_ Git - youtube-dl/blob - youtube_dl/extractor/bilibili.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     int_or_none,
   9     unified_strdate,
  10     ExtractorError,
  11 )
  12
  13
  14 class BiliBiliIE(InfoExtractor):
  15     _VALID_URL = r'http://www\.bilibili\.(?:tv|com)/video/av(?P<id>[0-9]+)/'
  16
  17     _TEST = {
  18         'url': 'http://www.bilibili.tv/video/av1074402/',
  19         'md5': '2c301e4dab317596e837c3e7633e7d86',
  20         'info_dict': {
  21             'id': '1074402',
  22             'ext': 'flv',
  23             'title': '【金坷垃】金泡沫',
  24             'duration': 308,
  25             'upload_date': '20140420',
  26             'thumbnail': 're:^https?://.+\.jpg',
  27         },
  28     }
  29
  30     def _real_extract(self, url):
  31         video_id = self._match_id(url)
  32         webpage = self._download_webpage(url, video_id)
  33
  34         if self._search_regex(r'(此视频不存在或被删除)', webpage, 'error message', default=None):
  35             raise ExtractorError('The video does not exist or was deleted', expected=True)
  36         video_code = self._search_regex(
  37             r'(?s)<div itemprop="video".*?>(.*?)</div>', webpage, 'video code')
  38
  39         title = self._html_search_meta(
  40             'media:title', video_code, 'title', fatal=True)
  41         duration_str = self._html_search_meta(
  42             'duration', video_code, 'duration')
  43         if duration_str is None:
  44             duration = None
  45         else:
  46             duration_mobj = re.match(
  47                 r'^T(?:(?P<hours>[0-9]+)H)?(?P<minutes>[0-9]+)M(?P<seconds>[0-9]+)S$',
  48                 duration_str)
  49             duration = (
  50                 int_or_none(duration_mobj.group('hours'), default=0) * 3600 +
  51                 int(duration_mobj.group('minutes')) * 60 +
  52                 int(duration_mobj.group('seconds')))
  53         upload_date = unified_strdate(self._html_search_meta(
  54             'uploadDate', video_code, fatal=False))
  55         thumbnail = self._html_search_meta(
  56             'thumbnailUrl', video_code, 'thumbnail', fatal=False)
  57
  58         cid = self._search_regex(r'cid=(\d+)', webpage, 'cid')
  59
  60         lq_doc = self._download_xml(
  61             'http://interface.bilibili.com/v_cdn_play?appkey=1&cid=%s' % cid,
  62             video_id,
  63             note='Downloading LQ video info'
  64         )
  65         lq_durl = lq_doc.find('./durl')
  66         formats = [{
  67             'format_id': 'lq',
  68             'quality': 1,
  69             'url': lq_durl.find('./url').text,
  70             'filesize': int_or_none(
  71                 lq_durl.find('./size'), get_attr='text'),
  72         }]
  73
  74         hq_doc = self._download_xml(
  75             'http://interface.bilibili.com/playurl?appkey=1&cid=%s' % cid,
  76             video_id,
  77             note='Downloading HQ video info',
  78             fatal=False,
  79         )
  80         if hq_doc is not False:
  81             hq_durl = hq_doc.find('./durl')
  82             formats.append({
  83                 'format_id': 'hq',
  84                 'quality': 2,
  85                 'ext': 'flv',
  86                 'url': hq_durl.find('./url').text,
  87                 'filesize': int_or_none(
  88                     hq_durl.find('./size'), get_attr='text'),
  89             })
  90
  91         self._sort_formats(formats)
  92         return {
  93             'id': video_id,
  94             'title': title,
  95             'formats': formats,
  96             'duration': duration,
  97             'upload_date': upload_date,
  98             'thumbnail': thumbnail,
  99         }