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     determine_ext,
  10     ExtractorError,
  11     int_or_none,
  12     xpath_text,
  13 )
  14
  15
  16 class AfreecaTVIE(InfoExtractor):
  17     IE_NAME = 'afreecatv'
  18     IE_DESC = 'afreecatv.com'
  19     _VALID_URL = r'''(?x)
  20                     https?://
  21                         (?:
  22                             (?:(?:live|afbbs|www)\.)?afreeca(?:tv)?\.com(?::\d+)?
  23                             (?:
  24                                 /app/(?:index|read_ucc_bbs)\.cgi|
  25                                 /player/[Pp]layer\.(?:swf|html)
  26                             )\?.*?\bnTitleNo=|
  27                             vod\.afreecatv\.com/PLAYER/STATION/
  28                         )
  29                         (?P<id>\d+)
  30                     '''
  31     _TESTS = [{
  32         'url': 'http://live.afreecatv.com:8079/app/index.cgi?szType=read_ucc_bbs&szBjId=dailyapril&nStationNo=16711924&nBbsNo=18605867&nTitleNo=36164052&szSkin=',
  33         'md5': 'f72c89fe7ecc14c1b5ce506c4996046e',
  34         'info_dict': {
  35             'id': '36164052',
  36             'ext': 'mp4',
  37             'title': '데일리 에이프릴 요정들의 시상식!',
  38             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
  39             'uploader': 'dailyapril',
  40             'uploader_id': 'dailyapril',
  41             'upload_date': '20160503',
  42         },
  43         'skip': 'Video is gone',
  44     }, {
  45         'url': 'http://afbbs.afreecatv.com:8080/app/read_ucc_bbs.cgi?nStationNo=16711924&nTitleNo=36153164&szBjId=dailyapril&nBbsNo=18605867',
  46         'info_dict': {
  47             'id': '36153164',
  48             'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
  49             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
  50             'uploader': 'dailyapril',
  51             'uploader_id': 'dailyapril',
  52         },
  53         'playlist_count': 2,
  54         'playlist': [{
  55             'md5': 'd8b7c174568da61d774ef0203159bf97',
  56             'info_dict': {
  57                 'id': '36153164_1',
  58                 'ext': 'mp4',
  59                 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
  60                 'upload_date': '20160502',
  61             },
  62         }, {
  63             'md5': '58f2ce7f6044e34439ab2d50612ab02b',
  64             'info_dict': {
  65                 'id': '36153164_2',
  66                 'ext': 'mp4',
  67                 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
  68                 'upload_date': '20160502',
  69             },
  70         }],
  71         'skip': 'Video is gone',
  72     }, {
  73         'url': 'http://vod.afreecatv.com/PLAYER/STATION/18650793',
  74         'info_dict': {
  75             'id': '18650793',
  76             'ext': 'mp4',
  77             'title': '오늘은 다르다! 쏘님의 우월한 위아래~ 댄스리액션!',
  78             'thumbnail': r're:^https?://.*\.jpg$',
  79             'uploader': '윈아디',
  80             'uploader_id': 'badkids',
  81             'duration': 107,
  82         },
  83         'params': {
  84             'skip_download': True,
  85         },
  86     }, {
  87         'url': 'http://vod.afreecatv.com/PLAYER/STATION/10481652',
  88         'info_dict': {
  89             'id': '10481652',
  90             'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
  91             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
  92             'uploader': 'dailyapril',
  93             'uploader_id': 'dailyapril',
  94             'duration': 6492,
  95         },
  96         'playlist_count': 2,
  97         'playlist': [{
  98             'md5': 'd8b7c174568da61d774ef0203159bf97',
  99             'info_dict': {
 100                 'id': '20160502_c4c62b9d_174361386_1',
 101                 'ext': 'mp4',
 102                 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!' (part 1)",
 103                 'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
 104                 'uploader': 'dailyapril',
 105                 'uploader_id': 'dailyapril',
 106                 'upload_date': '20160502',
 107                 'duration': 3601,
 108             },
 109         }, {
 110             'md5': '58f2ce7f6044e34439ab2d50612ab02b',
 111             'info_dict': {
 112                 'id': '20160502_39e739bb_174361386_2',
 113                 'ext': 'mp4',
 114                 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!' (part 2)",
 115                 'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
 116                 'uploader': 'dailyapril',
 117                 'uploader_id': 'dailyapril',
 118                 'upload_date': '20160502',
 119                 'duration': 2891,
 120             },
 121         }],
 122         'params': {
 123             'skip_download': True,
 124         },
 125     }, {
 126         # non standard key
 127         'url': 'http://vod.afreecatv.com/PLAYER/STATION/20515605',
 128         'info_dict': {
 129             'id': '20170411_BE689A0E_190960999_1_2_h',
 130             'ext': 'mp4',
