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 import json
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_etree_fromstring,
  11 )
  12 from ..utils import (
  13     int_or_none,
  14     unified_strdate,
  15     ExtractorError,
  16 )
  17
  18
  19 class BiliBiliIE(InfoExtractor):
  20     _VALID_URL = r'http://www\.bilibili\.(?:tv|com)/video/av(?P<id>[0-9]+)/'
  21
  22     _TESTS = [{
  23         'url': 'http://www.bilibili.tv/video/av1074402/',
  24         'md5': '2c301e4dab317596e837c3e7633e7d86',
  25         'info_dict': {
  26             'id': '1074402_part1',
  27             'ext': 'flv',
  28             'title': '【金坷垃】金泡沫',
  29             'duration': 308,
  30             'upload_date': '20140420',
  31             'thumbnail': 're:^https?://.+\.jpg',
  32         },
  33     }, {
  34         'url': 'http://www.bilibili.com/video/av1041170/',
  35         'info_dict': {
  36             'id': '1041170',
  37             'title': '【BD1080P】刀语【诸神&异域】',
  38         },
  39         'playlist_count': 9,
  40     }]
  41
  42     def _real_extract(self, url):
  43         video_id = self._match_id(url)
  44         webpage = self._download_webpage(url, video_id)
  45
  46         if '(此视频不存在或被删除)' in webpage:
  47             raise ExtractorError(
  48                 'The video does not exist or was deleted', expected=True)
  49
  50         if '>你没有权限浏览！ 由于版权相关问题 我们不对您所在的地区提供服务<' in webpage:
  51             raise ExtractorError(
  52                 'The video is not available in your region due to copyright reasons',
  53                 expected=True)
  54
  55         video_code = self._search_regex(
  56             r'(?s)<div itemprop="video".*?>(.*?)</div>', webpage, 'video code')
  57
  58         title = self._html_search_meta(
  59             'media:title', video_code, 'title', fatal=True)
  60         duration_str = self._html_search_meta(
  61             'duration', video_code, 'duration')
  62         if duration_str is None:
  63             duration = None
  64         else:
  65             duration_mobj = re.match(
  66                 r'^T(?:(?P<hours>[0-9]+)H)?(?P<minutes>[0-9]+)M(?P<seconds>[0-9]+)S$',
  67                 duration_str)
  68             duration = (
  69                 int_or_none(duration_mobj.group('hours'), default=0) * 3600 +
  70                 int(duration_mobj.group('minutes')) * 60 +
  71                 int(duration_mobj.group('seconds')))
  72         upload_date = unified_strdate(self._html_search_meta(
  73             'uploadDate', video_code, fatal=False))
  74         thumbnail = self._html_search_meta(
  75             'thumbnailUrl', video_code, 'thumbnail', fatal=False)
  76
  77         cid = self._search_regex(r'cid=(\d+)', webpage, 'cid')
  78
  79         entries = []
  80
  81         lq_page = self._download_webpage(
  82             'http://interface.bilibili.com/v_cdn_play?appkey=1&cid=%s' % cid,
  83             video_id,
  84             note='Downloading LQ video info'
  85         )
  86         try:
  87             err_info = json.loads(lq_page)
  88             raise ExtractorError(
  89                 'BiliBili said: ' + err_info['error_text'], expected=True)
  90         except ValueError:
  91             pass
  92
  93         lq_doc = compat_etree_fromstring(lq_page)
  94         lq_durls = lq_doc.findall('./durl')
  95
  96         hq_doc = self._download_xml(
  97             'http://interface.bilibili.com/playurl?appkey=1&cid=%s' % cid,
  98             video_id,
  99             note='Downloading HQ video info',
 100             fatal=False,
 101         )
 102         if hq_doc is not False:
 103             hq_durls = hq_doc.findall('./durl')
 104             assert len(lq_durls) == len(hq_durls)
 105         else:
 106             hq_durls = itertools.repeat(None)
 107
 108         i = 1
 109         for lq_durl, hq_durl in zip(lq_durls, hq_durls):
 110             formats = [{
 111                 'format_id': 'lq',
 112                 'quality': 1,
 113                 'url': lq_durl.find('./url').text,
 114                 'filesize': int_or_none(
 115                     lq_durl.find('./size'), get_attr='text'),
 116             }]
 117             if hq_durl is not None:
 118                 formats.append({
 119                     'format_id': 'hq',
 120                     'quality': 2,
 121                     'ext': 'flv',
 122                     'url': hq_durl.find('./url').text,
 123                     'filesize': int_or_none(
 124                         hq_durl.find('./size'), get_attr='text'),
 125                 })
 126             self._sort_formats(formats)
 127
 128             entries.append({
 129                 'id': '%s_part%d' % (video_id, i),
 130                 'title': title,
 131                 'formats': formats,
 132                 'duration': duration,
 133                 'upload_date': upload_date,
 134                 'thumbnail': thumbnail,
 135             })
 136
 137             i += 1
 138
 139         return {
 140             '_type': 'multi_video',
 141             'entries': entries,
 142             'id': video_id,
 143             'title': title
 144         }