5ce506c4996046e',
  38         'info_dict': {
  39             'id': '36164052',
  40             'ext': 'mp4',
  41             'title': '데일리 에이프릴 요정들의 시상식!',
  42             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
  43             'uploader': 'dailyapril',
  44             'uploader_id': 'dailyapril',
  45             'upload_date': '20160503',
  46         }
  47     }, {
  48         'url': 'http://afbbs.afreecatv.com:8080/app/read_ucc_bbs.cgi?nStationNo=16711924&nTitleNo=36153164&szBjId=dailyapril&nBbsNo=18605867',
  49         'info_dict': {
  50             'id': '36153164',
  51             'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
  52             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
  53             'uploader': 'dailyapril',
  54             'uploader_id': 'dailyapril',
  55         },
  56         'playlist_count': 2,
  57         'playlist': [{
  58             'md5': 'd8b7c174568da61d774ef0203159bf97',
  59             'info_dict': {
  60                 'id': '36153164_1',
  61                 'ext': 'mp4',
  62                 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
  63                 'upload_date': '20160502',
  64             },
  65         }, {
  66             'md5': '58f2ce7f6044e34439ab2d50612ab02b',
  67             'info_dict': {
  68                 'id': '36153164_2',
  69                 'ext': 'mp4',
  70                 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
  71                 'upload_date': '20160502',
  72             },
  73         }],
  74     }, {
  75         'url': 'http://www.afreecatv.com/player/Player.swf?szType=szBjId=djleegoon&nStationNo=11273158&nBbsNo=13161095&nTitleNo=36327652',
  76         'only_matching': True,
  77     }, {
  78         'url': 'http://vod.afreecatv.com/PLAYER/STATION/15055030',
  79         'only_matching': True,
  80     }]
  81
  82     @staticmethod
  83     def parse_video_key(key):
  84         video_key = {}
  85         m = re.match(r'^(?P<upload_date>\d{8})_\w+_(?P<part>\d+)$', key)
  86         if m:
  87             video_key['upload_date'] = m.group('upload_date')
  88             video_key['part'] = m.group('part')
  89         return video_key
  90
  91     def _real_extract(self, url):
  92         video_id = self._match_id(url)
  93         parsed_url = compat_urllib_parse_urlparse(url)
  94         info_url = compat_urlparse.urlunparse(parsed_url._replace(
  95             netloc='afbbs.afreecatv.com:8080',
  96             path='/api/video/get_video_info.php'))
  97
  98         video_xml = self._download_xml(
  99             update_url_query(info_url, {'nTitleNo': video_id}), video_id)
 100
 101         if xpath_element(video_xml, './track/video/file') is None:
 102             raise ExtractorError('Specified AfreecaTV video does not exist',
 103                                  expected=True)
 104
 105         title = xpath_text(video_xml, './track/title', 'title')
 106         uploader = xpath_text(video_xml, './track/nickname', 'uploader')
 107         uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id')
 108         duration = int_or_none(xpath_text(video_xml, './track/duration',
 109                                           'duration'))
 110         thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail')
 111
 112         entries = []
 113         for i, video_file in enumerate(video_xml.findall('./track/video/file')):
 114             video_key = self.parse_video_key(video_file.get('key', ''))
 115             if not video_key:
 116                 continue
 117             entries.append({
 118                 'id': '%s_%s' % (video_id, video_key.get('part', i + 1)),
 119                 'title': title,
 120                 'upload_date': video_key.get('upload_date'),
 121                 'duration': int_or_none(video_file.get('duration')),
 122                 'url': video_file.text,
 123             })
 124
 125         info = {
 126             'id': video_id,
 127             'title': title,
 128             'uploader': uploader,
 129             'uploader_id': uploader_id,
 130             'duration': duration,
 131             'thumbnail': thumbnail,
 132         }
 133
 134         if len(entries) > 1:
 135             info['_type'] = 'multi_video'
 136             info['entries'] = entries
 137         elif len(entries) == 1:
 138             info['url'] = entries[0]['url']
 139             info['upload_date'] = entries[0].get('upload_date')
 140         else:
 141             raise ExtractorError(
 142                 'No files found for the specified AfreecaTV video, either'
 143                 ' the URL is incorrect or the video has been made private.',
 144                 expected=True)
 145
 146         return info
 147
 148
 149 class AfreecaTVGlobalIE(AfreecaTVIE):
 150     IE_NAME = 'afreecatv:global'
 151     _VALID_URL = r'https?://(?:www\.)?afreeca\.tv/(?P<channel_id>\d+)(?:/v/(?P<video_id>\d+))?'