 131             'title': '혼자사는여자집',
 132             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
 133             'uploader': '♥이슬이',
 134             'uploader_id': 'dasl8121',
 135             'upload_date': '20170411',
 136             'duration': 213,
 137         },
 138         'params': {
 139             'skip_download': True,
 140         },
 141     }, {
 142         'url': 'http://www.afreecatv.com/player/Player.swf?szType=szBjId=djleegoon&nStationNo=11273158&nBbsNo=13161095&nTitleNo=36327652',
 143         'only_matching': True,
 144     }, {
 145         'url': 'http://vod.afreecatv.com/PLAYER/STATION/15055030',
 146         'only_matching': True,
 147     }]
 148
 149     @staticmethod
 150     def parse_video_key(key):
 151         video_key = {}
 152         m = re.match(r'^(?P<upload_date>\d{8})_\w+_(?P<part>\d+)$', key)
 153         if m:
 154             video_key['upload_date'] = m.group('upload_date')
 155             video_key['part'] = int(m.group('part'))
 156         return video_key
 157
 158     def _real_extract(self, url):
 159         video_id = self._match_id(url)
 160
 161         video_xml = self._download_xml(
 162             'http://afbbs.afreecatv.com:8080/api/video/get_video_info.php',
 163             video_id, query={'nTitleNo': video_id})
 164
 165         video_element = video_xml.findall(compat_xpath('./track/video'))[1]
 166         if video_element is None or video_element.text is None:
 167             raise ExtractorError('Specified AfreecaTV video does not exist',
 168                                  expected=True)
 169
 170         video_url = video_element.text.strip()
 171
 172         title = xpath_text(video_xml, './track/title', 'title', fatal=True)
 173
 174         uploader = xpath_text(video_xml, './track/nickname', 'uploader')
 175         uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id')
 176         duration = int_or_none(xpath_text(
 177             video_xml, './track/duration', 'duration'))
 178         thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail')
 179
 180         common_entry = {
 181             'uploader': uploader,
 182             'uploader_id': uploader_id,
 183             'thumbnail': thumbnail,
 184         }
 185
 186         info = common_entry.copy()
 187         info.update({
 188             'id': video_id,
 189             'title': title,
 190             'duration': duration,
 191         })
 192
 193         if not video_url:
 194             entries = []
 195             file_elements = video_element.findall(compat_xpath('./file'))
 196             one = len(file_elements) == 1
 197             for file_num, file_element in enumerate(file_elements, start=1):
 198                 file_url = file_element.text
 199                 if not file_url:
 200                     continue
 201                 key = file_element.get('key', '')
 202                 upload_date = self._search_regex(
 203                     r'^(\d{8})_', key, 'upload date', default=None)
 204                 file_duration = int_or_none(file_element.get('duration'))
 205                 format_id = key if key else '%s_%s' % (video_id, file_num)
 206                 formats = self._extract_m3u8_formats(
 207                     file_url, video_id, 'mp4', entry_protocol='m3u8_native',
 208                     m3u8_id='hls',
 209                     note='Downloading part %d m3u8 information' % file_num)
 210                 file_info = common_entry.copy()
 211                 file_info.update({
 212                     'id': format_id,
 213                     'title': title if one else '%s (part %d)' % (title, file_num),
 214                     'upload_date': upload_date,
 215                     'duration': file_duration,
 216                     'formats': formats,
 217                 })
 218                 entries.append(file_info)
 219             entries_info = info.copy()
 220             entries_info.update({
 221                 '_type': 'multi_video',
 222                 'entries': entries,
 223             })
 224             return entries_info
 225
 226         info = {
 227             'id': video_id,
 228             'title': title,
 229             'uploader': uploader,
 230             'uploader_id': uploader_id,
 231             'duration': duration,
 232             'thumbnail': thumbnail,
 233         }
 234
 235         if determine_ext(video_url) == 'm3u8':
 236             info['formats'] = self._extract_m3u8_formats(
 237                 video_url, video_id, 'mp4', entry_protocol='m3u8_native',
 238                 m3u8_id='hls')
 239         else:
 240             app, playpath = video_url.split('mp4:')
 241             info.update({
 242                 'url': app,
 243                 'ext': 'flv',
 244                 'play_path': 'mp4:' + playpath,
 245                 'rtmp_live': True,  # downloading won't end without this
 246             })
 247
 248         return info
 249
 250
 251 class AfreecaTVGlobalIE(AfreecaTVIE):
 252     IE_NAME = 'afreecatv:global'
 253     _VALID_URL = r'https?://(?:www\.)?afreeca\.tv/(?P<channel_id>\d+)(?:/v/(?P<video_id>\d+))?'
