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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import time
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     determine_ext,
  10     js_to_json,
  11 )
  12
  13
  14 class IPrimaIE(InfoExtractor):
  15     _VALID_URL = r'https?://play\.iprima\.cz/(?:.+/)?(?P<id>[^?#]+)'
  16     _GEO_BYPASS = False
  17
  18     _TESTS = [{
  19         'url': 'http://play.iprima.cz/gondici-s-r-o-33',
  20         'info_dict': {
  21             'id': 'p136534',
  22             'ext': 'mp4',
  23             'title': 'Gondíci s. r. o. (34)',
  24             'description': 'md5:16577c629d006aa91f59ca8d8e7f99bd',
  25         },
  26         'params': {
  27             'skip_download': True,  # m3u8 download
  28         },
  29     }, {
  30         'url': 'http://play.iprima.cz/particka/particka-92',
  31         'only_matching': True,
  32     }, {
  33         # geo restricted
  34         'url': 'http://play.iprima.cz/closer-nove-pripady/closer-nove-pripady-iv-1',
  35         'only_matching': True,
  36     }]
  37
  38     def _real_extract(self, url):
  39         video_id = self._match_id(url)
  40
  41         self._set_cookie('play.iprima.cz', 'ott_adult_confirmed', '1')
  42
  43         webpage = self._download_webpage(url, video_id)
  44
  45         video_id = self._search_regex(r'data-product="([^"]+)">', webpage, 'real id')
  46
  47         playerpage = self._download_webpage(
  48             'http://play.iprima.cz/prehravac/init',
  49             video_id, note='Downloading player', query={
  50                 '_infuse': 1,
  51                 '_ts': round(time.time()),
  52                 'productId': video_id,
  53             }, headers={'Referer': url})
  54
  55         formats = []
  56
  57         def extract_formats(format_url, format_key=None, lang=None):
  58             ext = determine_ext(format_url)
  59             new_formats = []
  60             if format_key == 'hls' or ext == 'm3u8':
  61                 new_formats = self._extract_m3u8_formats(
  62                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
  63                     m3u8_id='hls', fatal=False)
  64             elif format_key == 'dash' or ext == 'mpd':
  65                 return
  66                 new_formats = self._extract_mpd_formats(
  67                     format_url, video_id, mpd_id='dash', fatal=False)
  68             if lang:
  69                 for f in new_formats:
  70                     if not f.get('language'):
  71                         f['language'] = lang
  72             formats.extend(new_formats)
  73
  74         options = self._parse_json(
  75             self._search_regex(
  76                 r'(?s)(?:TDIPlayerOptions|playerOptions)\s*=\s*({.+?});\s*\]\]',
  77                 playerpage, 'player options', default='{}'),
  78             video_id, transform_source=js_to_json, fatal=False)
  79         if options:
  80             for key, tracks in options.get('tracks', {}).items():
  81                 if not isinstance(tracks, list):
  82                     continue
  83                 for track in tracks:
  84                     src = track.get('src')
  85                     if src:
  86                         extract_formats(src, key.lower(), track.get('lang'))
  87
  88         if not formats:
  89             for _, src in re.findall(r'src["\']\s*:\s*(["\'])(.+?)\1', playerpage):
  90                 extract_formats(src)
  91
  92         if not formats and '>GEO_IP_NOT_ALLOWED<' in playerpage:
  93             self.raise_geo_restricted(countries=['CZ'])
  94
  95         self._sort_formats(formats)
  96
  97         return {
  98             'id': video_id,
  99             'title': self._og_search_title(webpage),
 100             'thumbnail': self._og_search_thumbnail(webpage),
 101             'formats': formats,
 102             'description': self._og_search_description(webpage),
 103         }