X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fvlive.py;h=540246c79d8900feaada9921eb600a2f6695828d;hb=2786818c3360bcadc21109a9f740fba8f698c8a8;hp=baf39bb2cea714fb1578d80b4b8a83c9cd67e568;hpb=d6712378e73951bede475569c887a1ac73f660a9;p=youtube-dl diff --git a/youtube_dl/extractor/vlive.py b/youtube_dl/extractor/vlive.py index baf39bb2c..540246c79 100644 --- a/youtube_dl/extractor/vlive.py +++ b/youtube_dl/extractor/vlive.py @@ -1,11 +1,16 @@ # coding: utf-8 from __future__ import unicode_literals +import re + from .common import InfoExtractor from ..utils import ( dict_get, + ExtractorError, float_or_none, int_or_none, + remove_start, + urlencode_postdata, ) from ..compat import compat_urllib_parse_urlencode @@ -13,17 +18,30 @@ from ..compat import compat_urllib_parse_urlencode class VLiveIE(InfoExtractor): IE_NAME = 'vlive' _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P[0-9]+)' - _TEST = { + _TESTS = [{ 'url': 'http://www.vlive.tv/video/1326', 'md5': 'cc7314812855ce56de70a06a27314983', 'info_dict': { 'id': '1326', 'ext': 'mp4', - 'title': "[V] Girl's Day's Broadcast", + 'title': "[V LIVE] Girl's Day's Broadcast", 'creator': "Girl's Day", 'view_count': int, }, - } + }, { + 'url': 'http://www.vlive.tv/video/16937', + 'info_dict': { + 'id': '16937', + 'ext': 'mp4', + 'title': '[V LIVE] 첸백시 걍방', + 'creator': 'EXO', + 'view_count': int, + 'subtitles': 'mincount:12', + }, + 'params': { + 'skip_download': True, + }, + }] def _real_extract(self, url): video_id = self._match_id(url) @@ -31,16 +49,87 @@ class VLiveIE(InfoExtractor): webpage = self._download_webpage( 'http://www.vlive.tv/video/%s' % video_id, video_id) - long_video_id = self._search_regex( - r'vlive\.tv\.video\.ajax\.request\.handler\.init\(\s*"[0-9]+"\s*,\s*"[^"]*"\s*,\s*"([^"]+)"', - webpage, 'long video id') + VIDEO_PARAMS_RE = r'\bvlive\.video\.init\(([^)]+)' + VIDEO_PARAMS_FIELD = 'video params' + + params = self._parse_json(self._search_regex( + VIDEO_PARAMS_RE, webpage, VIDEO_PARAMS_FIELD, default=''), video_id, + transform_source=lambda s: '[' + s + ']', fatal=False) + + if not params or len(params) < 7: + params = self._search_regex( + VIDEO_PARAMS_RE, webpage, VIDEO_PARAMS_FIELD) + params = [p.strip(r'"') for p in re.split(r'\s*,\s*', params)] + + status, long_video_id, key = params[2], params[5], params[6] + status = remove_start(status, 'PRODUCT_') + + if status == 'LIVE_ON_AIR' or status == 'BIG_EVENT_ON_AIR': + return self._live(video_id, webpage) + elif status == 'VOD_ON_AIR' or status == 'BIG_EVENT_INTRO': + if long_video_id and key: + return self._replay(video_id, webpage, long_video_id, key) + else: + status = 'COMING_SOON' - key = self._search_regex( - r'vlive\.tv\.video\.ajax\.request\.handler\.init\(\s*"[0-9]+"\s*,\s*"[^"]*"\s*,\s*"[^"]+"\s*,\s*"([^"]+)"', - webpage, 'key') + if status == 'LIVE_END': + raise ExtractorError('Uploading for replay. Please wait...', + expected=True) + elif status == 'COMING_SOON': + raise ExtractorError('Coming soon!', expected=True) + elif status == 'CANCELED': + raise ExtractorError('We are sorry, ' + 'but the live broadcast has been canceled.', + expected=True) + else: + raise ExtractorError('Unknown status %s' % status) + def _get_common_fields(self, webpage): title = self._og_search_title(webpage) + creator = self._html_search_regex( + r']+class="info_area"[^>]*>\s*]*>([^<]+)', + webpage, 'creator', fatal=False) + thumbnail = self._og_search_thumbnail(webpage) + return { + 'title': title, + 'creator': creator, + 'thumbnail': thumbnail, + } + def _live(self, video_id, webpage): + init_page = self._download_webpage( + 'http://www.vlive.tv/video/init/view', + video_id, note='Downloading live webpage', + data=urlencode_postdata({'videoSeq': video_id}), + headers={ + 'Referer': 'http://www.vlive.tv/video/%s' % video_id, + 'Content-Type': 'application/x-www-form-urlencoded' + }) + + live_params = self._search_regex( + r'"liveStreamInfo"\s*:\s*(".*"),', + init_page, 'live stream info') + live_params = self._parse_json(live_params, video_id) + live_params = self._parse_json(live_params, video_id) + + formats = [] + for vid in live_params.get('resolutions', []): + formats.extend(self._extract_m3u8_formats( + vid['cdnUrl'], video_id, 'mp4', + m3u8_id=vid.get('name'), + fatal=False, live=True)) + self._sort_formats(formats) + + info = self._get_common_fields(webpage) + info.update({ + 'title': self._live_title(info['title']), + 'id': video_id, + 'formats': formats, + 'is_live': True, + }) + return info + + def _replay(self, video_id, webpage, long_video_id, key): playinfo = self._download_json( 'http://global.apis.naver.com/rmcnmv/rmcnmv/vod_play_videoInfo.json?%s' % compat_urllib_parse_urlencode({ @@ -62,27 +151,21 @@ class VLiveIE(InfoExtractor): } for vid in playinfo.get('videos', {}).get('list', []) if vid.get('source')] self._sort_formats(formats) - thumbnail = self._og_search_thumbnail(webpage) - creator = self._html_search_regex( - r']+class="info_area"[^>]*>\s*]*>([^<]+)', - webpage, 'creator', fatal=False) - view_count = int_or_none(playinfo.get('meta', {}).get('count')) subtitles = {} for caption in playinfo.get('captions', {}).get('list', []): - lang = dict_get(caption, ('language', 'locale', 'country', 'label')) + lang = dict_get(caption, ('locale', 'language', 'country', 'label')) if lang and caption.get('source'): subtitles[lang] = [{ 'ext': 'vtt', 'url': caption['source']}] - return { + info = self._get_common_fields(webpage) + info.update({ 'id': video_id, - 'title': title, - 'creator': creator, - 'thumbnail': thumbnail, - 'view_count': view_count, 'formats': formats, + 'view_count': view_count, 'subtitles': subtitles, - } + }) + return info