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