youtube_dl/extractor/indavideo.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     int_or_none,
  10     parse_age_limit,
  11     parse_iso8601,
  12     update_url_query,
  13 )
  14
  15
  16 class IndavideoEmbedIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:(?:embed\.)?indavideo\.hu/player/video/|assets\.indavideo\.hu/swf/player\.swf\?.*\b(?:v(?:ID|id))=)(?P<id>[\da-f]+)'
  18     _TESTS = [{
  19         'url': 'http://indavideo.hu/player/video/1bdc3c6d80/',
  20         'md5': 'c8a507a1c7410685f83a06eaeeaafeab',
  21         'info_dict': {
  22             'id': '1837039',
  23             'ext': 'mp4',
  24             'title': 'Cicatánc',
  25             'description': '',
  26             'thumbnail': r're:^https?://.*\.jpg$',
  27             'uploader': 'cukiajanlo',
  28             'uploader_id': '83729',
  29             'timestamp': 1439193826,
  30             'upload_date': '20150810',
  31             'duration': 72,
  32             'age_limit': 0,
  33             'tags': ['tánc', 'cica', 'cuki', 'cukiajanlo', 'newsroom'],
  34         },
  35     }, {
  36         'url': 'http://embed.indavideo.hu/player/video/1bdc3c6d80?autostart=1&hide=1',
  37         'only_matching': True,
  38     }, {
  39         'url': 'http://assets.indavideo.hu/swf/player.swf?v=fe25e500&vID=1bdc3c6d80&autostart=1&hide=1&i=1',
  40         'only_matching': True,
  41     }]
  42
  43     # Some example URLs covered by generic extractor:
  44     #   http://indavideo.hu/video/Vicces_cica_1
  45     #   http://index.indavideo.hu/video/2015_0728_beregszasz
  46     #   http://auto.indavideo.hu/video/Sajat_utanfutoban_a_kis_tacsko
  47     #   http://erotika.indavideo.hu/video/Amator_tini_punci
  48     #   http://film.indavideo.hu/video/f_hrom_nagymamm_volt
  49     #   http://palyazat.indavideo.hu/video/Embertelen_dal_Dodgem_egyuttes
  50
  51     @staticmethod
  52     def _extract_urls(webpage):
  53         return re.findall(
  54             r'<iframe[^>]+\bsrc=["\'](?P<url>(?:https?:)?//embed\.indavideo\.hu/player/video/[\da-f]+)',
  55             webpage)
  56
  57     def _real_extract(self, url):
  58         video_id = self._match_id(url)
  59
  60         video = self._download_json(
  61             'http://amfphp.indavideo.hu/SYm0json.php/player.playerHandler.getVideoData/%s' % video_id,
  62             video_id)['data']
  63
  64         title = video['title']
  65
  66         video_urls = []
  67
  68         video_files = video.get('video_files')
  69         if isinstance(video_files, list):
  70             video_urls.extend(video_files)
  71         elif isinstance(video_files, dict):
  72             video_urls.extend(video_files.values())
  73
  74         video_file = video.get('video_file')
  75         if video:
  76             video_urls.append(video_file)
  77         video_urls = list(set(video_urls))
  78
  79         video_prefix = video_urls[0].rsplit('/', 1)[0]
  80
  81         for flv_file in video.get('flv_files', []):
  82             flv_url = '%s/%s' % (video_prefix, flv_file)
  83             if flv_url not in video_urls:
  84                 video_urls.append(flv_url)
  85
  86         filesh = video.get('filesh')
  87
  88         formats = []
  89         for video_url in video_urls:
  90             height = int_or_none(self._search_regex(
  91                 r'\.(\d{3,4})\.mp4(?:\?|$)', video_url, 'height', default=None))
  92             if filesh:
  93                 if not height:
  94                     continue
  95                 token = filesh.get(compat_str(height))
  96                 if token is None:
  97                     continue
  98                 video_url = update_url_query(video_url, {'token': token})
  99             formats.append({
 100                 'url': video_url,
 101                 'height': height,
 102             })
 103         self._sort_formats(formats)
 104
 105         timestamp = video.get('date')
 106         if timestamp:
 107             # upload date is in CEST
 108             timestamp = parse_iso8601(timestamp + ' +0200', ' ')
 109
 110         thumbnails = [{
 111             'url': self._proto_relative_url(thumbnail)
 112         } for thumbnail in video.get('thumbnails', [])]
 113
 114         tags = [tag['title'] for tag in video.get('tags') or []]
 115
 116         return {
 117             'id': video.get('id') or video_id,
 118             'title': title,
 119             'description': video.get('description'),
 120             'thumbnails': thumbnails,
 121             'uploader': video.get('user_name'),
 122             'uploader_id': video.get('user_id'),
 123             'timestamp': timestamp,
 124             'duration': int_or_none(video.get('length')),
 125             'age_limit': parse_age_limit(video.get('age_limit')),
 126             'tags': tags,
 127             'formats': formats,
 128         }