_ Git - youtube-dl/blob - youtube_dl/extractor/vshare.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_chr
   8 from ..utils import (
   9     decode_packed_codes,
  10     ExtractorError,
  11 )
  12
  13
  14 class VShareIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?:www\.)?vshare\.io/[dv]/(?P<id>[^/?#&]+)'
  16     _TESTS = [{
  17         'url': 'https://vshare.io/d/0f64ce6',
  18         'md5': '17b39f55b5497ae8b59f5fbce8e35886',
  19         'info_dict': {
  20             'id': '0f64ce6',
  21             'title': 'vl14062007715967',
  22             'ext': 'mp4',
  23         }
  24     }, {
  25         'url': 'https://vshare.io/v/0f64ce6/width-650/height-430/1',
  26         'only_matching': True,
  27     }]
  28
  29     def _extract_packed(self, webpage):
  30         packed = self._search_regex(r'(eval\(function.+)', webpage, 'packed code')
  31         unpacked = decode_packed_codes(packed)
  32         digits = self._search_regex(r'\[((?:\d+,?)+)\]', unpacked, 'digits')
  33         digits = digits.split(',')
  34         digits = [int(digit) for digit in digits]
  35         key_digit = self._search_regex(r'fromCharCode\(.+?(\d+)\)}', unpacked, 'key digit')
  36         chars = [compat_chr(d - int(key_digit)) for d in digits]
  37         return ''.join(chars)
  38
  39     def _real_extract(self, url):
  40         video_id = self._match_id(url)
  41
  42         webpage = self._download_webpage(
  43             'https://vshare.io/v/%s/width-650/height-430/1' % video_id, video_id)
  44
  45         title = self._html_search_regex(r'<title>([^<]+)</title>', webpage, 'title')
  46         title = title.split(' - ')[0]
  47
  48         error = self._html_search_regex(
  49             r'(?s)<div[^>]+\bclass=["\']xxx-error[^>]+>(.+?)</div', webpage,
  50             'error', default=None)
  51         if error:
  52             raise ExtractorError(error, expected=True)
  53
  54         unpacked = self._extract_packed(webpage)
  55         video_urls = re.findall(r'<source src="([^"]+)', unpacked)
  56         formats = [{'url': video_url} for video_url in video_urls]
  57         return {
  58             'id': video_id,
  59             'title': title,
  60             'formats': formats,
  61         }
  62
  63     @staticmethod
  64     def _extract_urls(webpage):
  65         return re.findall(
  66             r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?vshare\.io/v/[^/?#&]+)',
  67             webpage)