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 import itertools
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     int_or_none,
  10     unified_strdate,
  11     ExtractorError,
  12 )
  13
  14
  15 class BiliBiliIE(InfoExtractor):
  16     _VALID_URL = r'http://www\.bilibili\.(?:tv|com)/video/av(?P<id>[0-9]+)/'
  17
  18     _TESTS = [{
  19         'url': 'http://www.bilibili.tv/video/av1074402/',
  20         'md5': '2c301e4dab317596e837c3e7633e7d86',
  21         'info_dict': {
  22             'id': '1074402_part1',
  23             'ext': 'flv',
  24             'title': '【金坷垃】金泡沫',
  25             'duration': 308,
  26             'upload_date': '20140420',
  27             'thumbnail': 're:^https?://.+\.jpg',
  28         },
  29     }, {
  30         'url': 'http://www.bilibili.com/video/av1041170/',
  31         'info_dict': {
  32             'id': '1041170',
  33             'title': '【BD1080P】刀语【诸神&异域】',
  34         },
  35         'playlist_count': 9,
  36     }]
  37
  38     def _real_extract(self, url):
  39         video_id = self._match_id(url)
  40         webpage = self._download_webpage(url, video_id)
  41
  42         if self._search_regex(r'(此视频不存在或被删除)', webpage, 'error message', default=None):
  43             raise ExtractorError('The video does not exist or was deleted', expected=True)
  44         video_code = self._search_regex(
  45             r'(?s)<div itemprop="video".*?>(.*?)</div>', webpage, 'video code')
  46
  47         title = self._html_search_meta(
  48             'media:title', video_code, 'title', fatal=True)
  49         duration_str = self._html_search_meta(
  50             'duration', video_code, 'duration')
  51         if duration_str is None:
  52             duration = None
  53         else:
  54             duration_mobj = re.match(
  55                 r'^T(?:(?P<hours>[0-9]+)H)?(?P<minutes>[0-9]+)M(?P<seconds>[0-9]+)S$',
  56                 duration_str)
  57             duration = (
  58                 int_or_none(duration_mobj.group('hours'), default=0) * 3600 +
  59                 int(duration_mobj.group('minutes')) * 60 +
  60                 int(duration_mobj.group('seconds')))
  61         upload_date = unified_strdate(self._html_search_meta(
  62             'uploadDate', video_code, fatal=False))
  63         thumbnail = self._html_search_meta(
  64             'thumbnailUrl', video_code, 'thumbnail', fatal=False)
  65
  66         cid = self._search_regex(r'cid=(\d+)', webpage, 'cid')
  67
  68         entries = []
  69
  70         lq_doc = self._download_xml(
  71             'http://interface.bilibili.com/v_cdn_play?appkey=1&cid=%s' % cid,
  72             video_id,
  73             note='Downloading LQ video info'
  74         )
  75         lq_durls = lq_doc.findall('./durl')
  76
  77         hq_doc = self._download_xml(
  78             'http://interface.bilibili.com/playurl?appkey=1&cid=%s' % cid,
  79             video_id,
  80             note='Downloading HQ video info',
  81             fatal=False,
  82         )
  83         hq_durls = hq_doc.findall('./durl') if hq_doc is not False else itertools.repeat(None)
  84
  85         assert len(lq_durls) == len(hq_durls)
  86
  87         i = 1
  88         for lq_durl, hq_durl in zip(lq_durls, hq_durls):
  89             formats = [{
  90                 'format_id': 'lq',
  91                 'quality': 1,
  92                 'url': lq_durl.find('./url').text,
  93                 'filesize': int_or_none(
  94                     lq_durl.find('./size'), get_attr='text'),
  95             }]
  96             if hq_durl:
  97                 formats.append({
  98                     'format_id': 'hq',
  99                     'quality': 2,
 100                     'ext': 'flv',
 101                     'url': hq_durl.find('./url').text,
 102                     'filesize': int_or_none(
 103                         hq_durl.find('./size'), get_attr='text'),
 104                 })
 105             self._sort_formats(formats)
 106
 107             entries.append({
 108                 'id': '%s_part%d' % (video_id, i),
 109                 'title': title,
 110                 'formats': formats,
 111                 'duration': duration,
 112                 'upload_date': upload_date,
 113                 'thumbnail': thumbnail,
 114             })
 115
 116             i += 1
 117
 118         return {
 119             '_type': 'multi_video',
 120             'entries': entries,
 121             'id': video_id,
 122             'title': title
 123         }