X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fafreecatv.py;h=b56abb1e604b78cfd5a0667e9703f216f8a9f8da;hb=HEAD;hp=e0a0f7c57b83c7a715e7a39c16a8a71df1cbb500;hpb=3d2c2752c5cd70fc7f9cebe8c4683a1de626017d;p=youtube-dl diff --git a/youtube_dl/extractor/afreecatv.py b/youtube_dl/extractor/afreecatv.py index e0a0f7c57..b56abb1e6 100644 --- a/youtube_dl/extractor/afreecatv.py +++ b/youtube_dl/extractor/afreecatv.py @@ -4,15 +4,13 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import ( - compat_urllib_parse_urlparse, - compat_urlparse, -) +from ..compat import compat_xpath from ..utils import ( + determine_ext, ExtractorError, int_or_none, - update_url_query, - xpath_element, + url_or_none, + urlencode_postdata, xpath_text, ) @@ -32,6 +30,7 @@ class AfreecaTVIE(InfoExtractor): ) (?P\d+) ''' + _NETRC_MACHINE = 'afreecatv' _TESTS = [{ 'url': 'http://live.afreecatv.com:8079/app/index.cgi?szType=read_ucc_bbs&szBjId=dailyapril&nStationNo=16711924&nBbsNo=18605867&nTitleNo=36164052&szSkin=', 'md5': 'f72c89fe7ecc14c1b5ce506c4996046e', @@ -43,7 +42,8 @@ class AfreecaTVIE(InfoExtractor): 'uploader': 'dailyapril', 'uploader_id': 'dailyapril', 'upload_date': '20160503', - } + }, + 'skip': 'Video is gone', }, { 'url': 'http://afbbs.afreecatv.com:8080/app/read_ucc_bbs.cgi?nStationNo=16711924&nTitleNo=36153164&szBjId=dailyapril&nBbsNo=18605867', 'info_dict': { @@ -71,6 +71,93 @@ class AfreecaTVIE(InfoExtractor): 'upload_date': '20160502', }, }], + 'skip': 'Video is gone', + }, { + 'url': 'http://vod.afreecatv.com/PLAYER/STATION/18650793', + 'info_dict': { + 'id': '18650793', + 'ext': 'mp4', + 'title': '오늘은 다르다! 쏘님의 우월한 위아래~ 댄스리액션!', + 'thumbnail': r're:^https?://.*\.jpg$', + 'uploader': '윈아디', + 'uploader_id': 'badkids', + 'duration': 107, + }, + 'params': { + 'skip_download': True, + }, + }, { + 'url': 'http://vod.afreecatv.com/PLAYER/STATION/10481652', + 'info_dict': { + 'id': '10481652', + 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'", + 'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$', + 'uploader': 'dailyapril', + 'uploader_id': 'dailyapril', + 'duration': 6492, + }, + 'playlist_count': 2, + 'playlist': [{ + 'md5': 'd8b7c174568da61d774ef0203159bf97', + 'info_dict': { + 'id': '20160502_c4c62b9d_174361386_1', + 'ext': 'mp4', + 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!' (part 1)", + 'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$', + 'uploader': 'dailyapril', + 'uploader_id': 'dailyapril', + 'upload_date': '20160502', + 'duration': 3601, + }, + }, { + 'md5': '58f2ce7f6044e34439ab2d50612ab02b', + 'info_dict': { + 'id': '20160502_39e739bb_174361386_2', + 'ext': 'mp4', + 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!' (part 2)", + 'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$', + 'uploader': 'dailyapril', + 'uploader_id': 'dailyapril', + 'upload_date': '20160502', + 'duration': 2891, + }, + }], + 'params': { + 'skip_download': True, + }, + }, { + # non standard key + 'url': 'http://vod.afreecatv.com/PLAYER/STATION/20515605', + 'info_dict': { + 'id': '20170411_BE689A0E_190960999_1_2_h', + 'ext': 'mp4', + 'title': '혼자사는여자집', + 'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$', + 'uploader': '♥이슬이', + 'uploader_id': 'dasl8121', + 'upload_date': '20170411', + 'duration': 213, + }, + 'params': { + 'skip_download': True, + }, + }, { + # PARTIAL_ADULT + 'url': 'http://vod.afreecatv.com/PLAYER/STATION/32028439', + 'info_dict': { + 'id': '20180327_27901457_202289533_1', + 'ext': 'mp4', + 'title': '[생]빨개요♥ (part 1)', + 'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$', + 'uploader': '[SA]서아', + 'uploader_id': 'bjdyrksu', + 'upload_date': '20180327', + 'duration': 3601, + }, + 'params': { + 'skip_download': True, + }, + 'expected_warnings': ['adult content'], }, { 'url': 'http://www.afreecatv.com/player/Player.swf?szType=szBjId=djleegoon&nStationNo=11273158&nBbsNo=13161095&nTitleNo=36327652', 'only_matching': True, @@ -85,166 +172,196 @@ class