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