X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fafreecatv.py;h=0f45358042bd4dbe1b2c768169c20ee01c51f539;hb=02f6ccbce3a50d8db3eac06a5820347cf674ca86;hp=2c58f4617770d3b48267a103b0dd96b2e4d33ad8;hpb=839728f5bfad3a6166be3839009b13963f00dfac;p=youtube-dl diff --git a/youtube_dl/extractor/afreecatv.py b/youtube_dl/extractor/afreecatv.py index 2c58f4617..0f4535804 100644 --- a/youtube_dl/extractor/afreecatv.py +++ b/youtube_dl/extractor/afreecatv.py @@ -175,10 +175,27 @@ class AfreecaTVIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + + if re.search(r'alert\(["\']This video has been deleted', webpage): + raise ExtractorError( + 'Video %s has been deleted' % video_id, expected=True) + + station_id = self._search_regex( + r'nStationNo\s*=\s*(\d+)', webpage, 'station') + bbs_id = self._search_regex( + r'nBbsNo\s*=\s*(\d+)', webpage, 'bbs') + video_id = self._search_regex( + r'nTitleNo\s*=\s*(\d+)', webpage, 'title', default=video_id) + print(video_id, station_id, bbs_id) video_xml = self._download_xml( 'http://afbbs.afreecatv.com:8080/api/video/get_video_info.php', - video_id, query={ + video_id, headers={ + 'Referer': url, + }, query={ 'nTitleNo': video_id, + 'nStationNo': station_id, + 'nBbsNo': bbs_id, 'partialView': 'SKIP_ADULT', }) @@ -187,10 +204,10 @@ class AfreecaTVIE(InfoExtractor): raise ExtractorError( '%s said: %s' % (self.IE_NAME, flag), expected=True) - video_element = video_xml.findall(compat_xpath('./track/video'))[1] + video_element = video_xml.findall(compat_xpath('./track/video'))[-1] if video_element is None or video_element.text is None: - raise ExtractorError('Specified AfreecaTV video does not exist', - expected=True) + raise ExtractorError( + 'Video %s video does not exist' % video_id, expected=True) video_url = video_element.text.strip() @@ -228,10 +245,19 @@ class AfreecaTVIE(InfoExtractor): r'^(\d{8})_', key, 'upload date', default=None) file_duration = int_or_none(file_element.get('duration')) format_id = key if key else '%s_%s' % (video_id, file_num) - formats = self._extract_m3u8_formats( - file_url, video_id, 'mp4', entry_protocol='m3u8_native', - m3u8_id='hls', - note='Downloading part %d m3u8 information' % file_num) + if determine_ext(file_url) == 'm3u8': + formats = self._extract_m3u8_formats( + file_url, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls', + note='Downloading part %d m3u8 information' % file_num) + else: + formats = [{ + 'url': file_url, + 'format_id': 'http', + }] + if not formats: + continue + self._sort_formats(formats) file_info = common_entry.copy() file_info.update({ 'id': format_id, @@ -271,107 +297,3 @@ class AfreecaTVIE(InfoExtractor): }) return info - - -class AfreecaTVGlobalIE(AfreecaTVIE): - IE_NAME = 'afreecatv:global' - _VALID_URL = r'https?://(?:www\.)?afreeca\.tv/(?P\d+)(?:/v/(?P\d+))?' - _TESTS = [{ - 'url': 'http://afreeca.tv/36853014/v/58301', - 'info_dict': { - 'id': '58301', - 'title': 'tryhard top100', - 'uploader_id': '36853014', - 'uploader': 'makgi Hearthstone Live!', - }, - 'playlist_count': 3, - }] - - def _real_extract(self, url): - channel_id, video_id = re.match(self._VALID_URL, url).groups() - video_type = 'video' if video_id else 'live' - query = { - 'pt': 'view', - 'bid': channel_id, - } - if video_id: - query['vno'] = video_id - video_data = self._download_json( - 'http://api.afreeca.tv/%s/view_%s.php' % (video_type, video_type), - video_id or channel_id, query=query)['channel'] - - if video_data.get('result') != 1: - raise ExtractorError('%s said: %s' % (self.IE_NAME, video_data['remsg'])) - - title = video_data['title'] - - info = { - 'thumbnail': video_data.get('thumb'), - 'view_count': int_or_none(video_data.get('vcnt')), - 'age_limit': int_or_none(video_data.get('grade')), - 'uploader_id': channel_id, - 'uploader': video_data.get('cname'), - } - - if video_id: - entries = [] - for i, f in enumerate(video_data.get('flist', [])): - video_key = self.parse_video_key(f.get('key', '')) - f_url = f.get('file') - if not video_key or not f_url: - continue - entries.append({ - 'id': '%s_%s' % (video_id, video_key.get('part', i + 1)), - 'title': title, - 'upload_date': video_key.get('upload_date'), - 'duration': int_or_none(f.get('length')), - 'url': f_url, - 'protocol': 'm3u8_native', - 'ext': 'mp4', - }) - - info.update({ - 'id': video_id, - 'title': title, - 'duration': int_or_none(video_data.get('length')), - }) - if len(entries) > 1: - info['_type'] = 'multi_video' - info['entries'] = entries - elif len(entries) == 1: - i = entries[0].copy() - i.update(info) - info = i - else: - formats = [] - for s in video_data.get('strm', []): - s_url = s.get('purl') - if not s_url: - continue - stype = s.get('stype') - if stype == 'HLS': - formats.extend(self._extract_m3u8_formats( - s_url, channel_id, 'mp4', m3u8_id=stype, fatal=False)) - elif stype == 'RTMP': - format_id = [stype] - label = s.get('label') - if label: - format_id.append(label) - formats.append({ - 'format_id': '-'.join(format_id), - 'url': s_url, - 'tbr': int_or_none(s.get('bps')), - 'height': int_or_none(s.get('brt')), - 'ext': 'flv', - 'rtmp_live': True, - }) - self._sort_formats(formats) - - info.update({ - 'id': channel_id, - 'title': self._live_title(title), - 'is_live': True, - 'formats': formats, - }) - - return info