yt_dlp/extractor/shared.py

   1 import urllib.parse
   2
   3 from .common import InfoExtractor
   4 from ..compat import compat_b64decode
   5 from ..utils import (
   6     KNOWN_EXTENSIONS,
   7     ExtractorError,
   8     determine_ext,
   9     int_or_none,
  10     js_to_json,
  11     parse_filesize,
  12     rot47,
  13     url_or_none,
  14     urlencode_postdata,
  15 )
  16
  17
  18 class SharedBaseIE(InfoExtractor):
  19     def _real_extract(self, url):
  20         video_id = self._match_id(url)
  21
  22         webpage, urlh = self._download_webpage_handle(url, video_id)
  23
  24         if self._FILE_NOT_FOUND in webpage:
  25             raise ExtractorError(
  26                 'Video %s does not exist' % video_id, expected=True)
  27
  28         video_url = self._extract_video_url(webpage, video_id, url)
  29
  30         title = self._extract_title(webpage)
  31         filesize = int_or_none(self._extract_filesize(webpage))
  32
  33         return {
  34             'id': video_id,
  35             'url': video_url,
  36             'ext': 'mp4',
  37             'filesize': filesize,
  38             'title': title,
  39         }
  40
  41     def _extract_title(self, webpage):
  42         return compat_b64decode(self._html_search_meta(
  43             'full:title', webpage, 'title')).decode('utf-8')
  44
  45     def _extract_filesize(self, webpage):
  46         return self._html_search_meta(
  47             'full:size', webpage, 'file size', fatal=False)
  48
  49
  50 class SharedIE(SharedBaseIE):
  51     IE_DESC = 'shared.sx'
  52     _VALID_URL = r'https?://shared\.sx/(?P<id>[\da-z]{10})'
  53     _FILE_NOT_FOUND = '>File does not exist<'
  54
  55     _TEST = {
  56         'url': 'http://shared.sx/0060718775',
  57         'md5': '106fefed92a8a2adb8c98e6a0652f49b',
  58         'info_dict': {
  59             'id': '0060718775',
  60             'ext': 'mp4',
  61             'title': 'Bmp4',
  62             'filesize': 1720110,
  63         },
  64     }
  65
  66     def _extract_video_url(self, webpage, video_id, url):
  67         download_form = self._hidden_inputs(webpage)
  68
  69         video_page = self._download_webpage(
  70             url, video_id, 'Downloading video page',
  71             data=urlencode_postdata(download_form),
  72             headers={
  73                 'Content-Type': 'application/x-www-form-urlencoded',
  74                 'Referer': url,
  75             })
  76
  77         video_url = self._html_search_regex(
  78             r'data-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
  79             video_page, 'video URL', group='url')
  80
  81         return video_url
  82
  83
  84 class VivoIE(SharedBaseIE):
  85     IE_DESC = 'vivo.sx'
  86     _VALID_URL = r'https?://vivo\.s[xt]/(?P<id>[\da-z]{10})'
  87     _FILE_NOT_FOUND = '>The file you have requested does not exists or has been removed'
  88
  89     _TESTS = [{
  90         'url': 'http://vivo.sx/d7ddda0e78',
  91         'md5': '15b3af41be0b4fe01f4df075c2678b2c',
  92         'info_dict': {
  93             'id': 'd7ddda0e78',
  94             'ext': 'mp4',
  95             'title': 'Chicken',
  96             'filesize': 515659,
  97         },
  98     }, {
  99         'url': 'http://vivo.st/d7ddda0e78',
 100         'only_matching': True,
 101     }]
 102
 103     def _extract_title(self, webpage):
 104         title = self._html_search_regex(
 105             r'data-name\s*=\s*(["\'])(?P<title>(?:(?!\1).)+)\1', webpage,
 106             'title', default=None, group='title')
 107         if title:
 108             ext = determine_ext(title)
 109             if ext.lower() in KNOWN_EXTENSIONS:
 110                 title = title.rpartition('.' + ext)[0]
 111             return title
 112         return self._og_search_title(webpage)
 113
 114     def _extract_filesize(self, webpage):
 115         return parse_filesize(self._search_regex(
 116             r'data-type=["\']video["\'][^>]*>Watch.*?<strong>\s*\((.+?)\)',
 117             webpage, 'filesize', fatal=False))
 118
 119     def _extract_video_url(self, webpage, video_id, url):
 120         def decode_url_old(encoded_url):
 121             return compat_b64decode(encoded_url).decode('utf-8')
 122
 123         stream_url = self._search_regex(
 124             r'data-stream\s*=\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
 125             'stream url', default=None, group='url')
 126         if stream_url:
 127             stream_url = url_or_none(decode_url_old(stream_url))
 128         if stream_url:
 129             return stream_url
 130
 131         def decode_url(encoded_url):
 132             return rot47(urllib.parse.unquote_plus(encoded_url))
 133
 134         return decode_url(self._parse_json(
 135             self._search_regex(
 136                 r'(?s)InitializeStream\s*\(\s*({.+?})\s*\)\s*;', webpage,
 137                 'stream'),
 138             video_id, transform_source=js_to_json)['source'])