AfreecaTVIE(InfoExtractor): m = re.match(r'^(?P\d{8})_\w+_(?P\d+)$', key) if m: video_key['upload_date'] = m.group('upload_date') - video_key['part'] = m.group('part') + video_key['part'] = int(m.group('part')) return video_key - def _real_extract(self, url): - video_id = self._match_id(url) - parsed_url = compat_urllib_parse_urlparse(url) - info_url = compat_urlparse.urlunparse(parsed_url._replace( - netloc='afbbs.afreecatv.com:8080', - path='/api/video/get_video_info.php')) - - video_xml = self._download_xml( - update_url_query(info_url, {'nTitleNo': video_id}), video_id) + def _real_initialize(self): + self._login() - if xpath_element(video_xml, './track/video/file') is None: - raise ExtractorError('Specified AfreecaTV video does not exist', - expected=True) + def _login(self): + username, password = self._get_login_info() + if username is None: + return - title = xpath_text(video_xml, './track/title', 'title') - uploader = xpath_text(video_xml, './track/nickname', 'uploader') - uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id') - duration = int_or_none(xpath_text(video_xml, './track/duration', - 'duration')) - thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail') + login_form = { + 'szWork': 'login', + 'szType': 'json', + 'szUid': username, + 'szPassword': password, + 'isSaveId': 'false', + 'szScriptVar': 'oLoginRet', + 'szAction': '', + } - entries = [] - for i, video_file in enumerate(video_xml.findall('./track/video/file')): - video_key = self.parse_video_key(video_file.get('key', '')) - if not video_key: - 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(video_file.get('duration')), - 'url': video_file.text, - }) + response = self._download_json( + 'https://login.afreecatv.com/app/LoginAction.php', None, + 'Logging in', data=urlencode_postdata(login_form)) - info = { - 'id': video_id, - 'title': title, - 'uploader': uploader, - 'uploader_id': uploader_id, - 'duration': duration, - 'thumbnail': thumbnail, + _ERRORS = { + -4: 'Your account has been suspended due to a violation of our terms and policies.', + -5: 'https://member.afreecatv.com/app/user_delete_progress.php', + -6: 'https://login.afreecatv.com/membership/changeMember.php', + -8: "Hello! AfreecaTV here.\nThe username you have entered belongs to \n an account that requires a legal guardian's consent. \nIf you wish to use our services without restriction, \nplease make sure to go through the necessary verification process.", + -9: 'https://member.afreecatv.com/app/pop_login_block.php', + -11: 'https://login.afreecatv.com/afreeca/second_login.php', + -12: 'https://member.afreecatv.com/app/user_security.php', + 0: 'The username does not exist or you have entered the wrong password.', + -1: 'The username does not exist or you have entered the wrong password.', + -3: 'You have entered your username/password incorrectly.', + -7: 'You cannot use your Global AfreecaTV account to access Korean AfreecaTV.', + -10: 'Sorry for the inconvenience. \nYour account has been blocked due to an unauthorized access. \nPlease contact our Help Center for assistance.', + -32008: 'You have failed to log in. Please contact our Help Center.', } - if len(entries) > 1: - info['_type'] = 'multi_video' - info['entries'] = entries - elif len(entries) == 1: - info['url'] = entries[0]['url'] - info['upload_date'] = entries[0].get('upload_date') - else: + result = int_or_none(response.get('RESULT')) + if result != 1: + error = _ERRORS.get(result, 'You have failed to log in.') raise ExtractorError( - 'No files found for the specified AfreecaTV video, either' - ' the URL is incorrect or the video has been made private.', + 'Unable to login: %s said: %s' % (self.IE_NAME, error), expected=True) - return info + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) -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, - }] + if re.search(r'alert\(["\']This video has been deleted', webpage): + raise ExtractorError( + 'Video %s has been deleted' % video_id, expected=True) - 