Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/xuite.py

   1 # -*- coding: utf-8 -*-
   2 from __future__ import unicode_literals
   3
   4 import base64
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_urllib_parse_unquote
   8 from ..utils import (
   9     ExtractorError,
  10     parse_iso8601,
  11     parse_duration,
  12 )
  13
  14
  15 class XuiteIE(InfoExtractor):
  16     IE_DESC = '隨意窩Xuite影音'
  17     _REGEX_BASE64 = r'(?:[A-Za-z0-9+/]{4})*(?:[A-Za-z0-9+/]{2}==|[A-Za-z0-9+/]{3}=)?'
  18     _VALID_URL = r'https?://vlog\.xuite\.net/(?:play|embed)/(?P<id>%s)' % _REGEX_BASE64
  19     _TESTS = [{
  20         # Audio
  21         'url': 'http://vlog.xuite.net/play/RGkzc1ZULTM4NjA5MTQuZmx2',
  22         'md5': 'e79284c87b371424885448d11f6398c8',
  23         'info_dict': {
  24             'id': '3860914',
  25             'ext': 'mp3',
  26             'title': '孤單南半球-歐德陽',
  27             'thumbnail': 're:^https?://.*\.jpg$',
  28             'duration': 247.246,
  29             'timestamp': 1314932940,
  30             'upload_date': '20110902',
  31             'uploader': '阿能',
  32             'uploader_id': '15973816',
  33             'categories': ['個人短片'],
  34         },
  35     }, {
  36         # Video with only one format
  37         'url': 'http://vlog.xuite.net/play/TkRZNjhULTM0NDE2MjkuZmx2',
  38         'md5': 'c45737fc8ac5dc8ac2f92ecbcecf505e',
  39         'info_dict': {
  40             'id': '3441629',
  41             'ext': 'mp4',
  42             'title': '孫燕姿 - 眼淚成詩',
  43             'thumbnail': 're:^https?://.*\.jpg$',
  44             'duration': 217.399,
  45             'timestamp': 1299383640,
  46             'upload_date': '20110306',
  47             'uploader': 'Valen',
  48             'uploader_id': '10400126',
  49             'categories': ['影視娛樂'],
  50         },
  51     }, {
  52         # Video with two formats
  53         'url': 'http://vlog.xuite.net/play/bWo1N1pLLTIxMzAxMTcwLmZsdg==',
  54         'md5': '1166e0f461efe55b62e26a2d2a68e6de',
  55         'info_dict': {
  56             'id': '21301170',
  57             'ext': 'mp4',
  58             'title': '暗殺教室 02',
  59             'description': '字幕:【極影字幕社】',
  60             'thumbnail': 're:^https?://.*\.jpg$',
  61             'duration': 1384.907,
  62             'timestamp': 1421481240,
  63             'upload_date': '20150117',
  64             'uploader': '我只是想認真點',
  65             'uploader_id': '242127761',
  66             'categories': ['電玩動漫'],
  67         },
  68     }, {
  69         'url': 'http://vlog.xuite.net/play/S1dDUjdyLTMyOTc3NjcuZmx2/%E5%AD%AB%E7%87%95%E5%A7%BF-%E7%9C%BC%E6%B7%9A%E6%88%90%E8%A9%A9',
  70         'only_matching': True,
  71     }]
  72
  73     @staticmethod
  74     def base64_decode_utf8(data):
  75         return base64.b64decode(data.encode('utf-8')).decode('utf-8')
  76
  77     @staticmethod
  78     def base64_encode_utf8(data):
  79         return base64.b64encode(data.encode('utf-8')).decode('utf-8')
  80
  81     def _extract_flv_config(self, media_id):
  82         base64_media_id = self.base64_encode_utf8(media_id)
  83         flv_config = self._download_xml(
  84             'http://vlog.xuite.net/flash/player?media=%s' % base64_media_id,
  85             'flv config')
  86         prop_dict = {}
  87         for prop in flv_config.findall('./property'):
  88             prop_id = self.base64_decode_utf8(prop.attrib['id'])
  89             # CDATA may be empty in flv config
  90             if not prop.text:
  91                 continue
  92             encoded_content = self.base64_decode_utf8(prop.text)
  93             prop_dict[prop_id] = compat_urllib_parse_unquote(encoded_content)
  94         return prop_dict
  95
  96     def _real_extract(self, url):
  97         video_id = self._match_id(url)
  98
  99         webpage = self._download_webpage(url, video_id)
 100
 101         error_msg = self._search_regex(
 102             r'<div id="error-message-content">([^<]+)',
 103             webpage, 'error message', default=None)
 104         if error_msg:
 105             raise ExtractorError(
 106                 '%s returned error: %s' % (self.IE_NAME, error_msg),
 107                 expected=True)
 108
 109         video_id = self._html_search_regex(
 110             r'data-mediaid="(\d+)"', webpage, 'media id')
 111         flv_config = self._extract_flv_config(video_id)
 112
 113         FORMATS = {
 114             'audio': 'mp3',
 115             'video': 'mp4',
 116         }
 117
 118         formats = []
 119         for format_tag in ('src', 'hq_src'):
 120             video_url = flv_config.get(format_tag)
 121             if not video_url:
 122                 continue
 123             format_id = self._search_regex(
 124                 r'\bq=(.+?)\b', video_url, 'format id', default=format_tag)
 125             formats.append({
 126                 'url': video_url,
 127                 'ext': FORMATS.get(flv_config['type'], 'mp4'),
 128                 'format_id': format_id,
 129                 'height': int(format_id) if format_id.isnumeric() else None,
 130             })
 131         self._sort_formats(formats)
 132
 133         timestamp = flv_config.get('publish_datetime')
 134         if timestamp:
 135             timestamp = parse_iso8601(timestamp + ' +0800', ' ')
 136
 137         category = flv_config.get('category')
 138         categories = [category] if category else []
 139
 140         return {
 141             'id': video_id,
 142             'title': flv_config['title'],
 143             'description': flv_config.get('description'),
 144             'thumbnail': flv_config.get('thumb'),
 145             'timestamp': timestamp,
 146             'uploader': flv_config.get('author_name'),
 147             'uploader_id': flv_config.get('author_id'),
 148             'duration': parse_duration(flv_config.get('duration')),
 149             'categories': categories,
 150             'formats': formats,
 151         }