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 compat_xpath
   8 from ..utils import (
   9     ExtractorError,
  10     int_or_none,
  11     xpath_text,
  12 )
  13
  14
  15 class AfreecaTVIE(InfoExtractor):
  16     IE_NAME = 'afreecatv'
  17     IE_DESC = 'afreecatv.com'
  18     _VALID_URL = r'''(?x)
  19                     https?://
  20                         (?:
  21                             (?:(?:live|afbbs|www)\.)?afreeca(?:tv)?\.com(?::\d+)?
  22                             (?:
  23                                 /app/(?:index|read_ucc_bbs)\.cgi|
  24                                 /player/[Pp]layer\.(?:swf|html)
  25                             )\?.*?\bnTitleNo=|
  26                             vod\.afreecatv\.com/PLAYER/STATION/
  27                         )
  28                         (?P<id>\d+)
  29                     '''
  30     _TESTS = [{
  31         'url': 'http://live.afreecatv.com:8079/app/index.cgi?szType=read_ucc_bbs&szBjId=dailyapril&nStationNo=16711924&nBbsNo=18605867&nTitleNo=36164052&szSkin=',
  32         'md5': 'f72c89fe7ecc14c1b5ce506c4996046e',
  33         'info_dict': {
  34             'id': '36164052',
  35             'ext': 'mp4',
  36             'title': '데일리 에이프릴 요정들의 시상식!',
  37             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
  38             'uploader': 'dailyapril',
  39             'uploader_id': 'dailyapril',
  40             'upload_date': '20160503',
  41         },
  42         'skip': 'Video is gone',
  43     }, {
  44         'url': 'http://afbbs.afreecatv.com:8080/app/read_ucc_bbs.cgi?nStationNo=16711924&nTitleNo=36153164&szBjId=dailyapril&nBbsNo=18605867',
  45         'info_dict': {
  46             'id': '36153164',
  47             'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
  48             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
  49             'uploader': 'dailyapril',
  50             'uploader_id': 'dailyapril',
  51         },
  52         'playlist_count': 2,
  53         'playlist': [{
  54             'md5': 'd8b7c174568da61d774ef0203159bf97',
  55             'info_dict': {
  56                 'id': '36153164_1',
  57                 'ext': 'mp4',
  58                 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
  59                 'upload_date': '20160502',
  60             },
  61         }, {
  62             'md5': '58f2ce7f6044e34439ab2d50612ab02b',
  63             'info_dict': {
  64                 'id': '36153164_2',
  65                 'ext': 'mp4',
  66                 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
  67                 'upload_date': '20160502',
  68             },
  69         }],
  70         'skip': 'Video is gone',
  71     }, {
  72         'url': 'http://vod.afreecatv.com/PLAYER/STATION/18650793',
  73         'info_dict': {
  74             'id': '18650793',
  75             'ext': 'flv',
  76             'uploader': '윈아디',
  77             'uploader_id': 'badkids',
  78             'title': '오늘은 다르다! 쏘님의 우월한 위아래~ 댄스리액션!',
  79         },
  80         'params': {
  81             'skip_download': True,  # requires rtmpdump
  82         },
  83     }, {
  84         'url': 'http://www.afreecatv.com/player/Player.swf?szType=szBjId=djleegoon&nStationNo=11273158&nBbsNo=13161095&nTitleNo=36327652',
  85         'only_matching': True,
  86     }, {
  87         'url': 'http://vod.afreecatv.com/PLAYER/STATION/15055030',
  88         'only_matching': True,
  89     }]
  90
  91     @staticmethod
  92     def parse_video_key(key):
  93         video_key = {}
  94         m = re.match(r'^(?P<upload_date>\d{8})_\w+_(?P<part>\d+)$', key)
  95         if m:
  96             video_key['upload_date'] = m.group('upload_date')
  97             video_key['part'] = m.group('part')
  98         return video_key
  99
 100     def _real_extract(self, url):
 101         video_id = self._match_id(url)
 102
 103         video_xml = self._download_xml(
 104             'http://afbbs.afreecatv.com:8080/api/video/get_video_info.php',
 105             video_id, query={'nTitleNo': video_id})
 106
 107         video_element = video_xml.findall(compat_xpath('./track/video'))[1]
 108         if video_element is None or video_element.text is None:
 109             raise ExtractorError('Specified AfreecaTV video does not exist',
 110                                  expected=True)
 111
 112         video_url_raw = video_element.text
 113
 114         app, playpath = video_url_raw.split('mp4:')
 115
 116         title = xpath_text(video_xml, './track/title', 'title', fatal=True)
 117         uploader = xpath_text(video_xml, './track/nickname', 'uploader')
 118         uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id')
 119         duration = int_or_none(xpath_text(video_xml, './track/duration',
 120                                           'duration'))
 121         thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail')
 122
 123         return {
 124             'id': video_id,
 125             'url': app,
 126             'ext': 'flv',
 127             'play_path': 'mp4:' + playpath,
 128             'rtmp_live': True,  # downloading won't end without this
 129             'title': title,
 130             'uploader': uploader,
 131             'uploader_id': uploader_id,
 132             'duration': duration,
 133             'thumbnail': thumbnail,
 134         }
 135
 136
 137 class AfreecaTVGlobalIE(AfreecaTVIE):
 138     IE_NAME = 'afreecatv:global'
 139     _VALID_URL = r'https?://(?:www\.)?afreeca\.tv/(?P<channel_id>\d+)(?:/v/(?P<video_id>\d+))?'