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'] + 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) - if video_data.get('result') != 1: - raise ExtractorError('%s said: %s' % (self.IE_NAME, video_data['remsg'])) + partial_view = False + for _ in range(2): + query = { + 'nTitleNo': video_id, + 'nStationNo': station_id, + 'nBbsNo': bbs_id, + } + if partial_view: + query['partialView'] = 'SKIP_ADULT' + video_xml = self._download_xml( + 'http://afbbs.afreecatv.com:8080/api/video/get_video_info.php', + video_id, 'Downloading video info XML%s' + % (' (skipping adult)' if partial_view else ''), + video_id, headers={ + 'Referer': url, + }, query=query) - title = video_data['title'] + flag = xpath_text(video_xml, './track/flag', 'flag', default=None) + if flag and flag == 'SUCCEED': + break + if flag == 'PARTIAL_ADULT': + self._downloader.report_warning( + 'In accordance with local laws and regulations, underage users are restricted from watching adult content. ' + 'Only content suitable for all ages will be downloaded. ' + 'Provide account credentials if you wish to download restricted content.') + partial_view = True + continue + elif flag == 'ADULT': + error = 'Only users older than 19 are able to watch this video. Provide account credentials to download this content.' + else: + error = flag + raise ExtractorError( + '%s said: %s' % (self.IE_NAME, error), expected=True) + else: + raise ExtractorError('Unable to download video info') - 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'), + video_element = video_xml.findall(compat_xpath('./track/video'))[-1] + if video_element is None or video_element.text is None: + raise ExtractorError( + 'Video %s does not exist' % video_id, expected=True) + + video_url = video_element.text.strip() + + title = xpath_text(video_xml, './track/title', 'title', fatal=True) + + uploader = xpath_text(video_xml, './track/nickname', 'uploader') + uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id') + duration = int_or_none(xpath_text( + video_xml, './track/duration', 'duration')) + thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail') + + common_entry = { + 'uploader': uploader, + 'uploader_id': uploader_id, + 'thumbnail': thumbnail, } - if video_id: + info = common_entry.copy() + info.update({ + 'id': video_id, + 'title': title, + 'duration': duration, + }) + + if not video_url: 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: + file_elements = video_element.findall(compat_xpath('./file')) + one = len(file_elements) == 1 + for file_num, file_element in enumerate(file_elements, start=1): + file_url = url_or_none(file_element.text) + if not file_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', + key = file_element.get('key', '') + upload_date = self._search_regex( + 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) + 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, + 'title': title if one else '%s (part %d)' % (title, file_num), + 'upload_date': upload_date, + 'duration': file_duration, + 'formats': formats, }) - - info.update({ - 'id': video_id, - 'title': title, - 'duration': int_or_none(video_data.get('length')), + entries.append(file_info) + entries_info = info.copy() + entries_info.update({ + '_type': 'multi_video', + 'entries': entries, }) - 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) + return entries_info + + info = { + 'id': video_id, + 'title': title, + 'uploader': uploader, + 'uploader_id': uploader_id, + 'duration': duration, + 'thumbnail': thumbnail, + } + if determine_ext(video_url) == 'm3u8': + info['formats'] = self._extract_m3u8_formats( + video_url, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls') + else: + app, playpath = video_url.split('mp4:') info.update({ - 'id': channel_id, - 'title': self._live_title(title), - 'is_live': True, - 'formats': formats, + 'url': app, + 'ext': 'flv', + 'play_path': 'mp4:' + playpath, + 'rtmp_live': True, # downloading won't end without this }) return info