 152     _TESTS = [{
 153         'url': 'http://afreeca.tv/36853014/v/58301',
 154         'info_dict': {
 155             'id': '58301',
 156             'title': 'tryhard top100',
 157             'uploader_id': '36853014',
 158             'uploader': 'makgi Hearthstone Live!',
 159         },
 160         'playlist_count': 3,
 161     }]
 162
 163     def _real_extract(self, url):
 164         channel_id, video_id = re.match(self._VALID_URL, url).groups()
 165         video_type = 'video' if video_id else 'live'
 166         query = {
 167             'pt': 'view',
 168             'bid': channel_id,
 169         }
 170         if video_id:
 171             query['vno'] = video_id
 172         video_data = self._download_json(
 173             'http://api.afreeca.tv/%s/view_%s.php' % (video_type, video_type),
 174             video_id or channel_id, query=query)['channel']
 175
 176         if video_data.get('result') != 1:
 177             raise ExtractorError('%s said: %s' % (self.IE_NAME, video_data['remsg']))
 178
 179         title = video_data['title']
 180
 181         info = {
 182             'thumbnail': video_data.get('thumb'),
 183             'view_count': int_or_none(video_data.get('vcnt')),
 184             'age_limit': int_or_none(video_data.get('grade')),
 185             'uploader_id': channel_id,
 186             'uploader': video_data.get('cname'),
 187         }
 188
 189         if video_id:
 190             entries = []
 191             for i, f in enumerate(video_data.get('flist', [])):
 192                 video_key = self.parse_video_key(f.get('key', ''))
 193                 f_url = f.get('file')
 194                 if not video_key or not f_url:
 195                     continue
 196                 entries.append({
 197                     'id': '%s_%s' % (video_id, video_key.get('part', i + 1)),
 198                     'title': title,
 199                     'upload_date': video_key.get('upload_date'),
 200                     'duration': int_or_none(f.get('length')),
 201                     'url': f_url,
 202                     'protocol': 'm3u8_native',
 203                     'ext': 'mp4',
 204                 })
 205
 206             info.update({
 207                 'id': video_id,
 208                 'title': title,
 209                 'duration': int_or_none(video_data.get('length')),
 210             })
 211             if len(entries) > 1:
 212                 info['_type'] = 'multi_video'
 213                 info['entries'] = entries
 214             elif len(entries) == 1:
 215                 i = entries[0].copy()
 216                 i.update(info)
 217                 info = i
 218         else:
 219             formats = []
 220             for s in video_data.get('strm', []):
 221                 s_url = s.get('purl')
 222                 if not s_url:
 223                     continue
 224                 stype = s.get('stype')
 225                 if stype == 'HLS':
 226                     formats.extend(self._extract_m3u8_formats(
 227                         s_url, channel_id, 'mp4', m3u8_id=stype, fatal=False))
 228                 elif stype == 'RTMP':
 229                     format_id = [stype]
 230                     label = s.get('label')
 231                     if label:
 232                         format_id.append(label)
 233                     formats.append({
 234                         'format_id': '-'.join(format_id),
 235                         'url': s_url,
 236                         'tbr': int_or_none(s.get('bps')),
 237                         'height': int_or_none(s.get('brt')),
 238                         'ext': 'flv',
 239                         'rtmp_live': True,
 240                     })
 241             self._sort_formats(formats)
 242
 243             info.update({
 244                 'id': channel_id,
 245                 'title': self._live_title(title),
 246                 'is_live': True,
 247                 'formats': formats,
 248             })
 249
 250         return info