 140     _TESTS = [{
 141         'url': 'http://afreeca.tv/36853014/v/58301',
 142         'info_dict': {
 143             'id': '58301',
 144             'title': 'tryhard top100',
 145             'uploader_id': '36853014',
 146             'uploader': 'makgi Hearthstone Live!',
 147         },
 148         'playlist_count': 3,
 149     }]
 150
 151     def _real_extract(self, url):
 152         channel_id, video_id = re.match(self._VALID_URL, url).groups()
 153         video_type = 'video' if video_id else 'live'
 154         query = {
 155             'pt': 'view',
 156             'bid': channel_id,
 157         }
 158         if video_id:
 159             query['vno'] = video_id
 160         video_data = self._download_json(
 161             'http://api.afreeca.tv/%s/view_%s.php' % (video_type, video_type),
 162             video_id or channel_id, query=query)['channel']
 163
 164         if video_data.get('result') != 1:
 165             raise ExtractorError('%s said: %s' % (self.IE_NAME, video_data['remsg']))
 166
 167         title = video_data['title']
 168
 169         info = {
 170             'thumbnail': video_data.get('thumb'),
 171             'view_count': int_or_none(video_data.get('vcnt')),
 172             'age_limit': int_or_none(video_data.get('grade')),
 173             'uploader_id': channel_id,
 174             'uploader': video_data.get('cname'),
 175         }
 176
 177         if video_id:
 178             entries = []
 179             for i, f in enumerate(video_data.get('flist', [])):
 180                 video_key = self.parse_video_key(f.get('key', ''))
 181                 f_url = f.get('file')
 182                 if not video_key or not f_url:
 183                     continue
 184                 entries.append({
 185                     'id': '%s_%s' % (video_id, video_key.get('part', i + 1)),
 186                     'title': title,
 187                     'upload_date': video_key.get('upload_date'),
 188                     'duration': int_or_none(f.get('length')),
 189                     'url': f_url,
 190                     'protocol': 'm3u8_native',
 191                     'ext': 'mp4',
 192                 })
 193
 194             info.update({
 195                 'id': video_id,
 196                 'title': title,
 197                 'duration': int_or_none(video_data.get('length')),
 198             })
 199             if len(entries) > 1:
 200                 info['_type'] = 'multi_video'
 201                 info['entries'] = entries
 202             elif len(entries) == 1:
 203                 i = entries[0].copy()
 204                 i.update(info)
 205                 info = i
 206         else:
 207             formats = []
 208             for s in video_data.get('strm', []):
 209                 s_url = s.get('purl')
 210                 if not s_url:
 211                     continue
 212                 stype = s.get('stype')
 213                 if stype == 'HLS':
 214                     formats.extend(self._extract_m3u8_formats(
 215                         s_url, channel_id, 'mp4', m3u8_id=stype, fatal=False))
 216                 elif stype == 'RTMP':
 217                     format_id = [stype]
 218                     label = s.get('label')
 219                     if label:
 220                         format_id.append(label)
 221                     formats.append({
 222                         'format_id': '-'.join(format_id),
 223                         'url': s_url,
 224                         'tbr': int_or_none(s.get('bps')),
 225                         'height': int_or_none(s.get('brt')),
 226                         'ext': 'flv',
 227                         'rtmp_live': True,
 228                     })
 229             self._sort_formats(formats)
 230
 231             info.update({
 232                 'id': channel_id,
 233                 'title': self._live_title(title),
 234                 'is_live': True,
 235                 'formats': formats,
 236             })
 237
 238         return info