diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 8235d42b6..2c2e2230f 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -20,6 +20,7 @@ from .auengine import AUEngineIE from .bambuser import BambuserIE, BambuserChannelIE from .bandcamp import BandcampIE, BandcampAlbumIE from .bbccouk import BBCCoUkIE +from .bilibili import BiliBiliIE from .blinkx import BlinkxIE from .bliptv import BlipTVIE, BlipTVUserIE from .bloomberg import BloombergIE diff --git a/youtube_dl/extractor/bilibili.py b/youtube_dl/extractor/bilibili.py new file mode 100644 index 000000000..45067b944 --- /dev/null +++ b/youtube_dl/extractor/bilibili.py @@ -0,0 +1,106 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + compat_parse_qs, + ExtractorError, + int_or_none, + unified_strdate, +) + + +class BiliBiliIE(InfoExtractor): + _VALID_URL = r'http://www\.bilibili\.tv/video/av(?P[0-9]+)/' + + _TEST = { + 'url': 'http://www.bilibili.tv/video/av1074402/', + 'md5': '2c301e4dab317596e837c3e7633e7d86', + 'info_dict': { + 'id': '1074402', + 'ext': 'flv', + 'title': '【金坷垃】金泡沫', + 'duration': 308, + 'upload_date': '20140420', + 'thumbnail': 're:^https?://.+\.jpg', + }, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + video_id = mobj.group('id') + + webpage = self._download_webpage(url, video_id) + video_code = self._search_regex( + r'(?s)
(.*?)
', webpage, 'video code') + + title = self._html_search_meta( + 'media:title', video_code, 'title', fatal=True) + duration_str = self._html_search_meta( + 'duration', video_code, 'duration') + if duration_str is None: + duration = None + else: + duration_mobj = re.match( + r'^T(?:(?P[0-9]+)H)?(?P[0-9]+)M(?P[0-9]+)S$', + duration_str) + duration = ( + int_or_none(duration_mobj.group('hours'), default=0) * 3600 + + int(duration_mobj.group('minutes')) * 60 + + int(duration_mobj.group('seconds'))) + upload_date = unified_strdate(self._html_search_meta( + 'uploadDate', video_code, fatal=False)) + thumbnail = self._html_search_meta( + 'thumbnailUrl', video_code, 'thumbnail', fatal=False) + + player_params = compat_parse_qs(self._html_search_regex( + r'