Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/eporner.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_str
   8 from ..utils import (
   9     encode_base_n,
  10     ExtractorError,
  11     int_or_none,
  12     parse_duration,
  13     str_to_int,
  14 )
  15
  16
  17 class EpornerIE(InfoExtractor):
  18     _VALID_URL = r'https?://(?:www\.)?eporner\.com/(?:hd-porn|embed)/(?P<id>\w+)(?:/(?P<display_id>[\w-]+))?'
  19     _TESTS = [{
  20         'url': 'http://www.eporner.com/hd-porn/95008/Infamous-Tiffany-Teen-Strip-Tease-Video/',
  21         'md5': '39d486f046212d8e1b911c52ab4691f8',
  22         'info_dict': {
  23             'id': 'qlDUmNsj6VS',
  24             'display_id': 'Infamous-Tiffany-Teen-Strip-Tease-Video',
  25             'ext': 'mp4',
  26             'title': 'Infamous Tiffany Teen Strip Tease Video',
  27             'duration': 1838,
  28             'view_count': int,
  29             'age_limit': 18,
  30         },
  31     }, {
  32         # New (May 2016) URL layout
  33         'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0/Star-Wars-XXX-Parody/',
  34         'only_matching': True,
  35     }, {
  36         'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0',
  37         'only_matching': True,
  38     }, {
  39         'url': 'http://www.eporner.com/hd-porn/3YRUtzMcWn0',
  40         'only_matching': True,
  41     }]
  42
  43     def _real_extract(self, url):
  44         mobj = re.match(self._VALID_URL, url)
  45         video_id = mobj.group('id')
  46         display_id = mobj.group('display_id') or video_id
  47
  48         webpage, urlh = self._download_webpage_handle(url, display_id)
  49
  50         video_id = self._match_id(compat_str(urlh.geturl()))
  51
  52         hash = self._search_regex(
  53             r'hash\s*:\s*["\']([\da-f]{32})', webpage, 'hash')
  54
  55         title = self._og_search_title(webpage, default=None) or self._html_search_regex(
  56             r'<title>(.+?) - EPORNER', webpage, 'title')
  57
  58         # Reverse engineered from vjs.js
  59         def calc_hash(s):
  60             return ''.join((encode_base_n(int(s[lb:lb + 8], 16), 36) for lb in range(0, 32, 8)))
  61
  62         video = self._download_json(
  63             'http://www.eporner.com/xhr/video/%s' % video_id,
  64             display_id, note='Downloading video JSON',
  65             query={
  66                 'hash': calc_hash(hash),
  67                 'device': 'generic',
  68                 'domain': 'www.eporner.com',
  69                 'fallback': 'false',
  70             })
  71
  72         if video.get('available') is False:
  73             raise ExtractorError(
  74                 '%s said: %s' % (self.IE_NAME, video['message']), expected=True)
  75
  76         sources = video['sources']
  77
  78         formats = []
  79         for kind, formats_dict in sources.items():
  80             if not isinstance(formats_dict, dict):
  81                 continue
  82             for format_id, format_dict in formats_dict.items():
  83                 if not isinstance(format_dict, dict):
  84                     continue
  85                 src = format_dict.get('src')
  86                 if not isinstance(src, compat_str) or not src.startswith('http'):
  87                     continue
  88                 if kind == 'hls':
  89                     formats.extend(self._extract_m3u8_formats(
  90                         src, display_id, 'mp4', entry_protocol='m3u8_native',
  91                         m3u8_id=kind, fatal=False))
  92                 else:
  93                     height = int_or_none(self._search_regex(
  94                         r'(\d+)[pP]', format_id, 'height', default=None))
  95                     fps = int_or_none(self._search_regex(
  96                         r'(\d+)fps', format_id, 'fps', default=None))
  97
  98                     formats.append({
  99                         'url': src,
 100                         'format_id': format_id,
 101                         'height': height,
 102                         'fps': fps,
 103                     })
 104         self._sort_formats(formats)
 105
 106         duration = parse_duration(self._html_search_meta('duration', webpage))
 107         view_count = str_to_int(self._search_regex(
 108             r'id="cinemaviews">\s*([0-9,]+)\s*<small>views',
 109             webpage, 'view count', fatal=False))
 110
 111         return {
 112             'id': video_id,
 113             'display_id': display_id,
 114             'title': title,
 115             'duration': duration,
 116             'view_count': view_count,
 117             'formats': formats,
 118             'age_limit': 18,
 119         }