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     _REGEX_BASE64 = r'(?:[A-Za-z0-9+/]{4})*(?:[A-Za-z0-9+/]{2}==|[A-Za-z0-9+/]{3}=)?'
  17     _VALID_URL = r'https?://vlog\.xuite\.net/(?:play|embed)/(?P<id>%s)' % _REGEX_BASE64
  18     _TESTS = [{
  19         # Audio
  20         'url': 'http://vlog.xuite.net/play/RGkzc1ZULTM4NjA5MTQuZmx2',
  21         'md5': '63a42c705772aa53fd4c1a0027f86adf',
  22         'info_dict': {
  23             'id': '3860914',
  24             'ext': 'mp3',
  25             'title': '孤單南半球-歐德陽',
  26             'thumbnail': 're:^https?://.*\.jpg$',
  27             'duration': 247.246,
  28             'timestamp': 1314932940,
  29             'upload_date': '20110902',
  30             'uploader': '阿能',
  31             'uploader_id': '15973816',
  32             'categories': ['個人短片'],
  33         },
  34     }, {
  35         # Video with only one format
  36         'url': 'http://vlog.xuite.net/play/TkRZNjhULTM0NDE2MjkuZmx2',
  37         'md5': 'c45737fc8ac5dc8ac2f92ecbcecf505e',
  38         'info_dict': {
  39             'id': '3441629',
  40             'ext': 'mp4',
  41             'title': '孫燕姿 - 眼淚成詩',
  42             'thumbnail': 're:^https?://.*\.jpg$',
  43             'duration': 217.399,
  44             'timestamp': 1299383640,
  45             'upload_date': '20110306',
  46             'uploader': 'Valen',
  47             'uploader_id': '10400126',
  48             'categories': ['影視娛樂'],
  49         },
  50     }, {
  51         # Video with two formats
  52         'url': 'http://vlog.xuite.net/play/bWo1N1pLLTIxMzAxMTcwLmZsdg==',
  53         'md5': '1166e0f461efe55b62e26a2d2a68e6de',
  54         'info_dict': {
  55             'id': '21301170',
  56             'ext': 'mp4',
  57             'title': '暗殺教室 02',
  58             'description': '字幕:【極影字幕社】',
  59             'thumbnail': 're:^https?://.*\.jpg$',
  60             'duration': 1384.907,
  61             'timestamp': 1421481240,
  62             'upload_date': '20150117',
  63             'uploader': '我只是想認真點',
  64             'uploader_id': '242127761',
  65             'categories': ['電玩動漫'],
  66         },
  67     }, {
  68         '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',
  69         'only_matching': True,
  70     }]
  71
  72     def _extract_flv_config(self, media_id):
  73         base64_media_id = base64.b64encode(media_id.encode('utf-8')).decode('utf-8')
  74         flv_config = self._download_xml(
  75             'http://vlog.xuite.net/flash/player?media=%s' % base64_media_id,
  76             'flv config')
  77         prop_dict = {}
  78         for prop in flv_config.findall('./property'):
  79             prop_id = base64.b64decode(prop.attrib['id']).decode('utf-8')
  80             # CDATA may be empty in flv config
  81             if not prop.text:
  82                 continue
  83             encoded_content = base64.b64decode(prop.text).decode('utf-8')
  84             prop_dict[prop_id] = compat_urllib_parse_unquote(encoded_content)
  85         return prop_dict
  86
  87     def _real_extract(self, url):
  88         video_id = self._match_id(url)
  89
  90         webpage = self._download_webpage(url, video_id)
  91
  92         error_msg = self._search_regex(
  93             r'<div id="error-message-content">([^<]+)',
  94             webpage, 'error message', default=None)
  95         if error_msg:
  96             raise ExtractorError(
  97                 '%s returned error: %s' % (self.IE_NAME, error_msg),
  98                 expected=True)
  99
 100         video_id = self._html_search_regex(
 101             r'data-mediaid="(\d+)"', webpage, 'media id')
 102         flv_config = self._extract_flv_config(video_id)
 103
 104         FORMATS = {
 105             'audio': 'mp3',
 106             'video': 'mp4',
 107         }
 108
 109         formats = []
 110         for format_tag in ('src', 'hq_src'):
 111             video_url = flv_config.get(format_tag)
 112             if not video_url:
 113                 continue
 114             format_id = self._search_regex(
 115                 r'\bq=(.+?)\b', video_url, 'format id', default=format_tag)
 116             formats.append({
 117                 'url': video_url,
 118                 'ext': FORMATS.get(flv_config['type'], 'mp4'),
 119                 'format_id': format_id,
 120                 'height': int(format_id) if format_id.isnumeric() else None,
 121             })
 122         self._sort_formats(formats)
 123
 124         timestamp = flv_config.get('publish_datetime')
 125         if timestamp:
 126             timestamp = parse_iso8601(timestamp + ' +0800', ' ')
 127
 128         category = flv_config.get('category')
 129         categories = [category] if category else []
 130
 131         return {
 132             'id': video_id,
 133             'title': flv_config['title'],
 134             'description': flv_config.get('description'),
 135             'thumbnail': flv_config.get('thumb'),
 136             'timestamp': timestamp,
 137             'uploader': flv_config.get('author_name'),
 138             'uploader_id': flv_config.get('author_id'),
 139             'duration': parse_duration(flv_config.get('duration')),
 140             'categories': categories,
 141             'formats': formats,
 142         }