 254     _TESTS = [{
 255         'url': 'http://afreeca.tv/36853014/v/58301',
 256         'info_dict': {
 257             'id': '58301',
 258             'title': 'tryhard top100',
 259             'uploader_id': '36853014',
 260             'uploader': 'makgi Hearthstone Live!',
 261         },
 262         'playlist_count': 3,
 263     }]
 264
 265     def _real_extract(self, url):
 266         channel_id, video_id = re.match(self._VALID_URL, url).groups()
 267         video_type = 'video' if video_id else 'live'
 268         query = {
 269             'pt': 'view',
 270             'bid': channel_id,
 271         }
 272         if video_id:
 273             query['vno'] = video_id
 274         video_data = self._download_json(
 275             'http://api.afreeca.tv/%s/view_%s.php' % (video_type, video_type),
 276             video_id or channel_id, query=query)['channel']
 277
 278         if video_data.get('result') != 1:
 279             raise ExtractorError('%s said: %s' % (self.IE_NAME, video_data['remsg']))
 280
 281         title = video_data['title']
 282
 283         info = {
 284             'thumbnail': video_data.get('thumb'),
 285             'view_count': int_or_none(video_data.get('vcnt')),
 286             'age_limit': int_or_none(video_data.get('grade')),
 287             'uploader_id': channel_id,
 288             'uploader': video_data.get('cname'),
 289         }
 290
 291         if video_id:
 292             entries = []
 293             for i, f in enumerate(video_data.get('flist', [])):
 294                 video_key = self.parse_video_key(f.get('key', ''))
 295                 f_url = f.get('file')
 296                 if not video_key or not f_url:
 297                     continue
 298                 entries.append({
 299                     'id': '%s_%s' % (video_id, video_key.get('part', i + 1)),
 300                     'title': title,
 301                     'upload_date': video_key.get('upload_date'),
 302                     'duration': int_or_none(f.get('length')),
 303                     'url': f_url,
 304                     'protocol': 'm3u8_native',
 305                     'ext': 'mp4',
 306                 })
 307
 308             info.update({
 309                 'id': video_id,
 310                 'title': title,
 311                 'duration': int_or_none(video_data.get('length')),
 312             })
 313             if len(entries) > 1:
 314                 info['_type'] = 'multi_video'
 315                 info['entries'] = entries
 316             elif len(entries) == 1:
 317                 i = entries[0].copy()
 318                 i.update(info)
 319                 info = i
 320         else:
 321             formats = []
 322             for s in video_data.get('strm', []):
 323                 s_url = s.get('purl')
 324                 if not s_url:
 325                     continue
 326                 stype = s.get('stype')
 327                 if stype == 'HLS':
 328                     formats.extend(self._extract_m3u8_formats(
 329                         s_url, channel_id, 'mp4', m3u8_id=stype, fatal=False))
 330                 elif stype == 'RTMP':
 331                     format_id = [stype]
 332                     label = s.get('label')
 333                     if label:
 334                         format_id.append(label)
 335                     formats.append({
 336                         'format_id': '-'.join(format_id),
 337                         'url': s_url,
 338                         'tbr': int_or_none(s.get('bps')),
 339                         'height': int_or_none(s.get('brt')),
 340                         'ext': 'flv',
 341                         'rtmp_live': True,
 342                     })
 343             self._sort_formats(formats)
 344
 345             info.update({
 346                 'id': channel_id,
 347                 'title': self._live_title(title),
 348                 'is_live': True,
 349                 'formats': formats,
 350             })
 351
 352         return info